Full text of "Educational psychology : an introductory text"

See other formats

••"•• I'h •" ~~'i in nr r" r~ ' r * ■.- -ri iii- <

: f

'%'■ BY TH ^o>mof ' 4

B. M . j.

i!_ AS--

■

' ' OF 5^

prt sf.'itE 0
ti». rut f jitoh of ,
B M.

" t/eoi :au v *'

EDUCATIONAL PSYCHOLOGY

Digitized by the Internet Archive
in 2017 with funding from
Wellcome Library

https://archive.org/details/b29820686

EDUCATIONAL PSYCHOLOGY

AN INTRODUCTORY TEXT

'vY

off

v> k —-— *

F i) i 10 In C

M, U.

'R. YAL Kjg

RUDOLF PINTNER

PROFESSOR OF EDUCATION, TEACHERS COLLEGE
COLUMBIA UNIVERSITY

^ tA

//'\\ s

\ * ( 3 1 »

1- »

/} /'

f»|! 1

in on

i \j L*. i

IdoU

'AA

V \ *V

0 i\._

.. y-"

LONDON

WILLIAMS & NORGATE LTD

1930

■""I

WELLCOMF INSTITUTE
LIBRARY

CoS!

we!MOmec

Cai!

No.

v . */'

PRINTED IN THE UNITED STATES OF AMERICA

PREFACE

Educational Psychology is frequently considered an
applied psychology, and strictly considered it undoubtedly
is that. As an application of psychology to education, we
may note that the earlier text-books of educational
psychology followed rather slavishly the conventional
order of subjects usually found in the text-books of pure
psychology. Thus we find many early educational psy¬
chologies, and indeed several recent ones, beginning with a
treatment of the nervous system, following on with a dis¬
cussion of the several senses, then going on to such topics
as perception, attention, emotion, memory, imagination
and generally ending up with the higher thought processes.
Frequently the applications to education in many of these
topics were difficult to make, and often they were very
far-fetched and had only the remotest connection with
education.

Thorndike’s Educational Psychology broke away very
definitely from these traditions. His treatment was an
investigation of original nature and its modifications.
Thus we have the emphasis laid upon the original tend¬
encies of the individual and the modifications of these
original tendencies which the school is trying to bring
about. This new departure of Thorndike’s has influenced
all educational psychologies written since the publication
of his books. Some writers have taken over the new ar¬
rangement completely, others have tried a compromise
between it and the older traditional sequence of topics in
pure psychology, so that in some books we still see much
space devoted to the nervous system and to sensation,
perception and so forth. It is interesting to note here
that this new departure of Thorndike’s is now beginning

PREFACE

to influence the text-books of pure psychology, and to
break down the traditional treatment referred to.

Since the appearance of Thorndike’s Educational Psy¬
chology, there has been a great development in the field
of psychological and educational measurement. Tests
of all kinds have been constructed. We are measuring
to-day things that did not appear measurable fifteen
or twenty years ago. This field of measurement has pro¬
duced its own specialists and separate courses in measure¬
ment have sprung up in our colleges. To a great extent
it has been kept apart from the regular course in educa¬
tional psychology.

My belief is that this measurement of original tendencies
and of educational modifications is an integral part of
educational psychology. A study of original nature is
helped by whatever measurements we can make, and a
study of the learning process implies the measurement of
educational achievement. It seems to me, therefore, that
the time has come to incorporate such work into our edu¬
cational psychologies, not as a separate topic added to the
traditional treatment in a closing chapter, but rather as an
integral part of the whole treatment. Some of the most
important contributions of modern educational psychol¬
ogy are to be found in the recent field of measurement, and
so it is only proper that an elementary text, such as this,
should lay stress on these things. Many of our students
do not have the time or inclination to take special courses
in measurement in addition to the regular courses in edu¬
cational psychology. It is my belief, therefore, that a
first course in educational psychology should contain the
general field of educational measurement as an integral
part of the course. Throughout the book my aim has
been to emphasize such topics as seem to me to be of most
value to the teacher and educator. Some topics of doubtful
value to the teacher, such as a study of the nervous sys¬
tem, have been omitted altogether. Curt and dogmatic

PREFACE

Vll

treatment has been meted out to many controversial
topics. Advanced students must go elsewhere, notably
to Thorndike’s books, for more extended treatment.

This book is designed for a one semester’s course for
students beginning a study of educational psychology. I
have given a few advised readings at the end of most
chapters. I have purposely kept these down to a mini¬
mum, with the hope that they will be read by all students.
A lengthy bibliography at the end of each chapter is more
likely to discourage rather than encourage further reading
on the part of the beginning student. A general bibliog¬
raphy is given at the end of the book, which the teacher
may use for additional reading assignments, if necessary.
At the end of each chapter I have also given the references
to specific pieces of work mentioned in the text. These
references are not intended as reading assignments for all
students, but merely as acknowledgments to the authors
from whom I have borrowed and also as indications to the
reader as to where he may go for original sources. I have
also added a few exercises, mainly true-false statements,
at the end of each chapter. These the student may use
to practice on for review purposes. Some of the true-false
statements will undoubtedly prove ambiguous and im¬
possible to mark plus or minus, and such might readily
stimulate discussion in the class.

As to acknowledgments, it seems needless for me to
mention the debt that all educational psychology owes to
Thorndike. The whole arrangement of this book and the
treatment of most of the topics show his influence. I
have made much use of pictures and diagrams borrowed
freely from others, and for such borrowings I wish here
to thank the following authors and publishers:

E. L. Thorndike, Leta S. Hollingworth, Daniel Starch,
E. S. Robinson, Antoinette Feleky, Arthur I. Gates, Don¬
ald G. Paterson, Lewis M. Terman, W. F. Book, W. B.
Pillsbury, H. E. Buchholz, Herbert Woodrow, W. H.

Vlll

PREFACE

Pyle, Peter Sandiford, Harold E. Jones, Elizabeth Irwin,
A. T. Poffenberger, E. M. Paulu, Henry H. Goddard,
George D. Strayer, John A. McGeoch, Mary Wentworth
McConaughy, Kathryn E. Maxfield, Carl E. Seashore,
Walter Scott, Charles H. Judd, Arnold Gesell, Rachel
Stutsman, Harry L. Hollingworth, William Healy, Ber¬
nard Muscio, Cyril Burt, C. H. Stoelting, The Macmil¬
lan Company, Stanford University Press, Psychological
Review, D. Appleton and Co., World Book Company,
Teachers College Bureau of Publications, Clark University
Press, Silver Burdett and Co., Houghton Mifflin and Co.,
A. W. Shaw Company, National Research Council, Har¬
vard University Press, Longmans Green, Warwick and
York, D. C. Heath and Co., Cleveland Foundation,
Archives of Psychology, National Society for the Study
of Education, Journal of Applied Psychology, George
Routledge and Sons.

Rudolf Pintner.

Yonkers, May, 1929.

,• \'y J - ~~~ ?

' . / PRESENTED
tP I DY THE EDITOR 0

CONTENTS

CHAPTER PAGE

I. The Field of Educational Psychology . 3

PART I—ORIGINAL NATURE

II. The Original Tendencies .... 15

III. The Functioning of Original Tendencies 40

IV. The Measurement of Non-Intellectual

Traits.67

V. The Measurement of Intellectual

Traits.89

VI. The Uses and Results of Intelligence

Testing .136

VII. The Inheritance of Mental Traits . 165

PART II—THE MODIFICATION OF
ORIGINAL NATURE

VIII. The Laws of Learning .... 185

IX. The Learning Curve. 197

X. Efficient Learning.224

XI. The Permanence of Modification . . 254

XII. The Transfer of Training . . . 263

XIII. Continuous Exercise without Rest . 287

XIV. Measuring Improvement in School Sub¬

jects .308

XV. The Uses of Educational Tests . . 335

XVI. New-Type Examinations .... 348
XVII. Conclusion ....... 357

CONTENTS

PAGE

General Bibliography.361

Appendix I. The Association Test . . . 363

Appendix II. Judgment of Intelligence from

Photographs .... 369

Index of Subjects.373

Index of Names.377

LIST OF ILLUSTRATIONS

F C' -

EDITOR

B. M. J.

■Cp/CA L

FIGURE PAGE

1. Expression of the Emotions .... 19

2. Vocalization Chart of Six-months Child . 23

3. Conditioned Reflex in the Dog ... 34

4. Graphic Rating Scale. 74

5. Photos of Children .... 90-91

6. (a) Ball and Field Test .... 97

6. (b) Ball and Field Test .... 98

7. Mare and Foal Test. 100

8. Five Figure Board Test .... 100

9. Casuist Form Board. 100

10. The Ship Test. 100

11. A Page from the Pintner-Cunningham Test 104

12. A Page from the Pintner-Cunningham Test 104

13. A Page from the National Intelligence

Test. 106

14. A Page from the Terman Group Test . . 107

15. A Page from the Pintner Non-Language

Test. 109

16. A Page from the Pintner Rapid Survey Test 110

17. A Page from the Army Alpha Test . .111

18. A Page from the Army Alpha Test . .112

19. A Merrill-Palmer Test.115

20. Musical Talent Chart.116

21. Distribution of I.Q.’s of Kindergarten

Children.117

22. Distribution of I.Q.’s of Grade IV Pupils . 117

23. Distribution of Gifted and Unselected

Children.118

24. Frequency Curves for High School Seniors 118

25. Composite Curve for Ninth Grade . .119

Xll

LIST OF ILLUSTRATIONS

FIGURE

26. Composite Curve for College Freshmen .

27. Distribution of Intelligence of Army Re¬

cruits .

28. Distribution of 93,965 Soldiers .

29. The Curve of Growth for Intelligence

30. Intelligence of High School Seniors .

31. Intelligence of Seniors in Different

Counties.

32. Occupational Intelligence Ratings

33. Occupational Standards on Intelligence

Test.

34. The I.Q.’s of Repeated Offenders

35. Comparison of Deaf and Hearing Children

36. Intelligence of Foreign-born Men in the

Army.

37. Stanford-Binet Examinations of Twins

38. Chart of Wedgwoqd-Darwin-GAlton Family

39. Chart of Kallikak Family .

40. Practice Curve for Alphabet Backwards .

41. Practice Curve for Alphabet Backwards .

42. Practice Curve for Alphabet Backwards .

43. Practice Curve for Alphabet Backwards .

44. Practice Curve : Subject A .

45. Practice Curve : Subject B .

46. Practice Curve: Subject C .

47. Practice Curve : Subject D .

48. Practice Curve: Subject E .

49. Practice Curve : Subject F .

50. Composite Learning Curve .

51. Practice Curve for Typewriting: Errors .

52. Practice Curve for Typewriting: Speed

53. Practice Curve for Cancellation

54. Practice Curve for Substitution .

55. Practice Curve for a Balancing Game

56. Learning Curves for Learning Braille

PAGE

120

121

122

123

145

147

150

153

155

157

159

170

173

175

199

200
200
201
201
202

203

204

205

207

208

209

210
211
212

LIST OF ILLUSTRATIONS

FIGURE

57. Learning Curve for Typewriting .

58. Learning Curve of a Blind Child

59. Learning Curves of Normals and Defec¬

tives .

60. Motivation in Penmanship .

61. Monthly Progress in Reading

62. Comparison of Length of Practice Periods

63. Distribution of Practice Periods .

64. Distribution of Practice in an Arithmetic

Text.

65. Speed and Accuracy in Reading .

66. Curve of Forgetting for Non-sense Mate-

RIAL «••«•«»«

67. Curve of Forgetting for Meaningful Ma¬

terial ........

68. Curve of Forgetting Material in Lectures

69. Work Curve for Mental Multiplication .

70. Relation Between Feelings and Output

71. Distribution of Output over Working Day

72. Time of Occurrence of Industrial Accidents

73. Improvement in Reading and Arithmetic .

74. Improvement in General Educational

Achievement.

75. Progress in Spelling.

76. Learning Curves for Addition of Fractions

77. Improvement in Educational Achievement

xiii

PAGE

213

214

215

216
217

240

241

243

247

254

256

260

292

297

299

300
311

311

312

313
315

LIST OF TABLES

TABLE

1. Average Score of Typical Groups

2. Writing the Alphabet Backwards .

3. Results of Learning Test for 88 Students .

4. Laboratory Experiments in Transfer of

Training.

5. School Experiments in Transfer of Training

6. Laboratory Fatigue Experiments .

7. Results of School Fatigue Experiments

8. Effects of Alcohol and Coffee

9. Achievement in Silent Reading in Tampa

Schools .

PAGE

151

198

206

271

276

290

295

304

339

XIV

EDUCATIONAL PSYCHOLOGY

CHAPTER I

]- J r : i-'Q cy re

THE FIELD OF EDUCATIONAL PSYCHOLOGY

The Contribution of Psychology to Education. — The

aim of all education from a psychological standpoint is
to change the individual in certain directions. Whether
we are thinking of the education of the child or the adult,
we always are thinking of effecting some change in his
reactions. Whatever be the instruments we may employ
— school, church, newspaper, library, club — in every
case we expect some change in the reactions of the individ¬
ual to take place as a result of their use. In every case,
then, our objective or goal is a changed reaction. In one
case we may wish to teach a child to modify his reaction
to the symbols 2X3 from saying or writing 5 or 4 or
laughing or saying “I don’t know” or turning away,
to the reaction of saying or writing 6. In another case
we may wish to modify the reaction of a voter on elec¬
tion day from using all the day for an outing, or think¬
ing that it does not matter whether he votes or not, to
the reaction of going and recording his vote. If we
achieve the desired modifications, we have been success¬
ful in our objectives.

Objectives not Determined by Psychology. — What the
objectives in education are at any given moment is not
the outcome of psychology. Objectives are determined
by the social philosophy of the age, by the actual needs
and wishes of a given community or group. Whatever
the objectives may be, whether desirable or undesirable,
specific or vague, psychology can, however, give very
decided help to the educator in attaining his aims. This
is the special function of educational psychology.

4 FIELD OF EDUCATIONAL PSYCHOLOGY

Definition of Educational Psychology. — General Psy¬
chology is often defined as a study of the behavior of
the individual in response to any and every kind of situa¬
tion that life presents. If now, we concentrate on those
situations in life that may be called specifically educa¬
tional, many problems arise that are not ordinarily dealt
with in general psychology. We may, therefore, define
educational psychology as a study of the behavior of the
individual in response to educational situations.

The Two Chief Divisions of Educational Psychology. —
The starting point for all and any kind of education is the
individual. It is the individual that we want to change.
Hence the first big division of our subject is an intensive
study of the individual. If we are to be good craftsmen,
we must know the material with which we have to work.
A good carpenter will know something about wood so
that he may know how to use the different kinds for the
various things he constructs. Some knowledge of the
raw material with which he works is necessary to the suc¬
cessful worker. If he is lacking in such knowledge he
will try to achieve impossibilities or he will bring forth
monstrosities. We cannot make silken purses out of
sows’ ears, although sometimes we try to do so.

The second division of our study is the problem of
modifying this raw material. What are the laws govern¬
ing the modification of the individual? We know that
we cannot make any changes that we wish. We cannot
hope to make any kind of modification that our desires
dictate or our fancy suggests. We may desire to make all
people intelligent and educate them all up to an LQ. of
150, but we know this to be impossible. Imagination
can conceive of individuals equipped with radio receiving
senses and radio transmitting mechanisms, but common
sense knows that education can never effect such modifica¬
tions. What are the changes that can be made? How
may we best accomplish these changes? Perhaps some

ORIGINAL NATURE

are easier to make at one time of life than others. What
are the best stimuli or instruments to effect the changes
we desire?

Original Nature and Its Modifications. — So we may-
divide the whole of Educational Psychology into two
parts:

I. The Original Nature of Man.

II. The Modification of Original Nature.

And you will notice that this book is so divided into
two parts. Educational Psychology tries to answer these
two questions: 1. What is the nature of the material
with which the educator has to deal? 2. How can this
material be modified? As in any other science, these two
divisions are arbitrary. They are made merely for the
sake of convenience. The individual is never stationary
or static. He is changing more or less all the time. Any
and every stimulus to which he reacts is causing more
modification of his original nature. The forces of envi¬
ronment are from birth continually working upon him
and changing him. But note the fact that they are work¬
ing upon him — upon a given type of material and they
can only make such modifications as this material, of
which he is made up, will allow. So we may very prof¬
itably study separately these two aspects, one at a time.

Original Nature. — If we watch the amazing mixture
of reactions which individuals in general make to the
countless objects in their environment, it would seem im¬
possible to bring any order out of the chaos. But as we
watch we note a certain similarity in reaction to certain
phases of our environment. An object such as a hand or
ball approaching the eyes suddenly generally causes the
eyelids to wink. The sudden appearance of a large un¬
known object in the dark generally causes the individual
to start and perhaps to tremble or even run, or make such
reactions as would be ordinarily called fear reactions.

6 FIELD OF EDUCATIONAL PSYCHOLOGY

The presentation to an individual of a board with differ¬
ently shaped cut-outs and a verbal or gestural stimulus
of “go on,” will generally lead to an attempt to fit the
blocks in the board. And so we could continue with many
other examples. This general similarity of response
among individuals is in the first place due to the common
inheritance of a certain type of physical body possessing a
certain type of nervous system. It is because our nervous
systems are all more or less alike that we find this general
similarity of response. Of course, it might be argued that
any of the sample reactions cited above might be modified,
if the environment be changed. Prolonged practice may
teach us to inhibit the winking reflex at the sudden ap¬
proach of an object to the eyes. We might learn to run
to a large unknown object seen approaching in the dark.
We might learn to throw the board and blocks away when¬
ever they are presented to us. But these are not the or¬
dinary reactions that take place, and it is the ordinary
reactions which arise in the ordinary environment of man,
without specific attempt at modification, that we are
going to consider his original endowment. For educational
purposes it is this stock of reaction tendencies, which
arise without specific training, that forms the foundation
for further modification.

This native equipment of man can be conveniently
divided up into four divisions:

1. Reflexes.

2. Instincts.

3. Emotions.

4. Capacities.

The first division containing the reflexes has to do with
those reactions limited to various parts of the body which
follow very regularly upon the presentation of a given
stimulus. The winking reflex is a good example. The
patellar or knee-jerk reflex is another. There are many

ORIGINAL NATURE

reflexes. In education we are not concerned directly
with them. The instinctive reactions are the more com¬
plex reactions made by the whole body. There are a
great many of them and they are difficult to classify.
Their significance for education is, however, great, be¬
cause they form the basis for much of our conduct. Our
character and to some extent our interests in life are
largely determined by the strength or weakness of various
groups of instinctive tendencies, and by the successful
modification of these tendencies in various directions.

An emotion is an hereditary type of reaction involving
changes of the bodily mechanism as a whole, but particu¬
larly of the visceral and glandular systems. The shock
of the stimulus throws the organism into a more or less
chaotic state for a brief period. This occasions the vague
indefinite stirred-up feeling characteristic of the emotions.
Emotions, as well as instinctive tendencies, are important
for character development. The proper modification of
our emotional tendencies leads to emotional stability
amid the shocks of life. Poor or faulty modification of
these tendencies results in emotional instability and its
resultant character defects.

In the last division, namely the capacities, we are con¬
cerned mainly with reactions of an intellectual type.
Individuals differ by original nature as to their ability to
learn. With some individuals learning takes place quickly
and easily and with others learning is slow and hard, in
spite of similar opportunities and incentives. Perhaps
there are special capacities for learning some things rather
than others, such as words or numbers or music or me¬
chanical things. More likely such differences are due to
different interests based upon instinctive tendencies, or
different amounts of sensitive acuity as in music, and
undoubtedly many of these differences exhibited by chil¬
dren and adults are due to environmental factors influ¬
encing the child from early life. At any rate differences

8 FIELD OF EDUCATIONAL PSYCHOLOGY

in the general ability to learn are obvious and such dif¬
ferences are all-important in education.

In this study of the original nature of the individual we
shall note that the great advance made by modern psychol¬
ogy has been the attempt to describe in quantitative
terms. If we can tell the educator how much general
learning ability each child possesses he can plan more
definitely for each child’s education. With this amount
of intelligence we can expect to make this amount of
modification and so on. Hence an important part of our
study will be to see what instruments of measurement
have been devised and what they can accomplish. Not
only in this field of general capacity, but also in the vaguer
field of character qualities, we shall note the growth of
quantitative work, and the help that this gives us in learn¬
ing more about the original nature of the child.

The Modification of Original Nature. — Having studied
the general nature of the individual, we must inquire
how modifications of this original nature are brought
about. Man is the most modifiable of all animals and
modifications of his reaction tendencies are numerous.
The school sets out very definitely to make certain types
of modifications. So we are next confronted with the
Laws of Learning, which attempt to describe how modifi¬
cations in general are achieved. A study of these laws or
principles will give us some insight into the general char¬
acteristics of learning and may give us hints as to how
people may study most effectively. It may help the
teacher to direct more economically the learning of his
pupils.

The best way to understand how modifications of
reactions take place is to have individuals learn different
things in different ways and to record carefully their learn¬
ing from day to day. When we do this we obtain practice
curves and a study of such curves will tell about the rate
at which modification or learning takes place for various

MODIFICATION OF ORIGINAL NATURE 9

things learned. We shall also find out something about
the amount that can be learned and this will lead us on
naturally to discuss what limits are set by the human
organism as to the changes that can be made.

When changes or modifications have been made, when
things have been learned, do they remain fixed? How
permanent are they? This raises the question of forgetting
and we shall have to study how forgetting takes place,
how modifications change or tend to disappear when
practice ceases.

We do not learn one thing at a time. We do not wait
until we have learned one thing before beginning to learn
another. We are always learning a great many things at
the same time. In school a pupil is studying several sub¬
jects at the same time. Does the study of one subject
have any influence upon another? What influence does
one modification that is taking place have upon another
modification that we are trying to effect? If we have
already learned something, what influence does this have
upon other things that we are now learning? We might
call this reciprocal modification, the influence of one
modification upon another. It is, of course, the important
problem of transfer of training.

If we go on learning, practicing, trying to make modifi¬
cations continuously without any pause for rest, we even¬
tually become tired. Continuous work leads to fatigue.
How long can we work efficiently before fatigue becomes
noticeable? What effect does fatigue, mental or physical,
have upon the rate and efficiency of our work. Closely
allied to this problem is the influence of the loss of sleep
upon work, and also the influence of drugs upon mental
efficiency.

Throughout all this discussion of modifications of reac¬
tions or learning, we shall note how psychologists have
devised various experiments to measure these modifica¬
tions. But, finally, we shall turn to a study of the instru-

10 FIELD OF EDUCATIONAL PSYCHOLOGY

ments which psychologists have devised to measure very
specifically some of the chief modifications that the school
is more particularly trying to effect. The school delib¬
erately tries to build up a set of reactions towards printed
words, which we call reading. It builds up another
set of reactions towards numbers, which we call arith¬
metic, and so on for the other school subjects. How
effective is the school in making these modifications?
To measure these specific improvements made by the
school there have arisen a number of measuring instru¬
ments called educational tests and they are of extreme
importance for the educator. They are objective meas¬
ures of the modifications of original nature, which the
school is attempting to make.

Summary

This chapter has given a brief outline of the field of educa¬
tional psychology. If you glance over the chapter headings
in the Table of Contents, you will see that the chapters of the
book follow this outline of the contents of the book. Let us
put in schematic form the field of our subject, so that we may
keep it definitely before us throughout our study.

I. Original Nature.

1. Reflexes. Of slight educational significance.

2. Instincts. More complicated reactions. Important

for interests and character.

3. Emotions. Important for character development.

4. Capacities. General term for more specifically intellec¬

tual reactions. Measurement of these
by means of intelligence tests.

II. Modification of Original Nature.

1. Laws of Modification. The laws of learning. The way

changes are made.

2. Making Modifications. Practice. Amount, rate and

limits. Economical ways of
learning.

SUMMARY

6 .

Permanence of Modifications. Forgetting.
Reciprocal Modification. Transfer of Training.

Tests of

Continuous Modification. Fatigue.

Measurement of Modifications in Educa- school sub-

tion.

jects.

12 FIELD OF EDUCATIONAL PSYCHOLOGY

Review

True-False Statements

As a review of what you have read, mark the following state¬
ments true or false. If you have any doubt, go back to the text.

1. We inherit a certain type of nervous system which deter¬
mines the kinds of responses we can make.

2. The modifiability of original tendencies is a measure of the
educability of the individual.

3. The primary purpose of educational psychology is to set up
certain aims and ideals towards which educators should
strive.

4. Educational practice is based upon the fact that most orig¬
inal tendencies are modifiable.

5. All methods of education should build on the original tend¬
encies of the child.

6. Education in general attempts to preserve intact the original
and natural tendencies of the child.

7. Emotions and instincts are the basic elements upon which
character is built.

8. There is no limit to the possible modifications of human
behavior.

9. A man is the result of his original nature and his environ¬
ment.

10 . Educational tests are quantitative measures of the modifi¬
cations of original nature achieved by the school.

PART I

£1

;V Ht EDITOR

B. M. J.

ORIGINAL NATURE

CHAPTER II

THE ORIGINAL TENDENCIES

Inheritance of Definite Nervous System. — Man in¬
herits from his parents a certain type of body which
includes a certain type of nervous system. This means
that he inherits a given set of receiving organs for re¬
ceiving impressions from the external world, a central
nervous system, and also a given set of end organs,
muscles, glands, and the like, by means of which he may
respond to the impressions which he receives from the
outside. Because of this structure which man inherits,
he tends to respond to certain stimuli in a given way.
Animals respond to stimuli in general somewhat differently
from man. The greater the difference between the nerv¬
ous system of an animal and that of man, the greater is
the difference between the responses of the man and the
animal likely to be.

Different Kinds of Original Tendencies. — These gen¬
eral tendencies to respond in a given way to a given stimu¬
lus are called original tendencies. As we saw in Chapter
I, some of these tendencies are called reflexes, when the
response is limited to a very small part of the body. The
larger responses, resulting in a more complex type of
activity, issuing in a rather vague set of movements are
generally called instincts. Still vaguer tendencies to
respond are sometimes called capacities. These do not
result in obvious physical movements. They may be
provisionally described as tendencies to think, to remem¬
ber, to learn and the like. Responses which result in an
internal general “ stirring-up ” of the organism and do
not lead to any very definite set of movements are called
emotions.

THE ORIGINAL TENDENCIES

No Sharp Division between these Tendencies. —

There is no sharp dividing line between these various
tendencies to reaction. We must be careful not to let
the names for these various tendencies become magnified
into powers or explanatory causes of our actions. If we
say that speech is an instinct, we are in danger of assum¬
ing that this is an adequate explanation of how an individ¬
ual learns to talk and to rest content with such an explana¬
tion. If we say that curiosity is an instinct, it affords an
easy explanation of all the new things we like to try and
do, and we tend to ignore the fact that there are many
new things we do not like to try or do. For this reason
some psychologists think it better to abandon the name
of instincts altogether. What we inherit is merely a
vague unrest, a vague and indefinite tendency to respond.
The modes of unrest depend upon the presence of specific
organs and mechanisms. The particular responses that
result in any case depend upon the environment that sur¬
rounds us. And from the very beginning this environment
gradually conditions or influences the specific responses
that are made. From the very beginning, therefore,
there is an active interplay between the organism and the
environment and no original tendency or instinct stands
out clearly and independently of the environment.

Continuum of Original and Acquired Reactions. —
Original and acquired reactions are merely two names
given to the two extremes of a continuous series of reac¬
tions. There can be no dividing line between them. If
we were to list a long series of responses to situations,
beginning with those that are most predictable and going
gradually over to those less and less predictable, we would
have those reactions ordinarily called reflexes and in¬
stincts at the one end and those called acquired reactions
at the other. Let us try to make such a list, putting down
the stimuli at one side and the probable reaction or reac¬
tions on the other.

LISTS OF INSTINCTIVE TENDENCIES 17

Stimulus

Response

increase of illumination
sudden loss of support
tap on patellar tendon
movements of stomach walls
bitter substance in mouth
sight of novel object
calling of one’s name
word association to “king”
hear telephone ring
two plus two

see acquaintance on street
receive invitation
see postage stamps
receive some money
look at a picture

contraction of pupil of eye

clutching, reaching, fear

knee jerk

hunger

spitting

reaching, handling

answering in some way

“queen”

pick up receiver

four

smile, nod, etc.
accept, reject
collect, ignore, etc.
spend, save, etc.
various thoughts

Original Tendencies very Vague. — As we read down
the column of stimuli, it is obvious that we are less able
to predict the reaction that would take place in human
beings in general. Even from the very start our probable
responses assume a healthy individual, because the pupil¬
lary reflex and the patellar reflex are impaired in certain
diseases, and some idiots do not spit out bitter substances
placed in the mouth. Only with a certain percentage of
educated individuals will the response “queen” be given
to the stimulus “king” (see page 56). Not all of us col¬
lect postage stamps and what we may think, say or
feel when looking at a given picture is very difficult to
predict.

All that we mean by an original tendency to reaction,
then, must be that a given stimulus is more likely to call
forth one kind of a response rather than another, because
of the particular make-up of our organism. These original
tendencies will be vague and general and subject to much
modification.

Lists of Instinctive Tendencies. — Because the human
being possesses a highly complicated nervous system,
there are a great number of these instinctive tendencies

THE ORIGINAL TENDENCIES

to action. Almost any text-book of psychology discusses
them thoroughly and many psychologists have attempted
to make lists of instincts or groups of instinctive tend¬
encies. There is naturally great difference of opinion as
to just which reactions are instinctive and as to how
those that are considered instinctive should be grouped.
We shall content ourselves here with giving two lists
of instincts, one suggested by Thorndike and the other
by Woodworth.

Original Tendencies in Human Behavior

Thorndike

1. Sensitivities

2. Attentiveness

3. Bodily control

4. Food, habitation

5. Fear, fighting

6. Social instincts, gregarious¬
ness

7 . Manipulation
Visualization
Vocalization

8. Expression of emotions

9. Learning, remembering

Woodworth

1. Responses to organic needs
Thirst, hunger, breathing, heat
and cold, shrinking from in¬
jury, crying, fatigue, sleep

2. Responses to other persons
Herd or gregariousness
Mating or sex

Parental or mothering

3. Non-specific or play instincts
Kicking, running, etc.
Locomotion
Vocalization

Manipulation, exploration

Laughter

Fighting

Self-assertion

Submission, docility

It is not necessary for the student of education to
memorize these lists or any other lists proposed by other
writers. Naturally psychologists will differ greatly in
their descriptions of these vague general tendencies to
respond.

The Emotional Responses. — In addition to these
modes of response, which are generally called instinctive,

<f \

PRESENTED s . •

( ' ’> v THE f OUOR Of ' _ j

B - M - J - <#

■ mrM

Figure 1 . — What emotions or feelings are indicated by these
pictures? (From Feleky’s Feelings and Emotions.)

THE EMOTIONAL RESPONSES

there are also other responses of a general nature, which
we call emotional responses. These responses are char¬
acterized by significant physiological changes in the or¬
ganism. There is increased activity in some glands and
restriction of activity in others. Thus in anger the secre¬
tion of the gastric juice into the stomach is restricted,
while the activity of the adrenal glands is increased. And
so it is with other emotional states. They are responses
of the organism to stimuli in its environment, and at the
same time they are physiological urges to further action.
When something makes us angry, we may go on to fight
or to overcome it; when something makes us afraid, we
experience an emotion of fear and this is preparatory to
further reactions of flight or hiding or crying or shouting
for help or braving it out and so on according to the en¬
vironment and our previous training.

Like the so-called instinctive modes of response, these
emotional modes of response are very general and vague.
A burst of tears may be the result of intense joy or anger
or fear. In ordinary life we can interpret the emotional
expressions of others fairly well, if we know the circum¬
stances leading up to the expression. In watching a
moving picture, we also have little difficulty because we
know the story, and the actors in addition generally exag¬
gerate certain conventionalized modes of expression so as
to leave us in little doubt about the emotions they are sup¬
posed to be experiencing.

But if we do not know what led up to an emotional out¬
burst, and if we are not familiar with the individual's
usual modes of expression it is by no means so easy.
Study the pictures in Figure 1, and jot down underneath
what emotion you think is portrayed. When you have
done this turn to the end of the chapter and check your
results. Emotional modes of response, just like any other
modes of response, are modified by our social environment.
Upon the original vague modes of response are built

THE GillGINAL TENDENCIES

different types of expression to different types of situa¬
tions, and there are considerable individual differences
in these modes of expression.

Original Tendencies Determine General Form of So¬
ciety. — The original tendencies determine the general
background of our civilization and, therefore, of our
educational aims. The fact that man is a social animal,
that he is gregarious and tends to feel satisfaction when
living and working with others of his species, is the basis
of our present form of civilization with its cities and
villages, its family groups and societies. In this he re¬
sembles gregarious animals such as the buffalo or the sheep,
rather than solitary animals like the tiger or the leopard.
If man had been predominantly a solitary animal, the
form of our civilization and the type of our educa¬
tional system would undoubtedly have had a different
sort of evolution than the one we are familiar with at
present.

Manipulation. — Manipulation is mentioned by most
writers as one of the important original tendencies. By
manipulation is meant the general tendency to touch
with the hands or feet objects in the immediate environ¬
ment of the individual. We reach forward to feel and
handle any strange or new object, unless we have been
taught not to do so by previous experience with certain
classes of objects. This general tendency to manipulate
or handle is one of great educational significance. By
means of it the baby learns about countless things in
its environment. Some of these things cause pain or
annoyance and it learns to avoid such things in the future.
All through life this tendency persists to a greater or less
degree. We feel we do not “know” an object unless we
have handled and manipulated it. The desirability for
doing things in school rather than merely reading or hear¬
ing about them is obvious. Merely reading about the Es¬
kimo in the second grade may teach the children something

VOCALIZATION

valuable, but if they are encouraged to make an Eskimo
village on their sand table, with igloos and boats and all
the rest, they will be vastly more interested and they
will undoubtedly “know” more about Eskimos when they
have done this. So in the higher grades and in the uni¬
versity, the laboratory and the workshop should supple¬
ment the recitation, the lecture and the book, in order
that we may make use of and foster this basic instinct
of manipulation. Perhaps the individuals who have in¬
herited a very strong tendency to manipulation and who
have been encouraged to develop this tendency, are those
who have become our great inventors and our master
mechanics.

If this tendency is fostered, encouraged and rewarded,
it will expand and flourish in many ways. A social or
educational system may do this. The history of the
United States shows a society in which the manipulative
tendency and habits built thereon have been highly
prized and highly rewarded. This was not only the out¬
come of the pioneer life, in general, but also the result of
the type of democracy which exalted the farmer and the
mechanic somewhat at the expense of the mental worker.
The great industrial and mechanical transformation in
the nineteenth century gave further encouragement to
the manipulative tendencies of man. The mechanical
genius and inventor is to-day the idol of our youth. On
the other hand there are civilizations and there have
been periods in history where tendencies to manipulation
and habits built thereon have been repressed and utterly
despised. To work with the hands was not fitting for a
gentleman. The ideal in life may be contemplation and
not action.

Vocalization. — The original tendency to vocalization
is at the basis of all speech. This tendency is seen very
early in the babbling and cooing of the infant. There is
evident satisfaction resulting from the sounds made.

THE ORIGINAL TENDENCIES

Certain sounds will be repeated over and over again.
The child soon learns that the making of certain kinds
of sounds brings very definite satisfactions, such as more
attention from the mother, more food, and the like, and
so these sounds tend to be repeated and retained and
other sounds tend to be neglected and forgotten. The
child does not learn by imitation. We have to teach the
child to imitate, if we want him to imitate. Indeed in
the early stages of language development, it might more
truly be said that the fond parent imitates the gurgling
and cooing of the child. So much so, that in some cases
children concoct words of their own which parents per¬
petuate, and in many cases the baby talk of the infant
is carried over into later childhood much to the conster¬
nation of the parent. Indeed we might readily build up
a new language out of the babbling of the infant by
rewarding some sounds and repressing others, by attach¬
ing certain sounds to certain situations in the same way
as we learn everything else in life.

Vocalization begins very early. The cry of the new
born infant is one of the first reactions after birth. “The
respiratory tricks, the cooing and babbling of even the
young infant are a preparation for language,” according
to Gesell (1). This author reports a twenty-four hour
count of the vocal reactions of a six-months-old child.
Figure 2 shows the number of times each separate
sound was used. Da, for example, was spoken 63
times, ngrr occurred 21 times, and so on. Altogether
104 separate vocal reactions were made during the
twenty-four-hour period. Considering how much time a
six-months’ baby usually spends in sleep, this number
of vocal reactions emphasizes the importance of this
tendency.

Vocalization is the basis for all language. What spe¬
cific language will be learned by a child depends, of
course, upon its environment. To the young infant be-

OTHER ORIGINAL TENDENCIES

fJUMBET? OF TIMES USED

ngnn uya

Figure 2. —Vocalization Chart of L. M., Age Six Months. Sixty-
four different sounds were distinguished and their frequency is
shown in the diagram. (From GeselFs The Mental Growth of the
Pre-School Child.)

ginning to vocalize no language possesses any terrors.
It would just as readily learn Chinese, or Polish or Greek
as English. Only after we have learned one set of lan¬
guage habits, does the acquisition of a new set present
any difficulties. When the child enters school this tend¬
ency to vocalize is used in the story-telling and singing of
the kindergarten, in the discussions about things and in
dramatics and so on in the upper grades. Attempts to
repress this tendency in school by rules against talking,
by attempting to enforce silence are generally futile. The
wise teacher will use this very strong tendency as a power¬
ful ally rather than struggle incessantly against it as an
enemy.

THE ORIGINAL TENDENCIES

Other Original Tendencies. — We might go on in this
way describing the original tendencies of the human
being, tracing their influence on the individual as he
grows older. Education makes use of these original
tendencies in various ways in its attempt to modify the
individual’s reactions. The fact that human beings pos¬
sess these original tendencies, rather than others, has on
the other hand influenced the type of educational methods
and institutions common to the human race. Instead of
tracing these influences in detail, suggestions are given
below for the student. Take the first column headed
“manipulation.” This tendency is enlarged in early
home training by giving the child toys. In the pre-school
education many things surround the child to encourage
this tendency. He is encouraged to try to put on his own
coat, use a spoon and fork and so on. In the kinder¬
garten, educational material to handle is common — cut¬
ting out things, stringing beads, building with blocks, etc.
In the elementary school we have shop work and sewing,
in the high school and college we have our various shops
and laboratories. The diagram on the following page is
merely suggestive, the student should work out others for
himself.

Original Tendencies Influence our Conduct. — All

through our lives these original tendencies, greatly mod¬
ified in many cases, are at work. Advertisers make skill¬
ful appeals to them. We all wish for the approval of our
fellows more or less. We shrink from ridicule. We want
to be as good as the other fellow, and so we are greatly
moved by the harrowing story of the young woman whose
husband lost forever his chance to become “ general
manager” because he did not know the proper way to
eat olives at the great banquet, and so we rush wildly to
our booksellers for a copy of The Great Book on Eti¬
quette lest we should suffer a like fate. We denounce
kings and potentates and all their evil ways, but thrill

TENDENCIES INFLUENCE CONDUCT 25

M anipu-
lation

Bodily

Control

Grega¬

riousness

Vocaliza¬

tion

Fear

Early

home

Toys

Crawling

Walking

Playing
with par¬
ents and
sibs

Crowing

Gurgling

Disap¬
proval of
mother

Pre¬

school

Toys, use¬
ful things

Stairs

Slides

Playing

with

others

around

Talking
to other
children

Disap¬
proval of
teacher

Kinder¬

garten

Educa¬

tional

material

Marching

Dancing

Plays

Stories

Songs

Disap¬
proval of
teacher

Ele¬

mentary

Shop

Sewing

Games

Stunts

Games

Poems

Songs

Dramatics

Disap¬
proval of
teacher
and
group

High

school

Shop

Lab¬

oratory

Athletics

Dancing

Organized

Games

Societies

Dramat¬

ics

Debates

Group

disap¬

proval

College

Labora¬

tory

Athletics

Organiza¬
tions of
all sorts

Dramatics

Debates

Speeches

Group

disap¬

proval

with a glow of pride when we are elevated to the high
position of Grand Master of the High and Sacred Order
of Elephants or Monkeys. We scorn the titles and deco¬
rations of an effete European aristocracy, but we eagerly
decorate our bosoms with and suspend from various parts
of our anatomy pins and keys and mystic symbols to
advertise our greatness and mark us off from the com¬
mon herd.

These general examples are given so that the student

THE ORIGINAL TENDENCIES

may realize how original tendencies form the background
of our lives. We must be careful not to imagine that one
instinct works alone and independently of all the rest.
When we suggested that a particularly strong tendency
to manipulation may be one of the factors producing the
great inventor, we must remember the presence and in¬
fluence of countless other tendencies. The inheritance of
a great capacity to learn would also be necessary, and a
suitable environment that would encourage and stimulate
all these inherited tendencies is of course pre-supposed.
The lack of a suitable environment, the emphasis of a
non-active type of education might easily modify or
weaken the instinctive tendency towards manipulation,
and a great inventor w r ould not be produced.

The Modifiability of our Original Equipment. — It is
important to remember how easily modifiable these orig¬
inal tendencies are, some undoubtedly more so than
others. Man, as contrasted with other animals, may be
thought of as possessing the most loosely organized, the
most readily modifiable, set of original tendencies. This
makes man the great learner among animals. This fact
of plasticity or modifiability makes man more susceptible
to his environment than other animals. This modifi¬
ability is due to the plasticity of his nervous system. The
connections are loosely organized, and, therefore, there
is great possibility of reorganization and modification.
This plasticity is, of course, greatest at birth and grad¬
ually decreases with growth. The adult is not so plastic
or modifiable as the child. Note also the long period of
childhood in man as contrasted with the relatively short
period of childhood in most animals. The shorter the
period of childhood, the less opportunity is there for
modification. In an insect we have a very highly organ¬
ized set of instinctive tendencies with a very short or
complete absence of a period of childhood. Certain
stimuli are responded to from the very beginning in a defi-

FEW EMOTIONAL RESPONSES

nite way and almost no modification is possible through¬
out life. The insect is mature from the very beginning,
busy at all times responding definitely to the appropriate
stimuli. Man grows slowly; he is not busy or intent upon
a task every moment of the day. He is, as Woodworth so
well expresses it, “the most pottering, the most hem-
and-hawing of all animals.” It is because he potters that
he discovers various ways of doing things. It is because
he “ hems-and-haws ” that he discovers an infinite vari¬
ety of modes of expression, and develops complicated
languages in place of the restricted means of vocal expres¬
sion of the dog or monkey or bird.

The Vagueness and Indefiniteness of Our Original
Equipment. — So modifiable are our hereditary modes
of response and so early in life are they susceptible to
modification, that some psychologists question whether
there are really any instincts at all. They believe that
the organism exhibits at birth a great number of bodily
movements which are unorganized and chaotic. These
are the units of reaction, and out of these, by means of
the stimuli of the environment playing upon the organ¬
ism, various reaction systems are built up. Whether
this be the case or whether we inherit more specific
modes of response as we have previously indicated, the
important fact for the student is to remember the great
possibility of modification that exists from the very be¬
ginning of the life of the organism. Upon this fact of
modifiability all our hope of great educational improve¬
ment depends.

Few Emotional Responses. — In a similar way, the
number of original emotional modes of response have
been limited to three, namely fear, rage and love.

In work with very young children, after trying many
different stimuli, three types of responses seem to be
common. These have been called by Watson fear, love
and rage.

THE ORIGINAL TENDENCIES

Rage: Stimuli

Tear: Stimuli 1. Sudden removal of support from under child

2. Loud sounds

3. Sudden disturbance when falling asleep
Responses 1. Sudden catching of breath

2. Random clutching with hands

3. Puckering of the lips

4. Crying

1. Hampering movements

2. Holding arms by side

3. Holding head

Responses 1. Crying and screaming

2. Slashing and striking movements

1. Stroking

2. Tickling

3. Rocking

4. Patting
Responses 1. Smiling

2. Gurgling

3. Cooing

4. Extension of arms (embrace)

Love: Stimuli

These may be the chief situations calling forth emo¬
tional reactions in the very young child. Undoubtedly as
the child grows older other situations will call forth origi¬
nal emotional tendencies to respond. But from the very
first months of life other stimuli become attached to the
original modes of response, and the modes of response
themselves become modified. Many of the complicated
emotional reactions of the adult are to be thought of as
modifications and mixtures of these three simple types.

How do these modifications arise? If there are only a
few specific situations which call forth the emotion of
fear in early life, how do we account for the numerous
fear reactions which the child shows as he grows older? A
study of some experimental work done with babies by Wat¬
son (4) may make this clearer, and at the same time have
important bearings upon our understanding of the develop¬
ment of character qualities in general in the individual.

CONDITIONED EMOTIONAL REACTIONS 29

Conditioned Emotional Reactions. — At nine months
of age the child started violently at the sudden sound of
striking a steel bar behind his back. The second stimula¬
tion produced a violent start again and in addition the
lips began to pucker and tremble. A third stimulation
caused the child to cry and show the common expressions
of fear. We may consider these reactions as native or
original with respect to the stimulus of a sudden loud
unrecognized sound. The question now is, are these
fear reactions also called forth by many other stimuli
when presented to the child for the first time. Evidently
not, for the infant was confronted suddenly and for the
first time with a white rat, a rabbit, a dog, a monkey,
masks with and without hair, cotton wool, burning news¬
papers, etc. To most of these stimuli the instinctive
response was one of reaching out to touch and manipu¬
late. None of them called forth the crying, shrinking
responses of fear, which were elicited by the loud sound.
How, then, does the emotional reaction of fear become
attached to other objects? How can we shift the in¬
stinctive reaction called forth by one stimulus over to
another stimulus which does not by original nature elicit
it in the first place?

Let us continue our description of the experiment. The
white rat was again presented to the infant. As before
he began to reach for the rat and just as his hand touched
the rat, the steel bar was struck immediately behind his
head. He jumped violently and fell forward. This was
done twice. A few days later the rat was presented again
and he started to reach forward, when the rat touched his
hand, the hand was immediately withdrawn. Five more
presentations of the rat with the loud sound were given, and
there occurred an increasing tendency for the child to show
fear and to cry. After this the rat was presented alone,
without any sound stimulus. Immediately the infant be¬
gan to cry and to crawl away as rapidly as possible.

THE ORIGINAL TENDENCIES

We see, therefore, how a given response (fear) attached
by original nature to a given stimulus (loud sound) has
been shifted over to another stimulus (rat). We may also
say that a given response (manipulation), attached by
original nature to a given stimulus (rat), has now been
modified or changed into a different response (crying and
crawling away).

Transfer of Fear to Other Stimuli. — The experiment
we are describing continues to show how the fear response
called forth by the rat was also called forth by a rabbit
when presented to the child. We must remember that
the very first response to the rabbit, as to the rat, was a
manipulation response. Now after the reaction to the
rat has been changed to a fear reaction, the rabbit also
calls forth a fear reaction. It is not necessary to go through
a series of experiments with simultaneous sound and
rabbit to change the original manipulation reaction.
There is evidently sufficient similarity between the rat
and the rabbit to effect a transfer from the one to the
other. There are evidently enough identities between
the two situations to cause a response learned in connec¬
tion with the first situation to transfer over to a second
similar situation. This is an important factor in learn¬
ing and we will take it up later in detail in its wider edu¬
cational significance. Here we may simply note that
once the child has been taught to fear a rat, he has also
been taught to fear similar animals, such as a rabbit. The
experiment goes on to to show that this was true also, to
some extent, for a dog and a fur seal coat, but not for a
package of cotton wool, a Santa Claus mask or the hair
on the head of the experimenters. Furthermore it was
shown that after an interval of thirty-one days, the
stimulus of the rat still called forth the fear reaction.

The Process of Unconditioning. — Now let us raise the
question as to whether we can reverse the process described
in the above experiment. Suppose a child has already

THE PROCESS OF UNCONDITIONING 31

learned to fear a rabbit, can we employ the same procedure
in order to change the fear response to the rabbit? In
this case we must reverse the process and associate with
the fear object a definite stimulus which arouses a pleasant
reaction. What kind of a stimulus is potent in calling
forth a pleasant reaction? Obviously the sight of food
when hungry is by original nature attached closely to
reaching and touching reactions and later, when the food
is being eaten, there are strong feelings of satisfaction and
pleasure.

Another experimenter, Jones (2), has shown that in
this way a child who exhibited fear of a rabbit could be
taught to like the rabbit. In this case the child was 2
years and 10 months old and at the beginning of the exper¬
iment had already formed fear reactions (conditioned
reactions) to a rabbit, white rat, fur coat, etc. The “ di¬
rect conditioning” consisted in introducing the rabbit to
the child while he was being given food which he liked.
Through the presence of the pleasant stimulus (food)
whenever the rabbit was shown, the fear was eliminated
gradually in favor of a positive response. Steps in the
degree of toleration listed by the experimenter show very
nicely how the fear was overcome:

1. Rabbit anywhere in room in cage causes fear reac¬

tions.

2. Rabbit 12 feet away in cage tolerated.

3. Rabbit 4 feet away in cage tolerated.

4. Rabbit 3 feet away in cage tolerated.

5. Rabbit close in cage tolerated.

6. Rabbit free in room tolerated.

7. Rabbit touched when experimenter holds it.

8. Rabbit touched when free in room.

9. Rabbit defied by spitting at it, throwing things at it.

10. Rabbit allowed on tray of high chair.

11. Squats in defenseless position beside rabbit.

THE ORIGINAL TENDENCIES

12. Helps experimenter to carry rabbit.

13. Holds rabbit on lap.

14. Stays alone in room with rabbit.

15. Allows rabbit to play in pen with him.

16. Fondles rabbit affectionately.

17. Lets rabbit nibble his fingers.

This list showing the changes in reaction tells in brief
how the baby became more and more accustomed to the
rabbit. The experimenter did what many a wise mother
might well do in order to get her child over a fear for an
animal. At no time was the child suddenly shocked by
the near presence of the animal. The procedure sounds
like good common sense. It is well, however, for us to
have the procedure worked out systematically, so that
We may understand fully how gradually such a resistance
to a fear is built up. Other mothers or teachers not so
Wise might have suddenly thrust the rabbit upon the
child, compelled him to hold it and touch it, and by so
doing increased rather than diminished the fear. Compul¬
sion, with reference to things a child dislikes, probably
very often increases the dislike rather than lessens it.
Association of the disagreeable with agreeable associates
or consequences may be the most effective means of re¬
moving the dislike or fear of a given situation.

Diagram of Conditioning. — The changes in reaction
which these experiments have demonstrated are techni¬
cally known as conditioning and unconditioning. The
conditioned response is the changed or modified response.
What has been happening in these experiments may be
represented diagrammatically, where S stands for stimulus
and R for response and —> for “is followed by.”

OTHER CONDITIONING EXPERIMENTS 33

Conditioning. First Experiment.
S. 1. Loud Noise-»

S. 2. Rat-■>

S. 1. + S. 2.-->

S. 2.-■>

Transfer.

S. 3. Rabbit->

S. 4. Fur Coat->

No Transfer.

S. 5. Cotton Wool->

S. 6. Mask-■>

Permanence.

S. 2.-»

R. 1. Crying (original tend¬
ency)

R. 2. Manipulation (original
tendency)

R. 1. Repeated seven times

R. 1. Original tendency condi¬
tioned

R. 1.

R. 2.
R. 2.

R. 1. (After 31 days.) How
long will it last?

Unconditioning. Second Experiment.

S. 2. Rabbit-> R. 1. Fear (conditioned re¬

sponse)

S. 7. Playthings -» R. 2. Manipulation (original

tendency)

S. 8. Food-» R. 2. Handling, satisfaction

(original tendency)

S. 2. + S. 7.- > R. 2. Several times

S. 2. + S. 8.-■> R. 2. Several times

S. 2. Rabbit-> R. 2. Manipulation

We have described these experiments in modifying
original tendencies in some detail because they illustrate
admirably how very modifiable our original tendencies
are, and at the same time they indicate how this modifica¬
tion begins at a very early age.

Other Conditioning Experiments. — These experiments
in conditioning or changing a response from one situa¬
tion to another have been stimulated by the Russian
physiologist, Pavlov, and his pupils. The classical exper¬
iment is the measurement of the flow of saliva in the dog

THE ORIGINAL TENDENCIES

as illustrated in Figure 3. The sight or smell of food in¬
creases the flow of saliva. If food is always presented
along with another stimulus, the sound of a bell or a
colored light, then in course of time the sound of the bell
or the colored light will act as an adequate stimulus for
the salivary flow.

Figure 3. — Pavlov’s Conditioned Reflex in the Dog.

Many other conditioning experiments have been car¬
ried out with animals, children and adults. Such experi¬
mentation suggests the ways in which children become
conditioned in everyday life. A slamming door as the
child gropes its way up a dark stair may be the beginning
of fear of the dark. A four-year-old child saw a snake
and ran in delight to her father to bring him to see the
pretty animal. As they went back to look for the snake
hand in hand, coming upon it suddenly, the swish of the
snake in the long grass made them both start back, the
father reenforcing the fear of the child, and so may have

IMPORTANCE FOR EDUCATION

been started a lifelong fear of snakes. Up to her fifth
year a little girl had never objected to eating chicken,
but after returning from the country she declared she
did not like the taste and from that time on always refused
it. She had seen a chicken killed while on her country
visit. Many years after she still “ hated chicken,” al¬
though she had long forgotten all about the killing of the
chicken. As a matter of fact the taste of chicken was not
distasteful to her, because she ate it once when she thought
it was something else. The transfer effect of the chicken
incident rapidly spread, so that she soon “hated” turkey,
duck, goose and other birds.

Importance for Education. — The fact that our hates,
loves, fears and the like are being formed at an early age
may be of extreme importance to education in general
and character education in particular. We are coming to
believe that a great deal of basic training can be achieved
in the pre-school age. At present society does not see fit
to compel children to enter school before the age of six.
Kindergarten education has extended rapidly and many
4 and 5 year olds are to be found in our schools. Perhaps
more deliberate conscious education is desirable even in
earlier life. It may make a great difference to a child’s
character and temperament, if he can be freed from a
great number of the fears, which accumulate in the ordi¬
nary uncontrolled environment; if he can be made to
love a great many of the things which originally call
forth neutral or negative responses. Original tendencies to
manipulation, vocalization and the like may be extended
or restricted, they may be turned into desirable channels
more easily at an early age than in later childhood.

The understanding of how emotional reactions become
conditioned or associated with situations which originally
did not call out an emotional response is not only of value
for early childhood, but it may also help us in many school
situations. No doubt many of us “hate” arithmetic

THE ORIGINAL TENDENCIES

and mathematics and, therefore, avoid them, and, there¬
fore, are not very good at them, because somewhere in
our school career disagreeable emotions were aroused by
the teacher or by circumstances surrounding the lessons.
The fear or dislike became attached to the arithmetic and
in course of time a habit of reacting in this manner grew
up. In studying some high school children who were
finding algebra very difficult, one investigator (Symonds,
3) found that several of them reported a dislike for
arithmetic dating back to their elementary school ex¬
perience. “Always hated arithmetic”; “cannot grasp
mathematics” are some of the expressions the students
themselves used. In a similar way, we may be able to
understand some hatreds or fears of spelling. The spell¬
ing lesson may have brought ridicule upon the child and
an emotional response of fear or anger is aroused and
attached to the subject itself. Similarly our great likes
and loves for certain subjects may arise because of the
agreeable reactions called forth by the circumstances
surrounding the lesson. Perhaps we “love” geography
and history, because we started in a sandpile.

Of course, these reasons for ability or inability to master
a subject are not the only ones at work. The general
capacities of the student are important, as are also the
methods of learning, but these are factors which will be
treated at length later on.

Summary

1. The inheritance of the same type of body and nervous sys¬
tem leads to common modes of response among individuals
of the same species.

2. These original modes of response are called instincts, emo¬
tions, reflexes, capacities, etc.

3. There is no sharp division between these different original
tendencies, nor can original reactions be sharply differen¬
tiated from acquired ones.

SUMMARY

4. All reactions are more or less acquired, but all are based
upon the original modes of response inherited by the in¬
dividual,

5. Lists of original tendencies made by psychologists vary very
much, because our original tendencies are very vague and
very modifiable.

6. Emotional modes of response are also very general and
unless very extreme are difficult to interpret without pre¬
vious knowledge of what led up to the emotional response.

7. The original tendencies, nevertheless, are important as they
form the basis of our general form of society.

8. Manipulation is seen very early in the child and can be
made use of at all stages of the educational process.

9. Vocalization is the origin of all language. It begins very
early and is made use of in many ways all through school.

10. Many other original tendencies can be made use of by the
teacher and they should be consciously so used.

11. The original tendencies are more modifiable in man than in
other animals and hence man is the great learner. The
plasticity of his nervous system and the long period of child¬
hood give great opportunity for education.

12. The process of conditioning is one way by means of which
we learn many things. This begins at a very early age and
many of our fears, likes and dislikes are undoubtedly formed
by this means in early life.

13. Hence the value of deliberate education in early life. The
importance of the early formation of right habits is the
psychological justification of the nursery school and other
forms of pre-school education.

Pictures of Emotional Expression

The nine photographs in Figure 1 are supposed to depict the
following emotional states:

1. Intellectual attention

2. Agreeable surprise

3. Horror

4. Contempt

5. Sneering

6. Laughter

7. Terror

8. Rage

9. Hate

THE ORIGINAL TENDENCIES

Review

True-False Statements

As a review mark the following statements true or false:

1. Manipulation can be called an instinctive tendency to
react.

2. Three basic emotional tendencies common to all infants are
fear, rage and love.

3. Man has a more definitely organized system of original re¬
sponses than most animals.

4. All instinctive tendencies are present and active during the
first week of life.

5. It is a fair assumption that all instinctive behavior is useful
at the present time.

6. Conditioning original tendencies always works for the good
of the individual.

7. The best original tendency to use for disciplinary purposes
in school is fear.

8. The toys of the nursery school appeal to the original tendency
of manipulation.

9. Reactions conditioned in early childhood cannot be uncon¬
ditioned.

10. The long childhood of man allows much opportunity for his
original tendencies to be modified.

11. Man is originally attentive to sudden changes and sharp
contrasts in his environment.

12. An original response may be transferred to other stimuli
which did not previously arouse such a response.

13. You must teach a baby to walk if you want it to walk.

14. Many likes and dislikes of school subjects are probably con¬
ditioned responses.

15. When two situations producing conflicting responses are re¬
peated simultaneously several times, the stronger response
tends to eliminate the weaker and become the established
response for either or both situations.

16. Fear of snakes is probably an original tendency.

17. We find bitter things disagreeable because our original na¬
ture tells us that they are harmful.

18. Instinctively he reached for the telephone to report the fire.

REFERENCES

19. Hampering or interrupting the individual in what he is set
to do almost always causes anger or rage.

20. The first few years of a child’s life are probably extremely
important for the formation of basic likes and dislikes.

21. Underline the best method of removing fear:

(1) verbal appeal; (2) disuse; (3) direct conditioning; (4) re¬
pression.

22. Complete this sentence:

Capacities, like reflexes and instincts, are inherited, but they
are more-and more-.

Advised Readings

Woodworth, R. S.: Psychology, New York, 1921, Chapters 6, 7,
and 8.

Thorndike, E. L.: Educational Psychology: Briefer Course, New
York, Teachers College, 1914, Chapters 1, 2 and 3.
Watson, J. B.: Psychology, from the Standpoint of a Behaviorist,
Philadelphia, 1919, Chapters 6 and 7.

References

1. Gesell, A.: The Mental Growth of the Pre-School Child, New

York, 1925.

2. Jones, M. C.: “A Laboratory Study of Fear: The Case of

Peter,” Pedagogical Seminary, 1924, Volume 31, pages
308-315.

3. Symonds, P. M.: “ Special Disability in Algebra,” Teachers

College Contributions to Education, No. 132, Columbia
University, 1923.

4. Watson, J. B., and Rayner, R.: “Conditioned Emotional

Reactions,” Journal of Experimental Psychology, Vol¬
ume 3, No. 1, February, 1920, pages 1-14.

CHAPTER III

THE FUNCTIONING OF ORIGINAL
TENDENCIES

In this chapter we shall describe the functioning of the
original tendencies in the life of the individual, and what
happens when the original modes of response are not
permitted to take their normal course. We include both
the emotions and the instincts in our general expression
“ original modes of response.”

Conflict. — Let us begin with the fact of conflict. The
infant responds to its environment directly by means of
such original modes of response as it possesses. When
stimulated to anger it screams and strikes at the offend¬
ing object, when a thing pleases it the tendency is to
grab and pull towards and probably put in the mouth. No
one would think of blaming the child for such conduct.
Indeed to most adults such behavior seems rather amus¬
ing. Similar behavior in an older child or adult is, how¬
ever, not amusing. We expect modifications of these
tendencies to take place as the child grows older and if
they do not, we censure them. Now, all through life,
tendencies to respond directly as the infant does, in ac¬
cordance with our original modes of response, exist
within us. We are taught to suppress them, to alter
them, to turn them into other channels. When aroused
to anger we very frequently feel a tendency to strike out
or scream, but we realize that such conduct would cause
disastrous results. But the tendency is there, neverthe¬
less, stirring us up, seeking some means of expression.
Some outlet seems to be necessary and so we gradually

EXTROVERSION

learn to modify the original modes of response and to
seek other modes of expression.

Extroversion. — When the outlet is found in a reaction
towards something in our external environment, we call
it extroversion, a turning to the outside. The teacher
reprimands the child, justly or unjustly. The child feels
that he is being treated unjustly, because the reason for
the reprimand is not clear to him or beyond his compre¬
hension. He feels enraged, ready to strike and scream.
But previous experience has taught him that such a re¬
action will only cause more trouble and unpleasantness.
Hence he represses the original responses and turns them
into making unjust remarks to others about the teacher,
refusing to do his work, bullying smaller boys, and in
extreme and longstanding cases, perhaps, playing truant,
organizing a gang for the purpose of stealing or commit¬
ting vandalism of some kind or other. It is the individual’s
way of justifying himself to himself. It is one way of
maintaining his own self-respect. If joy and pleasure
and success do not come legitimately to them in the
course of their normal school activities, some children
will find these devious paths to obtain the satisfactions
to which they have a right. The teacher, therefore, must
see to it that punishments and reprimands are under¬
stood by the child. She must give opportunity to each
child to experience joy and success in his work.

Swearing, when it has not become an habitual language
response, is often an outlet for an emotional stress. The
complaining type of individual, the grouchy man, is often
letting off steam by means of his complaints, because he
has been blocked in the direct expression of his emotions.
Alcohol and drugs very often act as outlets for suppressed
tendencies to reaction. When a superior blames us for
something in our teaching, we vent our indignation upon
the helpless children in the class. We blame them for
what is obviously our fault. When we cut our finger

42 FUNCTIONING OF ORIGINAL TENDENCIES

because of our unskillful handling of the knife, we throw
the unoffending knife away. When we miss our drive
at golf, we hurl the innocent club to the ground.

Introversion. — Instead of responding directly to the
things in our immediate environment, our search for an
outlet for repressed modes of response may take another
direction. We may turn in upon ourselves and find sat¬
isfactions in certain less explicit modes of response. The
offending remark is made by the teacher and the child
sits down, apparently calm and undisturbed. But if we
could read his thoughts, we might be astounded at what
is happening to the teacher there. He is making up a
story in which he is being justified and the teacher properly
and thoroughly humiliated. Defeated, unsuccessful and
despised in school, the child may take refuge in his day¬
dreams, where without disagreeable effort, he marches
forth resplendent in glory, conquering all his enemies and
doing marvellous and noble deeds. The petty slights and
wrongs, the frequent unfairnesses of the world, “the
oppressor’s wrong, the proud man’s contumely,” all dis¬
appear as by magic in our day-dreams. Released from
external necessities, from the toil and exertion of over¬
coming obstacles, we attain our desires in imagination.
If the child has no companions or unsatisfactory com¬
panions, he very often creates imaginary ones and plays
with them and holds lengthy conversations with them.
Discouraged or disgruntled with our position, we may
become Napoleons and Alexanders in a “cleaner, greener
land.”

Desirable and Undesirable Outlets. — Outlets of some
sort are necessary. Our original modes of response are
very powerful and all through life they play their part,
ready to burst forth in all their primitive intensity in new
situations to which we have not learned to adjust our¬
selves. Many of the excesses common in war-time may
be explained by the newness of situations to which men

SUBLIMATION

have not acquired modes of adjustment, and also to the
lack of many of the customary restraining factors present
in a normal peace-time environment. Introversion and
extroversion are simply two types of adjustment. In both
these types we find desirable and undesirable outlets. In
the extroversion type of response obviously such outlets
as drink, drugs, burglary and the like are undesirable.
In the introversion type day-dreaming may not be unde¬
sirable unless carried to excess. A pronounced tendency
to introversion has been picturesquely called a “ flight
from reality.” If this flight from reality keeps us out of
touch with our surroundings too long and if it paralyzes
our active life, it becomes undesirable. If our day-dreams
never spur us to action, if they tend to become completely
satisfying in themselves, then we become more and more
shut-in, more and more independent of the external
w T orld. The extreme case of this type is the form of in¬
sanity known as Dementia Praecox, where the patient
may for days be totally oblivious to his surroundings, but,
nevertheless, living an active imaginary life. Other ab¬
normal forms of day-dreaming lead to delusions of gran¬
deur and power.

Sublimation. — The question then remains as to what
are the desirable outlets for the tendencies that must
necessarily be repressed in any form of civilization. They
may be either of the extroversion or introversion type
and in most cases these types will be mixed. In general
the desirable form of outlet will be such as to lead the
individual to a mode of response that will be not only
harmless, but will contribute something of positive value
to himself or society. Finding solace in a pipe of tobacco,
various kinds of swearing, a moderate amount of day¬
dreaming — all these are very common and harmless or
neutral ways of finding an outlet. They may not be
sufficient for the more dominant instinctive urges, such
as the sex urge, and they certainly do not turn these

44 FUNCTIONING OF ORIGINAL TENDENCIES

tendencies to action into constructive channels. If a
reprimand makes us grit our teeth and try hard to do
better, if failure makes us attack the task again more
doggedly, if our day-dreams result in ambitions which we
actually plan and work to realize, then we may sajr that
we have found desirable outlets for our instinctive modes
of response. Failure as well as success may be a potent
stimulus to further action. We must see to it that chil¬
dren’s failures do actually lead to renewed attempts at
mastery. Day-dreams may result in pictures of what
we should like to be and act as a real spur to ambition.
If the day-dream results in a concrete plan, in a poem or
in a story, the outlet that our emotions have thus found
is a desirable one. This turning of our instinctive urges
into desirable channels has been called sublimation. It is
a refinement, elevation or purification of the original
mode of response. When the original tendency to rage
or anger at some specific injustice done to an individual,
say a case of child labor, makes a man devote his life to
bettering the condition of children, to help pass laws for
their protection, we may say that a sublimation of the
instinctive tendencies to rage, anger and fighting has
taken place. When a teacher rebels at the drudgery and
poverty of her lot, and, instead of complaining about the
unfairness of her superintendent or board of education,
converts such energy into improving her own professional
training or in working for the general betterment of the
profession, we have the same process at work.

To be sure the necessity for outlets for the original
tendencies of children have been known by teachers in
all ages. Stern and repressive systems of education have
paid less attention to them than more liberal and en¬
lightened ones. A wise control of the sex life of the
adolescent is not to be found in the strict segregation of
the sexes and the treating as obscene all reference to sex.
The modern educator attempts to give an opportunity

RATIONALIZATION

for the sublimation of these tendencies by means of soci¬
eties, dances, games and the like, where the two sexes
may have adequate opportunities to meet and form the
right attitudes towards each other. If the teacher is con¬
sciously aware of the ways in which our original tenden¬
cies function, it should be possible to arrange better op¬
portunities for desirable outlets.

As one psychologist (Watts, 4) dramatically expresses
it: “The human soul is a battlefield upon which the irra¬
tional impulses of the personality strive with the rational
and ethical interests of the personality for the supremacy.
The educator must justify his vocation by effectually
assisting the latter forces to establish such a mastery
that the conquered systems of desire will function not as
rebels but as willing citizens, in the little kingdom of the
mental life.”

Rationalization, — Our original tendencies are among
the strongest drives to action that exist. They form the
underlying motives for much of our behavior. All through
our lives, they are insistent. Even although man prides
himself on being the rational animal par excellence, never¬
theless, much of his behavior is directly motivated by his
original tendencies. At the same time we are taught
continually to act according to reason. We try to be
reasonable beings. We search continually for the reasons
for all happenings. We are not content until we can
assign a reason for most things, and so it comes about
that we try to find reasons for all of our own conduct.
This necessity for finding reasons for his conduct is in¬
stilled in the child from an early age. He is constantly
being asked, “Why did you do that?” The “why” is
insistent and he is given no peace, until a plausible reason
is given by him or suggested to him.

Now, as a matter of fact, many of our acts are the
direct consequence of original tendencies and we fre¬
quently do not ourselves know the motive that has been

46 FUNCTIONING OF ORIGINAL TENDENCIES

at work, or, knowing it vaguely, we are not willing to
delve back to this real motive and bring it forth into light.
We are ashamed of many of these real motives and would
rather not recognize them at all. They frequently do not
agree with our own ideals, with the picture of our self as
we should like to have it. The necessity for explaining
our conduct either to others or to ourselves makes us,
therefore, find some plausible reason for our conduct.
This process of finding a reason for our behavior, which
is not the real reason, is called rationalization. We ra¬
tionalize or make reasonable our conduct, either because
we do not know the real reason, or because we do not
wish to acknowledge the real reason. We do not ration¬
alize when we attempt to find the real mainspring of
our action, and, finding it, bring it clearly before us, or,
not finding it, acknowledge the fact that we do not know
the reason.

Rationalization is seen in its simplest form in the
excuse-making of the child. The child does not want to
eat his oatmeal, and so it becomes “too hot” or “too
cold” or “too lumpy.” He does not want to do his arith¬
metic and so his pencil is “too long” or “too short” or
his “pen won’t write” or he “must do his spelling first.”
The student does not want to study and so he decides he
“must call up that friend” or he “must take more exer¬
cise and have a walk.” The father wants to live out of
the city near the golf course and so he buys that suburban
house, so that the family “may get more fresh air,” and
the children “have more space to play.”

Particularly in the realm of moral conduct do we find
many rationalizations for deviations from a strict moral
code. The Prohibition Amendment has caused countless
rationalizations among people who will not admit that
they are law breakers. Hart (1) says very aptly, “Thus
it is a familiar fact that people of otherwise irreproach¬
able honesty will swindle the government or a railway

RATIONALIZATION

company with untroubled equanimity. If they are taxed
with the incongruity between their principles and their
conduct, a varied crop of rationalizations will be imme¬
diately produced. They will point out that a company
is not the same thing as an individual, that nobody really
loses anything, that the fares or taxes are so inequitable
that it is justifiable to evade them, and so on. The dis¬
tinction between the real and apparent causes of mental
processes is well illustrated in the advice given to the
newly created judge, ‘Give your decision, it will probably
be right. But do not give your reasons, they will almost
certainly be wrong.’”

As the child grows older we notice how the crude and
unabashed “I want,” springing from the original tend¬
encies of his nature, gradually changes to “I should like
this because” or “if I had or did this, I could. . . . ”
Now a development out of the crude statement of our
wants or the direct satisfaction of our wants without
regard to others, is of course necessary. We all rationalize,
some more and some less. We must be careful, however,
that the practice should not go too far. We must be on
the watch, lest we build up a fictitious personality assign¬
ing only such reasons as this personality will allow, and
at the same time satisfying our crude wants and desires
under the protection of the ingenious rationalizations
which we devise. The danger is that we shall become less
and less able to face reality. We shall permit all our crud¬
est and undesirable wants to find satisfaction under the
guise of rational conduct, and at the same time imagine
ourselves as strong, virtuous and rational beings. We must,
so far as possible, therefore, face the real reasons for our
own actions and attempt in like manner to help the child
do so. This does not mean incessant lecturing and “rea¬
soning” with the child. From much of the lecturing and
reasoning done by the adult, the child obtains splendid
suggestions for his own rationalizations.

48 FUNCTIONING OF ORIGINAL TENDENCIES

The Complex. — When a great number of stimuli,
which in other people lead to varied reactions and ideas,
tend in a particular individual to arouse a given set of
reactions or call up a given constellation of ideas, we say
that the individual in question has a complex. Put in
another way, a complex has been described as a system
of emotionally toned ideas. When a boy says he is “ crazy
about radio,” when numerous stimuli, unconnected with
radio as far as the average person is concerned, all tend
to make him think and talk about radio and when he
spends all his money and time on radio, we say he has a
“ radio complex.”

Complexes may be of all sorts, narrow or wide, tempo¬
rary or relatively permanent. The fanatical reformer
may be said to have a complex with reference to the
particular reform which he is incessantly urging upon
everyone he meets. The enthusiastic scientist may have
a complex centering around a particular scientific hypoth¬
esis which he has propounded. In all these cases certain
dominant ideas seem to take possession of the mind. The
individual is set to respond in a certain particular direc¬
tion to the innumerable stimuli crowding upon him. The
individual may imagine that he is thinking and reasoning
logically, when he is really selecting only those stimuli
that fit in with the dominant trend of thought. In this
way he becomes illogical, for he is selecting his evidence,
and refusing to consider anything that may contradict
his beliefs. The anti-vaccinationist feels so strongly that
vaccination is wrong that he cannot think about the sub¬
ject without getting emotionally stirred up. He cannot
reason logically about it, because he steadfastly refuses
to consider all the evidence. He thinks he is logical and
rational, but arguments of the sort that would convince
him in other matters are utterly wasted upon him in this
particular case.

An “inferiority complex” is a term that has been ap-

FREUDIANISM

plied to a general feeling of inferiority to others. Thoughts
and expectations of failure are continually present. The
individual with this complex is always willing to let others
lead. Each new task is begun with the expectation of
failure. In all plans and hopes for the future, the dif¬
ficulties and dangers and chances of failure loom up
more clearly than the thoughts of success. Similarly, we
can speak of a “superiority complex,” which makes the
individual regard himself as vastly better than those
around him, even although his actions belie his thoughts.

In several forms of insanity we find complexes more
clearly present and more potent in their influence on be¬
havior than is ordinarily the case with sane people. Every
stimulus may set the patient off on his particular complex.
For years he may have been relating everything to this
one set of ideas, so that now it forms his entire universe.
Complexes may be harmful or harmless, according as
they dominate the individual's life. Any one set of
thoughts that tends to crowd out all others will make for
a badly balanced personality. The teacher must be on
his guard against fostering a superiority or inferiority
complex in any child. A certain amount of success is
necessary for every child, as well as a certain amount of
failure.

Freudianism. — A great many of the concepts and
ideas we have been using so far in this chapter have been
directly or indirectly introduced into our subject by
Freud's system of psychology. Freud started with an
interest in abnormal phenomena and gradually worked
out a scheme of psychology always from the point of
view of its direct value in the treatment of mental disease.
This system of psychology has become so popular that a
brief description of it may be valuable here. Some of its
concepts are so valuable that they have found a permanent
place in the academic psychology of the present time.
On the other hand the whole system has by no means

50 FUNCTIONING OF ORIGINAL TENDENCIES

been accepted and is not likely to be so accepted by
psychologists in general. References to Freud and his
theories are, however, becoming so common in educa¬
tional literature, that it is well for the teacher to have
some understanding of the main scheme of Freud’s theory.

Briefly, then, we may say that Freud postulates the
existence of original tendencies in man, as do most pres¬
ent-day psychologists. He differs, however, from the
majority of psychologists in maintaining that the supreme
instinctive tendency is the sex instinct. Indeed, as he
develops his system it would seem as if no other instinct
had much motive power, except the sex instinct. He
uses this term “sex” in a very broad sense, and would
seem to include under it the affection of the parent for the
young, and of the young for the parent and also that
general desire to be with others of the same species which
we have called gregariousness.

Now the varied manifestations of the sex instinct, or
desires or wishes as Freud calls them, continually seek
expression. A great many such expressions are permitted
in our present civilization, but a great many are not.
Hence a great many of our desires or wishes must be re¬
pressed. When they are repressed, they go down into
the subconscious. They do not cease to exist but remain
as forces in the subconscious. Here they seethe around
seeking a way out into consciousness. Barring their
way stands the censor, in Freud’s picturesque language,
who will not permit them to rise into consciousness in
their crude natural forms, because as such they will offend
and shock the individual. The more cultured we are, the
higher are the standards of our censor. Hence the crude
naked desires must clothe themselves in varied disguises
in order to find an exit. This they do and so they emerge
in various ways. In dreams they come to us with the
least disguise, because when we sleep the censor is not so
vigilant as when we are awake. Our dreams, therefore,

PSYCHO-ANALYSIS

are important as indicators of the repressed wishes and
desires seething in our subconsciousness. If the desires
assume a better disguise they emerge as driving forces in
our lives and with the normal person they fit in or are
sublimated and hence the idea of sublimation which we
have already discussed. If sublimation does not take
place we have the abnormal adjustment, and all the symp¬
toms of the neurotic, hysterical or psychopathic individual
are to be conceived of as modifications of repressed wishes.

Psycho-analysis. — If we wish to cure the patient of
the mental disturbance, we must analyze his mental
condition. We must go back from the present symptom
to the initial wish or desire. Make clear to him what the
real undisguised wish or desire was in the first place.
Once he understands this and is allowed to give the wish
free expression in some desirable way, the nervous dif¬
ficulty will disappear. The analysis leads frequently
back to early infancy and the Freudians emphasize the
importance of this period of life for the beginning of ab¬
normal tendencies.

This brief statement of Freudianism as a system of
psychology and psycho-analysis as a method of treat¬
ment for psychical disorders must suffice in this place.
There is no reason for an extended treatment in an educa¬
tional psychology. Like all brief treatments of a com¬
plicated subject it fails to do justice to the real contribu¬
tion of Freud. However, enough is given to enable the
reader to follow the main lines of thought and also to
indicate why the psychologist is not ready to accept
Freud’s theory. The complete dominance of the sex
instinct is doubtful. Other instincts are important mo¬
tives to action, particularly in childhood. It seems mis¬
leading and needlessly confusing to merge all childish
loves for parents and siblings and friends into one sex
instinct. Granted that the sex instinct is one of those
which needs great modification throughout life, there is

52 FUNCTIONING OF ORIGINAL TENDENCIES

yet no need for a subconsciousness within which repressed
desires remain as active forces. All this with the pic¬
turesque censor is utterly alien to the teachings of science.
And there is no need for such hypotheses. The modifica¬
tions of instinctive activities by means of habit are
adequate to explain the building up of abnormal modes of
behavior. And, lastly, although psycho-analysis may
claim many cures, yet so also may Christian Science,
Mesmerism, Our Lady of Lourdes, patent medicines,
electro-therapy, blood-letting, homeopathy, and the
King’s Touch. The superior person may scorn the ig¬
norant peasant plodding along to the confessional, while
he himself is going to the consulting room of the psycho¬
analyst.

In place of analysis we must think of training. The
hysteric, the psychopathic, is to be thought of as an in¬
dividual whose reactions are undesirable and who may by
proper training be taught to act and think normally.
Belief in the possibility of relief is necessary and we may,
therefore, think of all the above-mentioned methods of
cure as being or having been potent means for training
with some patients at some times, and yet they remain
mere makeshifts which cover our ignorance of the real
causes of the trouble. In its attempt to find causes, the
method of psycho-analysis is to be commended.

Methods of Experimentation. — Various methods have
been tried in order to study more objectively our original
tendencies. We shall give a brief description of some of
the more important of these and thereby, show the very
different lines of approach which have been tried.

I. Physiological Methods . — The stirred-up feeling that
we call an emotion is due to very decided physiological
changes in the body, particularly in the glands. Exciting
stimuli increase or decrease the normal secretions of these
glands. The measurement of such increase or decrease is
generally not possible in human beings, but much useful

METHODS OF EXPERIMENTATION 53

experimental work with animals has been accomplished.
Stimuli causing the disturbances that we call fear or
anger diminish very markedly the gastric secretions of
the stomach and tend to check the churning movements
of the stomach that pass the food along. Individuals
under the influence of fear and rage are not able to digest
their food.

Exciting stimuli also increase the secretion of the adrenal
glands. The secretion, adrenalin, releases the supply of
sugar stored in the liver. Testing for sugar in the blood
is, therefore, used as a means of detecting the emotional
effect of a stimulus. So far, this method has not been
used very extensively, but it may contribute valuable
information for psychology.

Methods of experimentation more directly applicable
to the human being deal with the changes in breathing,
in the pulse and in blood pressure. Emotional disturb¬
ances cause measurable changes in all these respects, but
the interpretation of such changes is difficult. The
psycho-galvanic reflex measures the changes in the re¬
sistance of the body to an electric current. Sudden
changes in amount of resistance seem to be symptomatic of
emotional disturbance.

II. Dreams. — This method is largely used by the
Freudians. A study of an individual’s dreams may help
to reveal those instinctive tendencies and emotions which
are more apt to be inhibited during his waking conscious¬
ness. The Freudians have worked out an elaborate sys¬
tem for the interpretation of the symbolism which occurs
in dreams. To the general psychologist their methods
appear very unscientific and unreliable. To Freud many
dreams show the fulfillment of wishes. The desires that
we cannot attain in our waking life are fulfilled in our
dream life. “The dream substitutes itself for action.”
Thus the starving man dreams of a gorgeous banquet;
the smoker who has stopped smoking dreams that he is

54 FUNCTIONING OF ORIGINAL TENDENCIES

enjoying a fine cigar; the lover dreams of his beloved.
The dreams of small children seem to be frequently of
this wish-fulfillment type. These straight-forward wish-
dreams seem easy of explanation, but the interpretation
of more complicated dreams is difficult and has not been
satisfactorily accomplished. However much we may
disagree with the wild speculations of the Freudians, we
must admit that they have at least presented a challenge
to psychology in insisting upon a psychological study of
dreams.

III. Word Reaction Methods. — These methods all de¬
pend upon giving the subject a stimulus in words and
asking him to react to it in words. They are usually called
Association Tests, and can be divided into two large
groups (a) Free Association and (b) Controlled Associa¬
tion.

A. Free Association. — Here we have three types: (1)
the method of continuous association in sentences. Here
the subject is just asked to relate his thoughts, the exper¬
imenter prompting and encouraging when necessary.
This method is used largely in psycho-analysis. The
starting point may be any complaint of the patient; it
may be a dream or it may be some incident that the
patient has previously related to the experimenter. What¬
ever the stimulus is the subject is asked to repeat any
thoughts that occur to him on the presentation of the
stimulus and to continue for a more or less indefinite
period.

(2) Continuous series of words. Here the subject is
given one word as a stimulus and is asked to repeat or
write down all the words that occur to him for a given
period of time. If the experimenter starts with the word
“dog,” the subject might respond by saying, “cat, house,
tree, bird, wood, builder, noise,” and so on. Here not
only the words themselves but their number and sequence
are of importance. A form of this test occurs in the Binet

METHODS OF EXPERIMENTATION 55

Scale, where the number of words a subject can repeat
in three minutes is used. As a written three minute test
it has been given to children of various ages, and we note
a great increase in the number of words as the child grows
older, from about 20 at age 8 to about double that at
age 15.

(3) Single word response. Here the subject reacts to
the stimulus word by a single word response. This is
generally done orally, the experimenter calling out the
word and taking the time which elapses between the
stimulus and the response. Any list of words may be
used, but a very common list to employ is a series of one
hundred words standardized by Kent and Rosanoff (2).
It is standardized in the sense that the frequencies of all
the different responses which occurred when the test was
given to thousands of people have been listed. Hence we
may tell how frequent or common any specific response
word is. The authors of the test point out the difference
between normal and insane subjects. Where the normal
subjects give on the average 92 per cent common responses
and only 7 per cent individual responses, the insane on the
other hand give only about 71 per cent common responses
and about 27 per cent individual responses. The insane
person is more likely to differ in his thoughts and ideas
from the average person around him and the association
test brings out this difference by the greater number of
unusual response words. The more eccentric a person is,
the more likely is he to show a high percentage of unusual
or individual responses. In no sense, however, must this
test be taken to be diagnostic of insanity. A high per¬
centage of individual reactions may occur in perfectly
sane individuals, and also there are many insane people
who show a high percentage of common reactions. Chil¬
dren below the age of eleven give fewer common and a
greater number of individual responses than adults. Above
age eleven the percentage of common and individual

56 FUNCTIONING OF ORIGINAL TENDENCIES

responses is much the same as among adults. Evidently
the younger children have not yet entered into that com¬
munity of ideas which is common to individuals brought
up in the same type of environment. Habits of thought are
as yet more fluid and have not become as fixed as with
adults.

B. Controlled Association . — In this type of test the
subject is called upon to respond by a word of a certain
sort. Only a limited number of words are deemed cor¬
rect responses. Thus, if the stimulus word is “good,”
and the subject has been instructed to give the first “op¬
posite” that occurs to him, he will obviously try to select
one of the limited number of words which may be con¬
sidered the opposite of good, such as “bad, evil, wicked,
rotten, harmful,” and the like. His reactions are “con¬
trolled” to this extent, in contradistinction to the giving
of any word in the free association test.

There are many different kinds of controlled associa¬
tion tests. Some of the common types are: opposites
(noisy—quiet); part-whole (leaf—tree); genus-species
(taste—sweet), but it is possible to use many other rela¬
tionships. Some of these tests, notably the finding of op¬
posites, have proved very valuable in intelligence testing.

An Association Experiment. — Let us try one of these
association tests so that we may become more familiar
with the procedure of the test, and so that from our own
results we may appreciate better the meaning and signif¬
icance of association tests in general. The reader may do
the test by himself, or else the test may be taken by the
class as a whole with the instructor of the class acting as
experimenter. If it is to be done as a class experiment,
do not read any further than the end of this paragraph.
Stop there and skip to page 58, continuing with the
paragraph headed “The Uses of Association Tests.” If
you are going to do the test yourself now, provide your¬
self with a sheet of paper and a pencil. In any case, be

DIRECTIONS FOR THE TEST

sure you do not read Appendix I at the end of this book.
Appendix I tells you how to score and gives you the re¬
sults of the test. If you read Appendix I before starting
on the test as described below, it will be useless for you
to take the test. Your results will be worthless and have
no significance whatever. Do not read or glance at
Appendix I until you have taken the test.

Directions for the Test. — When all students have been
provided with a sheet of paper, tell them to write the
numbers from 1 to 18 vertically down the left hand side
of the paper, so that they may write a word after each
number. Then say, “You will be asked to close your
eyes and then to think of something that belongs to a
general class, for instance, musical instrument, amuse¬
ment, algebraic symbol. If I should ask you to think of
“a musical instrument/’ the first thing of that kind that
might occur to you might be “piano” or “violin” or
“cornet” or “flute” and so on. When the class of thing
to be thought of is announced, notice what particular
thing of that kind comes first to your mind. Do not
search for anything else, but at once open your eyes and
write down the particular thing you thought of. Re sure
to write down the first one that occurs to you. Please
pay no attention to what your neighbor is writing. Does
everyone understand? Get ready.

1. Number 1. Close your eyes. Write down the name of a
color—the name of a color.

(To the instructor—spell “color” and any succeeding
stimulus word, if it is obvious that some of the students do
not understand the word. Speak clearly and deliberately.
Allow time for all to write — ordinarily a few seconds is
sufficient. Proceed immediately tc the next word.)

2. Number 2. Close your eyes. Write down the name of an
article of furniture — article of furniture. (Pause.)

3. Number 3. Close your eyes. Write down the name of a
flower — the name of a flower. (Pause.)

58 FUNCTIONING OF ORIGINAL TENDENCIES

4. Number 4. Close your eyes. Write down any letter of the
alphabet — letter of the alphabet. (Pause.)

5. Number 5. Close your eyes. Write down the name of a
metal — name of a metal. (Spell metal.) (Pause.)

6. Number 6. Close your eyes. Write down the name of an
historic personage — an historic personage. (Pause.)

7. Number 7. Close your eyes. Write down a part of speech
— part of speech. (Pause.)

8. Number 8. Close your eyes. Write down the name of a
geometrical figure — geometrical figure. (Pause.)

9. Number 9. Close your eyes. Write down any verb — any
verb. (Spell) v-e-r-b — verb. (Pause.)

10. Number 10. Close your eyes. Write down the name of a
tool — the name of a tool. (Pause.)

11. Number 11. Close your eyes. Write down an article of
food — article of food. (Pause.)

12. Number 12. Close your eyes. Write down a part of the
body — part of the body. (Pause.)

13. Number 13. Close your eyes. Write down the name of a
day of the week — day of the week. (Pause.)

14. Number 14. Close your eyes. Write down the name of a
room in a house — room in a house. (Pause.)

15. Number 15. Close your eyes. Write down the name of an
animal — an animal. (Pause.)

16. Number 16. Close your eyes. Write down the name of a
book — name of a book. (Pause.)

17. Number 17. Close your eyes. Write down a girl's name —
a girl's name. (Pause.)

18. Number 18. Close your eyes. Write down the name of a
country — a country. (Pause.)

Now turn to Appendix I at the end of the book and
score the papers according to directions. The experiment
is described fully by Pintner (3).

The Uses of Association Tests. — As we have seen
from our experience with the Community of Ideas Test,
which we have just taken, there is a great deal of similarity
in our word responses to given stimulus words. Where
there are many marked deviations from the common

PSYCHIATRICAL

response, there must evidently be a reason for these devia¬
tions. Emotional disturbances are likely to cause such
deviations. If we study a number of the deviations from
the common responses on a long free association test,
like the Kent-Rosanoff Test, we may be able to find some
connection between them. They may all refer to some
general idea or activity. In other words they may lead
us to the discovery of a “ complex, ” a group of emotionally
toned ideas. If something is weighing on our mind, if
we are worried about something, if we have recently had
an experience that has caused us considerable emotional
up-set, it is very likely that the stimulus words of an
association test will call forth response words that are
more or less directly connected with the worry or exciting
experience. In this way two general uses of the associa¬
tion type of test have arisen.

Psychiatrical. — In attempting to understand the men¬
tal difficulties of a patient, the association test has been
employed. The responses of the patient may cluster
around one or two central ideas or complexes and in this
way a beginning of the study of the difficulty may be
made. The physician will prepare his list of words so as
to tap the complex which he suspects may be present.
He will take into account the length of time taken to
react to each word.

A very brief example given by Jung and Peterson may
illustrate the method. Below r are given the stimulus
words, the responses given by the patient and the reac¬
tion times:

Stimulus Word

1. Head

2. Green

3. Water

4. Stick

5. Long

6. Ship

Reaction Word

Hair

Meadow

Deep

Knife

Table

Sink

Time in Seconds

1.4

1.6

5.0

1.6

1.2

60 FUNCTIONING OF ORIGINAL TENDENCIES

Stimulus Word

Reaction Word

Time in Seconds

7. Ask

Answer

1.6

8. Wool

Knit

1.6

9. Spiteful

Friendly

1.4

10. Lake

Water

4.0

11. Sick

Well

1.8

12. Ink

Black

1.2

13. Swim

Can Swim

3.8

Stimulus words 3, 6, 10 and 13 all result in lengthened
reaction times, and 13 also shows a peculiar response.
All these words center about water and pointed to a
suicide complex which dominated the patient.

Legal. — Some attempt has been made to use the
association test for the detection of guilty knowledge, on
the theory that the guilty person will have some group
of emotionally toned ideas surrounding the crime he has
committed. The experimenter, knowing the nature of
the crime, inserts significant words among the standard
list of stimulus words. The significant words are words
which might have some connection with the crime under
investigation, but they would in general appear to have
little connection with such crime by the innocent in¬
dividual. The stimulus words are then presented and the
time taken to react to each word carefully noted. Next
day, without previous warning, the association test is
repeated.

The experimenter has four clues to guilty knowledge.
The first clue consists of significant reactions to words
showing a knowledge of the crime. Such naive reactions
would be rarely given by the cautious criminal. The
second clue is the time taken to respond. If certain words
call up as a first response a word connected with the crime,
the guilty individual may suppress such a response and
go on in his mind seeking a second or third word which
is apparently innocent. Now this suppression and search
for other words takes time, when time is measured in

LEGAL

fifths or tenths of a second. All words that take ap¬
preciably longer than the average time of response, are,
therefore, brought together and scrutinized to see if they
point to a knowledge of the crime. The third aid to the
experimenter is the repetition of the list, for if during the
first test the individual has suppressed the first response
and gone on seeking another apparently innocent re¬
sponse, he has thereby disobeyed the directions for the
test and he will not in general be able to remember until
the next test what particular responses he made in these
cases. The great drawback about lying is that we cannot
remember the lies we have told.

The fourth clue is the number of stereotyped responses,
that is, similar responses or responses of a definite type,
indicating that the subject is not following the instruc¬
tions accurately, but has prepared himself beforehand for
the sort of thing he will say. He is not giving just the
first word that comes into his mind and hence suspicion
is cast on him.

This is a brief description of the association test as
applied to the detection of falsehood or crime. It de¬
pends upon the emotional state of the guilty man. Ob¬
viously where the thief is pursuing his work in the spirit
of an occupation or profession, his day’s work may not
arouse any emotional response, just as the ordinary rou¬
tine of the honest man does not, and in this case the asso¬
ciation test will be of no use. Again the association test
can only tell us where a suspicion of guilt lies, it cannot
prove guilt. As a matter of fact, it has not been used in
the practical investigation of crime, but the possibilities
of such practical use are obvious.

The association test is a good example of one of the
means used in experimental work with our emotions and
instinctive tendencies. We shall see in the next chapter
how it can be modified for use in the measurement of
non-intellectual traits. As we have described it in this

62 FUNCTIONING OF ORIGINAL TENDENCIES

chapter, it is of no particular practical value in the school.
It is interesting, as I have suggested, to give the Com¬
munity of Ideas Test to school children, and one may
learn things from so doing, but the test itself is of no im¬
mediate practical value.

Summary

1. This chapter supplements the preceding chapter by de¬
scribing what happens when the original tendencies are at
work.

2. Conflict frequently arises between the original tendencies
and the demands of the environment.

3. When a conflict arises an outlet is sought, either of the ex¬
troversion or introversion type.

4. Extroversion finds an outlet in a reaction to the environ¬
ment external to the individual.

5. Introversion is the type of outlet that drives the individual
back upon himself.

6. Outlets, either of the extroversion or introversion type, may
be desirable or undesirable, and it is the business of educa¬
tion and of mental hygiene in particular to help the indi¬
vidual find desirable outlets.

7. Sublimation is the refining of undesirable original modes of
response, and is the name used for the finding of desirable
outlets.

8. Rationalization is the process whereby we invent, con¬
sciously or unconsciously, motives or reasons for our ac¬
tions, so that our actions may conform to a rational mode
of life.

9. The complex is a constellation or group of ideas clustering
around some central idea. This group of ideas is emotion¬
ally toned and dominates much of our behavior.

10. Freudianism is a system of psychology built up by Freud
to explain the behavior of man, and particularly to throw
light on his abnormal behavior.

11. Psychoanalysis is a method of treatment for certain forms
of abnormal behavior.

12. Physiological methods of studying the emotions have in¬
creased our knowledge of the bodily changes during emotion.

SUMMARY

13. The analysis of dreams is supposed by Freudians to give an
insight into many hidden motives of behavior.

14. Word reaction methods are used to gain information per¬
taining to our emotions and instinctive tendencies.

15. Such methods are used to unearth complexes and may be
of help to the psychiatrist. They may also help in the dis¬
covery of guilty knowledge but they cannot be commonly
used in legal procedure.

64 FUNCTIONING OF ORIGINAL TENDENCIES

Review

True-False Statements

As a review mark the following statements true or false:

1. The best method of adjustment to the thwarting of our
fundamental impulsions is the substitution of some whole¬
some but vigorous activity.

2. A timid person, lacking in ability to make adequate social
adjustments, frequently takes refuge in day dreams as a
substitute for action.

3. One is always aware of the fact when he is rationalizing his
conduct.

4. Most cases of mental conflict are due to the sublimation of
instinctive tendencies.

5. The free association test may be of use in the discovery of
complexes.

6. By the term “ censor,’’ Freud means the various conven¬
tions and taboos which restrain the free expression of many
original tendencies.

7. When we honestly endeavor to find the real motive for our
acts, the process is called rationalization.

8. Substitutions for thwarted impulses never lead to great
achievements.

9. Repressed wishes in the 11 subconscious ” are probably in the
interior of the brain.

10. Parents and teachers should always insist upon a child giv¬
ing a reason for his misbehavior.

11. The best suggestion for mental health is the intelligent fac¬
ing of reality.

12. A group of emotionally toned ideas is known as a complex.

13. The subjects on which we are “touchy” probably indicate
complexes.

14. Association tests give positive proof of the guilt of criminals.

15. Sublimation is the turning of a repressed original tendency
into a socially desirable and useful channel.

16. Extroversion and introversion, being unnatural modes of
reaction, are invariably undesirable.

17. In the functioning of original tendencies, conflicts arise be¬
tween our original tendencies and our environment.

REVIEW

18. A child’s imaginary companion is an example of introver¬
sion.

19. The Freudian subconscious seems to be a region into which
we thrust our repressed desires.

20. Dream interpretation is for Freud a method for the dis¬
covery of complexes.

21. People who give few common responses in an association
test are undoubtedly insane.

22. Some of us who “hate mathematics” may have been con¬
ditioned against it by having a teacher whom we disliked.

23. An intelligent child should always be expected to give a
good reason for his acts.

24. Self-pity is a good example of extroversion.

25. As all original tendencies have evolved in the slow and long
struggle for existence, they are valuable and should be
thwarted as little as possible.

26. Extreme outward behavior of one sort may be an attempt
to hide an inward state which is directly opposite.

27. Indicate the probable outlet involved in each of the follow¬
ing reactions by putting the appropriate number before
each. (1) Extroversion; (2) Rationalization; (3) Sublima¬
tion; (4) Introversion.

( ) a. A slow pupil imagining himself to be some day a

great leader.

( ) b. A pupil making excuses to his parents for a low

mark in Arithmetic.

( ) c. A pupil muttering in a class.

( ) d. A boy, holding a grudge against his class-mate,

working off the grudge in a vigorous game of
tennis.

( ) e. A pupil playing the role of a bully on the play¬

ground.

( ) f. A man scrubbing floors in an asylum claims he is

a millionaire.

( ) g. Why struggle for mastery and leadership? — the

supreme virtues are inherent in meekness.

( ) h. A woman having lost her husband and child in a

terrible accident throws herself wholeheartedly
into her work as head of an orphanage.

66 FUNCTIONING OF ORIGINAL TENDENCIES

( ) i. A man low in the scale of authority in his place

of work rules his wife and children with an iron
rod.

( ) j. People extraordinarily competent along one line

must be deficient in other lines.

( ) k. A man of insignificant physique walks with dig¬

nified gait and manner and talks in a loud mas¬
terful voice.

( ) 1. A pretty girl has no brains. (Spoken by a homely

girl.)

28. Underline the best response:

When a suppressed original tendency seeks a desirable out¬
let, it is called: (1) introversion; (2) extroversion; (3) sub¬
limation; (4) rationalization.

29. Underline the best response:

A group of emotionally toned ideas is called: (1) repression;
(2) a complex; (3) the libido; (4) psycho-analysis.

Advised Readings

Morgan, J. J. B.: The Psychology of the Unadjusted School Child,
New York, 1924.

Wells, F. L.: Mental Adjustments, New York, 1917.

Gates, A. I.: Psychology for Students of Education, New York,
1923, Chapters 6, 7, 8 and 9.

References

1. Hart, B.: The Psychology of Insanity, Cambridge University

Press, 1912, Chapter 5.

2. Kent, G. H. and Rosanoff, A. J.: 11 A Study of Association

in Insanity,” American Journal of Insanity, Volume 67,
1910.

3. Pintner, R.: “Community of Ideas,” Psychological Review,

Volume 25, No. 5, Sept., 1918, pages 402-410.

4. Watts, F.: Abnormal Psychology and Education, New York,

1924, page 146.

CHAPTER IV

THE MEASUREMENT OF NON-INTELLECTUAL

TRAITS

Personality. — When we talk about the personality of
an individual we evidently mean those traits, character¬
istics or qualities of an individual which are found in him.
Psychologically his personality would be a catalog of all
the different responses made to all the different situa¬
tions that confront him. It would be a complete picture
of the man’s reactions, and under the term reactions we
include all the things he says and all the thoughts he
thinks. It is, of course, impossible to make such a cat¬
alog. All that we can ever hope to do is to note some of
the common tendencies to respond to certain types of
situations. In ordinary life and in literature what we do
when we describe the personality of a man is to note
those characteristics which are particularly strong in him,
those that stand out unequivocally, those that mark him
off from his fellows.

Ordinary Descriptions. — Our description of a man
might be as follows: He is a medium-sized man, rather
stockily built, with broad shoulders and large head. He
is not physically strong and does not enjoy good health.
He has made more than an average success in his profes¬
sion and is keenly interested in his professional work.
He does not, however, enjoy a very widespread reputa¬
tion among his professional co-workers, although some
rate him very highly. He is clear-cut and decisive in his
professional thinking and shows courage in defending his
own particular views and opinions. In other matters he
appears rather timid and is decidedly timid where physical

NON-INTELLECTUAL TRAITS

danger is concerned. He enjoys social gatherings and
has a great number of friends and acquaintances. He
enjoys indoor games, but rarely plays any outdoor ones.
He is poor in administrative matters, hesitating to come
to a final decision and is likely to let such things slide. He
is punctual in attendance to his duties, and shows high
ability in handling his work. He is widely read and pos¬
sesses a vast amount of well-coordinated knowledge. He
is slow to anger and keeps his emotions well under con¬
trol at all times.

This is an unorganized random description of an indi¬
vidual such as we find frequently in literature, but it
brings out very well the heterogeneous mass of character¬
istics that spring up when we talk of the personality or
character of an individual.

It is an interesting class exercise to ask students to jot
down four or five phrases descriptive of well-known men,
say Roosevelt and Wilson, and then have the class dis¬
cuss the possibility of grouping these descriptive phrases
into broad categories. Taking the replies of one class at
random, we have such phrases as these:

Roosevelt

altruistic
big stick
rough and ready
aggressive

physically weak when young

writer of books

fat

very informal
typical blusterer
extrovert

overcame obstacles
great knowledge of birds
big physically
an opportunist
quick to learn

Wilson

writer of interesting letters

man of letters

high idealism

an introvert

courageous

died broken-hearted

thin

independent
brilliant mind
leader
aristocratic
a scholar
liked to travel
very self-centered
single track mind

ORDINARY DESCRIPTIONS

These are samples of phrases written by different stu¬
dents and it is not our purpose here to judge the correct¬
ness of their opinions. The lists are merely indicative of
the sort of thing that results from general descriptions of
people. If now we try to group these phrases we get
words like “fat,” “thin” and the like referring to the
physical appearance of the men. Then there are certain
phrases like “brilliant mind,” “quick to learn,” indica¬
tive of the intelligence of the individual. Then there are
phrases such as “man of letters,” “scholar,” “great
knowledge of birds,” pointing to the knowledge or skill
possessed by the individual. Lastly and by far the most
numerous we find a heterogeneous group of phrases, such
as “introvert,” “extrovert,” “leader,” “informal,” “died
broken-hearted,” “aggressive,” “liked to travel,” and so
on, having reference to the emotional make-up of the
individual, his moral and personal qualities, his likes and
interests.

Classification of Traits. — We may, therefore, make a
first rough tentative classification as follows:

1. Statements with reference to physical qualities or
physical appearance.

2. Statements with reference to ability to do things, to
think, to learn.

3. Statements with reference to knowledge, skills, or
things learned.

4. A heterogeneous group of statements referring to
emotions, such as anger, moral qualities, such as con¬
scientiousness, and such qualities as punctuality,
courage, and also such things as likes, dislikes, in¬
terests and ambitions.

With the first set of statements we have, as psychol¬
ogists, little or nothing to do, except in so far as they may
be significant for conduct or the explanation of conduct.
The second set of statements refer to what the psychol-

NON-INTELLECTUAL TRAITS

ogist has set apart as intelligence, the innate ability of the
individual to learn, to be modified or changed along cer¬
tain lines, and a discussion of the measurement of intel¬
ligence as these qualities have come to be called, is given
in another chapter. The third set of statements deals
with the individual's acquired knowledge or skill; it is
concerned with what he has learned, what modifications
in an intellectual sense have been acquired, whether such
acquirements are abstract, like ability to read Virgil in
the original, or concrete, like ability to drive an automo¬
bile or paint a house. The last or heterogeneous group of
statements has not yet found one word to describe it. It
is with the measurement of these qualities that this chap¬
ter is to deal, and hence we have called them non-intellec¬
tual traits to distinguish them from the intellectual traits
and the different kinds of knowledge and skill which are
now so well measured by psychology.

This group of non-intellectual traits contains all the
emotions and all those traits commonly designated char¬
acter qualities. It includes also personal traits that are
not generally dignified by the name of character traits.
Moral qualities in a narrow sense are also included, and
so also are such things as likes, dislikes, interests, ambi¬
tions, ideals, hopes and aspirations. This jumble of
traits has not as yet been clearly analyzed by psychology.
Many a theoretical attempt at classification has been
made, but no one such classification has been accepted.
Undoubtedly a clearer picture will emerge when once we
are able to measure such qualities. Measurement will
bring with it sharpness of definition and clarity in classi¬
fication.

Dependence on Original Tendencies. — All these traits
help to make up what is commonly known as the char¬
acter of an individual. Psychologically we may think of
character as the total mass of conduct tendencies of the
individual. What a man is likely to do in various situa-

MEASUREMENT OF SUCH TRAITS 71

tions is the index of his character. And what he is likely
to do, can of course only be surmised from a study of
what he is doing now and what he has done in the past.
When we study the conduct of an individual we find a
set of habits of action, more or less loosely organized.
These modes of reaction are based upon his original tenden¬
cies, but they have been conditioned by his environment.
We cannot get rid of the habits or modifications that have
been built upon the original tendencies, but the varying
strengths of the original tendencies have undoubtedly
influenced the modifications that have arisen, and hence
the diversity of characters that occurs among individuals
brought up in the same environment. The characters of
brothers and sisters may differ greatly in spite of the
similarity of their environments. The problem in the
measurement of character qualities, then, is to arrange
test situations by means of which the measurement of any
trait or traits may be made. In arranging these we are
not concerned as to whether the reactions called forth are
instinctive or acquired. All that we want is a measure¬
ment of the responses which the individual now makes.
If there is a close agreement or correlation between the
responses made on the test and a certain type of reaction
in everyday life, then we say that the test is valid, that
it is an index of the type of reaction in question.

Measurement of Such Traits. — The growth of meas¬
urement of character traits has been very slow and we
are only now in the experimental stage of such work.
There is little that can be of immediate practical value to
the teacher, but a survey of what is being done will be of
interest. There is a tremendous research activity in this
field at the present time, and we cannot here refer to all
the work that is being done. All that we can do is to
pick out a few samples for purposes of illustration. In
making such a survey we will begin first with some de¬
scriptions or inventories of the conduct reactions of young

NON-INTELLECTUAL TRAITS

children. Here we shall see character qualities in the mak¬
ing, without any definite attempt to measure or evaluate
them, merely to note their appearance in the young child.
Secondly, we shall pass on to the rating scale which is a
device for estimating how much of a given trait an indi¬
vidual possesses. This is merely an attempt at a more
accurate description. We watch an individual or we
depend upon our previous knowledge of him, and then
say that he possesses so much of this trait and so much
of that, or that he is better than this other individual
in this or that trait. Thirdly, we arrive at the test proper
in which the individual makes certain responses, which
are evaluated and scored. The score is the measure of
the amount of the trait in question.

Descriptions or Inventories. — The descriptions we
shall mention here are not the usual literary descriptions
which we discussed at the beginning of this chapter.
Such descriptions are of comparatively little interest to
the psychologist. They depend too much upon the per¬
son making the descriptions, upon his subjective judg¬
ment and his individual standards, and they are also
made in very general terms. There are, however, some
very careful and detailed descriptions of the reactions of
infants which are helpful in a beginning of any study of
character. By studying a number of infants, we can
arrive at a common or general tendency. Gesell (2) has
made such summaries for young children. In giving such
summaries of the reactions of young children he empha¬
sizes the individual differences which are present. “Even
the four-months-old child appears to have his individu¬
alities and idiosyncracies. Indeed almost every infant
is likely to have some trait of character, some trick, habit,
achievement, or peculiarity which expresses this indi¬
viduality and which occasionally may be symptomatic of
a defect, weakness, or excellency.”

Some of the items in GeselFs summary for a four-

RATINGS

month-old baby are: Notices large objects; shows selec¬
tive interest in animated face; not much affected by
strange persons, new scenes or solitude; laughs aloud,
etc. At six months we note: expresses recognition of
familiar persons; may show consciousness of strangers;
enjoys presence and playfulness of persons, etc. At
twelve months we note: cooperates while being dressed;
inhibits simple acts on command; imitates simple acts
like scribble and spoon-rattle. These are only a few items
out of many at definite stages of growth from birth to
five years. These general conduct tendencies may serve
in a way as character tests for the infant. They show
types of response that are in general developed at specific
stages of growth. These summaries are not exactly
tests but they serve as standards in our examination of
infants.

Ratings. — Many rating scales for all sorts of charac¬
teristics have been attempted. Some depend on rank¬
ing individuals in order of merit. Others depend on the
comparison of the individual to be rated with individuals
well-known to the rater, which individuals have already
been arranged in an order of merit. This type of rating
scale has been called the man-to-man rating scale and it
was largely used in the army for the rating of officers.
Another type of rating scale, frequently called the graphic
rating scale, makes use of the distance between the op¬
posites of a given trait. Terman (7) has made use of
this type of rating scale for comparing gifted and con¬
trol children. Twenty-five separate traits are involved,
but there can be just as many or few as the investigator
desires. Each trait is represented on a long line as in
Figure 4.

The rater is asked to compare the child he is rating
with the average of the same age. He has to make a
cross anywhere on the line where he thinks the child
belongs. These positions on the line are then turned into

Trait 4- Self-Confidence

NON-INTELLECTUAL TRAITS

"B ~-a "s
H & Q .2.

* pH

44>

bJO

• -4

O •

02 rH

O fl ^ ' O

»i * i d

3d d

m 02

® Cl

d Q +3

i—B © <r-(

>> d“

T5 ^3

02 tO

-C! 3

' ^

0^+3
« S «

P 02 3d
02 r O

f-4

4a
02
d
i fi
uL +a
■=3 03

02 a
02 P

«+-<

c v

C4 §

2 P

^ i «'

d d3 g

rr> <V O

Ph to o

• rH
02
02

44>

CD biD O
^73
02 d

® Cl

435

Cl
02
P
i y=i

«+h cd

ID O

02 02

9? 3 £ & £

£ «g

* s

f-4

CO 03

4J dj

a . <

02 4i> 03

S'S $

d be r „

(3d®

d 2 ■ r/

H °S

>5 2

=4* I

02 P
02 o3

• rH

CQ
<D
-4^>

* i*"i

to p- 02

8 -j# M

d to

2 02

.5 -a

O! 44

bfl

4-5>

1 >

£ A

Tj *4-3

a> s

3 £

C2 to

< 4-1

bC-

, ri

44 d 02
*h d d bJO

£ £ £ $

d gg 3

- 02 cd

f -1
02
t>
d

bfl

t-i

-g

o H 03

02
C 4
O

4=1

a £

3 ^ C3 O
© 43 d bD

3 02 £ g
8 |
q a 3 d__

W § g <5 O O 3

rd 02
d d
d o

d 4j

P -

o 3
S gS

§ £ 43

Jh ^ ^

^ w ^

£ S

4^ °
d P

Ci d
4-1 d)

xs 'w

& '3

d d

44)

-d

44)

»s

445

455

C-

b£) d

C4 .2

02 *43

>; d

(-4

02 445

Figure 4. — Graphic Rating Scale. (From Terman’s Genetic Studies of Genius. Stanford University Press.)

RATINGS

scores from 1 to 13, 1 being the highest possible rating,
7 the average and 13 the lowest. With this method one
can score just as finely as he likes, but he should be care¬
ful not to score more minutely than differences in the
trait which can be appreciated by the average rater. A
finer scale than 1 to 13 would be of doubtful value. In
Terman’s investigation both parents and teachers rated
many children on many traits, and there is a slight tend¬
ency for them to agree in their judgments of children.

Undoubtedly such a graphic rating system is one of the
best methods devised, but, nevertheless, it suffers as do
all rating schemes from the subjectivity of the perform¬
ance. Our judgments of individuals are at best hap¬
hazard and influenced by all sorts of prejudices and pecu¬
liarities. The precise meaning attached to a trait varies
from rater to rater. What I have in mind when I think
of self-confidence is not exactly the same as what you
have in mind. Furthermore, we are influenced by our
likes and dislikes. If we like a person, we are more likely
to rate him in general high and, conversely, if we dislike
him. We naturally think of individuals as “wholes”
rather than as conglomerations of separate traits, and so
we judge one person to be “good and fine” and an¬
other to be “poor and weak.” This general impression
casts a “halo” over the individual and all our specific
judgments of his traits are influenced by the “halo” of
the general impression which surrounds him. All this
makes the rating of individuals difficult and unsatisfac¬
tory. In lieu of anything better, we use rating scales,
but they will disappear at once as soon as adequate
objective tests of character qualities appear. In actual
school work rating scales may be used to good advantage
by the teacher to make him study and think over the
personal characteristics of each child in the class, but the
teacher should be warned not to place much emphasis
upon the results of such ratings.

NON-INTELLECTUAL TRAITS

Tests. — Tests for the measurement of personal and
moral traits are the outcome of very recent research, so
recent, indeed, that the results and value of such tests as
exist are much in doubt. It would be useless here to
attempt a survey of all the different sorts of things that
have been tried. We shall merely choose a few sample
tests of different aspects of character and try to describe
what they are and how they propose to measure character
qualities.

Emotions. — People are conceived as differing in the
amount of emotional response that will be caused by the
same situation and also in respect to the number of situa¬
tions that will evoke an emotional response. The Wood-
worth Personal Data Sheet is an attempt to measure the
general emotionality or the emotional stability of the
individual. A modification of this questionnaire has been
prepared for the use of children by Mathews (4). It
consists of a number of questions to which the child re¬
sponds by underlining “yes” or “no.” These questions
may be roughly classified as dealing with:

1. Fears, worries and the like.

2. Physical symptoms, pains, weariness.

3. Unhappiness, unsocial and anti-social moods.

4. Dreams, phantasies, sleep disturbances.

Some of the actual questions are:

Do you like to play with other children?

Would you rather play by yourself alone?

Do people find fault with you much?

Are you afraid of the dark?

Do you dream of robbers?

Do you ever stutter or stammer in your speech?

There are 75 to 100 questions of this kind. No signif¬
icance is attached to any one answer, and, of course, the
psychologist knows that he cannot expect the truth to all

CHARACTER TRAITS

questions from every child. The total number of “ un¬
favorable ” responses does, however, seem to be signifi¬
cant. The greater the number of such responses, the
more likely is the child to be emotionally unstable. We
cannot as yet say that there are specific standards or
norms for this test, nor can it be regarded as a sure diag¬
nosis of emotional instability. It seems to have decided
value as a survey of large groups of children. Those
who make particularly high scores may then be given a
detailed examination by the psychologist and psychia¬
trist. This test is a good sample of one of the attempts to
penetrate into the difficult field of testing the emotional
make-up of the individual.

Character Traits. — Among the large number of traits
under this heading, we find psychologists attempting to
measure such specific traits as aggressiveness, suggestibil¬
ity, confidence, sociability, decisiveness or speed of de¬
cision, self-assurance and the like. Many different meth¬
ods of attack have been tried. Let us take as a sample
the tests for suggestibility by Otis (5). A group test for
use with children has been devised. Each child is given
a test sheet and the examiner gives instructions as to
what is to be done with each item, just as if it were a test
of following directions, indeed many items are pure direc¬
tions items and the suggestibility items are scattered at
random throughout. The suggestibility items run all
the way from things that are very obvious, as with a
picture that might well be the head of a horse where the
child is told to underline “yes” if it looks like a race horse,
to questions that are absurd, such as, “Draw a line under
one of these words (pine-trees, maples), if it answers this
question, “Do acorns come from pine-trees or from ma¬
ples?” or to such direct suggestions as, “Here are names
of different fruits. Draw a line under the name of the
one you like best. Most people like plums best.”

This test has been given to many children and it would

NON-INTELLECTUAL TRAITS

seem to show that ability to resist suggestion increases
as the child grows older. The adult shows a much greater
resistance than do children in general. It would seem
also that the more intelligent child is less open to sug¬
gestibility than the less intelligent one. This trait of
resistance to suggestion is not a simple clear-cut trait
but is obviously closely akin to such qualities as self¬
assertiveness, self-confidence, good judgment, self-reliance
and auto-criticism.

Moral Traits. — Certain tests have been made to test
more specifically the moral and ethical qualities. There
have been many workers and many tests. The recent
work of Hartshorne and May (3) will be our best example.
Their work on the measurement of deceit, or the tendency
to deceive, or the trait of honesty is the most extensive
piece of research in character measurement that we pos¬
sess up to the present time.

They have constructed a battery of tests, both individ¬
ual and group, to measure the tendencies to deceit as
exhibited by school children. Some of their methods are:

Duplicating Technique. — An intelligence or educational
achievement test is given, taken to the office and a record
of the child’s performance made. Then the unscored
paper is returned to the child, together with a scoring key
and he is asked to score his own paper. The number of
corrections and additions made is a measure of his honesty.

Improbable Achievement Technique. — Here we have
tests where the child can easily fake the answer and in so
doing show his dishonesty. If a difficult puzzle is given
with a time limit within which an expert cannot solve it
and if a child solves it easily within this time limit, it is
very probable that he has not been honest. The peg
solitaire puzzle was one used in this way. Weight dis¬
crimination of seven small pill boxes, marked in order on
the bottom, but differing so slightly in weight that no
one could arrange them in order, except by chance, was

MENTAL TRAITS

another test used with this technique. Here also belong
the various “ peeping” tests, where mazes are to be traced
or numbers put in circles with the eyes closed—feats im¬
possible of accomplishment if honestly done. Only those
who “peep” can do them.

Double Testing Technique. — Here the children are
tested twice; once under strict supervision, and the sec¬
ond time under lax conditions where copying answers
from the key or changing answers can be accomplished.
For this purpose duplicate forms of reliable tests are used
and the difference between forms under no cheating con¬
ditions must be known for the population to be tested.
If now, differences more than three times the standard
deviation of the non-cheating differences are found, then
the fact of cheating is probable and the amount of cheat¬
ing can be measured. Many different types of tests were
used with this technique, such as arithmetic, information
and other educational tests, as well as cancellation,
substitution and similar simple speed tests.

In addition to these several methods used in school,
they have tested honesty in different situations:—

Honesty in Home Work. — One form of a word knowl¬
edge test is given to the pupils to take home with instruc¬
tions not to get help from any person or dictionary.
Another form of the same test is then given in school
under strict supervision. Amount of deception is meas¬
ured as described above.

Honesty in Athletic Contests. — A contest was arranged
for the school. Four tests were given: (1) dynamometer
or strength of grip; (2) spirometer or lung capacity; (3)
chinning or pulling up on the horizontal bar; (4) the stand¬
ing broad jump. In all cases practice trials were given,
followed by other trials which the pupils recorded them¬
selves. The deception score is the difference between the
highest of the practice trials and the highest of the self-
recorded trials.

NON-INTELLECTUAL TRAITS

Honesty in Parlor Games. — Here certain parlor games,
such as Pinning the Tail to the Donkey, Bean Relay, The
Mystery Man and the like were used in carefully con¬
trolled situations to measure the children’s honesty.

In addition there were methods to measure the stealing
and the lying types of deception.

This brief summary of the types of tests used must
suffice here. The group tests for school children in the
classroom were used very extensively by Hartshorne and
May. As to the validity of such tests, they say, “In
general and test for test, we are able to measure deception
almost as consistently and with almost as little error as
we are able to measure intelligence.” One such test of
deception is of little value, but “if we use ten tests of
classroom deception, however, we can safely predict
what a subject will do on the average whenever ten simi¬
lar situations are presented.” If we want to measure the
average deceptive behavior of large groups of children,
a much smaller number than ten tests will suffice.

These tests were given by the investigators to many
school populations and the results have been presented in
great detail. No other moral trait has been so thoroughly
measured and reported upon up to the present time. We
may very briefly summarize the authors’ chief conclu¬
sions: Older pupils are slightly more deceptive than
younger children. Sex seems to make no difference.
Honesty is positively related to intelligence; in groups of
children of approximately the same age, those of higher
levels of intelligence deceive definitely less than those of
lower levels. Children who show symptoms of emotional
instability are more likely to deceive than those with
fewer such symptoms. Deceit is definitely associated
with the economic level of the home and also with
the cultural background. Children of North European
ancestry show less tendency to deceive than those of
South European ancestry. Colored children deceive more

INTERESTS

than most of the white groups. Little difference was
found between children belonging to the three main
religious groups, Catholics, Jews, and Protestants. De¬
ception runs in families to about the same extent as eye
color, length of forearm and other inherited structures.
In most tests there are no grade differences. Children
over-age for their grade cheat less than those under-age.
Those getting high marks cheat less than those getting
low marks. Pupils rated high in deportment cheat less
than those rated low. There is considerable resemblance
in amount of cheating between friends in the same class¬
room. There is less cheating where the relations between
teacher and pupil are free and cordial than where this is
not the case. Progressive schools show less cheating than
conventional schools. Children who attend Sunday-
School regularly cheat in day school about the same as
those who rarely or never attend Sunday-School. De¬
ceit is not a unified trait. Most children will deceive in
certain situations and not in others. The motives for
cheating, lying and stealing are complex and inhere for
the most part in the general situations themselves. The
most common motive for cheating on classroom exercises
is the desire to do well.

Interests. — In addition to personal characteristics and
moral traits, individuals differ markedly in their interests.
What we are interested in influences our total outlook
upon life, our activities and in a general sense our future.
Older people obviously differ markedly in their interests.
Children, as all teachers know, differ greatly in this re¬
spect also, and it would seem as if such differences in
interests began at a very early age and were somewhat
influenced by the inherited tendencies and capacities of
the child. However that may be, the emergence of defi¬
nite interest trends is well known and their significance
for educational and, later in life, for vocational guidance
is great. By the time a child comes to school and increas-

NON-INTELLECTUAL TRAITS

ingly so during his school life, definite interests are pres¬
ent and it is important for education to measure them,
if possible.

Many attempts have been made to tabulate children’s
interests by having the child check fists of things or indi¬
cate his preferences or indirectly show his interests by
means of information tests. All of these schemes have
given us some indication of the different interests of chil¬
dren, but they have been very unreliable and subjective.
Recently an attempt has been made (Terman, 6) to
measure interests more objectively by means of the asso¬
ciation type of test such as we have described in the
previous chapter. The test consists of a very carefully
compiled list of 120 stimulus words. Not any word will
do, as in the ordinary free association test, but each word
must be equally adapted to provoke responses due to
intellectual, social or activity interests, for these were
the three general classes of interests which it was proposed
to measure. The children were simply asked to write
down in one word what the words shown them made them
think of. All the response words given to these stimulus
words were then given a score value for each of the three
types of interest. These weights or values were deter¬
mined by the responses given by a group of children who
were well known by their teachers. The method is too
complicated to describe in detail here. In this way, for
example, the response “diamond” to the stimulus word
“gem” is rated 20 for intellectual, 11 for social, and 15 for
activity interest. If the stimulus word “grand” evokes
the response “piano,” then this is rated 8 for intellectual,
9 for social, and 11 for activity interests. Each response
word receives a score for each of the three types of interest.
A child, therefore, receives as a result of giving responses
to all the stimulus words three total scores—one for each
type of interest. This is as it should be, for no one is
solely interested in one aspect of fife. The three totals

GENERAL TEMPERAMENT

give a measure of the amount of interest in the three differ¬
ent directions, and a comparison of the three scores with
each other shows the main or dominant interest. Thus a
boy, who is described thus “ would rather read than play—
likes history—no leadership—dramatic ability—enjoys
music/’ has a high score for intellectual interests, and
only a moderate score for social and activity interests.
A girl described as “musical—quiet and unassuming—■
extraordinary amount of physical energy” has a high
activity interest score and average scores for social and
intellectual interests. If this test proves as successful in
general use as it has with the originator of the test, it would
seem as if we are on the way to a very satisfactory and ob¬
jective means for the measurement of children’s interests.

General Temperament. — The most ambitious attempt
to measure the total personality is the carefully con¬
structed Downey Will Temperament Test (1). It is an
attempt to give a picture of a great many personality
traits, particularly those that are included under the
term temperament. The temperament of an individual
is presumed to show itself in various patterned forms of
activity. These patterns are determined by (1) the
amount of nervous energy at the disposal of the indi¬
vidual and (2) the tendency of such energy either to dis¬
charge itself immediately into motor areas or to find a
round-about pathway of discharge. Temperament shows
itself, then, in reactions of all sorts, in our speech, in our
habits of walking, writing and doing all sorts of things.
For test purposes walking or talking are difficult to con¬
trol and measure and, therefore, writing has been chosen
as the type of activity for use in a test. We must note
carefully that this is not an attempt to judge character
from handwriting, as in the pseudo-scientific study of
graphology. Writing is merely used as a convenient
type of response which will be indicative of the general
habits of response of the individual.

NON-INTELLECTUAL TRAITS

The tests are constructed both for individual and
group testing. The subject is asked to write his name in
his usual style and speed, as rapidly as possible, as slowly
as possible. He is asked to write other things in these
three ways. He is asked to disguise his handwriting as
much as possible, and to copy various styles of writing,
sometimes as quickly as possible, sometimes at his own
speed. He is made to write with his eyes open, his eyes
closed, while counting and with other forms of distrac¬
tion. There is also a memory test, a speed of decision
test and a suggestibility test. In this way a number of per¬
sonality traits are tested and each one is measured on a
scale running from 0 to 10. Each trait is scored separately
and a psychograph, or picture in the form of a graph,
showing the strength or weakness of each trait is con¬
structed. This is called the will-profile and is presumed
to be a measure of the temperament of the individual.
Some of the items of this profile are: speed of decision,
self-confidence, finality of judgment, interest in detail,
perseveration, motor impulsion and the like. A strong
dominating personality shows high scores in most of the
traits, and the reverse is true of a weak vacillating in¬
dividual.

At present it is difficult to appraise this attempt to
measure the general temperament. The reliability of the
test is doubtful. The validity of the test is difficult to
determine. In the hands of some workers it has shown
very low correlations with estimates of character as made
by other persons, but of course these character ratings
are probably in themselves very unreliable. It has been
tried with school children, but it is not suited for younger
children. At present, therefore, we must regard it as in
the experimental stage and not suited for practical use
in schools. It is well worth consideration, however, as
being one of the first attempts to measure objectively the
total temperament of an individual.

SUMMARY

Conclusion. — As a final comment to this chapter, let
it be said again that mere samples of the vast amount of
research work have been taken. Furthermore, we have
not attempted to review all the work that has been done.
For such a review, see Watson’s book mentioned in the
Advised Readings at the end of this chapter. The most
promising type of work seems at present to be that done
by Hartshorne and May. The general importance of all
the work that has been done lies in the fact that we will
in the future be able to measure those more subtle changes
in reaction tendencies, subsumed under the general head¬
ing of character traits and moral qualities, which educa¬
tors have always been attempting to improve. At the
present time, however, there are no easily administered
tests which the schoolman can use.

Summary

1. This chapter deals with the attempts of psychologists to
measure non-intellectual traits. A few sample studies are
cited. It is not an attempt to review all the research work
that has been done in this field.

2. General descriptions of character are quite useless for scien¬
tific purposes, hence the need for tests.

3. Useful inventories of the reactions of infants are a beginning
of character measurement at this early stage.

4. Rating scales are an attempt to make our judgments of char¬
acter qualities more objective.

5. Rating scales, however, are still very subjective and their
reliability is not great.

6. The Woodworth Personal Data Sheet is an attempt to
measure the general emotional stability of the individual.

7. The Otis test for suggestibility is an example of an attempt
to isolate and measure a specific character trait.

8. The best example of the measurement technique in this
whole field is to be found in the work of Hartshorne and
May with their battery of Deception Tests.

9. These tests have shown high reliability and validity in the
measurement of deception for groups of school children.

NON-INTELLECTUAL TRAITS

10. These tests have been given to large populations and im¬
portant relationships between the tendency to deceive and
other factors have been discovered.

11. An attempt at the objective measurement of children’s
interests by means of the association test technique has
been described.

12. An attempt to measure the general temperament of an
individual has also been described.

REVIEW

Review

True-False Statements

As a review mark the following statements true or false:

1. In making a rating scale, it is essential that the scale be
divided into minute divisions in order to care for individual
differences.

2. Two of the Hartshorne-May classroom deception tests will
give a good measure of a child’s general tendency to cheat.

3. It is possible now to predict accurately whether or not a
boy will become a delinquent.

4. The Downey Tests attempt to measure such traits as hon¬
esty and truthfulness by means of an analysis of handwriting.

5. The “halo” effect is eliminated by the man-to-man type of
rating scale.

6. A sure sign of marked aggressiveness is to be able to look
another in the eye without flinching.

7. To measure a person’s honesty would require a very large
battery of tests.

8. The Woodworth Personal Data Sheet attempts to measure
such traits as perseverance and flexibility.

9. Temperament is supposed to show itself in various habitual
forms of activity.

10. Association tests have been used in an attempt to discover
the dominant interest trends of the child.

11. Honesty seems to be positively correlated with degree of
intelligence.

12. Children who come from poor homes are just as likely to be
honest in school work as children who come from wealthier
homes.

13. A person’s character can be very well determined by his
handwriting.

14. The resemblance between siblings in deception is about the
same in amount as the resemblance in eye-color.

15. Fill in the blanks:

Three well-known types of rating scales are called:

( 1 )-

(2 ) -

( 3) -

NON-INTELLECTUAL TRAITS

16. Complete this sentence:

The character of a man can be thought of as the total num¬
ber of his-.

17. Underline the best response:

There exist now fairly objective tests for the measurement
of: generosity, sociability, initiative, honesty.

18. Complete this sentence:

The Woodworth-Mathews questionnaire was designed to
measure-.

Advised Readings

Hartshorne, H., and May, M. A.: Studies in Deceit , New York,
1928. Introduction and Chapters III and XXIII.

Watson, G. B.: Experimentation and Measurement in Religious
Education, Association Press, New York, 1927, Chapters
III and IV.

References

1. Downey, J. E.: The Will Temperament and Its Testing ,

Yonkers, 1923.

2. Gesell, A.: The Mental Growth of the Pre-School Child, New

York, 1925, Chapter 32.

3. Hartshorne, H., and May, M. A.: Studies in Deceit, New

York, 1928.

4. Mathews, E.: “A Study of Emotional Stability in Chil¬

dren,” Journal of Delinquency, Volume 8, 1923, 1-40.

5. Otis, M.: “A Study of Suggestibility of Children, Archives of

Psychology, No. 70, 1924.

6. Terman, L. M.: Genetic Studies of Genius, Stanford Univer¬

sity Press, 1925, Volume I, Chapters 16 and 18.

CHAPTER V

THE MEASUREMENT OF INTELLECTUAL

TRAITS

Intelligence. — We saw in the last chapter in our general
description of an individual that in addition to character
and moral qualities, there are a certain number of traits
that we may call intellectual traits. It is with these
traits that we are now concerned, and we shall be solely
concerned in this chapter with such traits from the point
of view of ability to accomplish, rather than from the
standpoint of what has been accomplished by the indi¬
vidual. A measurement of the knowledge acquired by
the individual will come later in the second part of this
book. Here we are concerned with the possibility of
measuring his general ability to acquire knowledge.
This general ability to learn has been called general in¬
telligence. We shall for the time being accept this term
without further description. Later in this chapter we
shall attempt a more detailed description and definition.

Judgment of Intelligence. — This ability to learn is
the most important trait of the child from the point of
view of the educator. Most things taught in school have
their intellectual aspect. A great many make their ap¬
peal solely to the intellect. We, therefore, are making
use of a child’s intelligence continually, and every teacher
knows that children possess all degrees of intelligence.
Some are slow in learning and some are quick; some learn
easily and some only with great effort, and to others some
things are forever barred. A teacher is constantly sizing
up her pupils from this general standpoint of ability to
learn. We are all doing this all through life, with reference

90 MEASUREMENT: INTELLECTUAL TRAITS

to our friends, our colleagues, our competitors. How suc¬
cessful we are, it is difficult to tell. We know that we
often make gross mistakes, as witness the frequent use
of the expression, “I didn’t think he had it in him,” and
the like. Nevertheless, a great many people have much
faith in their ability to size up the intelligence of a person
even after a very brief interview. We cannot here inter¬
view real persons. Let us, however, make our problem
more concrete by making judgments of the intelligence
of children from their photographs.

A Practical Experiment. — This experiment may be
done either by the reader himself or it may be arranged
as a class experiment by the instructor. In either case
no reference to Appendix II should be made until after
the experiment, otherwise the results will be worthless.

See that each member of the class is provided with a
sheet of paper. Tell them to begin at the top left hand
side and write as heading of the first column “child” and
under this write down vertically, using one line for each
letter, the letters “A” to “L” inclusive. Now tell them
to draw vertical lines down their paper making five
columns in all. These five columns are to be headed as
follows: Column 1. “Child”; Column 2. “My Rank”;
Column 3. “Test Rank”; Column 4. “Difference”; Col¬
umn 5. “d 2 ”. Now tell them to turn over the page where
the photos of the children are printed, each child being
designated by a letter. Say to the class, “Here are
twelve photographs of children. You are to rank these
children in the order of their intelligence. The children
range in intelligence from very bright to very dull or
stupid. Study them carefully. Decide which one you
think is the brightest and put the number “1” opposite
the letter for this child in Column 2, under “My Rank,”
on your paper. Now look for the next brightest child and
record him as number “2,” and so on until you reach the
dullest child, who will be recorded as number “ 12.” Allow

Fic4ure 5(a)

Figure 5(b)

A RANK CORRELATION

sufficient time for all to complete the ranking, but do
not allow the students to discuss during the ranking and
do not allow them to ponder over it too long. This is un¬
necessary and tedious for everyone concerned.

While the students are doing the ranking the instructor
should write on the board the same kind of table for judg¬
ing as the students have on their papers. He will use
this as a sample of what the students are to do in working
out a rank correlation. He should take the paper of
one student who has completed the ranking and copy on
the board the ranks in Column 2 under “My Rank.”

At this point it is well for the instructor to go over the
judgments quickly by calling for a show of hands. Ask
who rated Photo A among the best three (i.e. ranks 1 , 2
or 3) and jot the number of hands down on the black¬
board, then ask who rated Photo A among the poorest 3
(i.e. ranks 10, 11 or 12). Do this rapidly for all the pho¬
tographs. In a large class it will be found that someone
has generally rated each photo either as among the three
best or the three worst. Here the point can be made that
people in general certainly differ tremendously among
themselves as to which child is most intelligent and which
is least intelligent, and that such judgments conflict
radically with each other.

The instructor should now turn to Appendix II and
write on the blackboard the ranks of the children accord¬
ing to the test. He should do this slowly having the
class copy them on their papers. He should begin with
rank 1 and comment on the intelligence ratings as he
goes along. Give the class time and opportunity to note
the differences between the test ranks and their own
ranks and to make the appropriate reactions of surprise
and amazement that are sure to follow, if the students
have not studied Appendix II previously.

A Rank Correlation. — After the test ranks have been
copied down by each student, tell them to subtract the

92 MEASUREMENT: INTELLECTUAL TRAITS

“test rank” and “my rank” columns, paying no atten¬
tion to signs. Demonstrate carefully with the example
on the board. These differences show by how much they
deviated from the true rank of each photo. Then in the
column “d 2 ” write the squares of the values in the “dif¬
ference” column. Add this last column and multiply
by 6.

Now write down the formula for rank correlation,

namely p

1 — 62d 2

and tell the students to copy this

n (n 2 — 1)’

on their papers, explaining that they are calculating a co¬
efficient of correlation according to this formula and that
they have each their own particular value for 62d 2 . Ex¬
plain what 2(sigma) and d 2 mean. Tell them that “n”
is the number of cases and that n(n 2 —1) is equal to
1716 and will be the same for every student. Now solve
the equation on the blackboard, explaining each step
carefully and then have the students solve their own
examples. If 62d 2 is greater than 1716, then a negative
correlation will result. Be sure to explain how to handle
such a case. Make a rough distribution of the coefficients
of correlation on the board, or in a large class have two
students prepare such a distribution, showing median
and quartiles, and report to the class the next day. Com¬
pare this distribution with the table of results given in the
appendix.

Value of Judgments of Intelligence. — Evidently our
judgments of intelligence obtained from photographs are
valueless (Pintner, 3). If any particular individual hap¬
pens to have obtained a high coefficient, he should not be
unduly proud of his achievement. If he were to repeat
this test again in six months or so, after he has forgotten
the ranking of these particular photographs, the chances
are that his coefficient of correlation this time would be
very low. If he tries another set of twelve pictures, the
correlation is likely to be very different. In other words

TESTS ARE STANDARDIZED INTERVIEWS 93

we have no evidence that any individual possesses the
ability of estimating the intelligence of individuals from
their photographs, although many people believe they
can tell a lot about an individual from his photo. Hence
the custom of having applicants for positions send a
photograph. Apart from telling something as to the
beauty or ugliness of the applicant, nothing can be dis¬
covered with reference to the intellectual or moral qualities.

Photos not a Fair Test. — Objection may be made to
our experiment, and rightly so, that teachers do not
judge the intelligence of the pupils from photographs
but from the living subject. If we were to conduct an
experiment with actual children standing in front of the
class, the correlations might be a, little higher. If the
children were interviewed and asked questions for a
minute or so each, the correlations would undoubtedly
increase. If each of us could have these children as
pupils in school for six months, then our ability to judge
their intelligence would be very greatly increased. And
finally, if we were to interview each child, using definite
standard questions, we should approach the best measure
of his intelligence, for in one sense, this is what an intelli¬
gence test is. The intelligence test presents the same
definite situations to each child, and in accordance with
his manner of response to these situations we judge his
intelligence.

Tests are Standardized Interviews. — The intelligence
test is a sort of standardized interview. It is standardized
in the sense that we know how children of various de¬
grees of intelligence are likely to respond to our questions.
There is nothing mysterious or magical about it. It is
simply doing in a better way, what we all do in daily
life, what the teacher, principal, or employment man¬
ager is doing when conducting an interview, namely,
inferring the amount of intelligence of the candidate
from the quality and number of responses he makes to

94 MEASUREMENT: INTELLECTUAL TRAITS

our questions or tests. The very important addition
made by the psychologist is in the standardization. It
tells him how to evaluate the different kinds of responses
that may be made to a given situation. Let us now take
a glance at the different kinds of tests which are used
to estimate intelligence.

Different Types of Tests. — We may divide the nu¬
merous intelligence tests that are now available into dif¬
ferent types. From the standpoint of the number of
subjects examined at one time, we have the useful divi¬
sion into individual and group tests. We may also dif¬
ferentiate tests according to the content of the test mate¬
rial, whether verbal or non-verbal, or again as to the
manner of giving directions for the test and making the
appropriate responses, whether with use of language or
without language (non-language). Other tests in which
the subject is called upon to do things with his hands are
called performance tests, to differentiate them from the
tests which are largely or entirely oral or written. Again
we might differentiate tests according to the age of the
subject for which they are best adapted. There are tests
suitable for the various ages, from the pre-school child
up to the university student. Finally we may divide
our measures into tests of general ability as contrasted
with tests of special capacities. The tests of general
abilities are what we think of as intelligence tests, whereas
the tests of special capacities are restricted in scope, such
as tests of musical talent, of artistic ability, of math¬
ematical ability, and the like. There are as yet very few
of such special capacity tests. Let us discuss some of the
characteristics of these different types of tests.

Individual Tests. — The individual method of testing
intelligence was the first method to arise and the most
commonly used scale to-day, the Stanford Revision of the
Binet-Simon Tests, is a revision of the original scale con¬
structed by the French psychologists, Binet and Simon.

INDIVIDUAL TESTS

The Stanford Revision (Terman, 7) is applicable to Amer¬
ican children brought up in the usual American environ¬
ment which assumes attendance at school from about
age six onwards. It consists of a miscellaneous group of
about ninety tests arranged in groups of about six or more
for most ages from age three to age fourteen with the
addition of tests for two higher levels, called average and
superior adult. These tests are given orally by the ex¬
aminer and the child is scored right or wrong. Only so
many tests are given as is necessary to determine the
intelligence level of the child. The whole scale is not
given to every individual. Beginning with a level well
within the child’s ability, the examiner proceeds as far
as possible up the scale to an age level where all the tests
are failed. Here are some samples of the tests at different
age levels:

Age Three: Naming Familiar Objects. The examiner
shows the child a key and asks, “What is this?”, and
similarly with a penny, a closed knife, a watch, a pencil.
If three responses are correct the test is passed.

Age Four: Comprehension. The examiner asks the
child (a) What must you do when you are sleepy? (b)
What ought you to do when you are cold? (c) What
ought you to do when you are hungry? Two correct re¬
sponses are required. What is to be called “correct” and
what “incorrect” has been standardized, and is therefore
not determined by each individual examiner’s own judg¬
ment.

Age Five: Comparison of Weights. Two weights, of 3
and 15 grams, are placed before the child about two or
three inches apart. The examiner says, “You see these
blocks. They look just alike, but one of them is heavy
and one is light. Try them and tell me which one is
heavier.” Three trials with position of blocks reversed are
given. Credit is allowed if two of three comparisons are
correct.

96 MEASUREMENT: INTELLECTUAL TRAITS

Age Six: Missing Parts. A standard card is shown with
faces having parts omitted. The examiner says, “ There is
something wrong with this face. It is not all there. Part
of it is left out. Look carefully and tell me what part of
the face is not there.” Credit is allowed if three out of
four are correct.

Age Seven: Description of Pictures. Certain standard
pictures are shown the child, and he is asked, “What is
this picture about? What is this a picture of?” Credit
is allowed if two out of three pictures are described or
interpreted. There are definite standards for enumera¬
tion, description and interpretation.

Age Eight: Finding Likenesses. The examiner says,
“I am going to name two things which are alike in some
way and I want you to tell me how they are alike. Wood
and coal, in what way are they alike? In what way are
an apple and a peach alike? In what way are iron and
silver alike? In what way are a ship and an automobile
alike?” Credit is given if two out of four are correct.
Again what is to be considered correct has been stand¬
ardized.

Age Nine: Four Digits Backward. The examiner says,
“ Listen carefully. I am going to read some numbers
and I want you to say them backwards. For example, if
I should say 5—1—4, you would say 4—1—5. Do you
understand? Ready now; listen carefully, and be sure
to say the numbers backward.” (a) 6, 5, 2, 8 (b) 4, 9, 3, 7
(c) 8, 6, 2, 9. Credit is allowed if one of the sets is correct.

Age Ten: Naming Sixty Words. The examiner says,
“Now I want to see how' many different words you can
name in three minutes. When I say ready, you must
begin and name the words as fast as you can, and I will
count them. Do you understand? Be sure to do your
very best, and remember that just any words will do, like
‘clouds/ ‘dog/ ‘chair/ ‘happy’; ready; go ahead.” If sixty
words are given within the time limit, the test is passed.

INDIVIDUAL TESTS

Age Twelve: Dissected Sentences. The examiner points
to the first group of words below and says, “Here is a
sentence that has the words all mixed up, so that they
don’t make any sense. If the words were changed around
in the right order they would make a good sentence.
Look carefully and see if you can tell me how the sen¬
tence ought to read.”

Figure 6(a). — Reactions to the Ball and Field Test of the Binet
Scale. Responses credited at Year VIII.

1. FOR THE STARTED AN WE COUNTRY EARLY
AT HOUR.

2. TO ASKED PAPER MY TEACHER CORRECT I
MY.

3. A DEFENDS DOG GOOD HIS BRAVELY
MASTER.

Credit is given if two out of three are correct, with a
time limit of one minute for each sentence.

98 MEASUREMENT: INTELLECTUAL TRAITS

Standard Procedure. — The examples of the Binet
tests given above show one test for each age level from
three to twelve. There are three higher levels. The
reader must remember that we have given only one out
of at least six tests for each level. It is important to
note further that the way in which each test is presented
to the child has been carefully standardized. To deviate
from the prescribed method of giving is to change the
test, and to change the test means that we cannot use
it in this scale because we do not know to which level
it belongs. Furthermore the scoring of the responses
has also been carefully standardized. Figures 6(a) and 6(b)
show samples of reactions to the Ball and Field Test, where
the child is told to trace out a path to show how he would
hunt for a ball lost in a round field. Credit is given at
year eight if the child more or less systematically covers
the field. So, in a similar manner, standards for all the
tests have been established.

Figure 6(b). — Reactions to the Ball and Field Test of the
Binet Scale. No credit at Year VIII.

OTHER BINET REVISIONS

Mental Age Rating. — From the examples of tests
given, we see the miscellaneous character of the Binet
Tests. The aim obviously is to obtain a sampling of
many different kinds of reactions, so that the final result
will be an index of all-round general ability. This final
result is expressed in terms of mental age. The mental
age of a child is an expression of that child’s ability in
terms of the average ability of children of like chrono¬
logical age. If a child is said to have a mental age or
M.A. of 8 years, we mean that he has accomplished on
our tests as much as the average eight-year-old child
accomplishes. All the tests have, therefore, been given
to many children of different ages in order to discover
the average performance at different age levels. It is
this procedure that is known as standardization. The
tests are valuable because they have been well standard¬
ized. It is only when tests have been standardized, that
we are able to attach significance to a particular score or
mental age made on a test.

The Intelligence Quotient. — Having obtained the
M.A. on a test, we know the amount of intelligence pos¬
sessed by the child, but we do not know how bright or
dull the child is. This will be determined by comparing
the child’s M.A. with his C.A. or chronological age. If
his M.A. is higher than his C.A., obviously he is more
advanced than the average child. If his M.A. is less
than his C.A., he is more retarded. This comparison of
M.A. with C.A. is best made by computing the ratio of
M.A. to C.A., and this ratio is called the Intelligence
Quotient or I.Q. Intelligence Quotients from 90 to
110 show normal intelligence development, those below
indicate dullness and those above indicate superior
ability.

Other Binet Revisions. — In addition to the Stanford
Revision of the Binet Test, there are several other revi¬
sions used in this country. Notable among such revisions

100 MEASUREMENT: INTELLECTUAL TRAITS

is the Herring Revision of the Binet-Simon Scale. It
tests the same sort of ability as the Stanford Revision
does, and it forms a valuable alternative to the latter.
If children have been frequently tested on the Stanford,
they become familiar with the tests. In such cases the
Herring Revision should be used. There are revisions of
the Binet Tests adapted for use in almost every civilized
country in the world, so valuable has the basic idea under¬
lying the tests proved to be.

Performance Tests. — Other individual scales of intel¬
ligence, departing radically from the Binet-Simon type
of test, have been constructed. The Pintner-Paterson
Performance Scale is most commonly used. This scale
instead of asking the children questions as in the Binet,
gives them something to do or perform, hence the name.
It consists of form boards of many types and the ability
of the child to solve these different problems presented
to him in concrete materials is measured.

In the Short Pintner-Paterson Performance Scale (4)
there are ten tests as follows:

1. The Mare and Foal Test.

2. The Seguin Form Board.

3. The Five Figure Board.

4. The Two Figure Board.

5. The Casuist Form Board.

6. The Manikin Test.

7. The Feature Profile Test.

8. The Ship Test.

9. The Picture Completion Test.

10. The Cube Test.

Figures 7, 8, 9, and 10 show four of the above tests
and from these pictures a good idea of the type of reaction
demanded can be formed. No verbal directions are re¬
quired.

-•

Figure 7. —The Mare and Foal Test. (From Pintner and
Paterson’s A Scale of Performance Tests.)

Figure 8. — The Five Figure Board Test. (From Pintner and Paterson’s A Scale of

Performance Tests.)

Figure 9. —The Casuist Form Board. (From Pintner and Paterson’s

A Scale of Performance Tests.)

INDIVIDUAL AND GROUP TESTS

101

Concrete Intelligence. — The Performance Scale is
measuring intelligence as shown in reactions to concrete
situations, as contrasted with the more abstract type of
situation presented in the Binet. The Performance Test
requires no language and hence can be used with sub¬
jects who cannot understand English. The use of the
Performance Scale along with the Binet enlarges our
view of the intelligence of the subject. This scale gives
the examiner a good opportunity to study the child at
work without the constant interruption necessary with
the Binet. The child’s attitude and methods of work
will frequently give the examiner valuable information
about the non-intellectual traits of the child.

Difference between Individual and Group Tests. — The
individual test method of examining a child requires
special training on the part of the examiner. Each test
must be given in exactly the same manner to each child,
and yet each child must be treated differently in order
to make sure that he is trying his best and giving his
fullest cooperation. In most individual tests there are
no rigid time limits, as we have in group tests. This
flexibility of the circumstances surrounding the individual
test, combined with the rigidity of the presentation of the
actual test itself, makes the individual test strikingly
different from the group test. An experienced psycho¬
logical examiner will follow any useful leads which the
reactions to the test situations on the part of the subject
may reveal. He will do this, however, without in the
least spoiling the standard presentation of the test itself.
At the end of the examination, therefore, he will not only
have a mental age determined in absolute conformity to
the requirements of the scale itself, but he will also have
obtained much valuable information about the subject
which the mental age itself cannot reveal. This wider
opportunity of observation that the individual test af¬
fords, makes it of paramount importance in the psy-

102 MEASUREMENT: INTELLECTUAL TRAITS

chological clinic. A mere routine examiner will gain from
his interview with the child nothing but a bare mental
age, but an experienced clinical psychologist will have
in addition, many valuable clues as to the various abil¬
ities of the child and as to his character and personality.

Group Tests. — In group tests this close contact be¬
tween the examiner and the subject is not possible. The
very nature of the examination prevents the possibility
of making individual adaptations to the particular id¬
iosyncrasies of the subjects. All the subjects, because
they are in a group, are treated exactly alike. Directions
for performing any particular test are given only once
and are not repeated for the benefit of any particular
subject, as may be done in many individual tests. If
any subject is unwilling or shy or frightened or nervous
or in a bad mood, he cannot be given special attention
by the examiner and the test halted until better rapport
has been established. Indeed the examiner is almost
always unaware of such individual differences in the
group that confronts him. Again the time limit for any
test must necessarily be rigidly adhered to. All of these
factors make the group test less flexible than the indi¬
vidual test, so that the mental rating resulting from a
group test may not always be as good a measure of what
a subject can do under the most favorable circumstances,
as is the case with a well-administered individual test.
On the other hand, the group examination spurs on many
subjects to maximum effort because of the feelings of
rivalry and competition which arise owing to the pres¬
ence of others working at the same task. And, further¬
more, just because it makes no allowance for individual
feelings of laziness, shyness and the like, the mental rat¬
ing resulting from a group test contains within itself a
measure of the general adaptability and conformability of
the individual to the school situation. In one sense, there¬
fore, the group mental age is not as pure and undiluted

KINDERGARTEN OR PRIMARY TEST 103

a measure of intelligence as the individual mental rating
is. Mixed in with the group intelligence rating is a rating
of the docility of the individual, his willingness to follow
directions, to cooperate in school work and work at his
maximum under orders. This is no drawback to the
group test for school purposes, because these non-intel¬
lectual factors enter into most school activities. The
group test is, therefore, just as valuable as the individual
test in helping us obtain a complete picture of the gen¬
eral intelligence of the individual.

There are all types of group intelligence tests suited for
all levels of intelligence from the kindergarten stage to
the university. The vast majority are verbal, either in
the sense that they require a knowledge of understanding
the spoken language of the directions, or else assume the
ability to read printed words, or both. There are a few
that are strictly non-verbal, not depending on language
either in understanding directions or in working the test.
Brief descriptions of some of these different types of
group tests will be given. Only one or two tests of each
type will be mentioned, the choice being made among
commonly used tests. There are many other tests just
as good as those selected as samples.

A Kindergarten or Primary Test. — The Pintner-
Cunningham Primary Mental Test can be given to groups
of children entering school for the first time, either in the
kindergarten or first grade. It consists of a booklet of
pictures and the children are told what to do with each
set of pictures. Figure 11 shows page 4 of the test book¬
let. The children are told to “Look at the three girls at
the top of the page. I want you to find the prettiest.
Don’t tell any one, but look at all the girls. Then put a
mark on the prettiest girl. Go ahead. (Wait until all
have finished). Now look at all the elephants. Mark
the prettiest elephant. (Pause not longer than 10 sec¬
onds). Look at all the houses. Mark the prettiest house.

104 MEASUREMENT: INTELLECTUAL TRAITS

Figure 11 . — Mark the prettiest in each row. From the Pintner-
Cunningham Primary Mental Test. (World Book Co.)

(Pause 10 seconds.) ” Figure 12 shows page 12 of the
test booklet, and here the problem is to find the head for
the man and mark it, and then the arm and then the leg.

ningham Primary Mental Test. (World Book Co.)

There are altogether seven sub-tests in the total test,
with many items in each sub-test. None of the items
require a knowledge of words or letters or numbers, but

A HIGH SCHOOL TEST

105

the children must be able to understand English in order
to follow the directions. The test is given in the spirit of
a game and young children are very much interested in
it. A group test at this period of school life is extremely
useful in helping the teacher to classify a new group of
children about whose abilities practically nothing is
known. In one sense this is the most crucial period of a
child’s school life, because his first year in school may
determine his attitude towards school in general, and
during the first year habits of industry or idleness are
readily acquired.

An Elementary School Test. — The National Intelli¬
gence Test can be used with children from about the
fourth to the eighth grade. Each test consists of five
sub-tests, and the sub-tests for Scale A are: (1) Arith¬
metic Reasoning. (2) Completing Sentences, such as,

“The dog - black.” (3) Common Attributes, e.g.

underline the two words which tell what a mouse always
has — back, cat, eyes, cheese, trap. (4) Similarities and
Differences, e.g. marking which of pairs of words have
similar or different meanings — new and old, fall and
drop, expect and anticipate. (5) Symbol-Digit Test, in
which the subject copies the digits for the appropriate
symbols as shown in a key at the top of the test page.

Figure 13 shows Test 3 of the National Intelligence
Test, Scale A.

The National Intelligence Test is very widely used in
schools and has proved an extremely useful test. There
are two scales and several forms of each scale.

A High School Test. — The Terman Group Test of
Mental Ability is one of the most frequently used tests for
high school pupils. It consists of ten tests as follows: (1)
Information; underline the right word in “Coffee is a kind
of bark, berry, leaf, root.” (2) Best Answer; check the
best answer to this question — Spokes of a wheel are
often made of hickory because (a) hickory is tough, (b)

106 MEASUREMENT: INTELLECTUAL TRAITS

lit each row draw a line under each of the two words that tell what the thing always
has.

Samples

man

dog

house

( body cane head shoes teeth)
(blanket chain collar legs nose )
(cellar paint room servants walls)

Begin here 1 hen (chickens corn feathers neck roost)

2 tiger (bones cage cubs fur jungle)

3 squirrel (acorn fur nest peanut tail)

4 kitten (ball claws eating eyes mouse)

5 bicycle (basket bell brake frame wheels)

6 stone (field hardness hurt throwing weight)

7 lion (cage head keeper mane prey)

8 face (cheeks eyebrow glasses mustache mouth)

9 forest (cone9 flowers grass soil trees)

10 paper (edges envelope printing surface watermark)

11 Borneo (airplanes land rivers saloons universities)

12 gully (flowers sand sides steepness stream)

13 piano (keys mulic pedals scarf stool)

14 satisfaction (conquering contentment money pleasure trouble)

15 illness (ailment discomfort doctor nurse recovery)

16 mob (confusion excitement hanging negro torches)

17 fire (ashes danger flame heat wood)

18 sea (coast reefs salt shoals submarines)

19 alley (cans fence narrowness passage pavement)

20 crime (death lawlessness punishment theft wrong)

21 pilot (cap chart knowledge license raincoat)

22 measles (discomfort doctor nurse rash recovery)

23 nun (beauty convent teacher vow woman)

24 citizen (city country male privileges vote)

Figure 13. — Common Attributes. The National Intelligence
Test. Scale A. (World Book Co.)

it cuts easily, (c) it takes paint nicely. (3) Word Mean¬
ing; mark whether pairs of words mean the same or oppo¬
site, e.g. expel —- retain; comfort — console; etc. (4)
Logical Selection; underline the two right words in “A
horse always has harness, hoofs, shoes, stable, tail.”
(5) Arithmetic Reasoning Problems. (6) Sentence Mean¬
ing; answer “yes” or “no” to such questions as: Does a
conscientious person ever make mistakes? Is an alloy

A HIGH SCHOOL TEST

107

ANALOGIES

Samples

Ear is to hear as eye is to

table see hand play
Hat is to head as shoe is to

arm coat foot leg

Do them all like samples.

1 Coat is to wear as bread is to

eat starve water cook. i

2 Week is to month as month is to

year hour minute century.. z

3 Monday is to Tuesday as Friday is to

week Thursday day Saturday. 3

4 Tell is to told as speak is to

sing spoke speaking sang. 4

5 Lion is to animal as rose is to

smell leaf plant thorn. ~. 5

6 Cat is to tiger as dog is to

wolf bark bite snap... .... 6

7 Success is to joy as failure is to

sadness luck fail work. 7

8 Liberty is to freedom as bondage is to

negro slavery free suffer. 8

9 Cry is to laugh as sadness is to

death joy coffin doctor.. 9

10 Tiger is to hair as trout is to

water fish scales swims.... 10

11 1 is to 3 as 9 is to

18 27 36 45... .. 11

12 Lead is to heavy as cork is to

bottle weight light Scat. 12

13 Poison is to death as food is to

eat bird life bad. 13

14 4 is to 16 as 5 is to

7 45 35 25. 14

15 Food is to hunger as water is to

drink clear thirst pure . 15

16 b is to d as second is to

third later fourth last..-. 16

17 City is to mayor as army is to

navy soldier general private. 17

18 Here is to there as this is to

these those that then.. 18

19 Subject is to predicate as noun is to

pronoun adverb verb adjective. 19

20 Corrupt is to depraved as sacred is to

Bible hallowed prayer Sunday. 20

Right .

Figure 14 . — The Analogies Test from the Terman Group
Test of Mental Ability. (World Book Co.)

108 MEASUREMENT: INTELLECTUAL TRAITS

a kind of musical instrument? etc. (7) Analogies; Coat
is to Wear as Bread is to eat, starve, water, cook. (8)
Mixed Sentences; mark true or false mixed up sentences,
such as, true bought cannot friendship be; etc. (9)
Classification; cross out the word that does not belong
in the series, e.g. Frank, James, John, Sarah, William; or
death, grief, picnic, poverty, sadness. (10) Number
Series; continue the series with two numbers, e.g. 8 7 6 5

4 3-; or 16, 8, 4, 2, 1, -. Figure 14 shows

Test 7 of this test.

A College Test. — The most thorough-going test de¬
vised for college students is the Thorndike Intelligence
Examination. It consists of a battery of four tests, the
first two containing much of the same type of material
we have already described and the last two emphasizing
strongly the ability to read and understand difficult
passages.

A Non-Language Test. — All of the tests so far de¬
scribed presuppose an English-speaking environment for
the child. They are testing intelligence by means of
verbal knowledge. The Pintner Non-Language Test is
a good example of a very usable test which makes no
such presuppositions. It does not use oral language in
the directions nor printed words in the test itself. It con¬
sists of six tests as follows: (1) Movement Imitation, i.e.
reproducing the movements of a pointer after it has been
moved from dot to dot in different ways on the black¬
board. (2) Easy Learning; (3) Hard Learning; these two
tests are tests of the digit-symbol type involving the
learning of new associations. (4) Drawing Completion,
i.e. drawing in the missing parts of pictures. (5) Re¬
versed Drawings, i.e. reproducing geometrical forms as
they would be when turned upside-down. (6) Picture
Reconstruction, i.e. indicating by numbers the positions
of the parts of pictures so as to make a complete picture.

This test is well adapted to children in Grades III to

A NON-LANGUAGE TEST

109

VIII. Figure 15 shows the drawing completion page of
this test, where the problem is to complete the essential
missing part of the picture.

Non-Language Mental Test.

The great value of this non-language test is that it can
be given to children of foreign parentage whose English
is poor. It can also be effectively used to compare chil-

110 MEASUREMENT: INTELLECTUAL TRAITS

dren of different nationalities, and it is one of the few
group tests that can be used with deaf children. It is
also very useful for all children, as it is testing a somewhat
different aspect of intelligence from the usual verbal test.

NUMBER SEQUENCE TEST

Put

Look at the sample that follows:

2 4 6 8 10 12—

What number should come next? (A) 17 (B) 13 (C) 14 (D) 15 (E) 9

The right answer is 14, so put the letter that goes with it, C, in the margin. —-

Read this sample:

1 8 2 8 3 8— (E)8 (F)6 <G)2

The right answer is 4, so you must put H in the margin.

Do the rest in the same way. Remember to put the letter in the margin.

6 —

(N)S

(0)2

(P)7

(0)3

(R)8

15-

(S) 10

(T) 5

(U) 14

(V) 15

(W)20

(W)l

(X)3

(0)9

(Y)2

(Z)4

(A) 8

(B) 11

(D)6

(E)10

* ...

25-

(F) 28

(G) 30

(H) 29

0)21

(K) 26

(K)4

(L) 10

(M)6

(N)2

(0)3

5 —

(K)3

(L)7

(M)6

(N)2

(0)3

(K)3

(L)5

(M) 2

(N)7

(0)4

103

63 —

(D) 64

(E)54

(F) 47

(G) 55

(H)51

1/3-

(X)2/3

(Y)l/9

(Z)0

(A) 1/27

(B) 1/81

.. *

Figure 16. — Test 3 of the Pintner Rapid Survey Intelligence
Test. (T. C. Bureau of Publications.)

A Survey Test. — A Survey Test is a short test designed
for survey purposes where large numbers of children are
to be tested and where the emphasis is laid upon the per¬
formance of groups rather than upon individuals. The
Pintner Rapid Survey Intelligence Test is a short test
made up of four sub-tests of the usual verbal type, namely,
(1) Opposites; (2) Analogies; (3) Number Sequence; (4)
Classification. Figure 16 shows the Number Sequence
Test in Form A of this test. Very rapid and accurate

(H) 4 0)9

here

TESTS FOR ADULTS

111

scoring is achieved on this test by the device of having the
child write a number or letter in the margin, the correct
numbers or letters forming a sequence easily remembered
by the scorer. For example the right answers in the test
shown in Figure 16 will spell the word PSYCHOLOGY.

FORM 5 GROUP EXAMINATION ALPHA GROUP NO_

Name------ Rant^-—_ Age...,.

Company---- RegimentArm...—... Division._

In what country or state horn?. --r—- Years in U. S,?. j.,,. Race. _

Occupation.---- Weekly Wages...,__

Schooling: Grades, 1.2. 3. 4. 5. 6. 7. 8: High or Prep. School, Year 1.2.3. 4: College, Year 1. 2. 3. 4

TEST 1

ooooo

2. ®©©©©®@(D®

5. O O O A/o

6. OOOOO

7. ABCDEFGKIJK.LMNOP

8. OOO MILITARY GUN CAMP

9. 34-79-56-87-68-25-82-47-27-31-64-93-71-41-52-99

12 . 1 2-3 4 5 6 7 8 9

Figure 17. — Directions Test. — The Army Alpha Intelligence Test.

Tests for Adults. — There are several group intelligence
tests suitable for adults. The Otis Self-Administering
Intelligence Test, Higher Examination, is one of the most

112 MEASUREMENT: INTELLECTUAL TRAITS

TEST 7

f sky-blue :: grass— table; green warm big
SAMPLES < fish— swims : : man —paper time walks girl
1 day —night: i white — red black clear pure *

In each of the lines below, the first two words are related to each other in some way. What

S u are to do in each line is to see what the relation is between the first two words, and under-
e the word in heavy type that is related in the same way to the third word. Begin with No. 1
and mark as many sets as you can before time is called.

1 shoe—foot :: hat—kitten head knife penny. 1

2 pup—dog :: lamb—red door sheep book. 2

3 spring—summer : : autumn— winter warm harvest rise'.. 3

4 devil—angel : : bad— mean disobedient defamed good. 4

5 finger—hand :: toe—body foot skin nail.... 5

6 legs—frog : : wings— eat swim bird nest. 6

7 chew—teeth :: smell—sweet stink odor nose.:. 7

8 lion—roar :: dog—drive pony bark harness., 8

9 cat—tiger : : dog— wolf bark bite snap. 9

10 good—bad : : long—tall big snake short. 10

11 giant—large :: dwarf—jungle small beard ugly. 11

12 winter—season : : January— February day month Christmas...... 12

13 skating—winter : : swimming— diving floating hole summer. 13

14 blonde—fight : : brunette— dark hair brilliant blonde. 14

15 love—friend : : hate—malice saint enemy dislike. 15

16- egg—bird :: seed—grow plant crack germinate. 16

17 dig—trench :: build—run house spade bullet. 17

18 agree—quarrel : : friend— comrade need mother enemy. 18

19 palace—king : hut— peasant cottage farm city.:. 19

20 cloud-burst—shower : : cyclone— bath breeze destroy West. 20

21 Washington—Adams :: first—president second last Bryan. 21

22 parents—command :: children—men shall women obey. 22

23 diamond—rare : : iron— common silver ore steel. 23

24 yes—affirmative :: no—think knowledge yes negative. 24

25 hour—day :: day—night week hour noon..•. 25

26 eye—head :: window—key floor room, door... 26

27 clothes—man :: hair—horse comb beard hat. 27

28 draw—picture :: make—destroy table break hard.. 28

29 automobile—wagon : : motorcycle— ride speed bicycle car.,.. .... 29

30 granary—wheat :: library—read books paper chairs. 30

31 Caucasian—English : : Mongolian— Chinese Indian negro yellow... 31

32 Indiana—United States :: part—hair China Ohio whole. 32

33 esteem—despise :: friends—Quakers enemies lovers men. 33

34 abide—stay : : depart— come hence leave late. 34

35 abundant—scarce :: cheap—buy costly bargain nasty... 35

36 whale—large :: thunder—loud rain lightning kill-.. 36

37 reward—hero : : punish— God everlasting pain traitor. 37

38 music—soothing :: noise—hear distracting sound report. 38

39 book—writer : : statue— sculptor liberty picture state. 39

40 wound—pain :: health— sickness disease exhilaration doctor. 40

Figure 18. — Analogies Test. The Army Alpha Intelligence Test.

useful of these. The Army Alpha Test is another very
useful test. It was one of the first group tests devised
and was used extensively during the World War in test¬
ing about two million men. As one of the first group

EVALUATION OF SCORES

113

tests to be devised, it has influenced many group tests
that have since been constructed. Almost all of the
verbal tests at the present time show distinct resem¬
blances to the original Army Alpha Test. Figures 17 and
18 show test 1 and test 7 of the Army Alpha. Test 1
shown in Figure 17 is a test of Following Oral Directions.
The first direction is “Make a cross in the second circle,
and also a figure 1 in the third circle.” The fourth direc¬
tion is, “Make a figure 1 in the space which is in the
triangle but not in the circle or square, and also make a
figure 2 in the space which is in the square and circle,
but not in the triangle.” The seventh direction is “Cross
out the letter just before D and also draw a line under
the second letter before I.” The last direction is, “If 6
is more than 4, then (when I say ‘go’), cross out the
number 5 unless 5 is more than 7, in which case draw a
line under the number 6.”

Evaluation of Scores. — We have described briefly
one test for each level of intelligence from kindergarten
to college, one non-language test and one survey test.
There are many other tests and we do not mean to indi¬
cate by the selection made that these are the best tests.
There are many others equally good. After any one of
these group tests has been given, there results a score
for each person examined. This score may be then con¬
verted into some type of mental rating which interprets
the score. The most common method is to convert the
score into a mental age by means of norms or standards
for that purpose. When a* mental age has been obtained,
an I.Q. may be calculated in the usual way. This use
of M.A. and I.Q. is a following of the method devised
in conjunction with the Binet-Simon Tests. It is un¬
doubtedly useful and easy to understand, and is, more¬
over, fairly well understood by teachers in general. There
is, however, a danger in presupposing that the M.A. on
all tests means exactly the same thing, and much sur-

114 MEASUREMENT: INTELLECTUAL TRAITS

prise and astonishment may occur when the same child
is found to have different mental ages on different tests.
As we shall see later this is perfectly possible, if the dif¬
ferent tests are measuring different aspects of intelligence.
Furthermore, the difference in the control over the indi¬
vidual exercised by the examiner in the individual and
group tests respectively may lead to differences in the
results. In general one feels that the M.A. resulting
from a thorough individual test is more indicative of the
all-round ability of the child than the M.A. resulting
from one group test. With high school and college stu¬
dents it is difficult, if not impossible, in most cases to
obtain a reliable M.A. by means of a group test. Mental
ages above 14 or 15 become fictitious. In these cases,
therefore, other types of mental rating are resorted to.
Probably the best method for all kinds of tests for chil¬
dren of all ages is to express the score of the child in terms
of the standard deviation of some known group. In this
way we get the T score proposed by McCall and the
Mental Index used by Pintner. All these kinds of ratings
show us how much above or below the average a particu¬
lar child stands. One may also convert the score on a
test into a letter rating such as A, B, C and so forth.
This was done in the Army Intelligence Testing and is
used in several group tests. Each letter rating is defined
as to its significance, usually in terms of the percentage
making similar scores. Each of these different ways of
expressing a mental rating possesses its special advan¬
tages for particular purposes.

Tests for Very Young Children. — None of the tests so
far described are suitable for very young children. The
group tests do not go below the kindergarten level and the
Stanford Revision of the Binet Tests stops at age three.

The Kuhlmann Revision of the Binet (2) is valuable
because it contains tests for three months, six months,
twelve months, eighteen months and two years.

presented

■' THE EDITOR OF

Figure 19. — The nest of cubes test from the Merrill Palmer Tests. (Courtesy of Rachel Stutsman of

the Merrill Palmer School.)

TESTS FOR SPECIAL CAPACITIES 115

The Merrill-Palmer Tests (6) consist of a great number
of tests, mostly of the performance type, which can be
given to children from 18 to 72 months. Some of the
tests are: (1) Commands. “See the box. Take the box
from the table and put it on the chair.” (2) Straight
Tower. Build a straight tower, three blocks high, making
sure child watches the process; then say, “Now, you
make one.” (3) Questions. “What does a doggie say?”
“What does a kittie say?” “What is this?” (pencil).
“What is it for?” and so on. (4) Nest of cubes. Figure
19 shows the cubes taken apart. The nest is shown the
child, it is taken apart, then put together again while he
is watching, and then he is allowed to try. In addition
there are form boards and picture puzzles.

Like the Binet Tests, we have a miscellaneous group
of tests, with the performance type predominating. They
are not non-language in character, however, as all of
the tests are dependent upon the oral directions of the
examiner. The tests are arranged in order of difficulty
and are grouped in six-month intervals. The mental de¬
velopment of the young child is very rapid during these
early years.

Tests for Special Capacities. — In addition to the
attempt on the part of the psychologist to measure the
general capacity or intelligence of the individual, there
exist also attempts to measure special capacities. If
special capacities exist and are to some extent at least
inherited, it would seem possible that we might measure
their strength in the individual. As to the number of
such special capacities or talents, there is much doubt.
We do not know whether such things as mathematical
ability, literary ability, artistic ability and the like, are
due to the interests and opportunities of the individual
or whether they are dependent upon the inheritance of
certain specific abilities in addition to the general ability
or intelligence With regard to musical aptitude, how-

116 MEASUREMENT: INTELLECTUAL TRAITS

ever, there are good reasons for believing that it is de¬
pendent upon the inheritance of a specific capacity, and
tests have been devised by Seashore to measure this
capacity apart from such modifications as may have been
produced by musical training. Figure 20 shows the re¬
sults of these musical tests given to a highly gifted indi¬
vidual, and from this chart we can also see the several
different characteristics that have been measured.

ooooooooooo

HiNCO^iOCONOOOiO

Sense of Pitch . . .

Sense of Intensity .

Sense of Time. . .

Sense of Consonance

Acuity of Hearing .

Auditory Imagery .

Memory ....

Motility ....

Timed Action. . .

Rhythmic Action

Singing Key . . .

Singing Interval . .

Voice Control. . .

Quality of Voice

Talent Chart of Mr. White.

Figure 20. — The Measurement of Musical Talent. (From Sea¬
shore’s Psychology of Musical Talent, Silver, Burdett and Company.)

Tests of mechanical ability have also been constructed,
both of the pencil and paper type and also of the perform¬
ance type, and these give us a measure of an individ¬
ual's knowledge of and ability to handle and understand
mechanical things.

The Distribution of Intelligence. — Whenever intelli¬
gence tests are given to a large unselected group of indi-

THE DISTRIBUTION OF INTELLIGENCE 117

.9% 4.5% 11.6% 19,6% 25% 20.5% 2.S% 2,6% 1.8% .9%

Figure 21.— I.Q. distribution of 112 Kindergarten Children.
(From Terman’s The Intelligence of School Children.)

Figure 22. — Showing distribution of National Intelligence Test
I.Q.’s of 1225 pupils above Grade IV, Public School 64. (From Irwin
and Marks’ Fitting the School to the Child.)

118 MEASUREMENT: INTELLECTUAL TRAITS

viduals and the mental ratings of the group plotted on
a frequency curve, we find that these curves tend to take
a certain form. If we study Figures 21 to 28, we note

Figure 23. — I.Q. distribution for 999 gifted and 905 unselected
children. (From Terman’s Genetic Studies of Genius , Vol. I. Stan¬
ford University Press.)

Percent

Figure 24. — Frequency curves for high school seniors. (From
Book’s The Intelligence of High School Seniors.)

that the general shape of the curve is more or less the
same in all of them. The curves begin low at the left, rise
gradually higher and higher until they reach their max-

THE DISTRIBUTION OF INTELLIGENCE 119

imum in the middle and then gradually sink again to a
low point at the right. In terms of an intelligence test
this means that there are a few people who make very
low scores and that as the scores increase the number of
individuals making such scores also increases up to a
given point. This highest point on the curve represents

-ut

, r r

,r-r*j

S.D -3:0 -2.0 —1.0 0 3.0 4.0

Figure 25. — Composite curve for the ninth grade based upon
eleven tests. The broken line indicates the theoretical normal curve.
(From Thorndike’s Measurement of Intelligence.)

average or normal ability, because a great number of
individuals cluster around this point. From this point
on, as the scores increase, the number of individuals be¬
comes less and less until we end up with few individuals
making extremely high scores. In other words, the fre¬
quency of very low and very high intelligence scores is
small. There are few idiots and few geniuses in the
world. The frequency of the intermediate levels is some-

120 MEASUREMENT: INTELLECTUAL TRAITS

what greater. There are more dullards and bright peo¬
ple. Most numerous of all are the people of average
intelligence, who cannot be called either dull or bright
and who are far from being idiots or geniuses. Normal
intelligence is that degree of ability possessed by the
median or average individual, and, because intelligence
is distributed as we have described, it follows, there¬
fore, that normal intelligence is the most common or
frequent.

Figure 26. — Composite curve for college freshmen, derived
from eleven single curves. The broken line indicates the theo¬
retical normal curve. (From Thorndike’s Measurement of Intel¬
ligence.)

The type of distribution we have been describing is
generally called a normal distribution and the curves
shown in Figures 21 to 28 are generally referred to as
normal curves. Whenever a large number of individuals

THE DISTRIBUTION OF INTELLIGENCE 121

Figure 27. — Distribution of intelligence scores of white enlisted
men. The dotted curve represents the average percentage distribu¬
tion for the four camps. (From Memoirs of the National Academy of
Sciences , Vol. XV.)

122 MEASUREMENT: INTELLECTUAL TRAITS

is tested the tendency is to get a normal curve, even
though the group is not a random selection of the popula¬
tion in general. For example the distribution of the intel¬
ligence of high school seniors shown in Figure 24 has the
characteristics of a normal curve, and it shows that an
unselected sampling of high school seniors will give us a
few seniors of poor intelligence for high school seniors, a
few of very exceptional intelligence with the great major¬
ity lying in between these two extremes. Figure 23
shows two curves, one for unselected children and the
other for gifted children. The latter curve shows how
the lower portion is suddenly cut off owing to the fact
that no child below an I.Q. of 135 was included in the
gifted group.

Figure 28. — Distribution of 93,965 soldiers on combined scale.

(From Memoirs of National Academy of Sciences , Vol. XV.)

The Growth of Intelligence. — Everyone knows that as
a child grows, he becomes able to do more and more dif¬
ficult things. Things which were impossible for him to do
or understand at three years, are readily done or grasped
at eight. The capacity for reacting to more and more
complex situations increases as the nervous system de-

THE GROWTH OF INTELLIGENCE 123

velops and grows. At the same time we know that the
rate of growth diminishes as the child grows older. This
general growth of intelligence is pictured by Thorndike
(8) in Figure 29.

Figure 29. — The general nature of the relation of altitude of
intellect to age in years, 0 to 20. (From Thorndike’s Measurement of
Intelligence.)

This curve shows that the increment of intelligence
added each year becomes gradually less and less as the
child increases in chronological age. Common observa¬
tion would seem to bear this out. From birth to one
year, the mental change, as well as the physical, in an
infant is enormous. One can almost literally see him
growing and learning to do new things every day. This
marvellously rapid rate gradually slows up. We do not
note such tremendous changes between ages 7 and 8, or
ages 9 and 10. Some have maintained that there is an¬
other definite spurt at adolescence, but this has not been
clearly demonstrated. Again it is easier to find intellec¬
tual tasks which will discriminate between the 5 and 6
year old than to find such as will discriminate between
the 12 and 13 year old. Such general considerations as

124 MEASUREMENT: INTELLECTUAL TRAITS

well as the results of actual intelligence testing lead us to
believe that the curve of growth of intelligence is of the
type pictured in Figure 29. The increment of intelligence
added each year becomes less and less until it ceases
altogether.

The Limits of Growth. — This brings us to the next
question as to when intelligence stops. When does the
average individual arrive at maturity of intelligence, as
distinct from the amount of knowledge he may possess.
There can be no doubt that such a maximum point must
occur, and it probably occurs relatively early in life, long
before the very marked decline in intelligence that usually,
if not always, takes place in old age. Very probably in¬
telligence ceases to grow in adolescence, just as the body
ceases to grow at that time, because actual growth of
intelligence is undoubtedly closely connected with the
growth of the brain and nervous system. This hypothesis
of the early cessation of the growth of intelligence does
not mean, that thereafter an individual cannot learn any¬
thing new, just as a cessation of the growth of the body
does not mean that thereafter no new physical accom¬
plishments may be learned. A man may learn to skate
or play golf or billiards later in life, and so also may he
learn a foreign language or take up a new study. In
each case, however, his success will be conditioned by
the adequacy of his physical organization and the ade¬
quacy of his mental organization.

At present our intelligence tests do not measure much,
if any, intelligence growth beyond the ages of 14 to 18
for the average individual. At just what age growth
ceases is still debatable. Furthermore, we must remem¬
ber that our intelligence tests are still rather coarse in¬
struments of measurement, and that there may be growth
after this period which is too slight to be detected by the
instruments which we possess. In the army testing no
difference in intelligence was found between the different

THE CONSTANCY OF GROWTH

125

age groups from twenty to thirty, so that intelligence
evidently is not increasing between twenty and thirty.
Furthermore average fourteen-year-olds score as much as
the average army recruit. Tests given to unselected sam¬
plings of fourteen- and fifteen-year-olds by one investiga¬
tor showed no increase in intelligence by the latter group.
Other investigators think, however, that they have found
a slight increase up to about age eighteen. A great diffi¬
culty arises because it is very difficult to obtain unse¬
lected samplings of children at ages above the elementary
school.

The high school acts as a selective agency for children
of higher intelligence. At present, therefore, we must
leave as unsettled the actual point at which intelligence
ceases to grow, the probability being that the point is
somewhere between 14 and 18. We must remember that
all this discussion has related to the average individual,
for there is reason to believe that individuals possessing
more than average intelligence, may continue growing in
intelligence for a longer period, and, conversely, indi¬
viduals below normal intelligence may cease growth
earlier.

The Constancy of Growth. — Closely connected with
the preceding question is the further one of the constancy
or regularity of growth. Does the growth of intelligence
of the average child continue steadily from year to year
or are there marked fluctuations from time to time? There
may be and probably are fluctuations, just as there are
fluctuations in physical growth, but in general the picture
would seem to be one of steady and constant growth.
The evidence here is obtained from repeated tests of
the same children. If the I.Q.’s from test to test remain
relatively constant, then we may say that growth con¬
tinues regularly. Here are the results for one child tested
nine times by the Stanford Revision of the Binet:

126 MEASUREMENT: INTELLECTUAL TRAITS

Test

C.A.

M.A.

I.Q.

2-3

3-6

155

3-1

4-6

146

4-0

5-8

142

5-2

7-4

142

6-0

8-6

142

7-2

9-11

138

8-2

11-2

137

8-11

13-0

146

10-0

14-3

143

Probably the tests at the earlier ages on the Stanford
are slightly too easy and hence the abnormally high I.Q.
when the child was two years old. Apart from this one high
I.Q., the others range from 137 to 146. This amount
of variation over a five-year period is relatively small.
In such cases, therefore, we say that the I.Q. is constant.
In sixteen investigations of the I.Q.’s of varying numbers
of children who had been given at least two tests, the
median difference in I.Q. from the first test to the second
is 4.6 points.

The amount of constancy or agreement from one test
to another can also be expressed as a coefficient of corre¬
lation. The median correlation of seventeen such corre¬
lations found by different investigators is .90, and this
means that in general there is a distinct tendency for the
I.Q. of a child to remain fairly constant from one test
to another. This is all that the psychologist means when
he talks about the constancy of the I.Q. He does not
mean that the I.Q. never shifts from year to year. He
does not mean that there are not cases of very great dif¬
ference from one test to the next. Cases showing changes
of 20 or more points occur. These may be due to poor
tests, poor testers or to actual changes that have taken
place in the child. As to the causes of such real changes
in the rate of growth of intelligence in some few children
we are in ignorance. In general, however, the picture
of constant steady growth with no violent fluctuations is

THE EFFECT OF PRACTICE

127

the common one that has been found, and this is what
the psychologist means when he says that the I.Q. is
constant.

The Effect of Practice. — Familiarity with intelligence
tests or practice or coaching leads, of course, to an in¬
crease in score. There is nothing magical about intelli¬
gence tests so as to make them impervious to the uni¬
versal laws of learning — the laws of exercise and effect.
If, therefore, we deliberately teach a child the items of an
intelligence test, or if we give him the same intelligence
test once every day for several successive days, we will
obtain a score much higher than would have been ob¬
tained without such practice or coaching. This score will
not be indicative of his intelligence, if it is interpreted in
terms of M.A. or I.Q. as derived from norms established
on uncoached children. We must remember here that
our mental ratings are always comparisons with what
the standard or normal group of children has done. If,
therefore, a child deviates markedly from this group,
such as our coached or practiced child, then we cannot
legitimately interpret his score in terms of the standard
group.

If children are given direct coaching on the Stanford
Revision of the Binet, their mental ages can be increased
about two years. This has been demonstrated by one
investigator (Graves, 1). Why then does this not inval¬
idate the Binet Scale as a measure of intelligence? Be¬
cause the test is rarely given to children more than once
a year. When so given the child does not always do the
same tests, but as he grows in intelligence new and harder
ones enter to test his ability. The relative stability of the
I.Q.’s of children so tested shows that the practice gained
by yearly testings is negligible. The example given above
of the child given nine annual tests shows no effect of
practice or else the I.Q. would have steadily increased.
With a little care, therefore, the error that would be in-

128 MEASUREMENT: INTELLECTUAL TRAITS

troduced by practice or familiarity or coaching may be
easily avoided in the administration of intelligence tests.

The Description of Intelligence. — As we suggested at
the beginning of this chapter, we have left any attempt to
describe or define intelligence until the end. Our survey
of the tests and the discussion of distribution and growth
have probably helped the reader to grasp the psycholo¬
gist's meaning of intelligence far better than any formal
definition. Through and by means of our work with in¬
telligence tests, we are slowly coming to a better and fuller
meaning of what intelligence is. We may at present
amplify our concept of intelligence by thinking of different
kinds of intelligence and by considering different aspects
of intelligence.

Kinds of Intelligence. — The use of different instru¬
ments of intelligence measurement has brought to our
attention the fact that there are different kinds of intelli¬
gence, such as abstract, concrete and social. Abstract
intelligence would signify the ability to respond to sym¬
bols of various sorts, such as words, numbers, letters and
the like. Abstract intelligence is required by all who
would progress in the ordinary academic subjects in
school, such as reading, writing, arithmetic and so on.
In its higher levels it is typified by the student and philos¬
opher who is dealing with the relations of things sym¬
bolized in words or numbers.

Concrete intelligence denotes our ability to respond to
concrete things. It is not mere skill in manipulation such
as may come through practice, but it is rather the ability
to understand concrete situations and react adequately
to them. Performance tests measure this kind of ability
and in everyday life such ability is shown in our reactions
to doors, chairs, tables, lamps, stoves, street-cars, auto¬
mobiles, steps, elevators, turnstiles, cash-registers, type¬
writers and the innumerable machines, instruments and
things that people in different walks of life handle every

THE DESCRIPTION OF INTELLIGENCE 129

day. In its higher levels it is typified by the surgeon, the
inventor, the scientist, in so far as they are reacting to
concrete things. No doubt these higher levels cannot be
reached without also having a high, if not very high,
level of abstract intelligence.

Social intelligence is probably a third kind of intelli¬
gence, although for its measurement we have as yet no
separate scales. The assumption of this third kind of
intelligence, is, therefore, hypothetical. We can, how¬
ever, readily think of different degrees of ability in react¬
ing to persons as distinct from concrete things and from
abstract symbols. Ability to understand and react ade¬
quately to persons would be the criterion of social intelli¬
gence. This would not include the feelings or emotions
aroused in us by other people, but merely our ability to
understand others, to react in such a way toward them so
that the ends desired should be attained. Some people
seem frequently to react unfittingly to the presence of
others and they may be thought of as possessing little of
this kind of intelligence. Others on the contrary are
very able in dealing with people. We say they are able
to handle men well. Good generals, executives, teachers,
politicians, preachers must all possess large amounts of
this kind of intelligence.

This three-fold division of intelligence is merely a con¬
venient scheme for dividing the numerous different reac¬
tions that may be taken as indicative of intelligence.
There is no reason why other divisions might not be
made, and, indeed, it is quite common to speak of verbal
as opposed to non-verbal intelligence, and when we do
this we are cutting across the previous scheme of classifi¬
cation and considering all reactions as divisible into verbal
and non-verbal. Or rather, more narrowly, we are think¬
ing of the difference between the ability to react to verbal
test items as contrasted with the ability to react to non¬
verbal test items. Furthermore, in the above suggested

130 MEASUREMENT: INTELLECTUAL TRAITS

three-fold classification of intelligence there is no sugges¬
tion of one kind of intelligence being contrary to or ex¬
cluding the other. It is most probable that there is a
high correlation between the three kinds. Great social
intelligence presupposes a much above average abstract
and concrete intelligence. The great executive or politi¬
cian cannot be an imbecile or moron in abstract intelli¬
gence. So the great surgeon or inventor must have high
abstract intelligence, and the great philosopher is much
above average in his reactions to people and things.

Aspects of Intelligence . — From a different point of
view we can extend our concept of intelligence by de¬
scribing three different aspects which it possesses, namely
speed, level and range. The conception of speed is simple
and easily measurable on our tests by taking the time or
by allotting a given amount of time for a certain number
of tasks. Level is the height or altitude or degree of
difficulty to which an individual can rise. If we imagine
all the tasks in the world arranged in order of difficulty,
then the higher up this scale of difficulty we can climb,
the higher would be our level of intelligence. The range
of intellect would be the number of tasks at each degree
of difficulty that we could solve. This range, of course, is
determined by our opportunities for learning, because we
cannot do many things unless we have had the chance to
learn, but whether we are able to learn a given task or
not will depend upon the level of our intelligence. In all
intelligence testing, therefore, we must keep within the
range of things common to all individuals that are to be
compared.

For example, if we wish to compare in intelligence a
child who has been to school with one who has not had
such opportunity, we must not choose a test such as the
National Intelligence Test which assumes a range such
as is ordinarily experienced by school children only.
Rather we must choose a test, like the Pintner Non-

SUMMARY

131

Language Test that falls within the range common to
both school and non-school experience.

Definition of Intelligence. — This discussion of intel¬
ligence has enlarged our concept of the word and given
us a fair idea of what the psychologist at present means
by it. Ordinarily when asked for a brief definition, he
will answer by saying that intelligence is the ability of
the organism to adjust adequately; or else he will stress
the factor of modifiability and say, “ Intelligence is the
ability to learn, or to improve through experience.” None
of these definitions seems adequate, but that is of minor
importance if we are cognizant of what is meant by means
of the fuller description we have given above.

Summary

1. Our ability to judge intelligence without objective tests is
very poor.

2. Experiments on judging the intelligence of children from
their photographs show practically no agreement between
the rankings of the judges and the test ranking.

3. Intelligence tests may be thought of as standardized inter¬
views.

4. Intelligence tests may be divided into group and individual
tests.

5. The most commonly used individual tests are revisions of
the Binet-Simon Tests.

6. These tests are a miscellaneous group of tasks of all types
arranged in levels of difficulty.

7. A mental age is obtained by comparing a given child’s per¬
formance with the average performance of unselected
samplings of children of each chronological age.

8. The intelligence quotient is the ratio of mental age to the
chronological age, and it gives us a measure of brightness
or dullness.

9. Performance tests are a second type of individual test for
the measurement of intelligence. They may be said to
measure concrete rather than abstract intelligence.

132 MEASUREMENT: INTELLECTUAL TRAITS

10. Group tests permit of less adjustment to each individual
child than is the case with individual tests, but they are
for that very reason well adapted for predicting success in
school.

11. There are adequate group tests for every level of intelligence
from the primary to the college level, as well as non-lan¬
guage tests and survey tests for special purposes.

12. It is important to remember that each type of test is testing
a different aspect of intelligence and, therefore, we must
not expect a child to obtain the same I.Q. on all tests.

13. Tests for pre-school children are now being rapidly developed
and we are able to obtain a rough measure of intelligence
even at 12 or 18 months.

14. In addition to tests of general intelligence there are several
tests for special capacities, such as musical and mechanical
ability.

15. The distribution of intelligence is normal.

16. The growth of intelligence is rapid at first, gradually de¬
creasing as the child grows older.

17. The limit of growth has not been determined, but it is
probably somewhere in the later adolescent period.

18. In the average child the growth is more or less constant
from year to year, but in some children marked fluctuations
seem to occur.

19. The effect of practice or coaching on intelligence tests is to
increase the score. Practiced or coached children cannot,
therefore, be compared with the norms based on unpracticed
children.

20. Intelligence may be thought of as divided into different
kinds, such as abstract, concrete and social.

21. Intelligence may be thought of as having different aspects
such as speed, level and range.

22. Intelligence may be defined as the ability to adjust or the
ability to learn.

REVIEW

133

Review

True-False Statements

As a review mark the following statements true or false:

1. Intelligence testing had its origin in the effort to separate the
feebleminded from the normal.

2. If a child’s I.Q. changes from 120 to 130 from one year to the
next, then either the first or second test must be wrong.

3. Binet’s first testing was done in order to discover superior
children.

4. Three kinds of intelligence have been called abstract, con¬
crete and social.

5. It is impossible to secure a valid measure of intelligence un¬
less the child can read.

6. One can generally discover a person of low intelligence by his
facial appearance.

7. The Stanford-Binet, the Pintner-Paterson Performance and
the National Intelligence Test all measure identical abilities.

8. The intelligence quotient is obtained by dividing the chrono¬
logical age by the mental age.

9. When we say that the distribution of intelligence is a normal
distribution, we mean that there is an equal number of feeble¬
minded, average and bright children.

10. The Terman Group Test is a good test for high school chil¬
dren.

11. The curve of growth of intelligence usually shows a sharp
rise at adolescence.

12. It is possible to raise a child’s score on the Binet by coaching.

13. An intelligence test is an adequate measure only as long as
the individuals who take it have approximately the same
background.

14. In testing foreign or deaf children, one should use a non¬
language test.

15. We can measure a great growth of intelligence after the period
of adolescence.

16. We cannot at present register any appreciable growth of in¬
telligence after adolescence.

17. Mental development usually reaches maturity at the age of
thirty years.

134 MEASUREMENT: INTELLECTUAL TRAITS

18. With children who develop normally the I.Q. increases with

age.

19. Intelligence ratings are always comparisons with those who
have been used as the standardization group.

20. Write in the three kinds of intelligence that have been sug¬
gested: (1)- , (2) --, (3)-.

21. Underline the best response:

The Binet-Simon Scale was first introduced into this coun¬
try by: Terman, Goddard, Yerkes, Thorndike.

22. Before the name of each test below, put K if suitable for
the kindergarten, E for the elementary school and H for
the high school.

-Terman Group Test

-Pintner-Cunningham

-Miller Mental Ability

-National Intelligence Test

-Pintner Rapid Survey

-McCall Multi-mental

23. Useful revisions of the Binet Scale for the United States
have been made by:

L-

2 .-

3.-

24. The intelligence quotient is obtained by dividing the-

by the-.

Advised Readings

Pintner, R .'.Intelligence Testing, New York, 1923, Chapters 1-6.
Freeman, F. N.: Mental Tests, Boston, 1926, Chapters 1, 2, 4,
5, 6, 7, 13, 17 and 18.

References

1. Graves, K. B.: The Influence of Specialized Training on

Tests of General Intelligence, Teachers College, Co¬
lumbia University, N. Y., 1924.

2. Kuhlmann, F.: A Handbook of Mental Tests, Baltimore,

1922.

3. Pintner, R.: “ Intelligence as Estimated from Photographs,”

Psychological Review, Volume 25, 1918, pages 286-296.

GROUP INTELLIGENCE TESTS

135

4. Pintner, R., and Paterson, D. G.: A Scale of Performance

Tests, New York, 1917.

5. Seashore, C. E.: Psychology of Musical Talent, New York,

1919.

6. Stutsman, R.: “Performance Tests for Children of Pre-

School Age,” Genetic Psychology Monographs, Volume
1, Number 1, January, 1926.

7. Terman, L. M.: The Measurement of Intelligence, Boston,

1916.

8. Thorndike, E. L.: The Measurement of Intelligence, Teach¬

ers College Bureau of Publications, New York.

Group Intelligence Tests Mentioned in

This Chapter

Pintner-Cunningham Primary Test: World Book Co., Yonkers,
N. Y.

National Intelligence Test: World Book Co., Yonkers, N. Y.
Terman Group Test of Mental Ability: World Book Co., Yonkers,
N. Y.

Thorndike Intelligence Test: T. C. Bureau of Publications,
N. Y.

Pintner Rapid Survey Test: T. C. Bureau of Publications, N. Y.
Pintner Non-Language Test: College Book Co., Columbus,
Ohio.

Otis Higher Self Administering Test: World Book Co., Yonkers,
N. Y.

Army Alpha Test: Stoelting Co., Chicago, Ill.

CHAPTER VI

THE USES AND RESULTS OF INTELLIGENCE

TESTING

Numerous Uses. — The uses to which intelligence tests
have been put are manifold, and a detailed mention of
them all is more suited to a special manual on intelligence
testing than to an educational psychology. We shall,
therefore, indicate in broad outlines some of their uses in
the educational field. They are rapidly becoming an essen¬
tial part of the organization of the modern school, and as
such they represent one of the most useful contributions of
modern psychology to the practical business of teaching.

Classification. — The most common use of intelligence
tests, group tests especially, is as an aid toward better
classification of pupils in grades or in the various sections
of a grade. Children differ greatly in general intelligence
and in the rate of growth of their intelligence. They
differ in mental age or the present amount of intelligence
possessed by them, and also in their intelligence quotients
or the rate at which their intelligence is probably grow¬
ing. In most schools where little attention has been paid
to homogeneous classification the differences in intelli¬
gence among the children of any one grade are very great.

In Grade VB in a certain school in New York the Na¬
tional Intelligence Tests were given and the frequency
distribution according to mental age of the three sections
of this grade was as follows:

Mental Age (N.I.T.)

Grade 7-o

8-0

9-0

10-0

11-0

12-0

13-0

14-0

15-0

Med.

Total

7-11

8-11

9-11

10-11

11-11

12-11 18-11 14-11

15-11

M.A.

Cases

5B1

10-7

5B2 1

10-9

5B3 2

10-0

136

MENTAL AGE TO DETERMINE GRADE 137

This means that of the forty-two children in the first sec¬
tion of the grade three have mental ages between eight
years no months and eight years eleven months; eight have
mental ages of 9; 13 have mental ages of 10; and so on.
The median mental age of that section is ten years seven
months. In these three sections there are children with
mental ages from seven to fifteen, presumably all doing
the same type of work. Undoubtedly there are many
misfits, many who are really unable to do the work, and
many who are capable of doing much more advanced
work. Those at the lower end of the distribution have
probably been promoted more or less automatically and
very likely are totally unfitted to do the work of the fifth
grade. Those at the upper end might well profit by
skipping a half grade or by an extra promotion. Consul¬
tation with the teacher about the child with reference to
his school work, health and so on, is of course assumed,
before such changes are made. The mental age, there¬
fore, helps us to determine the grade or level at which
the child might best be working.

Mental Age to Determine Grade. — In the specific
example we have cited above, the teacher or principal
would begin by finding out whether those with the lowest
mental ages of seven and eight could really do the work of
the class. As this is very unlikely, they would either be
put into some special class or held in the same class at the
end of the term, or, as a last resort, demoted. Many of the
nine-year-olds should be dealt with in the same way, but as
these form a goodly number of the total class, the chances
are that the instruction of the group has been more or
less adjusted to their level, and, therefore, many of them
would remain. Those having mental ages of 10, 11 and
12 would be left in the class and they would be expected
to do normal work. The nine children with mental ages
of 13 would probably profit by being promoted at once
to the next higher half grade, regardless of their actual

138

INTELLIGENCE TESTING

educational performance in the present class. They
would be expected to do better work, and in most cases
they would come up to expectation. Lastly, the two
children with mental ages of 15 would be skipped a half
grade or be given two promotions each half year until
they arrived at a grade more suitable to their mental
level. In some such way as this the great spread of
mental ages in one grade, due to promotion by chrono¬
logical age, can be gradually restricted. The wise super¬
intendent or principal will not attempt a sudden revolu¬
tion in grade placement, but will in the way suggested
above work gradually from year to year.

The I.Q. for Sectioning. — Assuming, then, that such
necessary shifts from one grade to another have been
made and that the remaining children are probably all
able to do the work of the grade, then the I.Q., which
tells us the rate of growth of intelligence, may be used to
help us make a division of the children into different
sections so that they may move along at different rates
of speed. At the same time the brighter sections will
probably do more extensive work than the slower sections.
Broadening the field of study and acceleration of pace
generally go together. Here is the distribution of the
I.Q.’s of a YA Grade tested on the National Intelligence
Tests: —

I.Q.

Frequency

130-139

120-129

110-119

100-109

90-99

80-89

70-79

60-69

50-59

Total 93

USE OF M.A. FOR CLASSIFICATION 139

For reasons of space these 93 children had to be divided
into three equal sections. Instead of having a random
selection of all I.Q.’s in each of the three sections as was
actually the case, it would be better to have a section of
34 with I.Q.’s above 100; another of 31 with I.Q.’s from
85 to 100 and a third of 28 with I. Q.’s below 85. No hard
and fast lines should be laid down and shifts from one
section to another should be readily made for any good
reason.

The number of children in a grade, the number of class¬
rooms and teachers available, and other actual adminis¬
trative conditions will determine in each case the number
of sections that can be formed. The actual intelligence of
any particular school population will determine where
the lines should be drawn between one section and an¬
other. What might be considered an advanced or bright
section in one school will very likely be considered quite
ordinary or indeed mediocre in a school having a highly
intelligent population. There are schools in which the
average I.Q. is barely 90 and others in which the average
I.Q. is 115. The extreme form of the recognition of indi¬
vidual differences in intelligence is found in those schools
which attempt to let each child go at his own rate. There
are several well-known “plans” of school administration
which might serve as examples of attempts to recognize and
provide for each individual’s different intellectual needs.

Example of Use of M.A. for Classification. — School
classification by mental age as suggested above has been
reported by many investigators. One such report (Pintner
and Noble, 7) deals with a school of 450 children in grades
I to V inclusive. All the children were tested with in¬
telligence tests. Allowing one and one half mental years
as normal for each grade, it was found that 56.7% were
normally placed according to mental age; 13.1% were
above normal and 30.2% below. A reclassification was
made in consultation with the teachers. Sixty-nine chil-

140

INTELLIGENCE TESTING

dren were given extra promotions, 41 were not promoted
at the mid-year and 7 were demoted. A small ungraded
class for very low I.Q.’s was formed. This reorganiza¬
tion worked well. At the end of the next semester the
percentage of normally placed pupils was 71%; below
normal 16.2%; above normal 12.7%. Reports from the
assistant superintendent, principal and teachers all ex¬
pressed satisfaction in the results achieved, not only
with respect to increased teaching efficiency, but also
with respect to the general morale of the school.

Under this general heading of classification we may say
that intelligence tests help in:

(a) grading or placing the child in the correct grade.

(b) sectioning or placing the child in the correct section
of a grade, such as fast, medium, or slow sections.

(c) acceleration or deciding whether a child should be
given extra promotions or should skip a grade and
thus accelerate his general progress through the
grades.

Selection of Dull and Feebleminded. — One of the

earliest uses of intelligence tests in school was as a means
for the selection of the very dull and feebleminded so that
they might be taught in special classes. Such special
classes were in existence before the coming of the intelli¬
gence test, but the latter has been of great help in the
selection of such children.

Assignment of children to a special class should not be
made on the basis of a group test. A thorough individual
examination by a competent person is essential, because
the practical consequences of such segregation are very
serious for a child. Furthermore children in such special
classes should be re-tested from time to time in order to
verify the first mental rating and to see if some improve¬
ment may not have taken place. This is unfortunately
rarely the case.

SELECTION OF BRIGHT OR GIFTED 141

The intelligence quotients of children in such classes
generally range from about 50 to about 70. Few chil¬
dren with I.Q.’s below 50 stay long in the public schools.
They are generally sent to some institution for the feeble¬
minded, because they need a type of training and care
which the public school can hardly give. Those above
70 are generally able to succeed in the ordinary classroom,
although most special classes have some cases of I.Q.’s
in the seventies and eighties. It is bad policy, however,
to put any and all misfits into the same special class.
Disciplinary cases with I.Q.’s above 80 or 90 do not be¬
long in the same class with those of low I.Q.

The distribution of the I.Q.’s of the special class chil¬
dren of one large New York public school is shown below
(Irwin and Marks, 4):

I.Q .

Per Cent

71-75

20.7

66-70

30.0

61-65

20.7

56-60

15.3

51-55

10.7

46-50

2.0

41-45

0.6

Total

150

100.0

In this distribution we note that the policy of the school
is not to admit any with an I.Q. above 75, but to include
a fair number between 71 and 75. Again we note the
very small number of cases with I.Q.’s below 50.

Selection of Bright or Gifted. — It was the coming of
telligence tests that really discovered the child of superior
intelligence, in the sense of the well-rounded individual
of great potentialities. The number of children of high
I.Q.’s above 130 or 140, is probably about 1% of the
total school population. Several schools are experiment¬
ing with special classes for such children. If brought
together in a special class they have more opportunity to

142

INTELLIGENCE TESTING

expand and develop. Certain difficulties inherent in a
policy of mere acceleration are avoided. Their influence
upon each other is generally very beneficial. As a rule
such children are taller, heavier, healthier than the child
of average I.Q. They are just as much interested in
games, they love to play just as much as the average child.
They generally read a great deal more extensively than
does the average child. They are not generally aware of
their superiority in intelligence and in the ordinary class
many are often found doing merely average work. From
the first year in school they have found such work com¬
mended. They have satisfied their teacher. They do
not know that they are capable of much higher achieve¬
ment, because they have never been stimulated to do
more. Many, therefore, form habits of intellectual work
much below their real capacities and these habits may
continue throughout their whole educational career. These
are the kind of people who sometimes do exceptionally well
in some great emergency or when suddenly called upon
to do a task out of their usual line of work. They sur¬
pass themselves, and we remark, “I did not know he had
it in him.” He did not know he had it in him either,
because he never had received an adequate stimulus that
could call forth all his capacity. Such people do not sur¬
pass themselves, they merely surpass the standard that
the school set for them and which they were willing to
accept. It is very important, therefore, that bright chil¬
dren should be recognized as early as possible in their
school career, before they have had time to acquire habits
of mental work below their level. Intelligence testing
should begin as soon as the child enters school. It is
highly important for society that these intellectual re¬
sources should be conserved and not wasted through
ignorance and neglect.

It has been shown that these very bright children re¬
spond readily to an accelerated and broadened curriculum.

GUIDANCE

143

They do not need the same amount of drill in the funda¬
mental subjects. They can initiate and plan much better
than those of average intelligence. When grouped to¬
gether in the same class they serve as stimuli to each
other and the competition of keen minds prevents the
rise of snobbishness or vanity. Just what is the best
course of study for such children we do not yet know.
That it is not the course of study now adapted to the aver¬
age child is obvious. Much experimental work will be
necessary to find the proper material and the proper
methods for these active minds.

There is no sharp dividing line between normal and
bright children. In the formation of classes for bright
children the lower limit in terms of I.Q. has been set
anywhere from 110 to 155 according to the number of
children available. Terman’s (9) extensive study was
based on children with I.Q.’s above 135. The special
conditions existing in any particular school will deter¬
mine what children should be segregated for special in¬
struction. In some schools having an average I.Q. of
90 or 95, it might be feasible to form special classes for
those with I.Q.’s above 110. In other schools having an
average I.Q. of 110 or 115, such a limit would include
fifty per cent of the children, and therefore more highly
selected groups with I.Q.’s above 130 or 140 might be
formed. In any large school the class or classes for bright
children should be made up of those possessing the high¬
est I.Q/s within that school.

Guidance. — In a broad sense all intelligence testing
may be used for general educational and vocational guid¬
ance. In the narrower sense of giving specific educational
advice as to whether to go to high school or college, or as
to the course of study to be chosen, intelligence tests
may help but they cannot be used in a mechanical or
arbitrary way. The wise counselor will always make use
of them but never depend upon them alone. Achieve-

144

INTELLIGENCE TESTING

ment in school or college and success in later life do not
depend solely upon intelligence. Character qualities are
very important as well as temperament, health and oppor¬
tunity. Intelligence, however, prescribes the limits
within which the individual must function.

Abstract intelligence, as tested by most of our verbal
intelligence tests, is required for most of the regular
school subjects. Unless, therefore, a child possesses a
certain amount of it, he is not likely to be successful in
mastering them. Thus a child with an LQ. of 80 is un¬
likely to be able to complete the eighth grade successfully.
Such a child ought to be given practical vocational train¬
ing in the last few years of his school life so that at the
age of 14 he may be in some way prepared for his future
work in the world. A child with an I.Q. below 90 will
probably derive little profit from the ordinary high school
course. Children with I.Q.’s between 90 and 100 will
probably profit from some of the work in high school, but
few are likely to graduate without difficulty. In almost
all cases with I.Q.’s above 100, the children will be able
to take the regular course, although difficulty may be
found in some of the more abstract work. To grasp the
symbolism, understand the proofs and make the general¬
izations required in algebra probably requires a minimum
I.Q. of 110. It is doubtful whether children with I.Q.’s
below 110 should be advised to take the customary course
at a liberal arts college.

College Entrance. — A great many colleges make use
of intelligence tests for all entering students and some
use the results of such tests in helping to determine who
shall be admitted. No college uses the intelligence test
alone in making such a decision.

During the past decade college education has become
extremely popular and great numbers of high school
graduates are seeking admission to the college or uni¬
versity. In the state of Indiana an intelligence survey

COLLEGE ENTRANCE

145

Percent

TOTAL

GROUP

RATED

A or B

GROUPS

TOTAL

GROUP

RATED

D,EorP

Figure 30. — Percentage of high school seniors possessing the
highest (A or B) and the lowest (D, E or F) grades of intelligence in
the following groups: (1) our total or standard group; (2) all who will
not attend college; (3) those expecting to attend college, no college
selected; (4) all who expect to attend college; (5) those expecting to
attend a technical or professional school; (6) those expecting to at¬
tend a college of liberal arts. (From Book’s The Intelligence of High
School Seniors.)

(Book, 1) of high school seniors showed that those ex¬
pecting to attend college were very much like the total
group of seniors tested. Figure 30 shows the percentages
of students with various college intentions as compared
with the total group. So far as the intention of the stu-

146

INTELLIGENCE TESTING

dent is concerned, it would seem as if the colleges were
threatened with a great number of inferior or mediocre
individuals, and conversely were losing a great number of
superior intelligence who were not planning to go to col¬
lege. Undoubtedly a great number of those high school
students of inferior ability, who do carry out their inten¬
tion of proceeding to college, will encounter difficulties in
college. A great many of them will drop out each year
and relatively few will remain to graduate.

Disciplinary Use of Tests. — Teachers, administrative
officers and others are finding many ways in which they
may make use of the results of intelligence testing for
disciplinary purposes. Particularly effective is this in
the case of brighter students who are lazy or doing ineffi¬
cient work. In college, tests are frequently used to help
in determining dismissal for low scholarship. They are
also used to help in determining the amount of work a
student shall be allowed to carry.

Survey Uses. — Intelligence tests are frequently used
for survey purposes, where the results are of little sig¬
nificance for the individual, but where comparisons of
large groups are made. Surveys of the high school stu¬
dents of a whole state may be made, or of the elementary
schools within a city or within a certain section of a state.
A survey of a university may be made in order to compare
the intelligence of the students in the different colleges
of the university. The largest survey by means of intelli¬
gence tests was that made during the World War of more
than a million drafted men in the United States Army.
A sample of the type of comparison of schools in a sur¬
vey is shown in Figure 31, which shows the middle fifty
per cent of high school seniors in various schools in va¬
rious counties in Indiana. The great differences between
schools are well brought out.

Measures of Progress and Efficiency. — Intelligence
tests are of importance in helping to measure the educa-

PROGRESS AND EFFICIENCY

147

TEST SCORES
90 100

PERCENTILES 5

UO 120 130 140 150 160 170

10 25

COUNTY C

75 90 95

COUNTY B

COUNTY A

COUNTY F

COUNTY E

Figure 31. — Scores made by the middle 50 per cent of seniors
representing individual schools in 7 representative counties selected
at random from the 92 counties in the state. Figures at right of
bars designate the individual schools. (From Book’s The Intelli¬
gence of High School Seniors .)

tional progress of pupils on the one hand and the effi¬
ciency of the teaching on the other. This is done by
means of a comparison between a child’s intelligence and
his school achievement. Obviously the child of low in¬
telligence cannot be expected to make the same amount

148

INTELLIGENCE TESTING

of progress as the child of average or high intelligence.
We must therefore take the intelligence into account
when evaluating any individual’s school progress. What
may be considered splendid progress for a child of average
intelligence may be very poor for a child of high intelli¬
gence. To rest content with average progress in the
case of children of superior ability is to countenance idle¬
ness and inefficiency. It means a waste of intelligence.

A teacher’s efficiency can only be justly evaluated in
terms of the intellectual material with which she has to
deal. It is manifestly unfair to compare the work of
two teachers without knowing the mental make-up of
their pupils. And yet this is frequently done. One who
is achieving excellent results is praised without further
question, whereas another teacher with an inferior group
of children may be blamed for the small improvement
made. In reality the latter teacher may be much more
able and efficient. No real comparisons between teachers
can be made without taking into consideration the in¬
telligence of the children they are teaching. The measure¬
ment of a teacher’s worth lies in the amount of change
she can produce in her pupils, and this can only be rightly
interpreted in terms of the intelligence of the children
themselves.

Experimental Uses. — Lastly we may mention the use
of intelligence tests in scientific work in education. In
countless studies during the past twenty years, intelli¬
gence tests have found a recognized place. One of the
most common ways in which they are used is to help the
experimenter in the formation of equivalent groups. We
wish to measure the effect of a new procedure, a new
method. We raise such questions as to which of two
methods of reading is best, as to whether the study of
Latin increases our knowledge of English words, whether
dental hygiene improves scholarship, and the like. If
our method is to compare an experimental with a control

INTELLIGENCE AND OCCUPATION 149

group, then we must make our two groups as nearly
alike as possible in all ways, and one of the most important
ways in which two such groups must be kept alike is
undoubtedly with respect to intelligence. If they are
not so equated, the increase in efficiency of the experi¬
mental group may not have been due to the method of
teaching arithmetic employed, or the fact that they have
had their tonsils removed, but simply and solely to the
fact that their I.Q.’s are higher and they have there¬
fore been developing intellectually faster than the control
group.

Intelligence and Occupation. — Different occupations
obviously require different amounts of intelligence. Just
how much is required for each occupation we do not
know, but we may get some idea by studying the intelli¬
gence of those now holding different positions. In the
rough and ready selection of everyday life, people will
tend to drop out of those occupations for which they do
not have the requisite amount of intelligence.

The most comprehensive survey of the intelligence of
various occupational groups was made in connection with
the army testing. Figure 32 shows the intelligence ratings
of the middle fifty per cent of each occupation, the short
vertical line marks the median intelligence and the total
scale of intelligence is pictured at the top and bottom of
the diagram. The median intelligence of laborers, for
example, is C —, i.e. rather mediocre intelligence. The
middle fifty per cent stretch from D, through C— to C.
In addition there was a lower group of twenty-five per
cent who made ratings of D and D—■, and an upper
group also of twenty-five per cent who made ratings of
C, C+, B and A. As a matter of fact of the 1453 labor¬
ers tabulated, 10.7 per cent made ratings of C+, 3.5
per cent ratings of B and 0.62 per cent ratings of A. In
a similar manner we may interpret all the other occupa¬
tions on the chart.

150

INTELLIGENCE TESTING

D -

C-

C +

€-

LABORER_

J GEN. MINOR
\ TEAMSTER _
BARBER_.

HORSESHOER_

BRICKLAYER_..

COOK_

BAKER__

PA! NTER_

GEN. BLACKSMITH_.

GEN. CARPENTER_

BUTCHER__

GEN. MACHINIST_

HAND RIVERTER_

TEL. & TEL. LINEMEN-.

GEN. PIPEFITTER_

PLUMBER_

TOOL AND GAUGE MAKER.
GUNSMITH_

GEN. MECHANIC_

GEN. AUTO REPAIR MAN _.
AUTO ENGINE MECHANIC-
AUTO ASSEMBLER.

SHIP CARPENTER_

TELEPHONE OPERATOR___

CONGPETE CONST. FOREMAN*
STOCK-KEEPER_____

c+<

PHOTOGRAPHER..
TELEGRAPHER —

R.R. CLERK _

FILING CLERK __

GEN. CLERK _

ARMY NURSE _

BOOKKEEPER _

DENTAL OFFICER.

MECHANICAL DRAFTSMAN.

ACCOUNTANT _

CIVIL ENGFNEER

MEDICAL OFFICER_

A<e

ENGINEER OFFICER

l D " 7

C + | B I A I

C -

Figure 32. — Occupational intelligence ratings. Letter grades
on horizontal scale. Length of bar for each occupation is mid-range
of 50 per cent; median point is shown by a cross line. (From Mem¬
oirs of National Academy of Sciences, Vol. XV.)

The chart should not be considered as absolutely accu¬
rate for all occupations in ordinary peace-time condi¬
tions. The draft was somewhat selective. Nevertheless
in its broad outlines it is undoubtedly significant. It
shows how selection according to occupation is taking
place all the time. From the point of view of general

INTELLIGENCE AND OCCUPATION 151

intelligence alone, it shows the wide selection of occupa¬
tions open to a person of a given intelligence rating. The
man of high intelligence has the whole range of occupa¬
tions to choose from. The man of C + intelligence has a
fairly wide selection, particularly if he is not ambitious
to surpass the average of his group. If he wishes to
belong to the upper twenty-five per cent of his chosen
occupation, his choice is considerably narrowed. For the
man of D intelligence, the choice is very narrow indeed.

Other rankings of occupational groups have been made
but none on so vast a scale as the army chart. Table 1
shows the average score on an intelligence test given to
various occupational groups (Scott and Clothier, 8). The

TABLE 1

Average Score of Typical Groups

Groups Average Score

Sales force, employed to assist in holiday rush (women).. 25

Sales force, department store (women).27

Students, commercial business college (women).28

Sales force, employed to assist in holiday rush (men)... .29

Office boys.31

Sales force, department store (men).33

Machine operators (men).33

Job foremen.38

All office employees (women).40

Foremen.41

Nurses.42

Rotary Club members...46

Executives of progressive firm.51

Supervisors (manufacturing plant).52

Sales executives.54

Students (college of arts and sciences).54

Students (medical school).56

Engineering students.57

College presidents (small colleges).58

Students (succeeding in examinations for internship) ... 60

difference between the upper and the lower groups is
quite marked.

152

INTELLIGENCE TESTING

The use of intelligence tests in a large tire manufactur¬
ing company led to a decision that individuals having
scores below certain points were not good risks. Figure
33 shows the average intelligence scores of four different
occupational groups in this factory and also the critical
scores below which an applicant for a position should not
fall without having definite compensating qualifications.

Tests of Specific Capacities. — In addition to such gen¬
eral tests, mention must also be made of tests for spe¬
cific capacities. The best example at the present time is
the Seashore Test of Musical Capacity, which measures
such traits as pitch, intensity, time, rhythm, consonance
and auditory memory. These tests predict musical abil¬
ity and agree well with the estimates of music teachers
as to their students’ abilities. The Eastman School of
Music of the University of Rochester requires that these
tests be taken by all candidates for admission to the
school. Those whose talent, as measured by the test,
does not warrant the expenditure of time and money on
the study of music are advised accordingly. It is obvi¬
ously to the interest of both the student and the school
to have such tests made.

These are only a few of the ways in which general in¬
telligence tests and specific capacity tests may be used
in the wide field of vocational guidance and industrial
placement. Intelligence tests cannot place the right man
in the right job, but they do help, and any help given in
this important work of steering the individual to a fitting
occupation is well worth while not only in terms of hap¬
piness and contentment to the individual, but in terms
of the general efficiency and moral health of society in
general.

Intelligence and Delinquency. — One of the interesting
results of intelligence testing has been the study of the
mentality of the delinquent individual. A great many
studies have been made, and, although the percentage

INTELLIGENCE AND DELINQUENCY 153

Figure 33. — Occupational Standards on Mental Alertness Test for women applicants and women e:
ployees in a large tire manufacturing company. (From Scott and Clothier’s Personnel Management.)

154

INTELLIGENCE TESTING

of feeblemindedness among delinquents varies greatly
from one worker to another, most are agreed that this
percentage is greater than the percentage generally found
among the non-delinquent population. The variations
in estimated amounts of feeblemindedness vary not only
because of different concepts of feeblemindedness held by
different workers, but also because of the different insti¬
tutions in which the psychologist may have gathered his
cases. In workhouses and county jails more dull and
feebleminded prisoners are likely to be found than in
state or federal prisons. Juveniles committed to a re¬
form school will on the whole be duller than the general
run of those found in a juvenile court detention home,
and these again be duller than those in the special school
for truants. Figure 34 shows the intelligence quotients
of 1212 juvenile repeated offenders tested by the Stan¬
ford Revision (Healy, 3). From this chart we see that
there are a great many children of normal or above nor¬
mal intelligence who are delinquent, but there is an ex¬
cess of cases testing below 90 LQ. From a study of 4000
repeated delinquents the same investigator concluded that
13.5 per cent are clearly feebleminded. This percentage
of feeblemindedness is very much higher than that found
in the general population.

Lack of intelligence is undoubtedly one of the numer¬
ous causal factors at the basis of delinquent conduct,
but it is by no means the sole cause of such conduct. A
careful comparison of a delinquent with a comparable
non-delinquent group of children living in the same envi¬
ronment has been made by Burt (2). He says,* “The fol¬
lowing proves to be the order of importance of the various
conditions we have reviewed: (1) defective discipline; (2)
specific instincts; (3) general emotional instability; (4)
morbid emotional conditions, mild rather than grave, gen-

* Quoted by permission from Burt, C., The Young Delinquent, D. Ap¬
pleton and Co., New York.

INTELLIGENCE AND DELINQUENCY 155

Figure 34. — Distribution of I.Q.’s of repeated offenders.
(From Healy’s Practical Value of Scientific Study of Juvenile
Delinquents.)

erating or generated by so-called complexes; (5) a family
history of vice or crime; (6) intellectual disabilities, such
as backwardness or dullness; (7) detrimental interests,
such as a passion for adventure, for the cinema, or for
some particular person, together with a lack of any up¬
lifting pursuits; (8) developmental conditions, such as
adolescence or precocity in growth; (9) a family history
of intellectual weakness; (10) defective family relation¬
ships — the absence of a father, the presence of a step¬
mother; (11) influences operating outside the home — as
bad street companions and lack or excess of facilities for
amusement; (12) a family history of temperamental dis¬
order— of insanity or the like; (13) a family history of
physical weakness; (14) poverty and its concomitants;
and last of all, (15) physical infirmity or weakness in the
child himself.

“ Heredity appears to operate, not directly through the
transmission of a criminal disposition as such, but rather
indirectly, through such constitutional conditions as a
dull or defective intelligence, an excitable and unbalanced
temperament, or an over-development of some single

156

INTELLIGENCE TESTING

primitive instinct. Of environmental conditions, those
obtaining outside the home are far less important than
those obtaining within it; and within it material condi¬
tions such as poverty, are far less important than moral
conditions, such as ill discipline, vice, and most of all, the
child’s relations with his parents. Physical defects have
barely half the weight of psychological and environ¬
mental. Psychological factors, whether due to heredity
or to environment, are supreme both in number and
strength over all the rest. Intellectual conditions are
more serious than bodily, and emotional than intellectual;
while psycho-analytic complexes everywhere provide a
ready mechanism for the direction of overpowering in¬
stincts and of compressed emotional energy into open
acts of crime.”

The Deaf and the Blind. — The education of these
children, deprived as they are of one or the other of the
two most important senses, is for that reason very diffi¬
cult. Intelligence tests would seem also to suggest that
on the average they are somewhat below the hearing or
seeing child. This is very marked in the case of the deaf.
With verbal tests the difference is very great, with non¬
verbal or performance tests the difference is not so great,
but still marked. This makes the education of such chil¬
dren a doubly difficult task. In the case of the deaf
especially the lack of hearing closes the most important
avenue for the learning of language and hence slows up
and makes more difficult the process of learning to read.
Thus those, who might hope to compensate for their
deficiency somewhat through the printed page, are handi¬
capped at the start by the great difficulty in the mastering
of reading and also by a slighter endowment of general
intelligence.

Figure 35 shows a comparison of 678 deaf fourteen-
year-olds with 1120 hearing fourteen-year-olds on the
Pintner Non-Language Test. The percentage of deaf

INTELLIGENCE AND RACE DIFFERENCES 157

Figure 35. — Distribution of scores on the Pintner Non-Language
Mental Test of Deaf and Hearing Children—Age 14. (From Day,
Fusfeld and Pintner’s Survey of American Schools for the Deaf.)

children making low scores is much greater than the per¬
centage of hearing children, and, conversely, the per¬
centage of deaf making high scores is lower than the
equivalent percentages for the hearing. The mean score
of the deaf fourteen-year-olds is about the same as
the mean score for eleven-year-old hearing children
(Pintner, 5).

Intelligence and Race Differences. — The interest of
man in his fellow men of different racial origin leads him
continually to make comparisons and contrasts. It is
only natural that many comparisons of different racial
groups by means of intelligence tests should have been
made. For the most part these comparisons have been
of school children of different racial stocks in this country.

158

INTELLIGENCE TESTING

The most extensive comparisons that have been made are
between American white and negro children. The white
children make a higher average score. In terms of I.Q.,
where the white child makes an average of 100, the negro
child makes an average of from 85 to 95. In terms of the
total distribution about 25 per cent of the negroes reach
or exceed the median score of the whites. These differ¬
ences hold both for very verbal and abstract types of
tests as well as for more concrete and non-verbal tests.
This difference in intelligence between negroes and whites
is not very great. The overlapping of the two distribu¬
tion curves shows, as we have said above, that 25 per cent
of the negroes have more intelligence than the average
white, and it furthermore shows that a great number of
white and negro children have the same amount of intel¬
ligence. In other words the two racial groups do not
form two widely divergent intelligence types. There can
be no justification from the standpoint of intelligence of
a policy of separate education for the two groups. If
such is deemed advisable it must be justified on other
grounds. Nor can race prejudice against the negro find
support in these test findings. Race prejudice springs
from and is nurtured by emotional reactions and is not
occasioned by scientific findings, although we frequently
make use of such findings to rationalize our prejudices.
To say that you discriminate against the negro because
he belongs to a race having inferior intelligence, is an
attempt on your part to rationalize your racial prejudice.

Other racial groups in the United States have been
compared by means of the results found among drafted
men in the army testing. A great many men of foreign
birth were found in the draft army. The distribution of
the intelligence grades for many national groups is shown
in Figure 36. The white shows A or B (i.e. superior
ratings), the shaded shows C or average ratings and the
black D or inferior ratings. These ratings, of course, tell

SCOTLAND

S.W.EDEN

CANADA.

WHITE DRAFT

NORWAY

AUSTRIA

IRELAND

TURKEY

ALL FOREIGN COUNTRLES

D C BA

Figure 36. — Percentage distribution of letter grades in intelli¬
gence by nativity of foreign-born men in army draft. (From
Memoirs of the National Academy of Sciences , Vol. XV.)

160

INTELLIGENCE TESTING

us nothing about the mental ability of the nations from
which these immigrants come, but they do indicate
roughly the mental calibre of the sampling which is com¬
ing to this country.

The results of many verbal intelligence tests given to
school children of different racial stocks are very difficult
to evaluate because of the fact that most of these chil¬
dren come from homes where the language spoken is not
English. Hence, although the child speaks English in
school and on the street, he is probably more or less
handicapped in the use of this language. The use of
non-verbal tests is the only solution of this question.

Children of Italian parentage have been tested by many
investigators and there is a surprising unanimity in their
results on the Binet Test in many different parts of the
country. The average I.Q. of various groups lies be¬
tween 85 and 90. When tested by non-verbal tests
Italian children also fall below, but not so far below
American English-speaking children.

On the Pintner Non-Language Test the mean score
of twelve-year-old Italians was found to be 305 as com¬
pared with a mean score of 329 for twelve-year-old Amer¬
ican children. If these scores are converted into approx¬
imate mental ages, the Italians have an M.A. of 11-6, as
compared with the American M.A. of 12-6. In terms of
I.Q. this would be an Italian I.Q. of 92 as compared with
an American of 100. It would seem probable, therefore,
that Italian immigrants to this country have on the
whole been below the average intelligence of the country
as a whole.

Somewhat the same situation seems to be true of
Polish and other Slavonic races, with the probable excep¬
tion of the Bohemians. With reference to the Jewish
race the results are not clear. On the whole the intelli¬
gence seems to be about the same as that of American
children, with a possibility of being slightly superior.

SUMMARY

161

American Indian children fall below the average on tests
of abstract intelligence.

All of these results of the testing of children of foreign
parentage give us merely a clue to the intelligence of the
immigrants from foreign countries. They cannot in any
sense give us any knowledge as to the average general
intelligence of the countries from which these immigrants
have come. Up to the present time no adequate com¬
parative tests have been made between peoples in the
countries where they live. One very inadequate sampling
of Belgian children made in Belgium showed no difference
between their scores and the American norms on the
Non-Language Test (Pintner, 6).

Summary

1. The most common use of group intelligence tests is an aid
to the classification of pupils.

2. The mental age helps to determine the grade for which the
pupil is best fitted.

3. The intelligence quotient helps to determine the section
for which the pupil is best fitted.

4. Intelligence tests are used to select the dull and feeble¬
minded for special instruction.

5. Intelligence tests are essential for the selection of bright
children for special classes.

6. It is very desirable that such children should be selected as
early as possible in their school careers.

7. In a negative sense the results of intelligence tests are useful
for general educational advice and guidance.

8. Many colleges are using intelligence tests as an aid in the
selection of their students.

9. Surveys of high school students show that many of high
intelligence do not intend to go to college, whereas many of
low intelligence have such ambitions.

10. Intelligence tests may be used in a disciplinary sense, to
motivate school w T ork, to determine the student load, to
help in determining dismissal for low scholarship.

162

INTELLIGENCE TESTING

11. Surveys by means of intelligence tests show very striking
differences between schools.

12. Intelligence tests are necessary in order justly to evaluate
pupil progress and teaching efficiency.

13. In much educational research intelligence tests are now con¬
sidered necessary, especially for the purpose of equating
groups.

14. The army testing showed marked differences in the intelli¬
gence ratings of men in different occupations.

15. Intelligence tests are of use in vocational selection, but they
alone cannot determine the right man for the right job.

16. In general the mentality of the delinquent is below that of
the normal child.

17. Lack of intelligence is one of the causal factors at the basis
of delinquent conduct, but is by no means the sole cause of
such conduct.

18. Deaf and blind children in general fall below our norms in
intelligence tests for ordinary children, even when compared
on tests making allowance for their special handicaps.

19. Racial or national comparisons of intelligence ratings are
difficult to make, and no adequate ones of unselected sam¬
plings in their respective countries have so far appeared.
Our best results are based on racial groups in this country.

20. In general the comparison between American whites and
American negroes shows that about 25 per cent of the ne¬
groes reach or exceed the median score of the whites.

21. Italian children in this country fall markedly below the
norms on verbal tests, but not so much so on non-language
tests.

22. Of other racial groups in this country we have less reliable
measures of their abilities.

REVIEW

163

Review

True-False Statements

As a review mark the following statements true or false:

1. After pupils have been classified into grades, they should be
divided into sections according to mental age.

2. A pupil with an I.Q. of 80 is not likely to succeed in high
school algebra.

3. Intelligence is the only factor making for success in school.

4. The higher the social status of parents in general the higher
the intelligence of the children.

5. A man of 35 years having a mental age of 16 on the Stanford
Revision is below normal in intelligence.

6. The average score of unselected American negroes is lower
than the average score of unselected American whites on
intelligence tests.

7. Intelligence tests are useful in school because they measure
the degree of success the teacher has achieved in getting
her subject across to her pupils.

8. The homogeneous grouping of children according to intel¬
ligence widens the educational opportunity for each child.

9. Even a child with an I.Q. of 120 or above may have a special
disability in a certain school subject.

10. Between intelligence test scores and general school success,
there is a substantial correlation.

11. A physical handicap, such as deafness or blindness, is usu¬
ally counterbalanced by keener intelligence.

12. If a child is found to have a mental age above that of the
grade he is in, he should be promoted at once.

13. In comparing the intelligence of pupils in several different
schools, the same tests should be used for all schools.

14. The I.Q. is adequate for accurate vocational guidance.

15. In a regrading program the M.A. is a more important con¬
sideration than is the I.Q.

16. Children with I.Q.’s of 80 and below should be encouraged
to take up practical vocational training.

17. Promotions should always be made if tests show a higher
degree of intelligence than the grade indicates.

164

INTELLIGENCE TESTING

18. So far as capacity is concerned most persons can pursue only
one occupation with a good chance of success.

Advised Readings

Irwin, E. A., and Marks, L. A.: Fitting the Child to the School,
New York, 1924.

Dickson, V. E.: Mental Tests and the Classroom Teacher , Yon¬
kers, 1923.

Pintner, R.: Intelligence Testing, New York, 1923. Chapters
8, 9, 11, 12.

Poffenberger, A. T.: Applied Psychology, New York, 1927.

References

1. Book, W. F. : The Intelligence of High School Seniors, New

York, 1927.

2. Burt, C.: The Young Delinquent, New York, 1925, Chapter

14.

3. Healy, W.: The Practical Value of Scientific Study of Juve¬

nile Delinquents, Children’s Bureau, Publication No. 96,
Washington, 1922, page 19.

4. Irwin, E. A., and Marks, L. A.: Fitting the Child to the School,

New York, 1924, page 134.

5. Pintner, R.: “Psychological Survey of Schools for the Deaf,”

American Annals of the Deaf, Volume 72, No. 5, pages
377-414.

6. Pintner, R.: “Non-Language Tests in Foreign Countries,”

School and Society, Volume 26, 1927, pages 374-376.

7. Pintner, R., and Noble, H.: “The Classification of School

Children According to Mental Age,” Journal of Educa¬
tional Research, 1920, Volume 2, pages 713-728.

8. Scott, W. D., and Clothier, R. C.: Personnel Management,

Chicago, 1923, Chapter 14.

9. Terman, L. M., et al.: Genetic Studies of Genius, Volume 1,

Stanford University Press, 1925.

CHAPTER VII

THE INHERITANCE OF MENTAL TRAITS

The General Problem of Inheritance. — That physical
traits such as height, color of eyes or hair, or skin, general
build and similar characteristics are inherited is taken
for granted by all of us. That they are inherited has been
demonstrated repeatedly. That children should not in¬
herit the physical characteristics of their parents would
be considered strange. When they do not, as in the case
of a Cretin or Mongolian Idiot, some special cause or
reason is sought for as an explanation.

To some extent, likewise, the inheritance of mental
traits has been assumed. A biographer may attempt to
find in the ancestry of his subject the clue to certain
character qualities and personality traits. This is fre¬
quently the case with very marked temperamental quali¬
ties or extreme character traits. The same thing is more
or less true in the common experience of daily life. It
is particularly true when we are at a loss to find a ready
explanation of these character qualities in terms of en¬
vironment. Where we cannot see an obvious causal
connection between a certain type of behavior and the
environment of the child, we are apt to throw upon
heredity the burden of the cause.

Nevertheless there is no such general and widespread
belief in the inheritance of mental characteristics as in
the case of physical ones. As we have suggested, we
frequently seek the cause first in the environment. With
regard to some mental traits, such as general intelligence,
there is frequently much doubt expressed as to its in¬
heritance in the same way as physical traits. Since knowl-

165

166 THE INHERITANCE OF MENTAL TRAITS

edge and skill are acquired in school, many people are
likely to suppose that the forces of education are sufficient
to account wholly for the amount of knowledge and skill
acquired by the child. However, the tremendous differ¬
ences in attainment revealed by children subjected to
the same kind of schooling make us doubt the supreme
potency of environment. We, therefore, shift to the home
and try to explain the differences by means of differences
in opportunities in home environment. But again we are
confronted by tremendous differences in achievement
among children brought up in seemingly similar home
environments.

We may consider our problem as centering around two
questions: (1) Are brothers and sisters of the same fam¬
ily (siblings) more like each other than children taken
at random? and (2) Are siblings even if brought up in
different environments, more alike than unselected un¬
related individuals? Answers to either or both of these
questions will help us to understand better the potency
of heredity in regard to mental traits.

Pioneer Work of Galton. — The pioneer work in this
field was done by Galton in the nineteenth century. He
mapped out the two general trends that have been fol¬
lowed since then. In some of his studies Galton was
interested in the characteristics of certain groups of re¬
lated families. If in the study of such a family history
a certain trait keeps appearing in one generation after
another, it may be that such a trait is inherited. The
chances that it is inherited become greater the more
persistent it is in a family group, and the less obviously
it can be accounted for by the environment in which the
family has lived. In another series of investigations,
Galton started with a certain number of eminent and a
similar number of average men and counted their eminent
relatives. The eminent group had 535 eminent relatives
and the average group only 4. He also compared parents

CORRELATIONAL STUDIES

167

and children in certain physical and mental traits, and
found as great a resemblance in the mental as in the
physical traits. In this way, therefore, Galton suggested
two lines of study: (1) Measurements of Resemblance
by means of Correlations; (2) Family History Studies.

Correlational Studies. — Correlational studies attempt
to show the amount of resemblance of any trait existing
between pairs of related individuals as contrasted with
the amount of such resemblance in random pairs of un¬
related individuals. Such random pairs of unrelated
individuals will show no correlation, just because by
chance any amount of one trait may be paired with any
amount of another trait. If, however, there are factors
at work, such as heredity or similar home training, that
make for resemblance between siblings, then the correla¬
tion between siblings should be greater than zero. Such
positive correlations as we have suggested do not prove
the heredity of the traits in question. If, however, the
correlations for mental traits resemble the correlations
for physical traits, not subject to environmental influences,
then the suggestion is strong that heredity is a causal
factor in the mental as in the physical traits. If, further¬
more, we can find a number of cases of siblings reared
apart- showing correlations as high as those of siblings
reared together, then our belief in the inheritance of such
traits is strengthened. And, negatively, a lack of correla¬
tion among unrelated children reared in the same en¬
vironment would seem to indicate the potency of heredity
as contrasted with environment.

There are many studies of the resemblance of traits
and we cannot here review them all. We shall merely
try to indicate the general results. These results are ex¬
pressed in terms of correlation coefficients and we must
remember that the nearer the coefficient approaches 1.00
the greater the resemblance and that zero or very low
coefficients indicate lack of resemblance. We must re-

168 THE INHERITANCE OF MENTAL TRAITS

member also that the amount of resemblance will not be
perfect because children in the same family (or siblings)
do differ from one another. And finally for the purposes
of comparison we may note that correlations of from .44
to .62 have been found for such physical traits as color
of eyes, color of hair, cephalic index, height, deafness and
the like—traits which are hardly, if at all, dependent
upon environmental influence.

Character Traits. — In this field we have little evidence,
because of the difficulty of securing reliable quantitative
measures. The ratings of children by teachers have been
resorted to in lieu of objective measures. Teachers rated
about 2000 pairs of siblings on such traits as vivacity,
self-assertiveness, introspection, popularity and the like.
The average of 18 correlations is .53. If the accomplish¬
ment quotient (that is the E.Q. divided by the I.Q.) can
be taken as an indirect measure of effort put forth by the
child, we find a correlation of .32 for 188 pairs of sibs
tested on intelligence and achievement tests.

Now that objective tests of character traits are becom¬
ing available, more reliable measures are beginning to
appear. Investigations (May and Hartshorne, 5) of
sibling resemblance by means of objective tests for the
measurement of deception show a correlation of .47. It
would seem, therefore, that the resemblance of siblings
in character traits is about as large as that found for
physical and for intellectual traits.

Intelligence. — Here we have a great many more
studies, and all of them made by means of objective tests.
The average of several correlations reported by early
workers in this field who used psychological tests such as
cancellation, memory and the like is about .30. The
average of twelve later reports using standard intelli¬
gence tests is about .51, based upon about 5000 pairs
of siblings, with a range of correlations from about .33
to .68. It would seem as if the correlation increases

COUSINS

169

with increase in accuracy and adequacy of the trait
measured.

School Achievement. — There are also many studies of
the resemblance of siblings in school work. The measures
in these cases are teachers’ grades, general scholastic
records or special tests of specific studies. The average
of the many correlations reported is again .50.

Parent-Child Relationship. — Two studies report intel¬
ligence tests given to parents and children, giving average
correlations of .35 and .53. These correlations are much
the same as those found between siblings.

Twins. — The resemblance of twins in intelligence has
been studied by several workers and if we sum up the
correlations obtained we find that the average is about
.78 for about 400 pairs. The correlations here are de¬
cidedly higher than for siblings. There seems also to be
some evidence that so-called identical twins resemble
each other on intelligence tests somewhat more closely
than do non-identical twins.

There are several detailed studies of the specific reac¬
tions of twins on a given series of tests. Figure 37 shows
the profile of the reactions to the Stanford Binet Test
by one pair of twins. The failures and passes on the test
occur at almost identical places throughout.

Cousins. — Studies of further degrees of relationship
are very scarce. A few correlations have been reported
for cousins with reference to school achievement and in¬
telligence. The average correlation for about 500 pairs
is .23.

If now we look at these three degrees of relationship,
twins, siblings, and cousins, we note that the correlations
decrease as the relationship becomes less close: about
.70 for twins, about .50 for sibs, about .20 for cousins and
zero for unrelated children. There is great likelihood that
heredity is a potent factor in causing these resemblances.
Furthermore, the resemblance for siblings in intelligence,

170 THE INHERITANCE OF MENTAL TRAITS

Jean Jeanette

June, 1923

I.Q. 112 I.Q. 117

Pass Fail Fail Pass

J ean Jeanette

January, 1924
I.Q. 106 I.Q. 110

Pass Fail Fail Pass

Figure 37. — Stanford-Binet examinations of twins. (From
Wentworth’s Individual Differences in the Intelligence of School
Children .)

character and school achievement is about the same as the
resemblance in physical traits. Since the resemblance in
physical traits cannot be considered as due to environ¬
ment, it is argued that the resemblance in mental traits,
as in physical traits, is primarily a consequence of heredity.

Dependent and Foster Children. — Nevertheless, these
correlations do not prove that heredity is the sole factor.
A similar environment might be influential in causing
these correlations, for we note that as the correlations
increase from unrelated children up through cousins and

FAMILY HISTORY STUDIES

171

sibs to twins, so also does the similarity of environment
for each pair increase. These correlational studies cannot
tell us how much of this resemblance is due to heredity
and how much to environment. Other studies (5) have,
therefore, been made to find out what influence on intel¬
ligence and achievement ratings has been produced by
changes in the environment of children. These studies
deal with the changes produced in dependent children
adopted into foster homes, with comparisons between
siblings reared apart in different homes and siblings reared
together, and with unrelated children reared together.
These studies are recent and somewhat contradictory.
It is difficult to evaluate them at the present time, but in
general they show that some increase in I.Q. on such a
test as the Binet takes place in young children when trans¬
ferred from a poor home to a better home. We must not,
therefore, think of a Binet I.Q. as a fixed unchangeable
measure of a child’s intelligence. By a change to a supe¬
rior environment this rating can be increased from 10 to
20 points, if the change takes place at an early age and if
it is from a very poor to a very superior environment.
Such changes, however, are rare. They are more likely
to take place if the child is moved to the good environ¬
ment when very young, and less likely to take place with
children of school age.

Family History Studies. — The second method of show¬
ing the inheritance of mental traits is to analyze the
characteristics of a family group and to note what mental
traits predominate. If there is a great similarity in cer¬
tain traits through several generations the chances are
that this is due to the inheritance of underlying basic
abilities or tendencies to reaction. Galton (2) stimulated
such studies by his early work and his own family itself
illustrates well the concentration of eminent individuals
within one family group. He himself was the cousin of
Charles Darwin and the grandson of Erasmus Darwin, a

172 THE INHERITANCE OF MENTAL TRAITS

distinguished botanist and author. Of the two sons of
Erasmus, one was distinguished in experimental research
and the other was a notable physician and father of
Charles Darwin, one of the greatest men in modern
science. He married Emma Wedgwood, a granddaugh¬
ter of Josiah Wedge wood of pottery fame. Charles Dar¬
win’s four sons all became notable, one as a botanist,
another as an astronomer, another as an engineer and
another as a political economist and eugenist. Figure
38 is a family history chart showing the eminent people
in these related family groups. This figure illustrates
the typical method followed in family history investi¬
gations.

Another famous family whose descendants have been
traced is the Edwards family. Johnathan Edwards was
born in 1703 and in 1900 there were 1394 descendants
accounted for. Among these we find 12 college presidents,
60 physicians, 65 college professors, about 100 clergymen
and missionaries, 75 army or navy officers, 60 prominent
writers, by whom 135 books of merit were written and 18
important periodicals edited, about 100 lawyers, 30 judges,
80 public officials, such as United States Senators, gover¬
nors, members of Congress, mayors, and ambassadors,
as well as many successful bankers, business men and
managers of large industrial enterprises. There is no
trace of any feeblemindedness in this group and no record
of any criminal.

On the other hand there are also records of families in
which the opposite type of individual predominates. One
of the earliest of such studies is that of the so-called Jukes
family (1). Max Juke was born in 1720. He was a shift¬
less, intemperate and illiterate backwoodsman, married
to a stupid worthless woman. Up to 1877 about 1200
descendants had been traced, and among these we find
310 paupers, 7 murderers, 60 thieves, 50 prostitutes and
130 convicted of crime. Only 20 of the family learned a

FAMILY HISTORY STUDIES 173

trade and 10 of these did so in prison. A later follow-up
in 1915 of the Jukes showed that the descendants still
manifested the same traits and that migration of some
branches of the family to a different environment had
not changed their original native tendencies.

Another striking picture is that offered by the Kallikak

174 THE INHERITANCE OF MENTAL TRAITS

family (Goddard, 3). Martin Kallikak, a soldier in the
Revolutionary War had an illegitimate son by a feeble¬
minded girl. Later he married into a good family. Thus
we have two lines of descendants from the same man.
In 1912 the illegitimate line showed 480 descendants
among whom were 143 feebleminded, 36 illegitimates,
33 prostitutes, 24 alcoholics, 3 epileptics and 3 criminals.
The legitimate line showed 496 descendants, mostly good
citizens including doctors, lawyers, educators, judges and
business men. Only five were reported as showing bad
traits, such as alcoholism, insanity and immorality. There
were no criminals. Figure 39 shows the descendants of
the legitimate and illegitimate lines of the Kallikak family.
In this picture we note the great number of feebleminded
individuals and the tendency for the feebleminded to
marry the feebleminded.

There are many other family histories pointing in the
same general direction as those we have mentioned. The
tendency of certain traits to persist in a family is a strong
presumption of heredity.

In spite of all the changes that take place in families,
we seem to see no marked change in their personal charac¬
teristics—the intelligent family remains intelligent, the
dull family dull. If opportunities came in the way of the
dull family, and probably some did, they were unable to
make use of them. If misfortunes overtook the intelligent
families, and probably some did, they were able to survive
and overcome them.

Relatives of Superior Children. — Similar but some¬
what different evidence of the inheritance of desirable
traits is shown in Terman’s (7) study of the relatives of
superior children. Of the 62 members of the Hall of Fame,
14 or 22.6 per cent are known to be related to one or more
of the superior children studied. Twelve of the 643
children have parents or grandparents in Who’s Who ,
even although the average age of the fathers is only 41

RELATIVES OF SUPERIOR CHILDREN 175

CASPAR.

4 . mo.

<s>-

?HE LAWFUL. WIFE THE NAMELESS FEEBLE-MINDED CISU.

■0 WOT MARRIED/*"^

nARTIN
KALLIKAK, SR.
<L 1837.

In] B

FREDERKfL MARTIN

9LALJUKAK JR,

-m

HJLLARD.

JUSTIN.

HARTHAt

•NOT MARRIC

DEBORAH

Figure 39. — The legitimate and illegitimate lines of the Kallikak
family. The black denotes feeblemindedness. (From Goddard’s
Kallikak Family.)

years, whereas the average age of first inclusion in Who's
Who is 49.9 years. There were 35 other relatives reported
in Who's Who. In addition there were 58 other relatives
of sufficient distinction to be named in standard cyclope¬
dias of biography. Parents and grandparents have held
many responsible positions, 46 holding important national
or state positions, 67 important religious or fraternal posi-

176 THE INHERITANCE OF MENTAL TRAITS

tions, 27 university positions, 74 important business posi¬
tions and 18 bank presidencies.

Social Status. — Many studies have been made of the
social status of the family in which the eminent man is
born. Thus the fathers of 885 leading American men of
science were distributed as follows:

Professional 43.1%; in general population 3.0%

Business 35.7%; in general population 34.1%

Agriculture 21.2%; in general population 41.1%

In England and France similar studies have shown
similar facts.

The distribution of the occupations of the fathers of
Terman’s gifted children compared with the proportion
in the population from which the gifted children came is
very instructive.

Percentages

Fathers of

General

Gifted

Population

Professional

29.1

2.9

Public Service

4.5

3.3

Commercial

46.2

36.1

Industrial

20.2

57.7

In this comparison notice the large percentage of the
professional group as contrasted with its percentage in
the general population, and conversely, the small propor¬
tion in the industrial group as contrasted with the large
proportion in the general population. Eminent men not
only tend to come from the professional classes, but these
professional classes show a much greater proportion of
gifted children. What makes for eminence in after life
evidently tends to show in early childhood.

Another way of looking at this problem is shown by a
comparison of the intelligence ratings of different occupa¬
tional groups. One study by Sandiford (6) includes
about 5000 high school and college students. Another
study by Haggerty and Nash (4) includes about 8000

SOCIAL STATUS

177

children of all grades. The tests used were not the same,
so we need pay little attention to the absolute values of
the I.Q’s. Below we have combined the results of these
two studies:

Median I.Q. of Children

pation of Fathers

High School

All Grades

Professional

105

116

Business

103

107

Skilled

103

Semi-skilled

102

Farmers

102

Unskilled

101

The important point to notice is that the medians, al¬
though different, decrease in the same fashion as we de¬
scend from the professional to the unskilled group. There
are many other studies of this nature all showing the same
general trend.

Of course this relationship between the occupation of
the parent and the intelligence of the child is no proof of
heredity, but taken along with the other facts of this chap¬
ter it fits in admirably. The more intelligent children
tend on the whole to have the more intelligent parents
and these again tend to be found in greater numbers in
the higher occupations. For the higher the occupation,
the more likely is it to require intelligence. The medians
or averages quoted merely show the general tendency of a
rise in intelligence from the unskilled to the professional
group. If the whole distributions were to be shown, we
would see that there is much overlapping from group to
group. There are some in the unskilled group with intel¬
ligence as good as, or better than, the average of the pro¬
fessional group and, conversely, some in the professional
group with no better intelligence than the average of the
unskilled.

This general hierarchy of occupations with reference to
the intelligence of the men in them can be seen in Figure

178 THE INHERITANCE OF MENTAL TRAITS

32 on page 150 of this book. It is the chart of the intelli¬
gence of occupational groups in the army, which we used
in the last chapter in another connection. The diagram
shows the median intelligence rating by a short vertical
line, and the length of the line shows the distribution of
intelligence for the middle fifty per cent. We must re¬
member, therefore, that twenty-five per cent of the cases
lie to the right and twenty-five per cent to the left of the
horizontal line. The letters, A, B, C+, etc., indicate the
intelligence ratings. There were a few laborers who
achieved an A rating. The middle fifty per cent, however,
ranged from C to D and the lowest twenty-five per cent
fell in the D and D — classes. At the upper end notice that
the middle fifty per cent of the medical officers lie in the B
and A classes, that the upper twenty-five per cent is wholly
in the A class, while the lower twenty-five per cent lies in
the C+ or lower classes. As a matter of fact less than one
per cent fell below C and there were none in the D and D—
classes. The diagram then illustrates well the general
tendency to increase in intelligence as we go from laborer
to professional man, and it also illustrates the wide range
of intelligence to be found at each step. Much more than
intelligence determines a man’s occupation, but intelligence
is one of the factors. If, then, intelligence is inherited we
should expect to find a similar hierarchy in the sons of
the men in these different occupations and such a general
hierarchy, as we have already shown, exists among school
children.

Summary

1. Mental characteristics would seem to be inherited in the
same manner as physical characteristics.

2. The pioneer work in the study of the inheritance of mental
traits was done by Galton.

3. He started both methods of investigation, namely (1) corre¬
lational study of resemblance and (2) the family history
study.

SUMMARY

179

4. Correlations of siblings for physical traits range from about
.44 to .62.

5. The average of many correlations for character ratings of
siblings by teachers is .53.

6. The average of many correlations for intelligence tests of
siblings is .51.

7. The average of many correlations for scholastic tests and
grades of siblings is .50.

8. The correlations for the mental traits of siblings are about
the same as the correlations for the physical traits.

9. There are very few studies of the resemblance of parent and
child. Correlations of .35 and .53 have been reported.

10. The correlations of intelligence tests of twins average about
.78.

11. A suggestion of the resemblance between cousins is indi¬
cated by one report of a correlation of .23.

12. Random samplings of unrelated children show zero corre¬
lations on intelligence tests.

13. The correlation would seem to decrease the further apart
the degree of relationship becomes.

14. A marked change in environment begun at an early age and
continued for several years may lead to a change in intelli¬
gence as rated by the Binet Scale.

15. Family history studies show desirable or undesirable traits
running through many generations.

16. The relatives of children testing high on intelligence tests
are on the whole very decidedly superior to the relatives of
the average run of people.

17. There is a decided correlation between the occupation of the
father and the intelligence rating of the offspring.

180 THE INHERITANCE OF MENTAL TRAITS

Review

True-False Statements

As a review mark the following statements true or false:

1. Galton found that a group of eminent men had no larger
number of eminent relatives than a group of average men.

2. In mental tests, the resemblance between twins is no greater
than that between siblings.

3. Ancestry reduces the variability of offspring and determines
the point about which they vary.

4. There is little correlation between the general intelligence
of children and the social status of their parents.

5. The resemblance between siblings in I.Q. is greater than
that between unselected, unrelated individuals.

6. Correlations between siblings in mental traits are in general
higher than the correlations in physical traits.

7. The resemblance between siblings in school achievement is
much greater than in intelligence tests.

8. Foster children seem to improve somewhat in I.Q. if adopted
early and brought up in a very good environment.

9. The Kallikak study of Goddard proves that feebleminded¬
ness is hereditary.

10. It was the poor environment surrounding the Jukes that
led to so much poverty, criminality and immorality.

11. If the Jukes had been transplanted to a better environment,
the chances are they would have proved perfectly normal.

12. Every child with an I.Q. above 140 is almost certain to have
eminent relatives.

13. Eminent men tend to come from the professional class, and
this class produces a relatively greater number of children
with high I.Q/s.

14. Underline the best responses:

The correlation between siblings in intellectual traits is
about 10, 20, 50, 75, 90.

15. The correlation between twins in intellectual traits is about
10, 20, 50, 75, 90.

16. The highest average I.Q. is likely to be found among chil¬
dren whose fathers are: (1) laborers (2) farmers (3) profes¬
sional men (4) business men.

REFERENCES

181

Advised Readings

Galton, F.: Hereditary Genius, New York, 1869.

Goddard, H. H.: The Kallikak Family, New York, 1914.

East, E. M.: Heredity and Human Affairs, New York, 1927.

References

1. Dugdale, R. L.: The Jukes, New York, Fourth Edition, 1910.

2. Galton, F.: Hereditary Genius, New York, 1869.

3. Goddard, H. H.: The Kallikak Family, New York, 1914.

4. Haggerty, M. E. and Nash, H. B.: “Mental Capacity of

Children and Parental Occupations,” Journal of Educa¬
tional Psychology, 1924, 15: 559-572.

5. National Society for the Study of Education, Twenty-

Seventh Yearbook, 1928.

6. Sandiford, P.: “Paternal Occupations and Intelligence of

Offspring,” School and Society, 1926, 23: 117-119.

7. Terman, L. M. et al.: Genetic Studies of Genius, Volume I,

Stanford University Press, 1925, Chapter V.

PART II-THE MODIFICATION OF
ORIGINAL NATURE

CHAPTER VIII

THE LAWS OF LEARNING

General Tendency of Modification. — So far, in the
first part of this book, we have been trying to find out
what the essential nature of the individual is, what his
original nature is. In doing this we have made a survey
of his capacities and tendencies, and we have laid especial
emphasis upon any instruments that have been con¬
structed which attempt to measure these capacities and
tendencies, because such attempts at measurement give
us a more exact knowledge of man’s original nature. Now
there is one aspect or tendency of man that has not been
specifically mentioned, but which has been taken for
granted in all our previous chapters, and this is his tend¬
ency to change or be modified within limits by his sur¬
roundings. It is with this aspect of man’s original nature
that all the rest of the book will deal. It is at once the
most common-place fact about human nature and at the
same time the most important, surprising and significant
fact. If this tendency of man’s conduct to be modified
or changed did not exist, there would be no education
and no such study as educational psychology. By original
nature man is a modifiable organism, his inherited tend¬
encies set the bounds within which this modification
shall take place and his immediate environment deter¬
mines the type of modification that will be made. If a
child is of normal intelligence he will learn to speak Eng¬
lish if born in England, Spanish if born in Spain, and
Chinese if born in China; but if he inherits the mentality
of an idiot, he will learn no language, and the three differ¬
ent language environments we have mentioned would not
cause this change or modification in his behavior. Some

185

186

THE LAWS OF LEARNING

kinds of environment will affect or modify some indi¬
viduals, while other kinds of environment are needed to
modify other individuals. There is thus a continual
interplay between the individual and his environment.
What type of environment shall be effective will be de¬
termined by the original nature of the individual, and
what specific modifications are to be made, granted an
effective environment, will be determined by the en¬
vironment. In what follows we shall assume the limita¬
tions imposed by original nature and emphasize merely
the effect of the environment in causing changes.

Life is a never-ending series of situations confronting
the individual, a continuous series of stimuli. To this
never-ending stream the individual is continually react¬
ing and each reaction that the individual makes leaves
its impression upon the individual. It changes or modifies
him even if ever so slightly. An “experienced” man is a
man who has had lots of “ experiences,” a man who has
reacted to many different stimuli. If you are experienced
in a certain field, you have made a lot of different reac¬
tions, and you have been changed by those reactions,
otherwise you would be just as green as the veriest “ green¬
horn” in that particular field.

Necessity for Reaction. — Life is reaction. If there
were no reactions, the individual would be dead. A con¬
tinual series of reactions is necessary in order to maintain
one’s equilibrium in the face of the ever-shifting environ¬
ment. When the situation changes suddenly the organism
is called upon to make a number of sudden reactions to
maintain its equilibrium. This is obviously true physi¬
cally, as, when one slips with his foot, a number of quick
physical adjustments may be necessary to maintain one’s
upright posture. It is just as true mentally for a situa¬
tion may arouse painful feelings or ideas and we react
vigorously in all directions in order to remove the dis¬
comfort and regain our equilibrium.

TRIAL AND ERROR

187

We keep reacting, therefore, in order to maintain a cer¬
tain equilibrium or satisfying state of affairs. So long as
a situation is annoying to us we are spurred on to repeated
reactions in order to remove the annoyance and return to
our former state of satisfaction. Satisfaction and annoy¬
ance, therefore, are the great spurs to action. They are
the supreme teachers. Unless we can remove the annoy¬
ance and attain a state satisfying to us, we have not
learned our lesson. When we attain a satisfying state,
we have learned how to meet a given situation and we
try to repeat this reaction again on the recurrence of the
same situation.

Tendency to Seek Satisfaction. — This tendency to re¬
turn to a certain equilibrium (Raup, 1) is common not
only to animals and plants, but to the inorganic world as
well. The equilibrium satisfying to any particular organ¬
ism will be determined by the inherited constitution of
that organism. In this way, then, we have the mechanism
set up for the ceaseless reactions of man and at the same
time the explanation why he will react again and again
to the same situation, varying his responses from one to
another until he finds one that allows a complete or
partial achievement of a satisfying equilibrium or until
the situation shifts again and he is called upon to react
in another way. It is in this repetition of reactions, this
multiple response to the same situation that opportuni¬
ties for modification occur. We are thrown into a state
of annoyance, we try to adjust, we are partially success¬
ful and then the situation changes. We think over what
has happened, trying out other possibilities in our imag¬
ination. The situation occurs again, we try these other
possibilities and are more successful and so on. Thus we
have a gradual learning, a trial and error in actual be¬
havior, a trial and error in the realm of ideas.

Trial and Error. — The occurrence of a situation, a
reaction to it followed by feelings of annoyance, another

188

THE LAWS OF LEARNING

slightly different reaction and so on until success or
partial success accompanied by satisfaction results, this
in general is the bare outline of the process of learning.
Thus in learning a complex physical activity such as
swimming, when confronted with the situation of stand¬
ing in the swimming pool, swimming instructor giving
directions and so on, the individual makes certain bodily
movements, feels himself sinking and this causes great
annoyance. He reacts vigorously by somewhat different
movements, if these are partially successful he will tend
to repeat them or similar ones the next time he is con¬
fronted with the “sinking” situation. If the particular
reactions are unsuccessful and he continues to sink, he
tends not to repeat those reactions when again con¬
fronted with the “sinking” situation. Thus gradually do
certain reactions tend to become connected with certain
situations. It is a gradual trial and error process until
the individual learns a reaction that is satisfying to him.

This trial and error procedure is not only true of the
learning of physical acts but is equally true of all kinds
of learning. Thus in learning to recite a poem we may
read it over several times and then try to repeat it from
memory. We come to the end of the first stanza and
try to begin the second. We say a few words and then
stop. It does not sound right. We are not satisfied. We
try again. Perhaps this time the words sound right and
we are satisfied and so proceed, or else we are not satisfied
and try some other words or give up and go back to the
book in order to get the right words and start all over
again.

So in playing chess we find a continuous mental trial
and error. To a given situation on the chess board we
react by imagining what would happen if we made a cer¬
tain move. We notice that if we made this move our
opponent might make either of three or four moves. We
follow this on by imagining or trying out in our mind a

TRIAL AND ERROR

189

counter move to each one of the three or four possibilities,
and so on, until we are unable to envisage the changed
situation. If we arrive at an imagined position of the
board satisfactory to us, we make the move in question,
if not we start out again on another imaginary move.
We try out mentally many moves until we reach one that
appears most satisfying or least annoying.

The baby says, “ta, ma, da,” and the mother shouts in
glee “mamma” and laughs and cuddles it, and the baby
has decided feelings of satisfaction. It may repeat the
sounds, “ta, ma, da” or “ta, ma, ma” and so on, be¬
cause they are a satisfactory response to a certain situa¬
tion. As time goes on more satisfaction (attention of the
mother) results when “ta, ma, ma” is said than when “ta,
ma, da” is said, and so “ta, ma, ma” becomes selected
as preferable in a given situation and gradually “mamma”
is found to be the most satisfying response to certain
types of situations. The baby will quickly learn to say
those things which lead to satisfyingness. If parents re¬
tain their great enthusiasm for the “cute” baby talk of
the early stages, then this baby talk may last for years,
and occasionally we meet a six or seven-year-old who
still retains many baby words and baby expressions, be¬
cause those around him have caused satisfaction to be felt
by him in such responses and they are therefore retained.

In all improvement due to such trial and error reac¬
tion, the individual stops at a level which is satisfying to
him. The mere repetition of a response will not cause
improvement in that response. Improvement will only-
come if the result is annoying to the individual. As long
as the response is satisfying to the individual, there is no
motive to change the response. Thus the baby will go on
talking baby talk, so long as such talk is approved by
those around him. He does not imitate the speech of
those around him unless he is taught to do so — taught
by the universal laws of satisfaction and annoyance.

190

THE LAWS OF LEARNING

Mere repetition does not lead to improvement. This
is well illustrated by the old story of Johnny who was
kept in after school by his teacher to improve his English
usage. Johnny had written “I have went” in a little
composition and so his task was to write out “I have gone,
I have gone” one hundred times. One hundred repeti¬
tions would surely stamp in the right reaction. So Johnny
wrote and when he had finished the hundredth repetition,
he looked up and found that the teacher had left the room.
Feeling that he had done his task and had a right to go,
he went up to the teacher’s desk and laid his work on it.
Then thinking that it would be more polite to explain his
going, he wrote a little note, “ Dear Teacher, I have done
my lesson and seeing you were not here, I have went.”

The Laws of Learning. — We are now ready to sum¬
marize our discussion into the two laws of learning: —

The Law of Exercise. — Whenever a modifiable con¬
nection between situation and response is exercised, the
strength of that connection is increased.

The Law of Effect. — Modifiable connections between
situation and response, causing satisfaction, are strength¬
ened, causing annoyance, are weakened.

The first law is sometimes called the Law of Use and
Disuse. The fact of disuse is, of course, implied in the
statement above of the Law of Exercise. Exercise of a
function strengthens it, disuse or lack of exercise weakens
it. The Law of Exercise (Use and Disuse) is, however,
impotent without the motive force of the Law of Effect.
Reactions which are accompanied or followed by satis¬
fying states of affairs are learned quickly or tend to be
repeated. Reactions which are accompanied or followed
by annoying states of affairs tend not to be learned and
tend not to be repeated. We are satisfied when we have
achieved our purpose or aim, we are annoyed when we
fail to do so. To reach for food on the table, seize it and
eat it is satisfying; to reach for food and be unable to

READINESS

191

seize it is annoying and will lead to another and modified
reaction on the part of the individual.

Annoyance resulting from a given response may be so
great that the individual is baffled and enraged. He
feels that it is impossible to proceed and so he turns away
from the given situation. He gives up. He refuses to
react. This is what happens when things are too difficult
for us. We are trying to solve a problem but reaction
after reaction takes place with no solution so that even¬
tually we acknowledge defeat and give up. We turn
away to make other easier reactions that are satisfying to
us. In the process of thinking, success is the great stimu¬
lant to the next step in the procedure. If stage one is
successful we go on gleefully to stage two and so no.
“ Nothing succeeds like success.” The glow of satisfaction
in achieving one step on the way to our final goal spurs us
on to take the next step. Problems set before the child
must not be too difficult. He must be able to gain suc¬
cesses along the road to spur him on to continue to the end.

Readiness. — We have stressed the fact that a reaction
must be satisfying to the individual, and it is satisfying
only when his purpose has been achieved. At one time
a given situation will provoke a certain response and at
another time a different response and both may be satis¬
fying to the same individual. What is one’s purpose or
aim at any given moment, what will cause satisfaction
and what annoyance, depends upon the state of readiness
of the individual to react. That an individual has an aim
or purpose, means simply that he is ready to react in a
given direction, that there are certain kinds of stimuli or
parts of stimuli to which he will react more readily than
to others, and with which he will be more satisfied. This
is sometimes called the “Law of Readiness.” It empha¬
sizes the condition or state of the reacting individual.
The individual is not the same from moment to moment.
The total bodily condition changes and the state of the

192

THE LAWS OF LEARNING

organism at any given moment is a state of readiness for
certain stimuli and unreadiness for others. To respond
to the stimuli for which the organism is ready is satisfy¬
ing, not to respond is annoying. Suspense or expectation
when long delayed is annoying, for the organism is ready
to respond to certain kinds of stimuli which do not appear.
The state of readiness at any given moment is determined
by the whole bodily condition of the organism plus the
reactions which have immediately gone before. To be
forced to change suddenly from one thing to another is
generally annoying because the organism is not ready for
the sudden shift.

One very common cause of annoyance, leading to angry
outbursts out of all proportion to the seriousness of the
affair, occurs very frequently in automobile driving. The
unanticipated movements of other cars cause decided
annoyance and flashes of anger arise quickly. We always
feel, of course, that we are in the right and that it is the
other driver’s fault. But a calm review of the situation
reveals that frequently the anger is there although we are
in the wrong. The acute feeling of annoyance or anger
is undoubtedly due in many cases to the sudden necessity
for changing our response. We are ready or set to con¬
tinue reacting in a certain way, driving straight ahead,
and when the need for sudden change arises, a feeling of
annoyance springs up. Note that on a crowded street
we are ready or set for frequent sudden changes, and we
do not have the feelings of annoyance as frequently as on
less crowded thoroughfares.

What reaction will follow the presentation of a given
stimulus cannot be foretold, until we know the state of
■ readiness of the subjects. In the school room what reac-

tion will follow the stimulus 2? It might be 5 or 1 or 6,

all depending upon the preceding work of the class. If
the lesson is one in addition, the subjects will add, if

BASIS OF ALL IMPROVEMENT

193

subtraction then they will subtract and if multiplication
they will multiply.

This readiness of the organism to react in a given direc¬
tion is at the basis of those states of mind that we call
motives or desires or purposes. In the field of education,
it has been called the will to learn. Without this will to
learn, improvement is impossible. This is the same as
saying that if the organism is not ready to respond, to be
made to do so is annoying and, therefore, improvement
does not take place. If the child does not wish to learn his
multiplication (if his organism is not ready), and we force
him to repeat it twenty times, no improvement takes place.
If he wishes to learn (if his organism is ready), and he
repeats it twenty times, improvement will take place.

Basis of All Improvement. — The Laws of Readiness, Ef¬
fect and Exercise give us the basic facts as to improvement.
Readiness selects the parts of the environment to which
the organism is most likely to respond. Exercise stresses
the necessity for repeated reaction in order to achieve im¬
provement. Effect selects the reactions which will survive.

Translated into terms of the school, we may say that
readiness means the arousal of certain desires or aims in
the pupil, exercise stresses the need for practice and effect
means that satisfaction must be attached to the reactions
we desire to have the pupils make. If the pupils are
ready, if they want very badly to do or learn a given
thing, they will make numerous reactions and they will
feel great satisfaction at reaching a successful conclusion.
Hence the importance of motivation in all learning. The
motives may be of different kinds, desire to beat one’s
neighbor, to avoid blame or punishment, to please the
teacher or parent, to make or construct a certain thing,
to help in being the best class in the school, to help the
school or community. There must be some motive or
else readiness will not exist and if it does not satisfaction
will not result from the desired reaction.

194

THE LAWS OF LEARNING

A teacher, therefore, must see to it that the pupils are
in a state of readiness for the subject-matter to be learned,
that they receive sufficient exercise in reacting to it with
a clear knowledge of the results of their reactions as they
are learning, and lastly, that the effect of their reactions
be immediately felt by them in terms of satisfaction or
annoyance. Try to get the pupils interested in their
work. Keep records of their improvement. Correct the
examination papers as soon as possible, and tell the pupils
their marks. Praise liberally for good work. Be sparing
of blame or punishment. If necessary punish right after
delinquency so that the annoyance may be connected
with the faulty reaction in question.

Summary

1. By original nature man is a modifiable organism; he possesses
the tendency to be changed by his reactions.

2. All education rests upon this modifiability of the human
organism.

3. Life is a never-ending series of situations to which the indi¬
vidual is called upon to react.

4. We keep reacting in order to obtain satisfaction and avoid
annoyance.

5. We react again and again (trial and error) until we arrive at
a state satisfying to us.

6. Improvement does not take place by repetition alone, but
only by seeking satisfaction.

7. The Law of Readiness states that when the individual is
ready to react in a certain direction, to do so is satisfying and
not to do so is annoying.

8. The Law of Exercise states that whenever a modifiable con¬
nection between situation and response is exercised, the
strength of that connection is increased.

9. The Law of Effect states that those reactions accompanied
by or followed by a satisfying state of affairs tend to be re¬
peated; and conversely, those accompanied by or followed
by an annoying state of affairs tend to be inhibited.

REVIEW

195

Review

True-False Statements

As a review mark the following statements true or false:

1. Readiness is determined by all one’s original tendencies and
all one’s experience up to date.

2. If we wish to strengthen a connection it is sufficient that we
exercise it.

3. If a child is not interested in a subject, he will not be ready
to respond.

4. Satisfaction is not equivalent to pleasure, nor annoyance to
pain.

5. Ordinarily individuals do not differ very much in readiness
to respond.

6. Reactions followed by satisfaction tend to be repeated and
learned.

7. Painful stimuli are always annoying.

8. A definite purpose makes us ready to respond to certain
stimuli rather than to others.

9. Much drill in school is the only guarantee for improve¬
ment.

10. Annoyance should be attached to a response if we wish to
eliminate it.

11. One of the chief functions of a good teacher is to make the
child ready to respond to desirable stimuli.

12. Teaching is the arrangement of situations which leads to
desirable bonds and makes them satisfying.

13. Regardless of much exercise, the beauty of our hand¬
writing does not increase because we are satisfied with it
as it is.

14. Knowledge of success or failure helps the learner in his
learning.

15. The “set” or attitude of the learner helps to determine which
reactions will be satisfying.

16. Practice always makes perfect.

17. Complete the following sentence:

Reactions followed by- tend to be -; those

followed by-tend not to be-.

196

THE LAWS OF LEARNING

Advised Reading

Perrin, F. A. C. and Klein, D. B.: Psychology , New York, 1926,
Chapter V.

Reference

1. Raup, R. B.: Complacency , New York, 1925.

CHAPTER IX

THE LEARNING CURVE

Making a Learning Curve. — In the previous chapter
we discussed the laws governing improvement. In this
chapter we shall study the way in which improvement
actually takes place when we practice or repeat a reac¬
tion. The best and quickest way to understand what
takes place is for the student to construct a learning curve
for himself. The student is strongly recommended, there¬
fore, to pause here in his reading and to carry out the
following exercise.

Writing Alphabet Backwards. — Get several loose sheets
of paper and a watch with a second hand (or better still
a stop watch). You are to write the alphabet backwards
30 times without allowing yourself to make any mistake.
If you write down the wrong letter, cross it out and cor¬
rect it. Your score will be the time taken to finish the
alphabet backwards correctly. Use small letters and do
not join them one to the other.

Put 1 on the first line of your paper, denoting the first
trial. Note the time of starting on the first line of your
paper and immediately write the alphabet backwards
beginning with z and going back to a. As soon as you
have finished note the time and write the time taken in
seconds at the right hand side. Then read through the
alphabet forwards to be sure you have made no mistake.
If you have, penalize yourself by adding five seconds to
the time, but try not to make any mistakes, correcting
as you go along. This is perfectly possible. Fold the
paper over backwards so that you cannot see what you
have written. Continue in this way for thirty trials with¬
out stopping.

197

198

THE LEARNING CURVE

Making a Graph. — Now make several graphs of your
results as in Figure 40. Figure 40 is constructed from
Table 2, which gives the number of seconds for each in-

TABLE 2

Time in Seconds for Writing the Alphabet Backwards:

Thirty Trials

Trial No.

Time

Trial No.

Time

dividual trial. The numbers along the base line show the
trials from 1 to 30. The numbers along the vertical line
show the time in seconds from zero to 45. On your chart
make allowance for more than 45 seconds, if your first
trial took longer than 45 seconds. Notice how the curve
in Figure 40 drops very quickly at the start showing con¬
siderable improvement for the first eight trials. The
ninth trial is poorer than the eighth, but the tenth trial
shows improvement again. From this point on the curve
is very irregular. There is no consistent gain from one
trial to the next. There is, however, a general tendency
towards gain up to trial 27. The last three trials, 28, 29,
and 30, show no improvement over trial 27. We can see

MAKING A GRAPH

199

1 2 3 4 5 6 7 8 9 10 12 14 16 18 20 22 24 26 28 30

Trial No.

Figure 40. — Practice Curve. Writing the alphabet backwards.

Time of each trial.

1 2 3 4 5 6 7 8 9 10 1<1 12 13 14 T5

Average of-2 Trials

Figure 41. — Practice Curve as in figure 40; average of two trials.

very definitely that it is much harder to make improve¬
ment towards the end than at the beginning of the prac¬
tice. Would improvement continue after the 30th trial?
Very probably it would, if the subject of the experiment
had continued, assuming that he retained interest in the

200

THE LEARNING CURVE

work. We do not really know, but it is most likely. We
can safely guess, however, that improvement would be
more and more difficult to attain and the curve would
tend to flatten out so as to become parallel with the base
line.

Figure 42. — Practice Curve as in figure 40; average of three

trials.

Figure 43. — Practice Curve as in figure 40; average of five

trials.

MAKING A GRAPH

201

1 2 3 4 5 6 7 8 9 10 1112 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Trials

Figure 44. — Practice Curve. Writing the alphabet backwards.

Subject A.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Trials

Figure 45. — Practice Curve. Writing the alphabet backwards.

Subject B.

Figure 40 shows all the irregularities in improvement
from trial to trial, and it is important to study our results
in this form, so that we may be aware of the way in which
learning takes place,—an irregular process from point to
point, with many mistakes and many back-slidings. We

202

THE LEARNING CURVE

may, however, also wish to consider the improvement in
a more general fashion without paying so much attention
to each individual trial. For this purpose we may take
the average time of two, or of three or more trials and
plot a curve of these averages. Figures 41, 42 and 43
show such curves. We tend to iron out the irregularities
as we average a greater number of trials and we call such
curves smoothed curves.

Figure 46. — Practice Curve. Writing the alphabet backwards.

Subject C.

Smoothed Curves. — If we examine the curves we see
how the irregularities slowly disappear as we proceed from
Figure 40 to Figure 43. Figure 43 shows us the general
trend of improvement from the first set of five trials to
the second set and so on. Progress from one set to another
is consistently marked by diminished time. Figure 43
shows well the general trend of the whole practice experi-

INDIVIDUAL DIFFERENCES IN CURVES 203

ment, indicating marked progress during the early stages
with a diminishing rate of improvement thereafter. It
also suggests the possibility of further improvement, had
the practice continued, because the curve is by no means
parallel to the base line. The detailed curve showing
every trial gives us the best picture of what actually
takes place, the smoothed curve allows us to see more
clearly the general trend of the whole improvement.

Individual Differences in Curves. — If the student has
carried out the experiment as described in this chapter,
he will be able to compare his results
with the curves given here. There may
be quite a difference. Each individual
has his own method and rate of im¬
provement.

I have given this exercise in learning
to many of my classes in educational
psychology. Figures 44 to 49 show six
curves for the same exercise for six dif¬
ferent individuals. Figure 44 shows very
irregular improvement with a very de¬
cided set-back before the final improve¬
ment takes place. Figure 45 shows con¬
tinuous and steady improvement all the
way along, a great contrast to the
erratic improvement of the preceding
individual. Figure 46 shows a learner
who starts out very slowly and then
makes very great and very rapid gains.

This learner takes 250 seconds for the
first trial as compared with the 45 and 43 seconds fcr the
two preceding learners. Figure 46 also shows a very rapid
increase in ability from the beginning of the learning
down to trial 18 and from then on very little further im¬
provement. Figure 47 shows a learner who starts out
very poorly and who is also rather erratic. Trial number

Figure 47. —
Practice Curve.
Writing the al¬
phabet back¬
wards. Subject D.

204

THE LEARNING CURVE

11 is almost as bad as the first two trials. Figure 48
shows a learner who makes little improvement after the
nineteenth trial. Figure 49 shows very steady improve¬
ment all the way along.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Trials

Figure 48. — Practice Curve. Writing the alphabet backwards.

Subject E.

These curves are good examples of individual differ¬
ences in learning. They are curves of university students,
all of high intelligence and all mature people. Even among
a homogeneous group of this kind, we find marked in¬
dividual differences in learning.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Trials

Figure 49. — Practice Curve. Writing the alphabet backwards.

Subject F.

A COMPOSITE CURVE

205

Figure 50. — Composite Learning Curve. Writing the alphabet

backwards.

A Composite Curve. — We might ask the question as
to how a university class as a group improves in the func¬
tion under consideration. What is the general learning
ability of the group? Figure 50 and Table 3 give the facts
for a class of 88 students. The curves show the median,
the upper and the lower quartile performances for each
of the thirty trials. The curves are much smoother than
those for any one individual. The table shows the tre¬
mendous range in individual ability from trial to trial.
At the very first trial we have individuals who are able
to write the alphabet backwards quicker than other
individuals at the last trial. The class as a whole, how-

206

THE LEARNING CURVE

ever, shows steady improvement from the beginning of
the practice to the end.

TABLE 3

Time in Seconds for Writing the Alphabet Backwards:
Medians and Quartiles for Each Trial for 88 Students

Trial

No.

Median

Trial

No.

Median

72.0

55.4

35.3

24.5

19.2

15.3

60.3

46.5

29.5

23.8

20.0

15.8

55.3

40.6

26.0

23.0

19.1

15.0

45.8

38.0

25.3

23.7

18.4

15.0

42.0

33.3

21.0

23.0

18.4

14.8

37.5

30.5

22.0

22.2

18.7

14.1

35.4

28.3

20.3

22.0

18.0

14.1

33.3

26.3

19.4

22.1

17.0

14.0

31.0

25.7

18.0

20.4

17.1

13.8

30.6

25.0

18.3

20.6

17.0

13.6

29.0

22.5

18.0

21.0

16.3

12.9

27.5

22.0

16.3

20.4

15.6

13.0

27.5

22.1

17.3

20.0

15.8

13.4

27.0

22.0

18.0

18.8

15.5

13.1

24.5

20.5

16.3

17.6

15.1

12.7

Curves of Different Functions. — So far we have been
discussing the practice curve of a certain specific function,
namely learning to write the alphabet backwards, and we
have learned something of the general nature of practice
curves, but we must not suppose that all practice takes
place exactly like this. Some things are learned more
slowly and laboriously than others. It would be well,
therefore, to examine other practice curves in order to
get some idea of the different shapes they may assume.

The curves shown in Figures 51 to 55 have all been
made by students of mine as class exercises. The students
themselves chose what things they would practice. Fig¬
ure 51 shows the decrease in errors from day to day made

CURVES OF DIFFERENT FUNCTIONS 207

by a student in using the touch system on the typewriter
when copying a given list of words. The errors decrease
rapidly from 16 to 4, but for the last five days the number
of errors remains about 4 or 5. Evidently most of the im¬
provement has been made
during the first five days’
practice. During the last
five days the student
makes little or no gain.

This cannot be a so-called
physiological limit, be¬
cause it is obviously per¬
fectly possible for the
student to learn to type
the list of words without
error, and he would have
been able to do this if he
had continued at the task.

He has merely reached a
place in his practice where
for some reason or other
little improvement takes
place. This makes the
curve appear flat and this
flat part is frequently
called a plateau. Plateaus
or periods of no improve¬
ment during the exercise of
a function are sometimes found. They do not always oc¬
cur, however. There are probably many reasons for their
occurrence. We all know what it means to “ go stale ” at a
task. We do not seem to make any progress, and we
become discouraged. We are on a plateau in our learning.
Sometimes our “ staleness ” is caused by sheer physical fa¬
tigue. Sometimes we lose interest and we don’t try so hard.
If we don’t want to improve we don’t (Law of Effect).

6
7

2 8
o

£ g

10
11
12

0123456 7 89 10

Days

Figure 51. — Practice Curve for
typewriting, showing the number of
errors in typing a list of 35 words by
the touch system: one trial a day
for ten days.

208

THE LEARNING CURVE

Sometimes, in spite of all our trying, we still make no
improvement, and then the reason probably is that we
have not yet formed certain habits which are necessary
before we can go on to more difficult tasks. Certain
bonds or connections have to be formed before we can go
on to more difficult bonds or connections. Until these
bonds or connections have been well exercised the learner

makes little or no im¬
provement. When
they are well formed
another spurt of im-
provement takes
place, so that if our
student had continued
with the task shown
in Figure 51, the prob¬
ability is that he
would have overcome
the difficulties hold¬
ing him back and
eventually have
reached a place where
no errors were made.

Figure 52 shows the
curve for speed in
typewriting for the
same student. He in-

Figtjre 52. Practice Curve for type- cre ases in speed fairly
writing as in figure 51, showing the time . . „

in minutes for each trial. regularly for the ten

days of practice. No¬
tice that this student has drawn his curves differently
from all the curves we have so far discussed for writing
the alphabet backwards. The curves in Figures 51 and
52 go up to show gain in improvement.

Figure 53 shows the improvement in cancelling the
letters B, L, V and 0 in a closely written magazine article.

CURVES OF DIFFERENT FUNCTIONS 209

630

620

610

600

590

580

570

= 560

The student worked for ten minutes each time for ten
days. The curve shows the number of letters crossed out
in each day’s work. During the first four days improve-
m e n t was relatively
slow. Evidently certain
habits or adjustments
were being made, such
as ability to keep the
four letters clearly in
mind. On days five
and six very rapid im¬
provement takes place.

Something has been
learned that facilitates

the work enormously. "

From then on to the end % 540
of the experiment, im- ^ 53 °
provement is marked, J 520
but irregular.

Figure 54 shows im¬
provement in substitut¬
ing numbers for letters.

Each letter of the al¬
phabet is given a num¬
ber and then, with the
key before him the stu¬
dent writes the numbers Figure 53. — Practice in cancelling

for the letters of a given the letters B, L, V, O in a magazine

c mi article: one trial of ten minutes per

passage of prose. The day fo J ten days .

curve shows each five

minute practice period, four such periods each day for six
days. The number of letters written increases during the
course of practice. The curve is very irregular, because it
shows each five minute period. If, however, we plot the
average number of substitutions for each day’s work we
have the smoothed curve superimposed on the irregular

510

500

490

480

470

460

450

5 6

Trials

210

THE LEARNING CURVE

curve, and this shows an almost steady and equal im¬
provement from day to day. Here there is no sign of a
plateau, no sign of going stale or losing interest or having
to form certain habits before other and more complicated
habits can be formed. The march of improvement is
steady and continuous.

Figure 54. — Substitution of numbers for letters: detailed and

smoothed curves.

Figure 55 shows the scores made in a certain balancing
game, a toy. The object of this game was to manipulate
a board with a ball on it in such a way that the ball would
not fall into certain holes on the board. Scores were
allotted for each hole, increasing in value for the difficulty
of the hole. Each point on the curve represents the total
for twenty trials per day. Improvement is shown from

CURVES OF DIFFERENT FUNCTIONS 211

2000

1.900

1800

1700

1.600

1500

1400

1300

1200

1100

g 1000
co

900
800
700
600
500
400
300
200
100

1 2 3 4 5 6 7 8 9 10

Day

Figure 55. — Improvement in a balancing game.

the first to the second day, but from then on to day ten
no distinct improvement takes place. Here, then, is a
curve very different from most of those we have previously
discussed. The learners maintained that they were equally
interested during the whole period of practice, so that lack
of improvement could mot be due to lack of interest. It
would, therefore, seem as if the function was very easy
and that a physiological limit was soon reached, varia¬
tions in score from day to day being thereafter due to
chance. We have no proof that a physiological limit has
been reached. To make certain of this a great deal more
practice would be necessary.

All these curves were made by students in classes in
educational psychology. The reader should plot a few
curves for himself for different functions. If you are a
member of a class in educational psychology, you can

212

THE LEARNING CURVE

) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

DAYS

Figure 56. — Learning Curves of two subjects. (From Foster’s

Experiments in Psychology.)

obtain a great many curves for different functions by
having each member plot a curve for a different function.
In this way you can compare different functions, as well
as individual differences in the same function if you have
carried out the exercise described at the beginning of this
chapter.

We shall now give a few curves selected from other
writers’ works. Figure 56 shows the learning curves for
two subjects. The general tendency of the two curves is

CURVES OF DIFFERENT FUNCTIONS 213

very much the same. Figure 57 shows a curve for learn¬
ing typewriting. It is very irregular, but there is a gradual
trend upward, showing steady improvement. Figure 58
shows the progress of a blind child, thirteen years old, in
learning to read Braille. The points on the curve do not
show T daily tests, but tests given now and again over a
period of about four months. The general trend is just
about the same as in the learning of a very narrow func¬
tion such as writing the alphabet backwards done all on
one day. Figure 59 is particularly interesting. It shows
composite curves for two groups of children, one normal
group and one defective group, of equal mental age. The
learning progress for the two groups is substantially the
same.

0 50 100 150 200

Courtesy of the Journal of Applied Psychology

Figure 57. — Learning Typewriting. The vertical axis represents the
score in a five-minute test in typewriting. The horizontal axis represents
the number of hours of practice. (From Chapman, Journal of Applied Psy¬
chology , 1919.)

214

THE LEARNING CURVE

Reading'

rlirwiC' ISyrs- &B

Figure 58. — Progress of a blind child in reading Braille. (From
Maxfield’s Present Status of Instruction in Primary Reading in Resi¬
dential and Day School Classes for the Blind.)

Motivation. — Drawing a curve of something we are
practicing shows us at any moment our proficiency in
that function. It is much more accurate than relying
upon our own feeling of how good we are. It is like keep¬
ing a record. Keeping records of one’s proficiency is very
common in a great many games, and indeed most games
depend upon the record in order to tell who won. With
many games it is possible to keep a record of one’s own
score in order to see whether one is improving or not, and
trying to beat one’s own record becomes just as exciting
as beating one’s opponent. The golfer is just as anxious
to lower his record as to beat his opponent.

MOTIVATION

215

Figure 59. — Learning Curves of Normal and Defective Children.
(From Woodrow, Journal of Educational Psychology .)

One reason why keeping records in intellectual tasks is
not as common as in physical tasks is probably that it is
not as easy to measure the former as it is the latter. Now
that we have measuring devices for all the school subjects,
many teachers are using them to measure the progress of
their pupils. Children should be taught how to keep such
records of their own progress and if this is not overdone
they will find a great incentive to improvement by keep¬
ing such records. They like the actual making of the
curves themselves and they experience satisfaction as the
curve ascends, and satisfaction as we have already learned
is the potent factor in the law of effect. It helps to
strengthen the right connections. Standard tests of
arithmetic, writing or reading may be given every week
or every month and the progress of each individual re-

216

THE LEARNING CURVE

corded by himself. The average of the class as a whole
should be charted by the teacher and this class chart may
well serve as an instrument for keeping up the general
morale of the class. It is well not to do this with too many
subjects at the same time. Concentration upon one sub¬
ject at a time is probably best and a reasonable goal of
attainment should be kept before the class. When this
goal is attained in any one subject, a change to another
subject is desirable. If the class as a whole or any
individual child reaches a plateau long before the expected
goal has been achieved, it should be the function of the
teacher to find out why progress has been arrested, and
to endeavor to remove the disturbing causes if possible.

Figure 60. — Motivation in Penmanship. (From Paulu’s Diag¬
nostic Testing and Remedial Teaching.)

CHARACTERISTICS OF LEARNING CURVES 217

Plateaus are seldom necessary and are most frequently
caused by lack of interest, or a decrease in desire for im¬
provement. A plateau in general should be a warning to
the teacher that something has gone wrong with the
learning ability of the class. Figure 60 shows the individ¬
ual record kept by the pupil for the purpose of motivating
his work. This chart has been used for handwriting. The
line marked “what I should do” is the standard for the
grade in question.

Figure 61 shows a class record. The continuous line
represents the average monthly score for a fourth grade
class in the Starch reading tests. The broken line rep¬
resents the standard scores. This class is above standard
and gains steadily during the school year.

iSept. Oct. Nov. Dec. Jan. Feb. March April May>

Figure 61. — Monthly progress in reading. The dotted line rep¬
resents standard scores. (From Starch’s Educational Psychology.)

Characteristics of Learning Curves. — Are there any
characteristics which are common to all learning curves?
In general the answer to this question is “no.” Apart
from the fact that all show an increase in the ability to
perform the function in question and that all would, if
indefinitely continued, reach a limit of improvement
sometime or other, there are no constant characteristics
to be met with in all curves. Learning curves may be of
all kinds, some showing rapid initial improvement and

218

THE LEARNING CURVE

others slow initial improvement; some show steady per¬
sistent gains, others irregular fluctuation from point to
point; some have very definite plateaus, others show no
indication of a plateau. Such differences depend both
upon the function practiced and upon the individual
practicing. It is therefore useless and undesirable to
talk of a typical learning curve.

Initial Spurt. — Some curves show what has been called
an initial spurt which means a period of relatively great
improvement appearing at the beginning of the practice,
but many curves do not show this. If it occurs, it is
probably due either to the ease of forming certain initial
habits, or else to the greater interest and enthusiasm of
the learner at the beginning of the learning.

End Spurt. — Some curves show what has been called
an end spurt, which means a period of relatively great
improvement at the end of the learning. This can occur
only if the learner knows that the end of the learning is
approaching. Sometimes under such circumstances the
learner makes a last vigorous effort to better his per¬
formance, and this is registered by a decided rise of the
practice curve. Initial spurt, end spurt, and plateau are
all useful terms to describe certain characteristics of some
learning curves, but we must be careful not to imagine
that they always occur in all curves.

Physiological Limit. — A physiological limit would
occur in all curves if they were indefinitely continued.
By a physiological limit is meant that stage of profi¬
ciency which cannot be surpassed owing to the physio¬
logical limitations of the organism itself. True physio¬
logical limits are probably rarely attained by anyone in
any function. We do not go on practicing things until
our physiological limit is reached. To do so would mean
very persistent and continuous work. We have so many
other things to learn in life that we have not time to
devote to reaching our physiological limit in any par-

PHYSIOLOGICAL LIMIT

219

ticular function. And more important than this, we
lack any incentive to further improvement in most things,
because in the thousands of daily habits we have reached
a level of proficiency with which we are satisfied, and so
long as we are satisfied with certain responses we con¬
tinue to repeat them when the appropriate stimuli recur.
I am satisfied with my present proficiency in writing. It
is good enough. It serves its purpose. There is, no
doubt, great room for improvement both in speed and
quality, but so long as I feel content with it as it is, no
improvement will take place, regardless of the number of
times I exercise the function. I have not attained any
physiological limit in writing, because I could easily set
myself the task of improving and if I thought this worth
while, my writing would immediately improve in speed
and quality. And so it is with rate of reading, ability to
add, subtract, multiply and divide, with my efficiency
in dressing, in using a lawn mower, in tending the fur¬
nace, in running a motor car, in skimming over a news¬
paper, in typing, in reading a novel, in peeling an apple,
and all the other thousands of habits which I have formed
and brought up to a certain degree of proficiency satis¬
factory to me at the present time. I do these things well
enough for my purposes, but certainly not as well as I
could, were I to concentrate on any one of them. None
of them is near my physiological limit.

Physiological limits are probably reached or nearly
reached in some functions by some few people who set
themselves very deliberately the task of excelling in some
particular thing. People who compete for champion¬
ships of various sorts are probably near their physio¬
logical limits, as well as other people striving hard to
excel in tasks for which no records are kept. The cham¬
pionship records for running and other physical exercises,
the records for typing and chess and so on represent the
approximate physiological limits for these functions for

220

THE LEARNING CURVE

the human race. We say approximate limits, because we
see the present records every now and then surpassed.
With the ordinary school tasks, however, the teacher may
feel pretty sure that children are nowhere near their
physiological limits. Furthermore, the teacher has no
real concern with physiological limits, because his task
is to bring his pupils to a reasonable degree of proficiency
rather than to strive for a limit.

Summary

1. A learning curve shows the rate and amount of improvement
in the function measured.

2. Even in such a simple function as writing the alphabet
backwards every learner shows marked improvement in a
few trials.

3. Curves may be drawn to show all the details of improve¬
ment or smoothed curves may be made to show the general
trend of improvement.

4. There are very decided individual differences in the learning
curves of university students practicing the same task.

5. A composite curve showing the average performance of each
trial for a group of individuals may be drawn. This shows
the general trend of improvement for the group as a whole.

6. Samples of curves of different functions show that learning
curves differ widely from function to function as well as
from individual to individual.

7. Learning curves may also be constructed to show the gen¬
eral improvement in any function over a long period of
time by taking sample measures of this function once a
week or once a month.

8. The learning curves for defective and normal children of
the same mental age seem to be very similar.

9. Keeping a record of one’s proficiency in a subject is an ex¬
cellent method for sustaining one’s interest in the work.

10. Individual and class curves of school work may be used to
great advantage as a means for motivation in school.

11. There are no general characteristics common to all learning
curves.

SUMMARY

221

12. Most learning curves are irregular, showing that improve¬
ment takes place unevenly.

13. A plateau is a period of little or no improvement, followed
by improvement later on.

14. Plateaus may be caused by lack of interest or incentive,
“going stale,” fatigue or the necessity for the formation of
simpler habits before more complicated habits can be formed.

15. Plateaus are not commonly found in learning curves.

16. A plateau should be a warning to the teacher that some¬
thing has gone wrong with the learning ability of the class.

17. Initial spurt and “warming up” are names applied to the
preliminary improvement found in some curves.

18. End spurt is a name applied to a sudden increase in improve¬
ment found at the end of some learning curves.

19. End spurts do not occur unless the end is known by the
learner himself to be near.

20. A physiological limit is the highest point on the curve that
the learner can reach.

21. Few of us ever achieve a physiological limit in any function.

222

THE LEARNING CURVE

Review

True-False Statements

As a review mark the following statements true or false:

1. Learning curves always show regular and constant accelera¬
tion from one trial to the next.

2. Most subjects in learning to write the alphabet backwards
arrive at a plateau at about the seventh or eighth trial.

3. A rapid initial rise in a learning curve is universal.

4. A rapid initial rise is sometimes called an initial spurt.

5. There is no one typical curve of learning.

6. Most of us remain on plateaus in regard to the greater num¬
ber of our daily habits.

7. If a child reaches a plateau on his learning curve for addi¬
tion it is an indication to his teacher that he is becoming
lazy.

8. To have children keep records of their improvement in dif¬
ferent subjects is a very poor incentive for them to strive
to do better.

9. University students practicing the same task are likely to
show very similar learning curves.

10. The most common characteristic of learning curves is their

irregularity.

11. Almost every learning curve shows a plateau.

12. When a curve suddenly accelerates for the last two or three
trials, we call this acceleration an end spurt.

13. The learning curve differs according to the person who does
the learning.

14. Plateaus are more likely to occur in learning complex tasks
rather than in learning simple tasks.

15. The plateau in a learning curve indicates a period of im¬
provement.

16. A great deal of our habitual activity is not as efficient as it
could be, because we are satisfied with it as it is.

17. The learning curve gives a graphic picture of the amount,
rate, and limit of improvement.

18. We soon reach our limits of efficiency in most tasks because
our organism is so unmodifiable.

ADVISED READING

223

19. Underline the best response:

A long period of no improvement followed by further im¬
provement is called: (1) a plateau; (2) adaptation; (3)
initial spurt; (4) warming up.

20. Check the best response:

We do not reach our physiological limit on most tasks, be¬
cause :

(a) of our lack of intelligence.

(b) our learning tends to decline.

(d) of lack of sufficient practice.

Advised Reading

Thorndike, E. L.: Educational Psychology: Briefer Course ,
Teachers College, Columbia University, New York, 1914,
Chapters 14 and 16.

CHAPTER X

EFFICIENT LEARNING

Effective Ways of Learning. — So far we have discussed
the general laws of learning and the general characteris¬
tics of improvement in simple functions. We may now
ask the question as to how such knowledge may help us
to learn more effectively. In this chapter we shall dis¬
cuss some of the consequences of the general principles
of learning and in addition add what knowledge we pos¬
sess as to effective learning gathered from experimental
work in this field. There are thousands of problems with
respect to the best means of learning and many specific
points have not yet been definitely settled by experiment.
Nevertheless, there is a growing mass of knowledge re¬
sulting from reliable experiment in the learning process,
and we will discuss here the most useful facts for the
teacher and learner.

The Learning of Arithmetic. — One of the most effec¬
tive methods of approach is to ask what help educational
psychology can give to the learning of a particular sub¬
ject. Thorndike does this very effectively in his “ Psy¬
chology of Arithmetic.” The problem with reference to
the learning of any school subject is, he says, “the de¬
velopment of a hierarchy of intellectual habits.” What
we have to do is to build up a great many habits. Many
desirable habits are very complex. We cannot form
them right away. We need other simpler habits first in
order that the more complex ones shall be able to func¬
tion. For instance, we want our children to learn to cal¬
culate interest on money saved. We cannot teach them
this right away, because it involves knowing how to add,

224

THE LEARNING OF ARITHMETIC 225

multiply, and so on. Hence the necessity for forming
simpler habits first. That is, as Thorndike says, we
must build up a series or hierarchy of habits. In order
to do this, three important considerations emerge, ac¬
cording to Thorndike:

1. The choice of the habits to be formed.

2. The best order in which to form them.

3. The best means for forming each in that order.

The first point means that we must go over all the
details of a subject and seriously consider just what we
want the child to learn. And, further, we must decide
what habits he must form in such learning. It is easy
to say that we want the child to be able to add 2 and 3
together. We must go further and decide whether on

presentation of the stimulus 2+3 or 3 we want the child

to respond by saying to himself “2 and 1 are 3 and 1
are 4 and 1 are 5,” or “two, three, four, five” or “two
and three are five” or simply “five.” Obviously the
last response is the most economical, namely to think
“five” on presentation of the stimulus 2+3. It is in
this detailed sense that we must determine what par¬
ticular habits should be formed.

When we have made such decisions, we must then find
the best order in which to form these habits. Thorn¬
dike’s rules for arithmetical bonds or habits are as
follows:

1 . Other things being equal, one new set of bonds
should not be started until the previous set is fairly es¬
tablished, and two different sets should not be started
at once.

As an example, first learn multiplication without carry¬
ing, then with carrying and no zero difficulties, then
introduce the zero. Or in teaching punctuation, teach
the use of periods and commas first, then the use of ques-

226

EFFICIENT LEARNING

tion and quotation marks and finally the use of colons
and semi-colons.

2 . Other things being equal, bonds should be formed
in such order that none will have to be broken later.

It is probably better to teach long division with re¬
mainders from the very start, rather than to teach it
without remainders first. The child will then not form
the bond of “coming out even,” which must later be
broken. In reading, diacritical marks are of doubtful
value because bonds are formed, which later on have to
be broken. Some teachers of public speaking believe
that students should be taught right away to speak with¬
out notes. If they learn to speak first with notes, they
find it difficult to break this habit later on.

3. Other things being equal, arrange to have variety.

Do not work on addition until all the necessary bonds

have been formed. After some have been fairly well es¬
tablished, move on to subtraction, and then to multipli¬
cation.

4. Other things being equal, use objective aids to
verify an arithmetical process or inference after it is made,
as well as to provoke it.

After a pupil has responded to 2+3 by 5, he may check
his work by actually counting three objects and two ob¬
jects placed together in a pile. This will serve as a con¬
crete check upon what he has already done abstractly.

5. Other things being equal, reserve all explanations
of why a process must be right until the pupils can use
the process accurately, and have verified the fact that it
is right.

A child can profitably learn how to add, subtract, mul¬
tiply, and divide with United States money long before
he knows anything about decimals, or the reason for
placing decimal points where we do in such operations.
The important point is to be able to get the right answer
and be able to verify it. The reasons for the particular

LEARNING TO READ

227

method of procedure will be appreciated better after the
procedure has been learned.

6 . Arrange the order of bonds with due regard for the
aims of the other studies of the curriculum and the prac¬
tical needs of the pupil outside of school.

This is obviously a general rule to keep in mind with
all studies so that they may be linked together and func¬
tion in the life of the pupil.

These are some of the general principles which Thorn¬
dike uses in his investigation of the most effective way of
learning arithmetic. Everyone of these and other prin¬
ciples is illustrated by numerous examples. The student
should read the whole book as the best example of the
application of the psychology of learning to a particular
subject.

Learning to Read. — The investigations of Gates (1 and
2 ) in reading may be taken as another good example of
the application of psychology to a particular subject.
Gates started with a study of the reading disabilities of
pupils of all ages. In order to understand these reading
difficulties, he constructed many kinds of reading tests.
These tests help to diagnose the specific difficulties of
each pupil. Hence we have a battery of valuable tests
measuring the many specific abilities which go to make
up the complex act of reading.

This work naturally led this psychologist to raise the
question as to whether reading might not be so taught
from the beginning as to avoid the difficulties found in
so many pupils later in life. Hence he made an analysis
of the ways in which young children learn to read without
specific teaching. He found that in the recognition of
words, different children seized upon different charac¬
teristics in order to enable them to remember the word,
thus “monkey” was remembered by the monkey’s tail
on the y; or by the “hole,” that is the letter o or by the
“funny chair,” that is the letter k and so on. Further-

228

EFFICIENT LEARNING

more words were differentiated by means of their length
as “cow” and “dandelion.” In other words, children
left alone by themselves use all sorts of devices, that is
they form all kinds of habits. Now some of these habits
are useful and help greatly, but others are useless and
may retard the learning. Hence we need to analyze out
the most useful habits and teach those, and at the same
time not allow the useless and wasteful habits to be
formed.

In these investigations Gates found that many teach¬
ing devices were training the children in bad habits, that
later on would have to be discarded. Examples of such
wasteful devices are the use of flash cards and numerous
phonetic devices used in the so-called phonetic method of
teaching. All these devices he calls extrinsic, i.e. devices
that must be learned in addition to the reading proper.
Hence he discards them and in his proposed method of
teaching reading he uses the intrinsic device only. The
intrinsic device is a help and aid to learning some specific
aspect of reading, but at the same time is an integral
part of the whole reading process. It is not something
that has to be learned and cast away later on. As an
example of an intrinsic device to make children “observe
closely certain characteristic parts of words and grad¬
ually to become aware of certain elements common to
many different word-forms,” the pupil is given a drawing
with the following instructions:

1. Color the cat blue.

2. Color the hat black.

3. Color the coat brown.

and so on with many similar exercises.

This, of course, merely gives a bare outline of Gates’
work. It is, however, a good example of a teaching
method which has been arrived at by a psychological
analysis of the learning process.

LEARNING IS REACTING

229

Learning is Reacting. — Since all learning consists of
reacting to various stimuli, we should as far as possible
see to it that during our learning we are reacting in the
same way as we later wish to react, on the presentation
of a specific stimulus. This is so obvious in motor learn¬
ing as in learning acts of skill that we do not need to
elaborate. We learn to drive a motor car by driving one,
not by studying a diagram. Of course the diagram may
help in certain ways, but we should certainly never think
of studying all the moves on paper and then expect to
be able to get into a car and start out driving perfectly,
however perfect our “ paper ” driving may have been.
Obvious as this is in acts of skill, we frequently forget it
in other kinds of learning. We learn grammar in order
to write or speak correctly. But we do not find that
we write or speak correctly, whereas we do know more
grammar. If we want to write and speak correctly,
we must practice writing and speaking. Study your
French lesson “out loud” if you want to learn to speak
French. Keep writing poems if you want to learn to
write poetry.

If you are going to be examined on a subject and you
wish to do well in the examination, consider how the
questions will be asked and practice answering in the
same manner that you will be called upon to react. If it
is an oral examination frame possible questions and talk
them out aloud extempore so that you can judge for
yourself how they will sound to the examiner. If it is
an old-type written examination, write down answers to
possible questions in essay form. If it is a new-type
examination construct true-false, alternative answer and
completion items and after several days answer them
yourself. You will in all probability in this last case,
get all your answers right, because you will find that
the construction of objective items is an excellent form
of review.

230

EFFICIENT LEARNING

We learn exactly what we practice. A great deal of
the controversy over the value of different methods of
teaching arises from the fact that we frequently forget
this rule — that we learn exactly what we practice. Is
one method better than another? Better for what pur¬
pose, must be our first consideration before we can an¬
swer the question satisfactorily. Is the lecture or lab¬
oratory method better? If our aim is the acquisition of
a certain number of facts, the lecture method will prob¬
ably achieve this result more economically. If our aim
is to acquire a certain amount of skill in solving the prob¬
lems of a subject, the laboratory method will be most
effective. To become efficient mental testers the lab¬
oratory method of teaching is necessary, to acquire a
knowledge of the development of mental testing and its
present significance and value in education, the lecture
method would be quite adequate.

Models are a help in learning just because they help
to check up our responses and let us know how far off we
are from the desired response. But models should not be
so perfect, as to make it hopeless for the child ever to
attain them. This was the fault of the old copy-book
models in writing. They were too far beyond any possi¬
bility of attainment by the child.

Reaction of some sort is necessary in all learning.
“ React—react,” must be the watchword of all learners.
Passivity is waste of time. We do not learn by absorp¬
tion. Sitting quietly gazing at the book as so many chil¬
dren do is waste of time. Write out the spelling words or
the foreign words to be learned. If you do not know what
to write for your theme, get pencil and paper and write
anything. You may have to throw away a lot later, but
the very reactions of writing will probably set you on the
right road in time.

Avoid Irrelevant Reactions. — What has just been
said about reacting in the way we will need our informa-

AVOID IRRELEVANT REACTIONS 231

tion or skill later implies in a negative way the principle
that we should as far as possible avoid making reactions
which we shall not need later. Sometimes this cannot be
avoided, but so far as possible, it is a safe principle for
our guidance. Sometimes the easiest road seems to be
a round-about way, but it very rarely, if ever, is. The
teaching of such a complicated task as writing to young
children has led teachers to seek all kinds of aids in the
task. Many methods of preliminary practice before
writing are advocated, such as tracing letters in grooves,
tracing on transparent paper, running the finger over
sand paper letters, practicing parts of letters and so forth.
It has been demonstrated (Hertzberg, 3) that children
who are taught to write without any such preliminary
aids make faster progress than children who are taught
by means of them. Those who begin by tracing the letters
learn to trace and they have to forget certain habits
learned in tracing, when they come to writing without
any aid. All aids in learning are not to be condemned
outright, but whenever the required reaction can itself be
directly learned from the very beginning, the chances
are that it is more economical in the long run to begin
with it.

In the teaching of arithmetic we find many examples
of irrelevant reactions. Adding by counting on the fingers
is a gross example, which few teachers now-a-days would
teach, yet at the same time they may feel that the child
is aided by writing down the number to be carried. But
this writing down the number to be carried is an irrelevant
reaction which must later be got rid of and it is almost
as doubtful a teaching device as counting on the fingers.
These indirect methods of teaching, which are supposed
to help the child, are continually cropping up. Here is a
method of addition that I came across recently in a very
good school.

232

EFFICIENT LEARNING

ADD

5 6 9 4 3
8 2 16 9
3 2 15 0

6 6 5 8 6
2 5 9 8 3
15 8 16

2 3 5 6 1
14 8 19

3 7
4 2
5 0

3 8
3 1

3 1 8 0 2 7

It is of doubtful value, especially if children have once
been taught the ordinary method. Some children ar¬
rived at the answer 3 5 3 4 5 7, obtained by adding all
the totals, showing how other habits were functioning.
But even if there are no old habits to interfere, we must
remember that later we shall have to break down the
habit of writing down the full total for each column and
teach the child to “-carry” mentally. Present-day psy¬
chology would recommend that he be taught to carry
mentally from the very beginning of his learning to add,
so that no useless habits will have to be discarded later.

Historically a good example of the gradual abandon¬
ment of preliminary aids in favor of direct learning is
shown in reading. It used to be the thing to start learning
the A B C’s. What the child learned was to say A B C in
sequence and to recognize the letters. Then came da,
ma, pa, ta, fa, and countless syllables. If the child had
learned his previous lesson properly, he would respond
on seeing “da” by saying “dee-a.” So the poor child had
now to forget that A called for the response A, and D
for the response Dee, and learn that d and a together
were called da or day or something like that and so the

MNEMONIC DEVICES

233

drill in syllables would go on da, ma, pa, and me, be, te,
le, and so on. Then came real words and so when “date”
was presented the child would likely respond by “da-tee”
as he had been taught in syllables. So these bonds had
to be all broken again and much that had been painfully
learned had to be forgotten. To-day we find it far more
economical to begin with real words or phrases or sen¬
tences and form direct bonds with these, bonds which
will not have to be broken later.

Mnemonic Devices. — The whole question of the value
of mnemonic devices in learning can be best solved from
this general standpoint. If the device helps in forming a
real link between the material to be learned it is probably
helpful, but if the device calls for the formation of a lot
of useless bonds, which are in themselves hard to form,
such a device is probably very uneconomical.

My schooling dates from the time when it was con¬
sidered the proper thing to learn a great many facts in
such subjects as geography and history. My first contact
with a mnemonic device is still vividly remembered.
The teacher wrote on the blackboard without any pre¬
vious warning or explanation the following words one
under the other, each with a capital:

Reaching

Eton

Lunched

Granley

With

Grandpapa.

Now this procedure with this peculiar sentence with the
peculiar name “Granley” was all very strange and there¬
fore very interesting. It caught and held the attention
of the class. Then he told us that this sentence would

234

EFFICIENT LEARNING

help us to remember the towns on the River Thames in
their order from the source to the mouth, the first letter
of each word being the same as the first letter of the town.
And so the mnemonic device was completed in this fashion:

Oxford

Reaching

Reading

Eton

Windsor

Lunched

London

Granley

Greenwich

With

Woolwich

Grandpapa

Gravesend

He then told us the story, an incident in his own life,
about a trip on the Thames and a picnic and who Granley
was and so on. All of this I have forgotten, but the device
and its significance has not been forgotten, partly because
it was the first of this type I encountered and partly
because of its peculiarity and the circumstances surround¬
ing its first appearance.

This teacher had a great fondness for memory devices
and was ingenious in their construction. He gave us a
great number to help in memorizing geographical, his¬
torical and literary facts. I have now forgotten practically
all of them. A slight memory of some of the mnemonic
sentences remains, but these sentences are not connected
with any facts. For example, there was a sentence about
“little Gwendolyn” which was to help us remember the
names of the Welsh mountain peaks, but this is all that
I remember of the device or the mountains. After the
initial novelty of the device had worn off, what I was
learning was both the sentence and the facts. What
started out to be an aid, became shortly a fact to be mem¬
orized. Unless somehow or other the mnemonic device
is intrinsically connected with the facts themselves, it is
of doubtful value.

MNEMONIC DEVICES

235

Such an intrinsic connection is shown in this device
which I remember from my school days:

“ Peccavi, I’ve sinned,” said Lord Ellen so proud;

Dalhousie, more modest, said, “ Vovi, I’ve vowed.”

Which tells me that Lord Ellen (and I have a faint mem¬
ory that his real title was Ellensborough or something
like that) acquired the province of Sindh and Dalhousie
the province of Oude in India. Whatever the facts may
be worth to me now, the device has preserved them for
many years. Psychologically speaking the bonds re¬
quired to learn the rhyme are easier than those required
to learn the bare facts, and when the rhyme itself has
once been learned it contains the necessary facts within it.

Devices we make up ourselves are much more effective
than those learned from others or from books, hence the
doubtful value of much of what is sold in systems of mem¬
ory training. The self-made device is learned, so to
speak, in the process of making it. The device of some¬
one else must be learned as a new fact and is not so read¬
ily retained by the learner. An amusing example of this
necessity for learning the device of another and the pos¬
sibility in so doing of making mistakes can be illustrated
by this experience. I found it hard in a certain city to
remember the sequence of cross streets in the down-town
section when riding down town on a street-car, there
being no street signs at the intersections. The streets
were in order named Spring, Long, Gay. So I made up
the sentence “ spring along gayly” and this gave me an
easily remembered device. I told this to a colleague, who
also claimed to have experienced the same difficulty.
He thought it an excellent little memory device. A few
days later I overheard him explaining it to another, but
now it had become, “ spring gayly along.”

Memory devices, therefore, form no exception to the
general rule in learning, namely, that we learn precisely

236

EFFICIENT LEARNING

what we practice. We must, therefore, be critical of
accepting them, unless the device to be learned contains
in itself or closely bound up with it the things we wish
to remember. Mnemonic devices are not in themselves
either good or bad. They do not strengthen or weaken
the memory. They are just like any other information
and the laws of learning function with them just as with
anything else. It is quite all right to use them at times,
but remember that the best are those which you make
up for yourself, and the funnier, sillier, more striking
they are, the more likely are they to be easily remembered.
We have discussed memory devices under our topic of
relevant and irrelevant reactions, because they fit in
just here. If they are relevant they may be helpful; if
they are irrelevant they most certainly are not helpful.

Emotion and Learning. — The great spurs to learning
are feelings of satisfaction and annoyance. We must be
careful to distinguish such feelings from strong emotional
states. In general it is safe to say that intense emotion
hinders rather than helps the learning process. Extreme
fear, rage, joy, anger, hatred are not good companions
for the learner. They distract the attention. They in¬
terrupt the sequence of reactions necessary for the learn¬
ing of the matter in hand. Children may be so “scared”
of making a mistake that they make nothing else but
mistakes. The examination situation may so terrify a
pupil that he makes all kinds of wrong reactions. Urging
children to hurry may cause flurry and excitement.
Rivalry between individuals or groups may go over into
intense excitement and worry. Scolding or ridicule for
failure may cause fear or anger in a child. Whenever
such emotions become fairly intense, they are sure to
interfere with our habitual reactions.

Emotions represent a general stirring-up of the visceral
and organic parts of the individual. They represent a
preparation for extreme physical exertion. They do not

WHOLE AND PART METHODS OF LEARNING 237

help in the finer reactions required for acts of skill or acts
of judgment and reasoning. They disturb the poise of
the learner.

The Whole and Part Methods of Learning. — Much
experimental work has been done in trying to solve the
question as to how much of a connected piece of work we
should try to learn at once. Should we learn a poem bit
by bit, line by line, or should we try to learn the whole
poem as one piece. The general answer from experi¬
mental work on this problem is that the whole method
is better than the part method, other things being equal.
This means that wherever we have a unit of connected
work to master, it is better to try to learn it as a unit,
rather than to split it up into parts. Of course, the unit
must not be too large. We must be able to envisage it
easily in our learning period. We must be able to go over
it several times at one sitting for practical school learn¬
ing. What may adequately constitute a “whole” for
one individual may be too much or too little for another.

The reason for the superiority of the “whole” method
is really dependent upon the principle we have repeated
so often in this chapter, namely, that we learn exactly
what we practice. In the “whole” method we practice
the poem or passage as a unit. Whereas, in the “part”
method we practice the bits in isolation. After the bits
have thus been learned, we have to learn them all over
again as a unit in order to make them hang together.
We all know how difficult it frequently is to remember
the beginning of the “next” stanza in a poem. This is
not only due to the difficulty inherent in the fact that
frequently a new theme or thought is introduced at the
beginning of a stanza, but the difficulty may also be in¬
creased by the fact that we have learned the stanzas as
separate units, and the bonds between stanza and stanza
are weak.

The practical exigencies of mass instruction lead us as

238

EFFICIENT LEARNING

teachers to break up our material into bits that can be
assigned from one day to the next. Much of this is no
doubt necessary, because we need a check-up from day to
day, as to what the child is learning. For some things
this is no misfortune, but for other units of work it is a
drawback and we should try as far as possible to give
long-time assignments of units that are “ wholes.” It
certainly is not necessary for all the children in a class to
memorize the same poems and to learn them by bits.
Why not let each child choose his own poem and be made
responsible for learning it by a given date, and so on with
other topics.

Of course the whole method must be used with common-
sense. Reading over the unit as a whole until learned is
best, provided each item in the whole is equal in difficulty
to all other items. Then all the items will all be learned
equally well at the same time. This is never the case
with material outside of the laboratory. Therefore, the
“whole” procedure must be used until a general knowl¬
edge of the whole has been gained. Then we must con¬
centrate on the difficult bits bringing them up to the
whole unit. Then we must go over the whole again, and
so on until learned.

The whole method is used by the young child learning
the songs sung to him by his mother. He hears them
over and over again and sings them over and over again
with her and ultimately he has learned the whole song
or nursery rhyme. He does not learn them line by line.
And so he learns to read following the words of the story
as his mother reads, going over the story again and again
from beginning to end until one day he can say the whole
story and keep the right place with the printed symbols.
And the whole method is the one used by most people
not in school. Folk-songs are handed down from one
generation to another by the frequent repetition of the
whole song. And so are stories and legends transmitted.

LEARNING PERIODS

239

We don’t sit down to learn these things bit by bit. The
minstrels, bards, troubadours and minnesingers picked up
by the “whole” method the songs, stories and legends of
their profession. We learn exactly what we practice.
The “whole” method is best for learning things by
“wholes,” and the “part” method is best for learning
things by “parts.” Now in most material we generally
want the thing to hang together as a unit; we want to
learn the whole thing, and so in most school work we
should approximate the whole method as far as we can.

Most of this discussion of the whole and part method
refers to memorizing things. We are not so sure about
the superiority of the whole method in the case of learning
complicated motor acts. It has been shown that one can
make a pencil maze so complicated that it is practically
impossible to learn by the whole method, whereas it can
be mastered by the part method. We are, therefore, at
present not certain of the respective merits of the whole
or part method where motor activity is concerned.

The Length and Distribution of Learning Periods. —
Short periods of practice or drill are better than long ones.
Of course, the subject studied and the individuals who
are studying will have to determine what is to be con-
sidered short or long in each case. We cannot say that
five minutes is always better than ten minutes. Five
minutes may be an adequate period for some kinds of
drill, but it may be totally inadequate to get a good start
in other kinds of work. The unit or “whole” which we
discussed above will help to determine the length of the
period. The period must be long enough to go over the
work several times if it is a small unit and at least once
if it is a long unit. In highly concentrated drill the periods
can be made very short to be effective.

Now it follows that if we make our practice periods
short, we shall have to distribute them over a greater
number of days than would be necessary for longer periods,

240

EFFICIENT LEARNING

assuming that we want to devote the same total amount
of time to the learning. The distribution of the periods
and the length of the periods are, therefore, closely re¬
lated. Arrange for short lengths of practice spaced out
over a long period of time.

1234 66789 101112 13 : 14.16 16.1718 19 20 2122 23 24
Successive Five Minute Periods

Figure 62. — Practice in writing letters for numbers according to
a key. (From Starch’s Educational Psychology.)

Work done in the laboratory shows this law very nicely.
The curve in Figure 62 is taken from an experiment by
Starch, and shows the efficiency of four groups of stu¬
dents in each successive five minute period in writing
numbers for letters according to a key. The meaning of
the curves is as follows:

10 min. curve = group working 10 min. twice a day.

20 “

<« _

20 “

once “ “

40 “

<< __

40 “

every other day.

120 “

U —

120 “

at one time.

The two groups showing most improvement are the
group which has the shortest practice period and the

LEARNING PERIODS

241

group which has fairly short practice periods well dis¬
tributed. The twenty minute period is almost as good as
the ten minute period because the distribution of the
periods, one a day, is probably a little better than the
distribution of the shorter period, twice a day.

Figure 63. — The lower graph shows the results of fourteen half-
hour practices on the same day in a letter-symbol substitution experi¬
ment. The upper graph shows the results of the same number of
practices, one a day for fourteen days. (From Pyle’s The Psychology
of Learning.)

If we bunch our practice very much, if we try to cram
it all in a day, we may find that after a certain period we
are making no progress. We may in some functions even
be getting worse. After a certain period we may lose inter¬
est, get bored and careless. Thus Pyle (6) found that
14 half-hour periods of practice in a letter-symbol sub¬
stitution experiment all on the same day led to virtually
no improvement after the third half-hour period and after
the eleventh half-hour period his efficiency actually de¬
creased. A very clear case of diminishing returns for his
efforts. This is shown by the lower curve in Figure 63.

242

EFFICIENT LEARNING

The upper curve in the same figure shows how improve¬
ment in the same task continues to the end of the
fourteenth practice period when these periods are dis¬
tributed one a day over fourteen days.

Cramming is bunching one’s work in a short period of
time just before it is needed. If we try to cram too much
into a given period, we run the risk of confusion, forming
bonds in the wrong order or forming one set of bonds too
quickly after another set of bonds, and so on. If we
lengthen the cramming period we get fatigued and bored.
Cramming is only justified if we have no real interest in
the thing to be learned, and if we wish to forget it as soon
as a certain examination or test has been held.

There are many other experiments which all point in the
same direction. Work with school children has verified
the laboratory findings. In most school work there is no
necessity for crowding the drill into a short period of
time. We can spread it over a semester or term. We
should naturally have more practice at the beginning
and then let it taper off more or less evenly toward the
end. This kind of review or re-learning from time to
time is necessary in order to refreshen and strengthen
old bonds. Thorndike has pointed out the desirability
of considering carefully these factors in the construction
of text-books in arithmetic, and he has analyzed the
amount and distribution of practice with different kinds
of number combinations as actually found in several arith¬
metic text-books. For example, he finds that in one set
of arithmetic texts, the amount of practice with 5X5 grad¬
ually increases from its first appearance in grade 3 until
the end of grade 6. In grade 6 the pupil would actually
get a greater amount of practice in multiplying 5 by 5
than he received in any preceding grade. Figure 64
shows the distribution of practice on 5X5 according to
Thorndike’s analysis. From Thorndike’s many diagrams
we may justly conclude that little thought has been given

KNOWLEDGE OF IMPROVEMENT 243

to this problem of the amount and distribution of prac¬
tice in most arithmetic texts. Now that this has been
pointed out, future books will probably be better planned.

Knowledge of Improvement. — Keep the pupil in¬
formed of just what improvement he is making. If he
knows he is succeeding, that will make further success
easier. It has been shown in many laboratory experi-

244

EFFICIENT LEARNING

ments that subjects do not improve so fast if they are not
told their scores from time to time. If you give a test
and do not tell your pupils their scores, you will find no
enthusiasm for the next test of the same type, and if you
keep on giving these tests and never saying anything
about them, the chances are that the pupils will actually
decrease in their scores. This all follows naturally from
our laws of exercise and effect. We must feel the effect
of our response, either as satisfaction or annoyance, in
order to influence our next response to a similar situation.
This is the explanation of the pedagogical value of going
over and explaining the errors made by pupils on their
examination papers. The sooner this can be done the
better. Immediately after a short examination is com¬
pleted, have it corrected. In this connection the new-
type examinations are very useful. If a true-false test
has been given to a class, the instructor should either
have the students correct their own or each other’s papers,
or he should collect the papers and read over the examina¬
tion with the class. Errors are thus immediately cor¬
rected and the student is set on the right path. Students
of college level, at least, appreciate this method very
much.

Errors. — Errors made by a pupil will not correct
themselves. Errors are reactions, just like other reac¬
tions, and if satisfaction attaches to them, we will be
likely to repeat them when the same situation recurs.
We may learn to make errors, just as we learn to make
right responses. The more frequently an error is made,
the more likely is it to be made in the future. Teaching
should help the child to avoid learning errors. Once an
error has been made by a child, there seems to be a possi¬
bility of its recurring. Only slowly is it stamped out.
Myers (5) recorded over a period of time the following
responses made by a child to the combination 4+6:—8, 8,
10 , 8 , 10 , 10 , 8 , 10 , 10 , 10 , 10 , 8 , 10 , 10 , 10 , 10 , 10 , 10 , 10 ,

SPEED AND ACCURACY

245

8, 10, 10, 10. In this series we see the persistence of the
wrong response 8, for a long time. We must attach satis¬
faction to 10 and annoyance to 8 in order to encourage
the one and discourage the other.

Speed and Accuracy. — Accuracy in most work is
desirable. In some work, such as arithmetic, accuracy
is all important. It is no use teaching a child to be 50 or
75 per cent accurate in his arithmetical responses. We
must strive for 100 per cent accuracy or as near to that as
we can attain, otherwise there is little value in the arith¬
metic we teach him. In other subjects 100 per cent accu¬
racy may not be as important. We can profit a great
deal from what we read, even if our comprehension is not
100 per cent accurate. We do not need 100 per cent accu¬
racy in writing, whatever that may mean. And so with
other subjects.

But speed, too, is valuable. The slow worker falls be¬
hind in the competition of life. He is ineffective because
he is slow. There is so much to learn, there are so many
modifications to be made, that speed in learning to make
them is desirable. The old proverb says, “Slow but
sure,’ 7 and many people believe this. They have further
taken for granted the converse, “Quick and inaccurate.”
Fortunately, psychology has demonstrated that both of
the statements are wrong. To be sure there are all types
of individuals, slow and accurate, slow and inaccurate,
quick and accurate, quick and inaccurate. In general,
however, those who are quick tend to be more accurate
than those who are slow, and those who learn quickly
retain more than those who learn slowly. Speed and
accuracy go together more frequently than slowness and
accuracy.

The quick reaction is the well-learned habitual reac¬
tion. The slow reaction is the partially learned, doubtful
reaction, and is more likely to be inaccurate. Compare
the quick accurate reactions of the practiced adult reader

246

EFFICIENT LEARNING

with the slow inaccurate reactions of the first-grade be¬
ginner. Slowness is generally indicative of hesitancy,
uncertainty. It is the symptom of the learner, whose
habits are not yet well-formed. Of course, we may learn
to react slowly, just as we may learn to react quickly.
Other things being equal, therefore, we should try from
the beginning to form quick reactions, rather than slow
ones. Once we have learned the right response, we should
speed it up before we get into the habit of reacting slowly.
Slow reactions are frequently indicative of less desirable
methods of response than quick reactions. The child
who reacts slowly in adding, for example, probably does
so because he is counting on his fingers, or counting to
himself. He is employing a less desirable form of reac¬
tion than the direct response between two numbers and
their sum. And this less desirable response not only takes
longer but is subject to more possibility of error.

Many workers have shown that in most school sub¬
jects, children can easily be taught to increase the speed
of their reactions and this without any loss in accuracy.
Some have found increased accuracy resulting from the
speeding up process, very probably because the children
have substituted more desirable for less desirable methods
of response.

Pupils in the grades can be taught to increase their
speed of reading about fifty per cent without disturbing
their ability to comprehend. This is an enormous gain in
speed without any counterbalancing disadvantage.

Figure 65 shows different speeds and qualities in
measuring the work in silent reading of 1831 pupils. The
numbers in the circles indicate percentages. The small¬
est percentages are for the “slow but good” and the
“rapid but poor.” High rate and good quality are gen¬
erally related.

Age and Learning. — As we grow older we feel that it
becomes more and more difficult to learn new things.

AGE AND LEARNING

247

Rapid speed m&
good quality

Medium speed
and good quality

Slow speed and
good quality

(©)

Rapid ©peed
medium quality

Helium speed and
seliua quality.

Slew @peei m£
medium quality

Rapid ©peed asd
poor quality

Midiu® speed
and poor quality

Slow speed and I
poor quality

Figure 65. — Per cent of 1,831 Cleveland pupils found in each of
nine speed and quality groups in silent reading. (From Judd’s
Measuring the Work of the Public Schools .)

“You cannot teach an old dog new tricks.” This is only
partially true. What happens to most adults is that they
do not want to learn; or better, they know more definitely
than the child does what they do and what they do not
want to learn. The child is more docile. Thorndike’s (7)
studies of adult learning show that ability to learn in¬
creases up to late adolescence. There is no indication of
a decrease in this ability up to age 25. From age 22 to
age 42 the decline is very slow, only about one half of one
to one per cent a year. This slight decrease is about the
same for inferior and superior intellects, although they dif¬
fer greatly as to the difficulty of the tasks they can master.
It is only when we reach the age of 55 or thereabouts
that ability to learn seems to be definitely on the wane.
Childhood is not the easiest period in which to learn a

248

EFFICIENT LEARNING

language. Given equal interest and equal native capacity
adults of ages 20 to 40 can learn better than children of
ages 8 to 12. To quote Thorndike’s own words, “In gen¬
eral, nobody under forty-five should restrain himself
from trying to learn anything because of a belief or fear
that he is too old to be able to learn it. Nor should he
use that fear as an excuse for not learning anything which
he ought to learn. If he fails in learning it, inability due
directly to age will very rarely, if ever, be the reason.”

Incentives for Learning. — What are the best incentives
for the teacher to employ in making children learn better?
As between praise and blame there is no doubt that praise
is much more effective for most children than is blame
or punishment (Hurlock, 4). University students re¬
porting their experiences as high school students believe
very decidedly that they worked better under such in¬
centives as praise, encouragement and commendation and
worse under such incentives as censure, ridicule, threats,
sarcasm and punishment. Censure and punishment may
at times be useful and necessary, but in general there is
no doubt that the teacher should rely on encouragement
and praise.

Mental Imagery. — In older text-books of educational
psychology a great deal used to be said about mental
imagery and its effect on the learning process. It used to
be thought that each individual was dominated by a
special type of imagery, and such expressions as “eye-
minded,” “ear-minded,” and “motor-minded” came into
use. As a matter of fact, we now know that pure types
rarely exist. All of us use all types of imagery. Children
are not sharply divided into visuals, audiles, motiles and
so on. It is best for the teacher to disregard these dis¬
tinctions, and it is almost impossible anyway to deter¬
mine whether a child uses predominantly one type of
imagery rather than another. Attempts at classifying
pupils into image types should be abandoned. Material

SUMMARY—HOW TO STUDY

249

should be presented in the most effective ways in order
to call forth the kinds of reactions which will be required
later on by the child. Only in cases of special inability to
learn need we try other modes of presentation than those
usually found to be effective and economical.

Summary — How to Study

All of this chapter has been devoted to a discussion of the most
effective methods of learning. From the point of view of the
student or pupil doing the learning we could summarize the
material of the chapter into rules for studying. These might be
somewhat as follows:

1. Learn to react in the way you will be called upon later to
react. This is the most important and fundamental of all
the rules.

2. Don’t learn irrelevant details. Avoid irrelevant responses.

3. Don’t set yourself an aim impossible of attainment. Don’t
use a model impossible to achieve.

4. Don’t sit and wait for inspiration. React, and inspiration
may come later.

5. Use only such memory devices as are a real help. Don’t
waste time learning the devices. The best are such as you
make up for yourself.

6. Intense emotion of any kind hinders learning. Try to work
in a calm manner. Don’t fuss and worry.

7. Learn whole units as wholes, so far as is feasible. Don’t
chop up natural units into unnatural bits.

8. Devote a reasonable amount of time to any particular task.
Let it be short rather than too long. Short periods are
better than long.

9. Distribute the periods of study over as many days as pos¬
sible. Ten minutes each day for six days is better than
sixty minutes all on one day.

10. Cramming is bunching your work all together just before
it is needed, and so it is uneconomical as we saw in rule 9.
But if you don’t want to retain it and have to learn it,
nevertheless, cramming is the quickest way to get done
with it.

250

EFFICIENT LEARNING

11. Keep a record of improvement. Watch your score. It will
stimulate you to do better.

12. Avoid error from the start. Do not be complaisant about
errors. Making errors means you are forming a habit of
error.

13. Speed and accuracy go together. Don’t dawdle. Keep
moving. Most likely you can increase your speed and in¬
crease your accuracy at the same time.

14. People are never too old to learn. Older people can learn
about as rapidly as younger people.

15. The best incentive for learning is praise rather than censure.
In your teaching use ridicule and sarcasm sparingly.

16. Don’t bother about your mental imagery. You probably
use all kinds. Forget it.

17. Don’t try to learn too many things at the same time. Wait
until you are well along in one thing before adding a new
subject.

18. After working intensely at one topic, leave it alone for some
time. Give it a chance to form connections with the rest
of your knowledge and thus become a real addition.

19. If you have an essay to write, jot down at once your ideas
about it. Keep thinking about it at odd moments and add¬
ing to your notes. You will be surprised how much you can
do in preparation before you actually sit down to write it.

REVIEW

251

Review

True-False Statements

As a review mark the following statements true or false:

1. Teaching young children to trace letters will speed up the
initial stages of writing instruction.

2. Strong emotion stimulates mental activity and leads to
greater achievement.

3. Distributed practice periods usually yield better returns for
time spent in learning than does concentrated practice.

4. The best way to learn a thing is exactly in the form it is to
be used later.

5. An undesirable response may be eliminated by setting up
in its place a more satisfying one.

6. The so-called “natural methods” of learning, which a pupil
discovers for himself, may not be the best for him.

7. Intense emotional excitement is not helpful to learning.

8. To learn by the “part” method is always more effective than
to learn by taking large “wholes.”

9. It is psychologically wrong to use memory devices as an aid
to memory.

10. Learning should be thought of as reaction rather than ab¬
sorption.

11. It is important to know what habits should be formed and
also the order in which they should be formed.

12. It is well to explain the theory of long division before giving
your pupils examples to practice.

13. Children who work rapidly usually make the greatest num¬
ber of errors.

14. Learning by doing is based on a false psychology.

15. Children can learn poetry better and more rapidly at age 9
than at age 16.

16. It is good to have frequent reviews of a subject at short
intervals throughout the semester.

17. Study your French lesson “out loud” if you want to learn
to speak French.

18. The laboratory method of teaching is always superior to
the lecture method.

252

EFFICIENT LEARNING

19. A self-made memory device generally aids retention less
than one of the standard devices.

20. Cramming is psychologically and morally wrong.

21. The teacher by not giving out the results of achievement
tests sustains the interest of his pupils.

22. After forty an individual is too old to begin learning a for¬
eign language.

23. Most pupils can be divided into two classes, strongly “ear-
minded” or strongly “eye-minded.”

24. Ridicule generally acts as a decided spur to further improve¬
ment.

25. Check the best response:

Speed and accuracy in the performance of a task:

(a) seem to have no connection with one another.

(b) generally go together.

26. Check the best response:

Learning is a process of:

(a) passive absorption.

(b) eliminating original tendencies to react.

(d) storing up knowledge.

Advised Readings

Whipple, G. M.: How to Study Effectively, Second Edi¬
tion, Public School Publishing Co., Bloomington, Ill.,
1927.

Kitson, H. D.: How to Use Your Mind, Philadelphia, 1916.
Kornhauser, A. W.: How to Study, University of Chicago
Press, 1924.

Thorndike, E. L.: The Psychology of Arithmetic, New York,
1922.

References

1. Gates, A. I.: The Improvement of Reading, New York,

1927.

2. Gates, A. I.: New Methods in Primary Reading, Teachers

College, Columbia University, New York, 1928.

REFERENCES

253

3. Hertzberg, 0. W.: A Comparative Study of Different Methods

Used in Teaching Beginners to Write, T. C. Contribu¬
tions to Education, No. 214, Teachers College, Columbia
University, 1926.

4. Hurlock, E. B.: “The Value of Praise and Reproof as

Incentives for Children,” Archives of Psychology, No. 71,
July, 1924, Pp. 78.

5. Myers, G. C.: The Prevention and Correction of Errors in

Arithmetic, Plymouth Press, Chicago, 1925.

6. Pyle, W. H.: “Economical Learning,” Journal of Educa¬

tional Psychology, 1913, 4, pp. 148-158.

7. Thorndike, E. L. et al.: Adult Learning, New York, 1928.

CHAPTER XI

THE PERMANENCE OF MODIFICATION

Disuse Weakens. — Use strengthens. Disuse weakens.
As soon as we stop practicing or exercising, we begin to
forget. This is the opposite or converse of the Law of
Use. Whatever habits we have formed, whatever things
we have learned, begin to weaken as soon as we stop mak¬
ing use of them. Some of our habits and skills have had
so much exercise, and some of our knowledge has been so
thoroughly learned that, even after long periods of dis¬
use, there is little forgetting. With things not so well
I00 —a learned, however, forgetting comes quickly and
is easily measurable.

Curves of Forgetting. — Just as our curves of
learning pictured the growth or improvement in
a function because of exercise, so we may draw
curves showing the decay or weakening of a
function through lack of exercise. The classi¬
cal curve of forgetting is that of Ebbing*

m —

€9

*3 60

•d

% 40

«L>

20 —

-*-»

<L>

0 ! 2 6 14 20

Interval between learning and relearning: in days.

3©

Figure 66. — A curve of forgetting for non-sense series by Ebbing-

haus (after Thorndike).

haus, shown in Figure 66. Ebbinghaus learned a series
of non-sense syllables until he could repeat it once cor¬
rectly, and then, after a definite interval of time, relearned

254

CURVES OF FORGETTING

255

the same series, measuring how long it took him to re¬
learn. Then he learned another similar series and after
a different time interval, relearned that series. And so on
for many different time intervals. In this way he arrived
at the curve in Figure 66, which shows that the longer
the interval before relearning, the smaller the percentage
of time saved in relearning, or the greater the effect of
forgetting. We see, therefore, that with this kind of
learning at least, forgetting is very rapid at first and
gradually becomes less and less rapid. Even within an
hour after learning the words, more than half had been
forgotten.

Other investigators have not found forgetting to pro¬
ceed quite so rapidly as Ebbinghaus did. Another curve
(4) for forgetting is given in Figure 67. This shows the
percentage retained after intervals of 30, 60, 90 and 120
days. The students who took the experiment observed a
card with a number of objects and photographs on it and
then wrote out what they had seen. The curve is not
nearly so steep as the Ebbinghaus curve.

We cannot take Ebbinghaus’ results and frame a gen¬
eral law. There are evidently all kinds of forgetting
curves, just as we found there were all kinds of learning
curves. The more thoroughly the thing has been exer¬
cised or learned the less steep will be the forgetting curve.
Ebbinghaus’ curve represents the loss for things just
learned to the bare point of retention. Not all that we
learn is learned in that way. Frequently we over-learn
quite a bit and in some complex functions there are parts
which we over-learn tremendously, so that they are re¬
tained for very long periods of time.

Over-learning is very obvious in complex acts of skill,
such as typewriting, swimming, driving an automobile
and so forth. In such complex functions many minor
habits are tremendously over-learned, so that they may
not be forgotten after years of disuse. The total complex

256 THE PERMANENCE OF MODIFICATION

Figure 67. — A curve of forgetting for meaningful material.
(From McGeoch and Whitely, Journal of Educational Psychology ,
1926.)

function does not drop out of existence like a series of
non-sense syllables. Much of it is still left. Enough of
it is left, so that we do not say, “I have forgotten how to
typewrite, or how to swim,” but rather, “Fm very rusty
in my typing or swimming.” Enough of the connections
will still function, so that after some exercise we may be
able to bring ourselves up to the level of efficiency at-

CURVES OF FORGETTING

257

tained before and in some cases to exceed that level.
Book (1) for example, found that after a period of seven¬
teen months without practice in typing he had lost only
a little of his skill, and that after ten daily practices of
10 minutes each, he was able to exceed his best previous
record. Here the loss from disuse was very rapidly
regained.

This ability to reinstate rapidly things partially for¬
gotten, which had been previously well exercised, is
familiar to all of us. Much of our school knowledge may
be forgotten, but how quickly it comes back to us if we
go over it again. Many a father or mother is surprised
at the amount of Latin or algebra he or she knows when
going over it again after many years of disuse with a son
or daughter. Much of it comes back quickly and the
parts that were frequently exercised come back the most
readily of all. The Mother Goose rhymes you learned in
your childhood may not come back readily just now, but
if you give them a little encouragement they can soon be
brought back to their former strength, as many a young
mother can testify.

The significance of all this is that we should not worry
too much about forgetting. Many reactions now for¬
gotten can easily be revived when needed. If we wish
to keep a reaction continually ready to function, after
once being thoroughly learned, just a little exercise now
and again, lest we forget. No improvement, habit or
knowledge is absolutely permanent, and yet on the other
hand neither is it ever absolutely forgotten. It has left
its trace and will be more easily revived than it was
originally learned.

We cannot say, however, that all practice always leads
to permanent modification or change. The effects of
some practice seem to be very evanescent. In an investi¬
gation (Gates and Taylor, 2) with a group of school chil¬
dren, 78 days of practice in memorizing digits was given.

258 THE PERMANENCE OF MODIFICATION

During this period of use the average score rose from 4.36
to 6.36. Then followed a period of disuse or no practice
for about five months. When tested after this period,
the average score dropped back to 4.71, or almost as low
as at the beginning of the whole experiment, before any
practice at all. A little later the practice was resumed
and in 22 days the score rose from 4.71 to 5.73. In the
first period of practice it took 44 days to arrive at this
score of 5.7, but this time it only took 22 days. At a
first glance this would seem as if the first long practice
had made the relearning easier. This was not the case,
however, for a control group which had no previous prac¬
tice raised their score from 4.83 to 5.92 within 22 days.
In other words, they did just as well as the experimental
group without the benefit of the long period of 78 days
of practice. In some things, therefore, skills or techniques
may be learned and readily forgotten, so that in be¬
ginning again we have practically to relearn them all over
again.

Supposed Contradictions to the Law of Disuse. — It

has been suggested that there are some exceptions to the
general Law of Disuse. Thus it has been said that “we
learn to skate in summer and swim in winter ” meaning
thereby that a period of disuse may lead to improvement
in a function. There is little experimental evidence for
anything like this. It may be true that there have been
cases where, after a period of disuse, an individual did
better at a task than he did just at the point of stopping.
It will generally be found, however, that this stopping
point was by no means the highest point reached by him
in his improvement. He had previously reached higher
points and his ability after disuse had not exceeded his
highest level of performance. When the golfer makes a
good score on his round after a period of lack of exercise,
he may compare it with his last score at the end of a sea¬
son’s play, when he was fatigued or “stale” from too

THE EFFECT OF LECTURES

259

much play. If he compared it with his best score during
the past season, he would find that disuse has weakened
the bonds somewhat.

If there are any authentic cases of improvement after
disuse, they must be explained on the theory that lack
of exercise caused the diminution of undesirable bonds
which were impeding improvement. When these had
been weakened by disuse, the desirable bonds were able
to function better than they ever did before.

School Work. — Forgetting of knowledge and skill
learned in school by the child goes on steadily, as soon as
lack of exercise takes place. Hence the necessity for fre¬
quent review. Hence the necessity for the distribution
of practice over long intervals. During the long vaca¬
tion from June to September forgetting takes place. The
actual loss in certain school subjects has been measured.
Every teacher knows the necessity for some relearning
in September. But if the previous learning has been
efficient and economical, a little relearning will bring it
all back.

One investigator (5) gave standard tests at the end of
the school year in June and then again at the beginning of
the next school year in September. He found a slight gain
on intelligence and reading tests but a slight loss on
arithmetic tests. Evidently the kind of reaction measured
by the intelligence and reading tests is being exercised
during the vacation and so these abilities continue to
increase as the child develops. Very little of the formal
arithmetic practiced in school is made use of during the
long vacation, and therefore it tends to diminish by
disuse.

The Effect of Lectures. — How much does a college
student retain as a result of listening to a lecture and how
long is it retained? Careful measurement (3) of many
groups gives the curve shown in Figure 68. Immediately
after the lecture 62 per cent can be recalled. Three or four

260 THE PERMANENCE OF MODIFICATION

days later about 50 per cent is retained. After one week
the percentage drops to 37, after two weeks to 30 and

after eight weeks to 23. Like the Ebbing-
haus curve we have here a relatively
sharp drop at the beginning and then a
more and more gradual decrease.

Speed of Learning and Retention. —
Does the rapid learner forget quickly?
Does the slow learner retain most? Evi-

uj 50

0 Vz 1

3 4 5

TIME IN WEEKS

Figure 68. — A curve of forgetting material in lectures. (From
Jones’ Experimental Studies of College Teaching.)

dently not. The indications are that those who learn
quickly in general retain most, and probably also retain
it for a longer period.

Over-Learning. — This is a phrase used to express the
fact that we may learn something past the point of just
being able to retain or accomplish. The longer we wish
to remember something, or the more permanent we wish
to make a modification, the more must we over-learn.
We have had a great deal of practice with the multiplica¬
tion tables. We have over-learned them tremendously.
If we were not to make use of 3 X3 = 9 for many years, we
would still be able to respond with 9 at the sight of 3X3,
because we have over-learned this very greatly. In cram¬
ming we do not over-learn. We learn just about to the
point of bare retention, and hence things we have learned
by cramming are soon forgotten.

SUMMARY

261

Summary

1. Disuse weakens. Forgetting begins as soon as we stop
practicing.

2. The rate at which we forget things varies greatly. Some
modifications are more permanent than others. There are
all types of forgetting curves.

3. Permanence is greatly influenced by the amount of over¬
learning. Some facts and some skills are very permanent,
because they have been so thoroughly over-learned.

4. What has been thoroughly learned and then forgotten or
partially forgotten, can generally be very rapidly reinstated.

5. During the summer vacation school children tend to forget
much of that which they do not make use of during this period
but they continue to gain in such subjects as reading.

6. The memory of things learned in a college lecture decreases
steadily for about two weeks. Nevertheless, after eight weeks
about 23 per cent is still retained.

7. The rapid learner retains most and also probably retains it
for a longer period.

262 THE PERMANENCE OF MODIFICATION

Review

True-False Statements

As a review mark the following statements true or false:

1. During a period of disuse a function remains stationary.

2. The slow learner retains what he gets better than the quick
learner.

3. The Ebbinghaus curve is the typical forgetting curve.

4. If we wish to retain for a long period of time, we must over¬
learn considerably.

5. In complex acts of skill, many parts are generally considerably
over-learned.

6. The more a thing is over-learned, the more readily can it be
reinstated once it has been forgotten.

7. There is some evidence to make us believe that some skills
go on developing or increasing during a long rest period.

8. In September after the summer vacation children are just
about as good in arithmetic as they were in the previous June.

9. “Easy come, easy go,” meaning, that those who learn quickly,
forget quickly, has been shown to be true by psychological
measurements.

Advised Readings

Thorndike, E. L.: Educational Psychology: Briefer Course ,
Teachers College, Columbia University, New York, 1914,
Chapter 17.

References

1. Book, W. F.: The Psychology of Skill, New York, 1925.

2. Gates, A. I. and Taylor, G. A.: “An Experimental Study

of the Nature of Improvement Resulting from Practice
in a Mental Function, ” Journal of Educational Psychol -
ogy, 1925, 16, 583-592.

3. Jones, H. E.: “Experimental Studies of College Teaching,”

Archives of Psychology, No. 68, 1923.

4. McGeogh, J. A., and Whitley, P. L.: “The Recall of Ob¬

served Material, ” Journal of Educational Psychology ,
1926, 17, 419-425.

5. Morrison, J. C.: “The Effect of the Summer Vacation on

Learning,” Ohio State University, Educational Research
Bulletin, Oct. 1, 1924, 245-249.

CHAPTER XII

THE TRANSFER OF TRAINING

Reciprocal Modification. — So far in all our discussion
of the modification of original nature, we have taken up
one modification at a time. All the learning curves in
Chapter IX dealt with the modifications of one function,
with the learning of a definite task, with the formation of
a specific habit. The learning of anything may last for
many days or months or years, as when one learns golf
or arithmetic or good manners. And in between the
practice of any of the specific functions we are learning
other things. We do not practice one thing until we are
perfect, and then begin another. In every-day life we
are practicing countless things at the same time. And so,
in this chapter, our problem is to discuss what effect
modifications of original nature have upon each other,
the reciprocal action of one modification upon the other.
If modifications in a boy’s reactions involved in golf,
arithmetic and table manners are taking place during
the same period of time, our problem is to discover
whether the modifications taking place in one of these
functions influence the modifications involved in any of
the other functions, and to what extent.

If we turn to specific school illustrations, the importance
of the problem will immediately become obvious. Does
the learning of French help the student in the learning
of another foreign language? Does Latin or grammar
help a student to write better English composition? Does
improvement in accuracy in arithmetic lead also to im¬
provement in accuracy in Latin? If we train a child to
be accurate in one thing, say Latin translation, does it

263

264

THE TRANSFER OF TRAINING

mean that he will automatically become accurate in all
other things in life? If we teach a child about honesty,
will he act honestly? If we train him to be honest at
home, will he be honest in school, and if so, will he be
honest later on in business or in his professional career?
Or in other words, what influence has any one modifica¬
tion upon any other?

General Assumptions in the World at Large. — Let us

see if a cursory glance over many of life’s activities will
give us a clue to what general experience has taught the
human race. Do we assume transfer or not? In many
cases it would seem that we do. The good teacher is
promoted to be a principal. The duties of a teacher and
those of a principal are widely different, and yet the
assumption is that efficiency in the first post is a good
indication of efficiency in the second. So the capable uni¬
versity professor, whether he excels as teacher or research
worker, is promoted to the administrative office of dean
or president. The good salesman is made sales manager.
The good lawyer, business man or politician is elected
mayor, or governor or congressman. Indeed the whole
theory of electing men to various offices, as in our demo¬
cratic state, assumes a great amount of transfer. Because
A has been a good business man and has shown ability in
looking after his own affairs, therefore, he will be able to
look after other people’s affairs equally well. B is a good
physician, and, therefore, should prove an excellent public
health officer. And so we might go on with numerous
illustrations of how the world at large believes and acts
upon the assumption that demonstrated ability or effi¬
ciency in some particular narrow sphere is a good index
of future efficiency in another sphere, even though the
latter may be rather far removed from the former. Ob¬
viously the world believes to a certain extent in the good
all-around man, and in his ability to function well in many
capacities.

ASSUMPTIONS IN THE WORLD AT LARGE 265

But the world at large does not carry this to extremes.
To demonstrate one’s efficiency in one job is not an open
sesame to every other job. The good lawyer is not ap¬
pointed public health officer. The engineer is not ex¬
pected to function as minister or judge. Because one
is good in any of these special lines, one is not expected
to be expert in any or every other special line. The in¬
creasing number of government, state and municipal
positions put under a civil service control, is evidence of
the lack of belief in transfer. The city manager move¬
ment may be interpreted as a doubt as to whether the
generally good all-around man has sufficient expert knowl¬
edge to manage the affairs of a city. School superin¬
tendents to-day are chosen with more care as to their
particular qualifications. Not any good man will do.
Obviously many positions require a special expertness for
success, and expertness in any other line is no guarantee
of success in general. Obviously, therefore, every-day
life affords plenty of examples where transfer is not as¬
sumed.

There are other conflicting experiences of mankind that
make one hesitate about assuming too much transfer.
We all know that individuals may be courteous, kind and
generous in company or in business relations, and at the
same time be rude, cruel and selfish at home. An un¬
scrupulous business man may be the very essence of
honesty in his relations with his friends, his club, his
church. The kind doting father may be an employer of
children under sweatshop conditions. So kindness, hon¬
esty, generosity are not traits that transfer from one set
of circumstances to all others.

It might be more correct to restate all this by saying
that we find in the world around us belief in a vast amount
of transfer from some positions to others, belief in mod¬
erate amounts of transfer in other cases down to a belief
in no transfer in many other cases. Belief in all degrees

266

THE TRANSFER OF TRAINING

or amounts of transfer could probably be found and such
belief is based upon the general experience of mankind.
Undoubtedly many such assumptions are erroneous or
distorted. We probably assume more transfer in many
cases than really exists.

But the general picture that this survey of everyday
life has shown is a true one, namely that different amounts
of transfer from one function to another exist. The sci¬
entific question is not whether there is or is not transfer.
We have nothing to do with the question, “Do you be¬
lieve in the transfer of training? ” The only real question
for psychology is, “What amount of transfer takes place
from this function to that function under these specific
conditions?” And the answer may be in terms of posi¬
tive, zero or negative amounts.

Historical Importance of Problem in Education. — The
general problem we are discussing has figured large in the
history of education. There it is known under the name
of the doctrine of Formal Discipline. The mind needs
training or disciplining. Certain subjects are better dis¬
ciplines than others, hence they should be preferred in
the training of children.

This disciplinary conception of education comes into
prominence in the seventeenth century, although the
general theory had appeared at various times previously.
It springs up to defend some well-established subject in
the curriculum whose usefulness is being attacked. Be¬
fore the Reformation, Latin was the most useful, util¬
itarian, practical subject in the curriculum for the majority
of children in school, because Latin was the language of
religion, of philosophy and of practically all culture.
With the rise of the several vernaculars with literatures of
their own, with the increasing spread of literacy beyond
the ranks of the clergy and the scholar, Latin became of
less practical use for the average boy in school. Hence
its usefulness was questioned and its answer was that it

PROBLEM IN EDUCATION

267

possessed more disciplinary value than any other study.
It trained the mind. Even if the content learned were
useless, the training was priceless.

The doctrine of formal discipline fitted in well with the
faculty psychology of an earlier period. Reasoning,
memory, imagination, perhaps accuracy, are faculties or
powers of the mind. What better training for these fac¬
ulties could there be than Latin translation and Latin
syntax? When once trained these faculties can cope
equally well with any other content.

Formal discipline assumes a very great amount of
transfer. The general argument of formal discipline is
always employed in one way or another, whenever a well-
entrenched subject is attacked by a new and more utili¬
tarian subject clamoring for admission into the curriculum.
Latin was so well entrenched and has been so well taught
and systematized for so many years that it has borne the
brunt of the attack. Latin has made claims to a greater
amount of transfer than most of the other subjects. But
really every subject makes use of the doctrine of formal
discipline to justify itself in the face of an attack from a
more utilitarian subject.

Science came into the curriculum as a utilitarian sub¬
ject in the face of much opposition. It split up into
several studies which are now being attacked by the more
utilitarian General Science. Modern Languages were
introduced because of their utilitarian value, but they
fall back to-day upon their general cultural value or their
transfer value in order to justify their place. Industrial
training is attacking manual training which justifies
itself on formal disciplinary lines.

As soon as a subject is introduced into the curriculum
of any school or college, it begins to be organized and
“formalized” by the teacher. To the teacher this subject
is necessarily important and it takes the center of the
stage. The teacher believes in the intrinsic value of the

268

THE TRANSFER OF TRAINING

subject itself and he sees also the multifarious connec¬
tions with other subjects. Hence the subject is not only
intrinsically valuable but has transfer value to other sub¬
jects and to life in general. No one can be a good teacher
of a subject without believing in the value of his subject
far beyond the mere content he is teaching. I believe a
course in educational psychology is not only valuable
for the immediate facts it teaches, but also because the
general principles developed in such a course may help
the teacher in many a concrete classroom situation. I
believe the point of view developed spreads over or trans¬
fers to many other situations in life. At the same time
I know that I am far too optimistic in my belief in the
transfer value of educational psychology. I feel sure that
actual measurements of the transfer effects would be
very disappointing to me as a teacher. And yet I go on
hoping for these transfer values because it is only in so far
as I do hope for them and believe in them that some trans¬
fer is attained.

Mental Discipline in English Public Schools. — This
doctrine of formal discipline has had and still has a tre¬
mendous hold on education. It still dominates much of
our education. In “A School Master’s Diary” (2), pub¬
lished in 1918, a teacher in one of the English Public
Schools records his futile attempts at innovations and
bursts forth in the following paragraph:

“Subjects are taught just in so far as they are distaste¬
ful; the fact that one can work hard at anything just be¬
cause it is interesting is regarded as impossible. If one
begins to argue, you are countered by the shibboleth of
‘mental discipline,’ which is supposed to be the final
word on any topic of controversy. If grammar grind
provides a mental discipline, grammar grind must there¬
fore be invaluable, quite apart from its utilitarian aspect.
Consequently boys are taught many things which serve
no useful purpose and lead nowhere, simply because it is

LACK OF TRANSFER NOTICED

269

good for them to have to perform arduous, pointless tasks
without asking the ‘why’ of them.”

Wide Claims for Transfer Still Current. — We must
not think, however, that this belief in wide transfer is
merely a remnant of a past or alien civilization. It con¬
tinually crops up in our own country in many forms and
in sweeping dogmatic statements. Thus we read in 1928,
“A knowledge of a foreign language contributes in an
unusual degree to the making of internationally-minded,
broad-thinking, intellectually resourceful and contented
citizens.” Would it were so, indeed! And again in an¬
other connection we read, “My claim would be that a
student trained in mathematics to distinguish necessary
from sufficient conditions would normally inquire in any
situation whether a thing known to be sufficient were or
were not a necessity.” And again about Greek: “It
forces students to think closely about the meaning of
words. In ordinary life people do not think about words.
. . . Many people live mentally in a sort of fog most of
the time. Greek forces one to express oneself accurately.”
These are just a few of the many modern examples of
claims for wide transfer that are still current.

Lack of Transfer Noticed. — Although belief in wide
transfer was general from the Reformation down to the
end of the nineteenth century, there arose every now
and again some doubt in the minds of educators. In
Boston in 1845 a committee made a survey of the educa¬
tional achievements of the school children. One para¬
graph of the report (1) is: “The questions in grammar
are the best proof that scholars may parse technically
and point out the relations of words, their mood, case,
person, number and gender; and yet, in the very sen¬
tences which they make use of to express these relations,
and in quoting rules in justification of what they write, be
continually making blunders; and may parse their sentences
grammatically in the most ungrammatical language.”

270

THE TRANSFER OF TRAINING

Here we have evidence that formal training in English
grammar is not having much, if any, effect upon English
composition, because pupils can “ parse their sentences
grammatically in the most ungrammatical language.’’

The Psychological Attack. — The decay of the old
faculty psychology and the rise of modern psychology
with its greater interest in practical problems led even¬
tually to the psychological attack on the problem of
transfer. The problem now becomes quantitative and
the question is, “How much influence does improvement
in one activity have on some other activity?”

We must, therefore, have an initial test of an activity,
a period of practice in a second activity and a final test
in the first activity to see if any improvement has taken
place in this activity because of the practice in the second
activity. The scheme for this experiment is roughly as
follows:

Experimental Group : Initial Test : Practice : Final Test

Control Group : Initial Test : No Practice : Final Test

You will notice that there are two groups, a control
as well as an experimental group. This control group
has no practice period, but takes the initial and final
tests, and is supposed in all other ways to be subjected
to the same influences as the experimental group. Why
is the control group necessary? Because if the experi¬
mental group made a large gain on the final test after
the practice period, we would not know whether this
large gain were due to the specific practice or to some
other influence. The control group will check this, be¬
cause if the gain is due to some other influence the con¬
trol group will also show it. The real influence of the
specific practice will therefore be the difference in gain
made between the control and the experimental groups.
This is the actual influence of the practice period, and is
called the residual gain.

RESULTS OF LABORATORY STUDIES 271

It will be obvious at once that our results will be more
accurate the more alike the control and experimental
groups are. Hence it is customary to equate these two
groups, to make them as nearly alike as possible. In
careful modern experimental work with school children
the control and experimental groups are equated with
reference to chronological age, mental age, grade, educa¬
tional achievement and so on.

If, after all these precautions have been taken, the
experimental group makes a greater gain in the final test
than the control group, we have a quantitative expres¬
sion of the amount of transfer of one activity upon an¬
other.

Results of Laboratory Studies. — The first experiments
in transfer were those made by psychologists in their
laboratories and were not directly connected with school
subjects. Many of them were not as carefully made and
controlled as we have suggested they should be in the
preceding paragraph. Let us, however, see what general
results were obtained.

Table 4 gives a very brief summary of some of these
earlier experiments carried on in the laboratory. The
results are grouped according to the kinds of functions

TABLE 4

Laboratory Experiments in Transfer of Training

Author

Date

Training

Transfer to

Result

Memory:

James

Memory pas-

Other memory

Slight or

sages

passages

none

Ebert and
Meumann

Non-sense syl-

Numbers, let-

22% resid-

lables

ters, words,

ual gain

poetry, etc.

acc. to
Dearborn

272

THE TRANSFER OF TRAINING

TABLE 4— Continued

Author Date

Training

Transfer to

Result

Fracker

Memory of

Similar stimuli

16% resid-

sound in-

ual

tensities

Dissimilar

3% resid-

stimuli

ual

Sleight

Poetry, prose,

Dates, sylla-

3-4% resid-

tables

bles, etc.

ual

Winch

Poetry

Historical

10% resid-

prose

ual

Reaction-Perception:

Gilbert and

Reaction to

Reaction-sound

52%

Fracker

sound

touch, color,
etc.

No controls

Thorndike and
Woodworth

Areas, weights,

Similar stim-

52% down

lines, letters,
etc.

uli

to zero

Coover

Marking words

Other words

75% as much
as in train-

ing

Kline

Cancellation

Nouns, verbs

Interference.

of e. s.

Less than

control

group

Bennett

Discrimina-

Other colors

60-23%

tion of shades
of red

gain

Coover and

Sound intensi-

Brightness

10% gain

Angell

ties

intensities

Sensori-Motor:

Judd

Theory of re-

Target under

Improve-

fraction

water

ment when
depth is
changed

Webb

Maze

Other mazes

19-77%

Coover and

Typewriter

Very little

Angell

Sorting Cards

reactions

or doubtful

RESULTS OF LABORATORY STUDIES 273

investigated. Under memory we have five experiments
dealing with this type of reaction. The first line tells us
that James in 1890 conducted some experiments. He
asked the question whether training in one kind of mem¬
orizing would affect, or influence or transfer to other
kinds of memorizing. He was raising the general ques¬
tion as to whether the memory can be trained. So he
tested himself on a certain passage of poetry (initial
test), then he practiced memorizing another kind of
poetry (training or practice period), then he tested him¬
self again on the first kind of poetry (final test), and
found that no improvement had taken place. He re¬
peated these experiments with some of his students and
found slight gains, so in the “Result” column of Table 4
we find “ slight (improvement) or none.” These early ex¬
periments of James were carried out in what we would
consider now-a-days a rather perfunctory manner. And
so we must not lay too much emphasis on them.

Take the next experiment which was more carefully
conducted by Ebert and Meumann in 1905. Here the
training was in memorizing non-sense syllables and the
question raised was whether practice in memorizing this
material would increase one’s ability in learning numbers
or letters or words or poetry. Initial and final tests were
given, but there was no control group, so Dearborn at a
much later period worked over the tests again with a con¬
trol group and estimated that the real amount of gain
due to the training period (residual gain) was about 22
per cent.

In a similar way we might discuss the results of each
one of the experiments listed in the table, but this is un¬
necessary. A general study of the table will be sufficient.
First we note that the kinds of reactions in which the
various subjects were trained are rather special and
definite. The functions trained are narrow rather than
broad. We find training in such narrow functions as

274

THE TRANSFER OF TRAINING

estimating areas, lengths of lines, running mazes, sorting
cards, while the broadest function is learning poetry. We
do not find such broad functions as learning Latin or
studying law or engineering. And again we note that
activities tested for transfer effect are also specific and
narrow, such as discriminating colors, typewriting reac¬
tions, memorizing dates or prose. We do not find the
experimenters asking whether any specific kind of train¬
ing influences the study of French or mathematics or
study in general or memory in general or one’s ability to
be a doctor or lawyer. The reason why narrow functions
have been chosen in the laboratory is of course clear.
Such narrow functions can be more easily measured.
Broad functions are difficult to measure accurately.

Another thing we note from the table is the great
amount of similarity between the training function and
the transfer function. There is similarity in the func¬
tions found in columns 2 and 3. Memory for non-sense
syllables is paired with memory for numbers, etc.; mark¬
ing certain kinds of words with marking other words;
discriminating shades of red with discriminating other
colors; learning some mazes with learning other mazes
and so on. We do not find the question raised as to
whether training in one function results in transfer to an¬
other widely different function, for example, whether
training in running a maze influences typewriting reac¬
tions, or whether marking words influences reaction to
sound or color; and so on. In other words we find in
general the experiments are limited to narrow measurable
functions and the transfer effects to similar narrow func¬
tions are measured.

When we run down the column headed “ Results,” we
note that in general there is an appreciable amount of
transfer effect. Almost all experiments find some trans¬
fer effect and several find a considerable amount. Hence
we may sum up by saying that the fact of transfer is ob-

RESULTS OF SCHOOL EXPERIMENTS 275

vious; that the amount of transfer is considerable with
closely allied functions; that the amount of transfer de¬
creases as the functions become more and more unlike.
If the functions are widely separated, much more so than
any in our table, there is likely to be little or no transfer.

Results of School Experiments. — For educational pur¬
poses the laboratory studies we have just described are
not particularly helpful, but they have established the
general method of experimentation and the general fact
of transfer under certain conditions. What shall we find
with school children working with the broader functions
of school subjects? Table 5 (p. 276), gives us a summary
of the chief experiments. Most of these experiments were
done on experimental and control groups of school children
working under class room conditions. Under foreign lan¬
guages the differentiation is generally between students
studying Latin and those not studying Latin. Rarely
have the control and experimental groups been equated
for intelligence and this is important when comparing
Latin and non-Latin groups, because of the selective
influence of the Latin curriculum. Students studying
Latin have been found to score slightly higher on intelli¬
gence tests than non-Latin students.

Keeping in mind the shortcomings of much of the ex¬
perimental work, we may study the results. In three
cases where the residual gain is high (29 to 34 per cent),
we notice that the functions are very closely allied. Thus
mental multiplication is closely allied to the other fun¬
damental operations in arithmetic. In all other cases
the transfer is slight or so small as not to be measurable.
The study of formal grammar has very little or no influ¬
ence upon correct usage of English or ability to reason
about words and definitions. You remember the Boston
Report in 1845 found that scholars “may parse their
sentences grammatically in the most ungrammatical lan¬
guage.” The study of Latin does seem to help a little in

276

THE TRANSFER OF TRAINING

TABLE 5

School Experiments in Transfer of Training

Author

Date

Training

Transfer to

Result

Arithmetic:

Starch

Mental mul-

Adding, divid-

29% resid-

tiplication

ing, etc.

ual

Winch

Computation

Reasoning

Slight or
doubtful

Grammar:

Briggs

Formal

Reasoning

None

Starch

Formal

Correct usage

Slight

Foreign Languages:

Swift

Latin and

Spanish

Latin most

non-Latin

progress

groups

Starch

Latin

Size of English

3-4% in

vocabulary

favor of

Latin stu¬
dents

Perkins

Latin

Spelling, defini-

Latin

tions, etc.

superior

Starch

Foreign lan-

Writing and

Transfer

guage

reading tests

(no transfer
for memory

Science:

tests)

Hewins

Biology

Biological tests

34% resid¬
ual

Non-biological

5% resid-

tests

ual

Rugg

Geometry

Geometrical

32% resid-

tests

ual

Non-geomet-

7% resid-

rical tests

ual

studying Spanish, in English vocabulary and spelling,
but certainly not very much, and certainly not enough to
justify the teaching of Latin for such transfer values alone.

MORE RECENT STUDIES OF TRANSFER 277

If we sum up these school experiments, we find that
they bear out the conclusions derived from the laboratory
experiments. Closely allied functions show a fair amount
of transfer, but as functions become less allied the amount
of transfer dwindles and soon approaches zero. This
takes us back to our discussion of learning. We learn
what we practice. Learning is reacting. We learn to
react to certain stimuli. Unless these stimuli appear in
the new situation, we have not learned how to react to
that new situation; there is no transfer. If the stimuli
are somewhat like the old stimuli, we may be able to
react to them; there is a little transfer. If they are very
similar to the old stimuli, we know how to react; there is
much transfer.

The effect of changing the stimulus to which we have
learned to respond has been shown very effectively by
Thorndike (4). In a class of college students only 6%
made errors when called upon to find the square of x+y;
but 28% made errors when called upon to find the square
of bi+b 2 . If the stimuli are still further changed, Thorn¬
dike shows that the lack of transfer becomes still greater.

More Recent Studies of Transfer. — Several recent
studies have dealt with the influence of the study of
Latin. These show that after one year’s study of Latin
a gain of 1.8 score points on the Thorndike-McCall Read¬
ing Test is made by the Latins above the gain made by
the non-Latins. On a vocabulary test Latin students in
a year gained roughly five words of Latin origin, while
non-Latin students gained only two; but both groups
made the same gain in words of non-Latin origin. The
study of Latin then was only helpful for English words of
Latin derivation. In spelling English words of Latin
origin Latin pupils gained one word more than non-
Latin pupils. All of these and similar investigations
merely confirm the earlier experiments. A study of Latin
does transfer, it does help in building up an English vo-

278

THE TRANSFER OF TRAINING

cabulary, in learning to spell English words, but the
amount of such transfer is very small. Symonds (3)
shows that if we want to build up a vocabulary, the thing
to do is to teach words directly and the amount of gain
so made is much greater than any to be expected indirectly
through the study of Latin.

Latin should not be taught for these indirect values.
They are too slight, too evanescent to justify the amount
of time and labor spent on the study of the subject. Latin
must justify itself on other grounds. It has other grounds.
It should abandon the appeal to formal discipline. What
we say about Latin is equally true of any other subject
that rests its case on disciplinary values.

General Ability to Reason. — But we have not yet
finished with disciplinary claims. Even if the gains for
any subject are slight in such concrete things as English
vocabulary, correct use of English and so forth, there
may be gains in general ability to reason, in ability to
generalize, to draw valid conclusions and other similar
higher capacities. In fact such are the claims made in
greater or less degree for Latin, for mathematics, for
science and indeed for many other subjects. Which sub¬
jects as taught in the present-day high school show the most
gain in general reasoning ability? Are there any subjects
that show greater gains than others? In an extensive in¬
vestigation covering 8564 high school pupils in Grades IX,
X and XI, Thorndike (5) has answered these questions.
As initial and final tests he used tests of selective and
relational thinking, generalization and organization. The
practice period was a year’s work in high school subjects
taught in the usual way. The gains in score in the final
tests were then computed for each type of study. French,
for example, gained 2.5 score points, i.e. taking French
“ increased a pupil’s gain by 2.5 score points over what it
would be if he had taken an average mixture of studies
instead of French.” French, as a matter of fact, led the

TRANSFER OF MORAL QUALITIES 279

list followed by chemistry and trigonometry. Latin
comes a little further down along with general science
and bookkeeping. Towards the bottom of the list come
stenography, economics, cooking and sewing. But we
shall not print the list because the differences between
subjects are indeed very slight and not reliable, and a
second similar investigation might change the places of
many of them. For, as Thorndike says, “the results are
in pronounced opposition to the traditional view that
certain subjects produce much more general improve¬
ment in ability to think than others.” Latin then is not
any better in this respect than French, nor is algebra
better than bookkeeping.

Another very important conclusion emerges from
Thorndike’s investigation, namely, the value of general
intelligence in making further gains in ability to think
and reason. After one year’s work in high school the
highest one per cent in general ability gained 20.5 points,
while the lowest one per cent gained 1.5 points, regard¬
less of the studies they had taken. It is not the studies,
then, that cause differences in gain, it is the abilities of
the students taking these studies. “Those who have
most to begin with gain the most during the year. What¬
ever studies they take, they seem to produce large gains
in intellect.” And Thorndike’s final conclusion, “The
intellectual values of studies should be determined largely
by the special information, habits, interests, attitudes
and ideals which they demonstrably produce. The ex¬
pectation of any large differences in general improvement
of the mind from one study rather than another seems
doomed to disappointment.”

Transfer of Moral Qualities. — Most of our evidence
for transfer rests upon experimental work with school
studies. There are, however, a few studies dealing with
the transfer of moral or non-intellectual values. These
all point in the same direction as the studies dealing with

280

THE TRANSFER OF TRAINING

intellectual values. Honesty learned in one situation
cannot be expected wholly to transfer to a different
situation; the greater the difference the less will be the
transfer. The teaching of good citizenship in school must
not be expected to transfer wholly to out-of-school situa¬
tions. We may learn to be unselfish and play for the
team in baseball or football, but it may affect very slightly
our attitude when away from the “play-ground” situation.

Method of Teaching Affects Transfer. — We must not
forget that the amount of transfer we may find in any
given school situation will be determined by the method
of teaching. We cannot say that Latin will transfer 10
per cent to English vocabulary; or if we say this, we mean
Latin as generally taught, or Latin as taught in a specific
way. By changing our method of teaching we may change
the amounts of transfer effect to other subjects. Any
teacher could easily increase the amount of transfer
effect on English vocabulary by teaching Latin with
special emphasis on and reference to growth of words.
All this is very obvious. In our psychological terminology,
what we are doing in altering the methods of teaching a
subject is making it more or less allied to other subjects.
Emphasizing Latin words and their evolution in other
languages is making this particular kind of Latin more
allied to English vocabulary, and hence as we have seen
from the psychological experiments in the laboratory we
are to expect more transfer effect. We learn exactly what
we practice. If we practice etymology, we learn etymol¬
ogy, and where Latin and English etymology overlap
the one helps the other or in other words we have transfer.
The method by which a subject is taught will determine its
transfer value. Beware, however, of teaching Latin in
order solely to teach English Better far teach English
directly.

The Mechanism of Transfer. — How does transfer
take place? We have suggested this in all that has gone

THE MECHANISM OF TRANSFER 281

before. We spoke about subjects overlapping, being
closely allied, having things in common. The more alike
two stimuli are, the more likely are they to call forth the
same reaction. “ Ratio ” is more like “ ratiocination ” than
“vernunft” is, so that a knowledge of “ratio’ 7 is more
likely to help us in understanding “ratiocination” than
a knowledge of “vernunft” is. Bicycling is more like
driving an automobile than skating is, hence knowledge of
how to ride a bicycle may help us more in learning to
drive an automobile than knowledge of how to skate.
The stimulus of mathematics work before one with the
reaction “must do it well,” “must be accurate,” is more
like the stimulus of chemistry work before one than it is
like the stimulus of English composition, and hence we
might expect more transfer of reaction “must do it well,”
“must be accurate” to one’s chemistry work than to
English composition. But we would not expect much
transfer in any case, unless the total situations of “doing
arithmetic” and “doing chemistry” were much alike.

The common elements in two situations determine the
amount of transfer from the one situation to the other.
These elements which are common to two situations may
be identities of content or identities of procedure, and the
method of teaching will determine which identities are
most emphasized. Latin may be taught with an empha¬
sis upon the words which are like English words (iden¬
tities of content), or it may be taught with an emphasis
upon accuracy in thought and expression in general (iden¬
tities of procedure).

Of course we do not react to the common elements
alone. We react to the total situation always. And
methods of training may make us see or appreciate iden¬
tities in certain situations which other methods do not,
and so some methods obtain more transfer than others,
because they emphasize identities in certain directions.
But most important of all evidently for general transfer

282

THE TRANSFER OF TRAINING

is the general ability of the individual who is reacting.
If one possesses much general ability he is likely to note
and appreciate more identities than if he has little gen¬
eral ability. If intelligence is the ability to react ade¬
quately to a novel situation, then we may say that the
intelligent individual is able to do this better, because
he is more likely than the unintelligent to see the identities
between the new situation and previous situations and
hence make an adequate transfer to the new situation.
He solves the problem because it is like a problem he has
solved before. He sees the identities of content or pro¬
cedure, which the less intelligent individual fails to see.

Application to Educational Practice. — No subject
should be kept in the curriculum for its transfer values
alone. It must have other and independent claims and
its place in the curriculum must be considered solely in
the light of these. If some subjects resulted in much
greater general improvement in ability to think than
others, they surely should receive greater consideration.
But all the evidence is against this, so that any claim in
this respect must be waived aside until established.
Each subject, therefore, must stand upon its own feet
and be taught for its own sake. Do not teach a subject
in such a way as to attain large transfer effects to another
subject. Teach the other subject directly and you will
attain better results with less time and less labor.

Be very sceptical of the general moral value of religious
education or of the general character-building qualities
of physical education, until these claims have been more
definitely demonstrated. If the battle of Waterloo was
won on the playing fields of Eton, where was the battle
of Austerlitz won? If the college athletes play fair on the
football field is that any guarantee that they will play fair
in the class examination?

In school administration do not jump to the conclu¬
sion that a good first grade teacher will naturally make

SUMMARY

283

a good sixth grade teacher or a good elementary teacher
will make a good high school teacher, or a good teacher
will make a good principal and so on. The recognition of
general goodness is not enough.

Summary

1. The problem of transfer is the problem of reciprocal mod¬
ification. What influence does a given modification have
upon other modifications?

2. The world at large has learned to assume transfer in some
cases, but not in others.

3. Formal discipline is the belief in a very general and wide¬
spread transfer. Belief in this theory of formal discipline
has influenced educational practice very considerably.

4. If the utilitarian value of a subject is attacked, it generally
lays claims to disciplinary values.

5. Belief in wide transfer is still common among many educa¬
tors and much that is taught is still defended on the grounds
of transfer values.

6. Psychological investigation of transfer values has introduced
a definite experimental procedure, with initial and final
tests flanking the practice period, with control and experi¬
mental groups equated.

7. The results of laboratory studies show considerable transfer
between narrow allied functions; but little transfer between
other functions.

8. The results of school experiments dealing with much broader
functions show small transfer effects. There may be trans¬
fer from one school subject to another. The fact of transfer
is obvious. The amount in general is very small.

9. No one high school subject, as at present taught, is markedly
better than any other in increasing a pupil’s general ability to
reason, to generalize, to carry on abstract and relational
thinking.

10. The most important factor in the increase of general ability
to think is the general intelligence already possessed by the
student and not the particular course of study he may
take.

284

THE TRANSFER OF TRAINING

11. The transfer of moral qualities seems to follow the same
trend as the transfer of intellectual qualities. In general
we greatly over-estimate the amount of transfer of such
qualities.

12. The method of teaching affects the amount of transfer to
any given situation. It may do this by emphasizing com¬
mon elements of content or common elements of procedure.

13. The mechanism of transfer is by means of the elements com¬
mon to two situations. And these elements may be either
of content or of procedure and the amount of similarity per¬
ceived is determined by the method of teaching and the in¬
telligence of the learner.

14. In school we should never rely on the transfer values of a
subject. To pin our hopes on much transfer is to court dis¬
appointment.

REVIEW

285

Review

True-False Statements

As a review mark the following statements true or false:

1. It is doubtful whether any particular subject rather than
another is an effective instrument for improving generally
the mind’s ability to think.

2. Individual differences in transfer of training depend upon
degree of intelligence.

3. The study of English grammar leads to improvement in
written composition and in speech.

4. The residual gain in a transfer experiment measures the
real amount of transfer.

5. Teaching the theory has never been found to cause improve¬
ment in the actual practice of a function.

6. If you wish to improve your memory, it is well to practice
memorizing lists of numbers or dates.

7. The study of mathematics increases one’s general accuracy.

8. Training in addition may show transfer to multiplication.

9. Training in Latin helps one in his knowledge of English
words.

10. There may be transfer of attitudes as well as transfer of
content.

11. The method of teaching a subject will not influence the
amount of transfer to other subjects.

12. The disciplinary value of Latin justifies its retention in the
curriculum.

13. Any change in the concrete particulars reasoned about will
interfere somewhat with our reasoning.

14. Pupils having different amounts of mental ability tend to
make about the same amount of improvement on general
reasoning tests after one year of high school work, regard¬
less of the course of study they follow.

15. Spread of improvement diminishes as functions become
more unlike.

16. The increase in knowledge of English words through the
study of Latin is an example of transfer by means of identi¬
ties of procedure.

286

THE TRANSFER OF TRAINING

17. To promote a professor, distinguished for research, to the
position of dean indicates a belief in a great amount of
transfer.

18. Underline the best response:

Transfer of training from one function to another would
seem to take place by means of: (1) generalization; (2) men¬
tal exercise; (3) identical elements; (4) ideals.

19. Check the best response:

The most important factor in general mental improvement
regardless of the subject studied is:

(a) mental set.

(b) intelligence.

(d) concentration.

Advised Readings

Starch, D.: Educational Psychology, New York, 1927, Chapter
14.

Thorndike, E. L.: Educational Psychology: Briefer Course,
Teachers College, Columbia University, New York, 1914,
Chapter 18.

References

1. Caldwell, O. W., and Courtis, S. A.: Then and Now in

Education, Yonkers, 1924, p. 52.

2. Mais, S. P.: A School Master’s Diary, London, 1918.

3. Symonds, P. M., and Penney, E. M.: “The Increasing of

English Vocabulary in the English Class, ” Journal of
Educational Research, 1927, 15, 93-103.

4. Thorndike, E. L.: “The Effect of Changed Data upon

Reasoning,” Journal of Experimental Psychology, 1922,
5, 33-38.

5. Thorndike, E. L.: “Mental Discipline in High School

Studies/’ Journal of Educational Psychology, 1924, 15,
1-22 and 83-98.

CHAPTER XIII

CONTINUOUS EXERCISE WITHOUT REST

Fatigue. — So far we have always assumed that in try¬
ing to make any improvement, the exercise has been car¬
ried on for given periods with rest, change of occupation
and sleep intervening. We now raise the question as to
what happens to a learning or improvement curve, if the
exercise is continuous. In other words how long can a
person work at a task without showing decrease in effi¬
ciency? If we practice anything continuously without
stopping, will the improvement curve continue to mount
indefinitely? Of course we know it will not. Something
happens that we call fatigue. At any rate individuals
stop working relatively soon, because they become rest¬
less, bored, uncomfortable and they seek other occupa¬
tions or rest. The length of time any given individual
will work continuously at any task varies enormously
according to his interest in the task, his desire to attain
a given end, the external pressures and compulsions
holding him down to the task and the like.

Mental and Physical Fatigue. — Mental and physical
fatigue are two names for the same phenomenon. There
is no real difference between them. Fatigue is something
that lowers the efficiency of the organism for any kind
of work. If the work has mainly to do with the large
muscles of the body, as in walking or sawing wood or
playing golf, the discomfort arising from continuous ex¬
ercise is called physical fatigue. If, on the other hand,
the work involves none of the larger muscles and is mainly
concerned with thinking or reacting with speech or sub¬
vocal movements, as in multiplication or studying Latin

287

288 CONTINUOUS EXERCISE WITHOUT REST

or writing a dissertation, the discomfort arising is called
mental fatigue. There is no sharp line dividing the one
from the other. The educational psychologist is mainly
concerned with the latter type of reaction, and decreasing
efficiency due to continuous exercise of this sort is what
we shall call mental fatigue.

Measurement of the work of a worker is difficult. There
are two ways in which this can be measured. The first
and simplest and most objective is to measure the quan¬
tity and quality of the product produced, as in tasks in
arithmetic we may measure the number of tasks attempted
and the correctness of these tasks. The second way is to
consider the amount of effort put forth by the worker,
the discomfort and obstacles overcome by him. One
pupil may work harder than another, but produce less
work. This is obvious, and in school the teacher may
take cognizance of this so far as he can. He may give a
dull child credit for effort put forth. Indeed, the A.Q.
technique is an attempt to provide a measure of this
kind, for it tries to evaluate the product in terms of the
general intelligence of the child. Two children producing
the same amount and quality of work may have very
different A.Q.’s. This same work may be good with
reference to one child’s capacity, but poor with reference
to another’s.

The Laboratory Experiments. — Let us glance now at
the results of the measurement of fatigue in the psycho¬
logical laboratory. Table 6 gives a condensed summary
of many experiments arranged in chronological order.
Notice that the function tested or the work done is re¬
stricted to the type of reaction which we popularly say
involves a lot of thinking. The reactions called for are
such as involve the finer musculature of the body. They
do not involve much overt muscular movement. They
are more mental than physical, because we are interested
in the effects of mental fatigue. The arithmetical opera-

THE LABORATORY EXPERIMENTS 289

tions of adding and multiplying are the most commonly
used reactions. Notice also that the duration of time is
generally rather short. This is particularly true of the
earlier experiments. The later experiments deal with
longer time periods and we shall see presently the reasons
for this. The subjects used in the experiments are gen¬
erally few in number and in several cases the subject is
the experimenter himself. It is difficult to obtain sub¬
jects who will consent to undergo long, monotonous and
hard work.

Now if we read down the column of results, we are
first of all struck by the contradictory character of the
results. Some experimenters report a considerable fatigue
effect and others none. One hour of addition is reported
as showing considerable fatigue whereas another experi¬
menter reports no fatigue effect. Some of this discrep¬
ancy is due to the elusive character of the thing that is
being measured, the differences in the amount of interest
in the task, and the differences in the amount of effort
put forth. If a subject knows that the experiment is for
the purpose of obtaining a measure of fatigue, if the sub¬
ject believes that mental fatigue shows itself quickly,
he is likely to be unconsciously influenced in his attitude
towards the work. We can readily produce a fatigue
curve if we are anxious to do so.

The earlier experiments were more or less ambiguous
as to the rapid onset of fatigue. Thorndike’s first ex¬
periment in 1900 with work periods of 3 to 8 hours showed
no measurable fatigue results. This made him sceptical
of the fatigue produced after short periods of work.
Hence when he returned to these fatigue experiments in
1910 and 1911 we note that he lengthens the work period
and also makes the work more difficult, i.e. by introduc¬
ing mental multiplication of three place by three place
numbers. When he does this he begins to find slight but
definite fatigue effects. This work of Thorndike cul-

290 CONTINUOUS EXERCISE WITHOUT REST

TABLE 6

Summary op Laboratory Fatigue Experiments

Experimenter Date

Reaction

Burgerstein

Adding

multiply¬

ing

Bergstrom

Translat¬
ing Ger¬
man

Holmes

Addition

Bettman

Reactions

Oehrn

Counting,

memory,

adding,

writing,

reading

Weygandt

Addition

Vogt

Addition

Vogt

Memory
for digits

Thorndike

Cancella¬
tion, areas
memory,
complex
mental
work

Duration

Subjects

Result

1 hour

Children

None; per¬
haps gain

4 hours

Self

None

36 min.

Children

Doubtful

or none

1000 trials

Adults

Slight in¬
crease in
time

2 hours

10 adults

Very slight
fatigue ef¬
fects. In¬
terference
in nonsense
syllables

90 min.

3 adults

Consid¬
erable. One
fourth bet¬
ter after
rest

90 min.
daily

Self

Consider¬
able. 24%
better after
rest

75 min.
daily

Self

9% better
after rest

3 to 8
hours

Wood-

worth

None

THE LABORATORY EXPERIMENTS 291

TABLE 6 —Continued

Experimenter Date

Reaction

Lindley

Addition

Bolton

Addition

Kafemann

Addition

Henmon

Addition

Specht

Addition

Whipple

Cancella¬

tion

Thorndike

Multipli¬

cation

3 place X

3 place

Thorndike

Addition

Arai

Multiplica-'
tion 2 place
X 2 place

Arai

Multiplica¬
tion 4 place
X 4 place

Starch & Ash

Addition

Ritchie

Multiplica¬

tion

Duration

Subjects

Result

1 hour

3 adults

None

1 hour

Adult

Consid¬
erable de¬
crease

90 min. 8
days

Self

2% better
after rest

1 hour

Adults

None

Adults

Marked
fatigue
even after

10 min.

2 hours

Self

None

4 to 12
hours

Adults

21-59%
more time
than after
rest

\y 2 to 2

hours

Fatigue
effect of

2 hours

Adults

Increased

efficiency

12 hours

Self

Increase in
time—
119%.
efficiency
equals 75%
of initial

2 hours

Adults

4.3% loss

10 hours

Adults

Slight drop
in number
done.
Marked
rise in er¬
rors

292 CONTINUOUS EXERCISE WITHOUT REST

Figure 69. — Work Curve for Mental Multiplication. The vertical
axis represents the number of 4-place by 4-place examples worked per
hour of working time. The horizontal axis represents the successive
hours of work from 11 a.m. until 11 p.m. (Drawn from date of Arai,
Columbia Contributions to Education, No. 54, after Robinson.)

minates in the work of his student, Arai, who still further
increased the difficulty of the work and the length of the
work period. Arai worked continuously for twelve hours
on four successive days at a hard mental task, namely
the mental multiplication of four place by four place
numbers. She reduced as far as possible every physical
reaction. She merely looked at the card with the two
series of numbers until she had memorized them, so as to
avoid eye strain. Then she made the mental calculation
and when this was finished she wrote down the answer.

In this way such physical reactions as writing and looking
were reduced to a minimum. After twelve hours of this
severe mental application, the subject was still able to
multiply although she was working much more slowly
and making a great many more errors. Figure 69 shows
the number of examples worked per hour during the
twelve-hour period. The number done decreases some¬
what at first, but towards the end remains surprisingly
constant for many hours. At the end of this long period

THE LABORATORY EXPERIMENTS 293

she was still able to multiply. She was not fatigued to
the extent of not being able to multiply anything at all.
Thorndike estimates her probable efficiency to be about
75% of her total multiplying ability. Evidently, then,
mental fatigue does not arise as readily and as quickly as
is popularly supposed. Given an adequate motive, an
individual can work for many hours at a mental task
without loss in efficiency and for very many hours without
great loss in efficiency. After twelve or sixteen hours the
problem will become complicated by the onset of sleepi¬
ness. But within the ordinary waking period we may
justly say that it is hard to obtain real mental fatigue
under experimental conditions.

We will note, however, the short-comings of our ex¬
periments, particularly as regards the type of work. This
always is a more or less mechanical type of work, doing
the same type of thing over and over again. This is
chosen so that efficiency from hour to hour may be accu¬
rately measured. Would fatigue show itself earlier and
more definitely with more creative work, such as planning
a campaign, writing a book, composing music and the
like? It is impossible to answer this. Such things are
hard to measure as they are being done from hour to
hour. Furthermore, twelve hours spent in writing poetry
is a mixture of many different kinds of activities, and how
much time is involved in pure creation is doubtful. Again
from accounts of authors, scientists, composers, who have
worked continuously for long periods of time when the
divine afflatus was upon them, we may surmise that the
amount of mental fatigue would not likely be greater
than such as would arise from the very dry and monoto¬
nous task of multiplying mentally four place by four place
numbers. The reader might well try how long it takes
him to multiply 6483 by 2795 without using pencil or
paper. If pure creative work could be isolated under
laboratory conditions and if a worker could be made to

294 CONTINUOUS EXERCISE WITHOUT REST

work at it continuously, without dropping down from
the creative level, it is possible that fatigue would show
itself earlier than it does in hard but routine mental
work.

To sum up the results of the laboratory experiments,
we may rightly say that mental fatigue is slow in onset
and that an individual can work several hours without
rest without any appreciable loss of efficiency. The usual
restlessness which arises after a few hours of mental work
is not mental fatigue, but discomfort, boredom, desire to
do something more attractive.

The School Experiments. — Fatigue in adults in the
laboratory may be hard to register. The situation may
be different with children in school. Does not the or¬
dinary work in school tire the child? Does not mental
fatigue increase during the school day and leave the
child less efficient at the end? Several experimenters
have raised this problem and the results of their work
are summarized in Table 7. The usual procedure is to
give similar tests at the beginning of the school day and
at the end, and then to measure any difference that may
be present. Sometimes similar tests are given at differ¬
ent periods during the school day for several days and
efficiency at 9 o'clock, 10 o'clock and so on up to the last
period measured.

Now if we read down the column headed “Results" in
Table 7, we are struck by the fact that little or no decrease
in efficiency is reported. Evidently children are able
to work (we do not say they actually do work) as effi¬
ciently at the end of the school period as at the beginning.
The work of the ordinary school day does not decrease a
child’s efficiency in such “mental” work as multiplica¬
tion, addition, dictation and the like.

Now this seems to contradict the experience of many
teachers. They report that children are tired out at the
end of the school day; that they cannot do “hard” sub-

THE SCHOOL EXPERIMENTS

295

TABLE 7

Results of School Fatigue Experiments

Author

Date

Tests

Results

Sikorski

Writing from dictation

Errors increase slightly

Bolton

Memory for digits

No loss in efficiency

Laser

Addition and mul-

Work improved (prac-

tiplication

tice)

Friedrich

Addition, multiplica¬
tion, dictation

Very slight decrease

Ebbinghaus

Memory for digits.

No difference. No

Addition, multiplica-

real difference.

tion. Completion ex-

Decrease in effi-

ercises

ciency (?)

Ritter

Marking letters. Dic¬
tation

No difference.

Dictation of unrelated
words

Slight decrease

Thorndike

Adding, multiplying,

No difference between

cancelling, memorizing,
etc.

late and early work

King

Addition, multiplica¬
tion, judgment

No difference

Heck

Arithmetic

Increase in amount;
decrease in accuracy

Dawson

Multiplication

No difference

jeets, such as arithmetic as well as they can in the fore¬
noon. The psychologist believes that what is happening
here is not fatigue at all, but physical restlessness due to
constraint, boredom due to uninteresting subject-matter,
increasing desire to play as the time for play draws near.
Many a teacher has been surprised to notice how effi¬
ciently children can and will work even at the end of the
day when the thing they are doing really interests them.
There is no question of fatigue then. Some of the more
progressive schools report a difficulty of getting children

296 CONTINUOUS EXERCISE WITHOUT REST

to go home, so much do they love their school work. No
question of fatigue there. By monotonous senseless drill,
by uninteresting subject matter, by stupid teaching we
may bore our pupils and drive them to hate school and
all learning, but even then we are not likely, during the
few hours of school each day, to occasion mental fatigue.
The chances are that few, if any, students in school or
college ever become mentally fatigued.

Nervous Breakdown. — It is frequently said that this
student or that student has had a nervous breakdown
because of studying too much, because he is mentally
fagged out. In all such cases it can readily be shown
that mental fatigue was not the cause of the breakdown.
In fact in most cases the amount of study done is woe¬
fully far from producing mental fatigue. It is not the
study that produces the breakdown, but the worrying
about the study. There are as many mental breakdowns
among indifferent students as among assiduous students.
A mental collapse may be brought on by physical dis¬
turbances, by emotional up-sets, by fears and worries,
but hardly by pure mental fatigue. Not study, but worry
about study and more likely worry about something else
causes the breakdown.

Feelings of Dissatisfaction. — Another interesting re¬
sult of the psychological experiments on fatigue is the
suggestion that the satisfyingness of a task is not directly
correlated with the efficiency of the worker. We give up
unsatisfying tasks if we can, not because they produce
mental fatigue, but because they are unsatisfying. If we
persist in spite of the unsatisfyingness, we are frequently
as efficient as when the task was satisfying. Figure 70
shows an attempt to measure efficiency at a task and
feelings of satisfaction at the same time. Several sub¬
jects worked for about five and a half hours at a mental
task divided into 15 units. The solid line of Figure 70
shows that their work did not decrease in efficiency as a

PerCent Maxtmum 5cone

SUGGESTIONS FOR THE SCHOOL 297

Extremq/y

1 Good

Extnenefy
7 Timtt

Figuee 70. — Relation between feelings and output. (From
Poffenberger’s Applied Psychology.)

whole. After each unit they rated their feelings from
“extremely good” to “extremely tired,” and the broken
line curve in the figure shows how more and more unsat¬
isfying and boring the work became. They felt tired, but
their efficiency was not decreased. If made to work, an
individual may be just as efficient as when he works
because he wants to. It is not a desirable state of affairs.
But it should be a suggestion to us, so that we do not
give up our own work as soon as we lose zest for it, and
excuse ourselves with the rationalization that it is not
efficient because it is unsatisfying. Although school work
should as far as possible be made satisfying to the pupils,
not all work in life can be so made. A pupil should learn
to work at unsatisfying tasks, if necessary, and be shown
that such work may be really efficient.

Suggestions for the School. — For school purposes then
we should adhere to our general policy of placing the less
interesting subjects at the beginning of the school day,
not because pupils are mentally fatigued at the end of
the day, but because they are less bored and less restless
at the beginning. Notice that we lay stress on interest
in a subject rather than hardness. A “hard” subject,
like arithmetic, may be successfully taught and learned
at the end of the day, if the pupils are vitally interested
in it, and occasionally a teacher can be found who knows
how to make arithmetic an interesting and vital subject.

298 CONTINUOUS EXERCISE WITHOUT REST

Amount of interest, freedom from boredom, rather than
the difficulty of a subject should be our guide in deter¬
mining the order of subjects studied during the day. This
order of interest will differ from teacher to teacher and
from school to school. The point that the psychological
experiments have made clear is that we need not take into
account the factor of mental fatigue in determining the
order of studies.

Intense Motivation and Fatigue. — Mental fatigue,
producing a loss of efficiency in mental work, does not
readily appear even after several hours continuous work
under normal pleasant conditions. Intense motivation
may postpone a drop in efficiency in mental work even
under abnormal conditions. Knight and Remmers (3)
have shown how the intense desire to make a college
fraternity may keep men at a high degree of mental
efficiency in spite of loss of sleep and unpleasant physical
conditions. During the five day’s initiation to a frater¬
nity, several college students underwent severe physical
exertion with only one or two hours sleep each night.
At the end of this period they were given addition tests
and told that unless they did well on these tests they
would not make the fraternity. The motive for doing
well on these tests was very strong. The tests consisted
of 120 minutes continuous work with a record of each of
the twenty-four periods of five minutes each. The re¬
sults show a very slight loss in efficiency at the end of this
two hour period. The average per cent correct for the
first 12 periods was 85; for the last 12 periods 83. The
total amount accomplished during the first 12 periods
was 256 examples; during the last 12 periods 249 exam¬
ples. Furthermore, these tired students under the pres¬
sure of this strong motive did twice as well as other col¬
lege students taking the test under ordinary class-room
conditions with no strong incentive. Evidently, there¬
fore, mental fatigue is not only slow in appearing under

FATIGUE IN INDUSTRY

299

ordinary conditions but it can be counter-acted by a
strong incentive under very abnormal conditions.

Fatigue in Industry. — We cannot in this book go into
the extensive investigations of fatigue which have been
carried out in industry. We may note, here, however,
that industry in general is not interested in the question
of mental fatigue, but rather in the general problem of
lack of efficiency, whether this be due to mental or physical
factors or feelings of boredom, restlessness or any mixture
of these. Physical factors are of much greater import
here than in school work.

p&vmmse

Figure 71. — Distribution of output over working day. Work of
Italian typesetters. (From Muscio’s Industrial Psychology.)

Measurements such as those shown in Figure 71 are
of value to industry. The work curve rises in the fore¬
noon and then gradually descends. During the afternoon
the descent is continuous. Does fatigue, boredom, or
loss of interest affect the attention and care a worker
gives to his work? Figure 72 shows the number of acci-

300 CONTINUOUS EXERCISE WITHOUT REST

dents according to the time of occurrence during the day.
The greatest number of these occur just before the noon-
hour rest and just before the close of work at the end of
the day. Fatigue, whether mental or physical, lack of
interest and feelings of unsatisfyingness increase as the
day wears on and the total effect of these factors leads
to a greater number of accidents.

&e*tcntage FATIGUE

Figure 72. — Shows the time of occurrence of industrial accidents.
I in Germany in 1887; III in Lancashire, England. (From Muscio’s
Industrial Psychology.)

The Effect of Loss of Sleep. — Closely allied to the
general question of what happens when we work con¬
tinuously without rest is the question of what would
happen if this lack of rest is taken to include lack of
sleep. Sleep is one of the ordinary phenomena of life
that we take for granted, but just what it is, is very diffi¬
cult to determine. There are numerous investigations,
physiological and psychological, with regard to it, and
there are many hypotheses as to how it is caused and as
to what it really is. Into all these we cannot enter here.

VENTILATION

301

We shall merely touch upon the question as to what
happens psychologically when people go without sleep
for longer or shorter periods.

With a group of college students it was found (5) that
one night's loss of sleep was not sufficient to interfere
with their efficiency on the ordinary intelligence test.
Although many said they were tired, this feeling of tired¬
ness did not make them do poorer work on the test. Even
two nights’ loss of sleep has been shown (4) not to cause
any decrease of efficiency in mental work, although all the
subjects report the necessity for putting forth greater
effort after such a deprivation of their usual sleep. How
much more loss of sleep an individual can sustain without
showing a decrease of efficiency we do not know. Much
will depend upon the subject’s general state of health and
what he is doing in the intervals between tests. The
strength of the motive to work will also be a determining
factor. What the experiments so far do demonstrate, is
that one or two nights’ loss of sleep is not a very serious
disturbance for the individual. Evidently the organism
has a great deal of ability to adapt itself to unusual con¬
ditions. In no sense, however, should these experiments
be interpreted to mean that students can study all night
without harm. This may be possible, but in general the
safe rule is to get as much sleep as seems necessary.

Ventilation. — We all know that it is much pleasanter
to work in a well-ventilated room, where the temperature
is not too high and the humidity is not too great. If the
room becomes hot and “ stuffy,” we begin to feel uncom¬
fortable. We feel drowsy, disinclined to work and fre¬
quently we complain of headaches and other minor dis¬
comforts. In a badly ventilated classroom we notice
the flushed faces of the children. They are frequently
restless and not inclined to work. Experts tell us that
optimum conditions in ventilation call for a temperature
of about 68 degrees Fahrenheit, a humidity of 60 per cent

302 CONTINUOUS EXERCISE WITHOUT REST

and 45 cubic feet of outside air per person per minute.
Nevertheless, it has been shown by Thorndike that indi¬
viduals can do efficient mental work under much less
desirable conditions, namely with a temperature of 86
Fahrenheit, a humidity of 80 per cent and with no fresh
air or movement of air. Under such uncomfortable con¬
ditions good work can be done, and furthermore this can
be kept up for several days. Manual work would seem
to decrease in efficiency under bad ventilation quicker
than mental work. We do not know what the effect on
the organism would be if work were continued under such
unfavorable conditions for several months or years at
a stretch.

This ability of the organism to work effectively under
adverse atmospheric conditions falls in line with its
ability to withstand the effects of fatigue and loss of
sleep. It is another indication of the great adaptability
of the human body. We must not, however, draw the
conclusion that ventilation is of no importance in school.
It unquestionably is. Rut, on the other hand, we must
not allow changes in temperature or uncomfortable at¬
mospheric conditions to be brought forward as excuses
for poor work. If such atmospheric conditions are not
too unpleasant, we should neglect them and go on work¬
ing with the belief that our work will not suffer because
of them. Thousands of summer schools testify that stu¬
dents can do good work when the thermometer is high
and the humidity great, although the conditions for
work may not be as comfortable as during the regular
semester.

These psychological experiments in ventilation must
not be used as an excuse for badly-ventilated school build¬
ings. It is more comfortable to work in a well-ventilated
room with a moderate temperature and with abundance
of fresh air, and we must insist upon optimum conditions
as far as we can. It would be as unreasonable to con-

THE EFFECTS OF DRUGS

303

elude that ventilation does not matter, as it would be to
conclude from our sleep experiments, that sleep does not
matter and that an individual ought to make a habit of
staying up all night every two or three days.

The Effects of Drugs. — Closely connected with the
effects of fatigue and loss of sleep are the psychological
effects caused by drugs.

Alcohol has a depressing effect upon the organism, con¬
trary to the popular belief that it is a stimulant. Accord¬
ing to the dose administered it retards in general the
reactions of the individual. It causes a loss in efficiency
in such tests as tapping, color naming and finding the
opposites of words. Individuals vary greatly in their
ability to tolerate alcohol. Holiingworth (1) found that
resistance to the drug is positively correlated with general
competence. The person of high general ability may
possess a more stable nervous system, which is less readily
affected than the less stable nervous system of the less
competent individual.

Caffein in small doses is slightly stimulating and has
been found to increase somewhat the efficiency of mental
work. Large doses, however, soon lead to a loss of effi¬
ciency. Coffee may be the cause of sleeplessness. Ordi¬
narily, however, this is caused more by expectation than
by the coffee. We are unable to go to sleep because we
expect not to go to sleep readily after drinking a cup or
two of coffee. Holiingworth showed that loss of sleep
was caused as much by giving syrup with no caffein as
with doses of 2 or 4 grains of caffein, which correspond
roughly to one or two cups of coffee. Larger doses of
caffein consisting of 6 grains, or three cups of coffee, do
interfere greatly with an individual’s normal sleep habits.

Table 8 shows Hollingworth’s comparison of the effect
of alcohol and coffee. The plus signs indicate increase in
ability on the tests and the minus signs indicate decrease.
The figures indicate per cent of the work gained or lost,

304 CONTINUOUS EXERCISE WITHOUT REST

TABLE 8

Comparing the Effect of Alcohol and Coffee on Various

Tests (after Hollingworth)

Bottles

Grains

3-4

5-6

6-9

3-4

Beer

Caffein

Pulse

+ 8

+ 10

+ 19

Steadiness

—68

—241

—370

—2.0

—583.0

Tapping

— 7

— 13

— 14

+3.1

+ 3.8

Coordination

— 6

— 10

— 20

+0.3

— 4.0

Color Naming

— 2

— 7

— 12

+7.0

+ 4.4

Opposites

— 5

— 12

— 23

+5.1

+ 6.2

Adding

—10

— 15

— 16

+1.8

+ 2.5

Substitution

Memory

Z/l

based upon records taken without drugs. Thus we note
that in all the tests, motor or mental, alcohol leads to
decreases in efficiency. The only increase caused by
alcohol is in the pulse rate. Coffee, on the other hand,
leads to a slight increase in efficiency in most of the
tests, the most noticeable decrease being in the steadiness
test.

Tobacco has been difficult to experiment with, because
it is difficult to introduce the drug without the knowledge
of the subject. In all these drug experiments, it is abso¬
lutely necessary to disguise as far as possible the purpose
of the work, because subjects are so much influenced by
preconceived notions as to what ought to happen to them.
One experimenter (2) has, however, overcome the diffi¬
culty with tobacco by means of an experimental pipe
with electrically heated hot air. When blindfolded none
of the subjects were able to detect the difference between
the experimental pipe with hot air and the real pipe with
tobacco, while someone else was smoking in the same

SUMMARY

305

room. The immediate effect after smoking a pipe of
tobacco on smokers seems to be slightly favorable in such
mental operations as addition, where old habitual bonds
are functioning; but the effect seems to be unfavorable
for learning, where it is a question of the formation of
new bonds. Non-smokers are affected unfavorably in
most tests. It would seem, therefore, that such mild
drugs as coffee and tobacco, as usually taken, exert very
little influence on mental operations.

We have no authentic experiments as to the influence
of tobacco or coffee on children. Presumably, such effects
would be less favorable than with adults. In general,
therefore, such drugs should not be given to children. As
children grow older the effects of such drugs, as far as
we know them, should be told to the adolescent. He
should not be frightened by monstrous stories of the evil
effects of tobacco, for example. Such exaggeration very
readily defeats its own purposes. He will soon find out
the untruth and his confidence in the teacher as an au¬
thority in other matters will be undermined. Tobacco
may not be very good for high school boys, but it may
not be very bad. It is, however, undoubtedly much
better for them to smoke openly in public than to dis¬
guise their practice by lying and subterfuge.

Summary

1. There is no real difference between mental and physical
fatigue.

2. All work, if carried on continuously without rest, tends to
become less and less efficient, but the point at which loss
of efficiency begins is often difficult to determine.

3. Feelings of dissatisfaction or of boredom and the like are
not good measures of the onset of fatigue or of decrease in
efficiency.

4. Laboratory experiments in fatigue show us that an individ¬
ual may work several hours at a difficult task without serious
loss of efficiency.

306 CONTINUOUS EXERCISE WITHOUT REST

5. The school experiments in fatigue show us that loss of effi¬
ciency in mental tasks does not develop because of fatigue
during the ordinary school day.

6. It is not necessary to consider the possible fatigue effects
of school subjects in arranging the daily schedule. Interest
and liking for a subject are more important than possible
fatigue effects.

7. Nervous breakdowns are not caused by mental fatigue.
The emotional factors in an individual’s life are the causes.
Worry about study, rather than too much study, is the po¬
tent factor.

8. Feelings of dissatisfaction or tiredness are not a good meas¬
ure of efficiency.

9. Loss of efficiency in work may be postponed for a long time
by intense motivation.

10. Industrial fatigue, as measured by output of work, includes
mental and physical fatigue as well as the dissatisfactions
and annoyances caused by continuous application to a task.

11. One or two night’s loss of sleep does not seem to have any
marked effect on mental tasks.

12. Drugs vary enormously with reference to their effects on
mental work. Individual differences are marked here; the
generally competent individual seems able to resist longer
than the less generally competent.

13. Alcohol, even in small doses, acts as a depressant rather
than as a stimulant.

14. Coffee in the customary small doses acts as a stimulant.

15. Tobacco seems to have very minor effects on habitual
smokers.

REFERENCES

307

Review

True-False Statements

As a review mark the following statements true or false:

1. Experiments show that children do best on arithmetic dur¬
ing the first period in the morning.

2. Mental “breakdowns” are generally caused by worry rather
than by excessive mental work.

3. After four hours of mental work it is impossible to continue
without serious loss of efficiency.

4. The effect of smoking on smokers in mental tests is always
unfavorable.

5. The curve of satisfyingness always follows the curve of
achievement.

6. The mental efficiency of school children on mental tests
seems to be fairly constant during the school day.

7. Two nights’ loss of sleep will not affect one’s efficiency, if
there is strong motivation.

8. Mental fatigue increases rapidly in the course of the usual
school day.

9. We never find increased efficiency after two or three hours
continuous work.

10. All work curves show an initial spurt.

11. Small doses of alcohol are stimulating and lead to increased
scores on psychological tests.

References

1. Hollingworth, H. L.: “The Influence of Alcohol,” Journal

of Abnormal and Social Psycholoqij, 1923-24, 18, 204-
237 and 311-333.

2. Hull, C. L.: “The Influence of Tobacco Smoking on Mental

and Motor Efficiency,” Psychological Monographs, Vol¬
ume 33, Number 3, 1924.

3. Knight, F. B., and Remmers, H. H.: “Fluctuations in Men¬

tal Production when Motivation is the Main Variable,”
Journal of Applied Psychology, 1923, 7, 209-223.

4. Laslett, H. H.: “An Experiment on the Effects of Loss of

Sleep, "Journal of Experimental Psychology, 1924, 7,45-58.

5. Robinson, R. B., and Robinson, F. R.: “Effects of Loss of

Sleep,” Journal of Experimental Psychology, 1922, 5,
93-100.

CHAPTER XIV

MEASURING IMPROVEMENT IN SCHOOL

SUBJECTS

School Improvement. — The improvement that the
school is most interested in is that which it can bring
about in its pupils. The school is an agency organized to
bring about certain changes in the behavior of individuals.
It is one of the many institutions trying to modify human
behavior. Society determines what modifications the
school should endeavor to make. The modifications
thought desirable change from time to time and from na¬
tion to nation, as the history of education shows. The
Greeks thought improvement in music and gymnastics was
important; in the middle ages the schools considered im¬
provement in Latin of first importance. None of these
improvements are considered extremely vital in our
schools to-day. Educational psychology attempts to
study the changes or modifications which the school is
trying to make in its pupils, whatever those modifications
may be.

So far our study of modification or learning has been
general, in an attempt to understand the universal prin¬
ciples at the basis of all learning. In this chapter we will
be concerned with the specific modifications which the
school is trying to make. The changes in pupil behavior
which the school is trying to make are generally thought
of as subjects which the pupil is learning. The modern
school is, however, not only interested in teaching certain
subjects in the narrow sense, but also in modifying the
behavior of the child in regard to all those things summed

308

ACHIEVEMENT TESTS

309

up under such a phrase as “desirable citizenship.” The
school is attempting to produce individuals who will
know certain things, who will act in moral and desirable
ways, who will have certain appreciations and so on.
Much of this is necessarily vague and particularly so
when we descend from the abstract generalization to the
concrete particular instance. Nevertheless, as the school
is organized to-day, we see it very specifically trying to
teach certain things, and so the psychologist has started
to measure those specific things, hoping to move along
to the measurement of the more intangible effects of
schooling later on.

We really should not call them intangible, because any
change made in a child will show itself in a change in his
reactions. Tests can be constructed to measure what¬
ever changes the school determines to make in a child.
At the present time schools emphasize the acquirement
of certain knowledges and skills. Hence the psychologist
has set about the measurement of such knowledges and
skills.

Achievement Tests. — In this way there have arisen
many tests for school achievement. They are measures
of the modifications brought about by the school. The
score of any one such test represents for any given child
a point on a learning curve for that particular function.
If the tests are repeated over a period of time, the scores
could be used to make learning curves for the function in
question. Educational tests measure the improvement
that the school has made up to a given time. They show
what has been accomplished with the methods and mate¬
rials used. They cannot show what might be accom¬
plished with better methods or better management. They
give a picture of the existing state of affairs. They have
arisen and are in widespread use to-day, because they
are accurate measures and because they are standardized
measures. They are more accurate than the ordinary

310 IMPROVEMENT IN SCHOOL SUBJECTS

school examination. They are free from the personal
bias of the teacher and their results can be compared
with the results obtained in other schools all over the
country.

The point we wish to emphasize here, however, is the
important fact that all such tests are attempts to measure
the modifications imposed upon original nature. They
answer the question, “How much change has been ef¬
fected?” Educational tests repeated from time to time
give us a modification curve or learning curve for a par¬
ticular individual. They show the rate and degree of
modification taking place. Norms are the average amounts
of modification achieved under present school conditions
for children in general.

Learning Curves of School Subjects. — If we test a
child again and again in a school subject, we will get a
learning curve for that subject, just as we obtained
learning curves for different functions, such as writing
the alphabet backwards, typewriting and so forth in
Chapter IX. Ordinarily exact records are not kept of a
child’s progress in any school subject. As the child moves
from grade to grade the examinations become increas¬
ingly difficult and such changes in examinations make it
impossible to construct a learning curve. However, by
the use of standard educational tests, we may measure
the progress of any child.

Figure 73 shows the progress curve in reading for a
child who had been given the Stanford Achievement Test
seven times at irregular periods during four years. The
units of measurement here used are the educational age
norms for the test. We notice an almost continuous in¬
crease in reading ability from July, 1923, to June, 1927.
On the same figure there is also given the curve for arith¬
metic. The improvement here is steady from July, 1923,
to June, 1926. There is a slight drop in the last test given
in June, 1927.

LEARNING CURVES OF SCHOOL SUBJECTS 311

Figure 73. — Improvement in Reading and Arithmetic for one

child.

Figure 74 shows two educational achievement curves
for two other children tested by the Stanford Achieve-

Figure 74. — Improvement in general educational achievement

for two children.

ment Tests over a period of four and a half years. Again
we note a fairly continuous and steady improvement
with the exception of one test. It is interesting to note
in the case of these three children the tendency to slow
up or drop back in the June, 1927, test. Up to June, 1926
these three children had a very individual type of instruc-

312 IMPROVEMENT IN SCHOOL SUBJECTS

tion which took into account their special needs and
peculiarities. This was followed by a year in the public
schools with the usual classroom instruction. During
this year no progress was made in the standard subjects
measured by the Stanford Achievement Tests. Perhaps
the individual attention given to the children had raised
their achievement beyond what was customary in the
ordinary school, or else the children were engaged in
other improvements and adjustments so that their im¬
provement in the subjects tested suffered thereby. The
second year of public school shows decided improvement.
Repeated tests given to children show how they are im¬
proving or learning, and we may note that such curves
are learning curves just the same as those discussed in
Chapter IX.

Sept. Oct. Nov. Dec. Jan. Feb. March April May

Figure 75. — Progress in spelling of third grade class. Dotted
line represents standard scores. (From Starch’s Educational Psy¬
chology.)

Composite Learning Curves. — Just as we have drawn
learning curves of school subjects for individual children,
so we can construct learning curves for groups of chil¬
dren. Figure 75 shows the learning curve in spelling, as
measured by the Starch Spelling Test, for a third grade
class. The class was tested each month during the school
year and the solid line represents the improvement made,
and we note an irregular but steady improvement from

COMPOSITE LEARNING CURVES

313

September till May. The broken line is a hypotheti¬
cal curve representing the probable improvement dur¬
ing the school year, as determined by the norms for the
test.

a «
2 §

O e
■+* ps

t-i

<+H

• rH
-»-=>
o

f-i

-ss

f-H

1-1

s-<

(-H

032

02
'g

'S3 °
. -S 3 00
CO o ^

O £

*C -S
02

a os

S3 >;
02 ^
. 02

-sS

W -to

Figure 76 is an interesting composite curve taken from
Hollingworth’s (1) investigations of superior children. It
shows the improvement in the ability to add fractions,
as tested by a specific test, for two groups of superior
children tested many times during a period of three years.
The solid line represents the progress of a very superior

314 IMPROVEMENT IN SCHOOL SUBJECTS

group with I.Q.’s around 165 and the broken line that
of a superior group with X.Q.’s around 146. Notice how
the very superior group keeps well ahead of the su¬
perior group, until the former approaches the maxi¬
mum score on the test, and is, therefore, not able to
show increased improvement. These two curves are
learning curves for a specific function for two groups of
children.

The norms of any well standardized educational test
will show the average learning curve for children in gen¬
eral. If we use age norms, then we get the average learn¬
ing curve by age, and if we use grade norms we get the
average learning curve by grade. These curves will rep¬
resent what children of various ages or grades are accom¬
plishing in the average school under present-day condi¬
tions. They do not represent maximum accomplishment
under ideal conditions, but they simply show us what
children are now doing.

Such a curve is shown in Figure 77, constructed from
the grade norms of the Stanford Achievement Test. It
shows us the average number of points scored for each
grade for the composite of school subjects included in the
Stanford Achievement Test. Each half year of instruc¬
tion shows a steady increase in learning the ordinary
school subjects. It is to be noted that this curve differs
from the other learning curves we have discussed in as
much as it is not derived from repeated measures on the
same children. The children tested in Grade V are not
the same as those tested in Grade IV, and so on. To make
the curve comparable with our other learning curves,
the same group of children should have been tested each
half year as they progressed through the grades. The
curve in Figure 77 takes no account of the children who
fall behind or drop out owing to lack of intelligence.
Nevertheless, we may regard it as a general learning curve
of the ordinary school subjects in the grades.

THE TYPES OF TESTS

315

Figure 77. •— Improvement in educational achievement from
grade to grade. Stanford Achievement Test grade norms.

The Types of Tests. — Let us now make a brief survey
of the numerous educational or achievement tests which
have been constructed in order to help in the measure¬
ment of improvement in school work. Any attempt to
describe these tests must take into account the many
different aspects of a subject that can be measured. No
single classification of educational tests is possible. They
can, however, be roughly arranged (A) according to
their purpose, or the aspect of any particular subject
with which they are concerned, and (B) according to the
field of knowledge or skill, that is the subject-matter,
which they measure.

A. Aspect or Purpose: 1. Amount , rate , or speed . — A
test may emphasize the amount accomplished in a given
period of time, or the rate or speed at which certain things

316 IMPROVEMENT IN SCHOOL SUBJECTS

can be done. Thus in arithmetic we may wish to teach
children to add at a given rate. Speed tests will measure
how quickly they can carry out the various operations in
arithmetic. If we are not interested in the speed at which
our pupils perform arithmetical operations we should not
give speed tests in arithmetic. An example of a standard
arithmetic test which measures how much can be done in
a certain time is the Courtis Standard Research Tests in
Arithmetic, Series B. Test No. 1 consists of 24 addition
examples, all of the same form and presumably of equal
difficulty. The amount done in eight minutes forms the
basis of the score. And so with other tests in other arith¬
metical operations. In like manner we have tests of
speed in reading, writing and other subjects.

2. Difficulty or Power. — A test may emphasize the de¬
gree of difficulty which a pupil can master, the power or
command he has over a subject. Such tests are made up
of items increasing gradually in difficulty from very easy to
very hard. The best “difficulty ” or “power” tests arrange
the items in steps of equal difficulty from easy to hard.

An example of such a test in Arithmetic is the Woody
Arithmetic Scales. The addition scale, for example, is
constructed of 38 addition examples, beginning with such

easy items as 3 and increasing gradually in difficulty

3 1

through items like ^+^ = to hard items like 25.091+

100.4+25+98.28+19.3614= . How far the pupil

can go up this ladder of difficulty will measure his power
in addition.

In Reading, the Thorndike-McCall Reading Scales give
a measure of difficulty, and so there are other tests in
other subjects.

Difficulty or power tests are most useful in diagnosing
the particular difficulties or mistakes made by a pupil.
When they are used for this purpose they are called

THE TYPES OF TESTS

317

“diagnostic” tests. Thus the Charters Diagnostic Lan¬
guage Tests are specially constructed to diagnose the lan¬
guage difficulties of the pupils. By means of them we can
find out or “ diagnose” the language troubles of a pupil.

3. Quality. — A test may try to measure how well
a person can do a certain thing, regardless of speed or
of the difficulty of the operation. Such tests empha¬
size quality. The best example of this type is the Thorn¬
dike Handwriting Scale. It consists of a series of exam¬
ples arranged in order from poor quality to good quality
and the specific example to be rated is compared with the
samples on the scale.

Quality again is important in English composition and
there are several scales for its measurement.

These three aspects are perhaps the most important.
There may be others. We might, for example, consider
range as an aspect and construct a test to see how wide a
range of information has been attained in such subjects
as history or geography. And so there undoubtedly are
other aspects. Any aspect which the school may from
time to time emphasize will undoubtedly be measured by
appropriate tests. Aspects overlap and most tests measure
more than one aspect at the same time, but generally they
emphasize one aspect in particular.

This brief treatment of the aspect of a subject measured
by a test is important. We must always keep in mind
precisely what we are measuring. There is no one test
for Arithmetic, nor for Reading, nor for any school sub¬
ject. We must measure one aspect at a time, and be
careful to remember what that aspect is when we come
to interpret our results. Furthermore we must not be so
stupid as to blame the test for not measuring what it makes
no claim to measure.

B. Subject-Matter. — Tests classified according to the
field of knowledge or skill tested can be divided into two
large groups: 1. General or composite tests including

318 IMPROVEMENT IN SCHOOL SUBJECTS

several fields of knowledge; 2. Special tests dealing only
with one field of knowledge or with a limited portion of
that field.

1. General or Composite. — Tests of this type include
a number of school subjects and it is possible by their
use to arrive at a general rating of the educational achieve¬
ment of the child. A good example of such a composite
educational test for the elementary school is the Stanford
Achievement Test. It includes sub-tests in the following
school subjects: — three different kinds of reading, two
kinds of arithmetic, nature study and science, history and
literature, language usage, and spelling. The total score
on this test is a good measure of a child’s general school
accomplishment but this score tells us nothing about his
ability in any specific subject. The total score of any
given child can be compared to the norms and we can
then tell how the child in question compares with other
children of his age or grade. Similarly the average scores
for classes or schools can be compared with each other
and with the general standards.

The Stanford Achievement Test is a fairly long and
thorough test and because of this it gives fairly good
independent measures of the several subjects entering
into the composite. So in addition to the general educa¬
tional standing of the child, we can obtain his standing on
particular subjects, such as reading, arithmetic and so on.

An example of a much briefer test covering the ordinary
subjects in the elementary field is the Pintner Educational
Survey Test. This also contains tests in reading, arith¬
metic, history, geography, language and so on. The
tests are very brief and little reliance can be placed on the
rating of any specific subject. Nevertheless, the total
score on the whole test gives a fair measure of a child’s
general educational standing. Tests of this type are gen¬
erally called survey tests, because they are designed for
a general survey of a school or school system, and they

THE TYPES OF TESTS

319

are not meant to give detailed information about specific
children.

At the high school level, we have the Iowa High School
Content Examination which consists of four sub-tests as
follows: (1) English, Literature, etc.; (2) Mathematics;
(3) Science; (4) History and Social Science. It is a thor¬
ough examination of the usual things taught in high
school. A general score for the whole test may be ob¬
tained as well as separate scores for the four separate
parts of the examination.

2. Special Tests. — The number of special tests for
separate subjects is very great and it is not our purpose
to attempt to describe them all. Still less is it possible
to evaluate them or contrast the merits of one with those
of another. Indeed the point of view we shall emphasize
is that most of them possess merit of one kind or another,
and the important thing for the teacher to remember is
the particular aspect of a subject that each one is measur¬
ing. No one test will measure all the objectives of arith¬
metic teaching in the school. We must carefully choose
the test that best fits the aim or objective we want to
measure.

We shall run over some of the more important sub¬
jects in the curriculum and make comments on some of
the tests in each field.

a. Arithmetic. — Arithmetic is well represented by
measuring instruments, because it is relatively easy to
construct tests in this field. The content is fairly uni¬
form in all schools and the scoring of such tests is definite
and easy. If we want to measure speed and accuracy in
the four fundamental operations we can use the Courtis
Standard Research Tests, Series B. These tests will not
tell us how hard or complicated examples in addition or
subtraction our children can solve, but merely how
quickly and how accurately they can add, subtract, mul¬
tiply or divide arithmetical examples of a given degree of

320 IMPROVEMENT IN SCHOOL SUBJECTS

difficulty. They will merely answer that question and
they should not be used for any other purpose.

If, however, we are interested in the problem of diffi¬
culty and wish to know how high on the scale of difficulty
or complexity a child can rise, then we will use the Woody
Arithmetic Scales or some similar type of measuring in¬
strument. The Woody tests are difficulty or “power”
tests in the four fundamental operations.

Difficulty or “ power” in general in the four funda¬
mentals can be measured by the Woody-McCall Mixed
Fundamentals, which as the name implies is the same
type of test as the Woody test except that here the four
fundamental operations are mixed together.

In addition to the fundamentals of arithmetic, prob¬
lem solving is very important, and none of the scales
dealing with the fundamentals will give us a measure of
this. Therefore, if we wish to measure problem solving
rather than dexterity in the fundamentals we must make
use of some such test as the Buckingham Scale for Prob¬
lems in Arithmetic. There are three scales for the differ¬
ent grade levels from III to VIII, and the use of this test
will give us a good measure of the child’s ability to solve
arithmetical problems as distinct from his ability in the
four fundamental operations. The diversity of tests in
arithmetic demonstrates very well the necessity for care¬
fully choosing the appropriate test for that specific thing
which we wish to measure.

As we go from the elementary to the high school, the
problem becomes still more complex, for here the field
widens into algebra and geometry. The Hotz First-Year
Algebra Scales cover the main operations in elementary
algebra.

The Columbia Research Bureau Plane Geometry Test
is a very thorough test in geometry for high school stu¬
dents. There are, in addition, several other tests in al¬
gebra and geometry, and each has its particular merit.

THE TYPES OF TESTS

321

b. Reading. — When we pass from arithmetic to read¬
ing, we pass from a relatively simple to a relatively com¬
plex school subject. For reading comprises such abilities
as being able to articulate printed words placed before
one, being able to perceive silently the symbols, remem¬
ber the general content, respond to questions about it,
understand the meaning of isolated words, carry out
printed directions and so forth. The functions measured
by “ reading ” tests may therefore, be basically very
diverse. Ordinarily reading tests are divided into three
groups, silent reading, oral reading and vocabulary tests.

Silent reading tests are numerous. They consist of
passages to be read followed by questions to be answered
in order to test the comprehension of the passages. The
Thorndike-McCall Heading Scales for Grades III to XII
are excellent examples of tests for the measurement of
comprehension in reading. For example, the pupil is
asked to read a short passage and then, with the passage
still before him, answer some questions:

“On Monday Dick saw a red fox, a gray squirrel and a black
snake in the woods. The next day he saw a brown rabbit and
five brown mice in the field. He killed the fox and all the mice,
but let the others live.

*‘ What did Dick do to the fox? .

“What color was the rabbit? .

“Where did Dick see the fox? . ”

The tests are primarily power tests in the sense that they
measure ability to comprehend. They do not measure
rapidity of reading. They have a high correlation with
general intelligence tests of the verbal type, partly be¬
cause the latter assume the ability to read with under¬
standing. The Burgess Picture Supplement Scale, de¬
signed for Grades III to VIII, consists of passages
containing directions to be carried out. These directions
refer to some picture and the child responds by making

322 IMPROVEMENT IN SCHOOL SUBJECTS

some mark or drawing on the picture. The child does
not have to respond by a word, phrase or sentence as in
the Thorndike-McCall Tests. The Monroe Standardized
Silent Reading Tests offer passages for reading with objec¬
tive type questions to measure comprehension. They give
measures both of comprehension and of rate of reading.

Oral reading tests measure the ability to read orally
without error. Errors consist of mispronunciations, omis¬
sions, substitutions, insertions and repetitions. The Gray
Oral Reading Tests cover this field very well. Measure¬
ment of oral reading must be done individually and it is
therefore very time-consuming. Again, since oral reading
is not very important in daily life, the need for such tests
is limited.

Vocabulary tests confine themselves to a measurement
of the knowledge of words. They generally have a high
correlation with verbal intelligence tests, many of which
contain vocabulary tests as sub-tests of the composite
test. Vocabulary tests are generally constructed on the
objective test pattern so that they are simple to give and
easy to score. Knowledge of words is undoubtedly one
of the most important factors in reading ability. The
Thorndike Test of Word Knowledge is a good sample of
this type of test. The test consists of one hundred graded
words, each of which has along with it five other words,
and the child has to mark the word which means the
same or nearly the same thing. Here are samples from
the list of 100:

No. 1. afraid: full of fear — possible — necessary — raid — ill.
No. 31. anon: year — sometimes — hitherto — again — now.
No. 54. ardor: anger — need — zeal — difficulty — pagan.

No. 97. madrigal: song — mountebank — lunatic — ribald —
sycophant.

Reading has so many aspects that no one test can
measure all of them. The most thorough-going attempt

THE TYPES OP TESTS

323

to construct tests for many of these aspects is seen in the
work of Gates. Pie has constructed special tests for the
primary grades; silent reading tests for Grades III to VIII
of four different types: Type A, to appreciate the general
significance of a paragraph; Type B, to predict the out¬
come of given events; Type C, to understand precise
directions; Type D, to note details; a graded word pro¬
nunciation test; and a phonetic ability test. These tests
of Gates cover the field of reading very thoroughly. They
are absolutely indispensable for diagnosing the reading
difficulties of pupils. Each test has its special function.
This whole battery of tests illustrates excellently the
necessity for deciding definitely just what we want to
measure before we start our testing. We cannot just
measure “reading” in general. We must decide spe¬
cifically what aspect of reading we are going to deal with.

c. Handwriting. — The essential factor in handwriting
is quality or legibility. A certain degree of legibility is
necessary as a minimum and a large amount of it is
highly desirable. Here we are dealing then with the
measurement of a product quite unlike reading or arith¬
metic. The scales constructed for this purpose consist
of a series of sample products ranging from bad to good
by gradual steps. The product to be rated is compared
with the samples of the scale and given the value of the
sample to which it most nearly approximates. The two
most useful handwriting scales are the Thorndike Scale
and the Ayres Scale. As can be readily imagined from
our description of these scales, the measurement of hand¬
writing cannot be as objective as the measurement of
arithmetic or reading. Raters may vary as to their judg¬
ment of the quality of any given sample of handwriting.
A certain amount of practice is necessary in order to use
these scales properly. Care must be taken in comparing
handwriting scores obtained by different raters. Never¬
theless, in spite of all these difficulties, inherent in any

324 IMPROVEMENT IN SCHOOL SUBJECTS

quality scale, scales for the measurement of handwriting
are of great value.

d. Spelling. — Ability to spell words according to the
existing convention of the day is, strangely enough, very
highly rated by educators and therefore many scales or
tests have been constructed for the measurement of this
function. These are generally based upon commonly
used words and are arranged in order of difficulty. The
Morrison-McCall Spelling Scales and the Iowa Spelling
Scales are useful in the elementary grades, while the
Brigg’s Scales are of value in the high school.

e. English. — Under this heading we have to consider
such diverse subjects as Composition, Language, Gram¬
mar and Literature. Composition, the ability to express
oneself properly in writing, is one of the most important
objectives in education. What is measured here is the
product produced by the child, and, as in Handwriting,
this is done by comparison with a standard scale of sam¬
ples arranged in order of merit. The Thorndike Exten¬
sion of the Hillegas Scale contains fifteen degrees of
merit from very poor to very good compositions. What
is judged by the rater is the general merit of the composi¬
tion and this general merit is compared with the stand¬
ard samples of the scale. The Van Wagenen English
Composition Scales consist of separate scales for exposi¬
tion, narration and description. Each composition is
rated with reference to thought content, structure and
mechanics. It is an attempt to analyze more specifically
the elements that enter into composition ability. The
rating of composition, like the use of any quality scale,
is difficult and what has been said above with reference
to the use of handwriting scales applies here with still
greater force, because of the complexity of the thing we
are rating. Ordinarily a fair amount of practice with
any scale is necessary before reliable judgments are se¬
cured. Even with the use of the scales, individual teach-

THE TYPES OF TESTS

325

ers may differ greatly in their judgments. Nevertheless,
the use of any such scale by a teacher will tend to make
his ratings more and more uniform, and for educational
research purposes the use of a scale by trained workers
is indispensable.

Language tests measure the ability to use correct Eng¬
lish and are of course more objective and easier to use
than are the composition scales. The Wilson Language
Error Test presents a short story and the pupil has to
correct all the mistakes in the story. The Charters
Diagnostic Language Tests expect the child to detect
and correct errors. The Kirby Grammar Test, in addi¬
tion, calls for the choice of the grammatical principle
concerned.

Attempts to measure what is taught under the heading
of English literature in schools are very diverse and very
limited. Probably this is owing to the very hazy and
disputed objectives of literature teaching in the schools.
There are such tests as the Abbott-Trabue Exercises in
Judging Poetry, and the Van Wagenen Reading Scales in
English Literature. Literature information can be tested
readily by the ordinary information type of test.

f. Geography. — Here the problem is largely one of
testing geographical information. The Gregory-Spencer
Geography Tests are a good example. They cover such
phases of geography as trade routes, causal geography,
description and map location, physical and commercial
geography. The Posey-Van Wagenen Geography Scales
attempt to cover a wider range, and include a Thought
Scale which is supposed to measure a child’s ability to
think about geographical problems.

g. History. — Here, as in Geography, the field is well
represented by tests. The Barr Diagnostic Tests in Amer¬
ican History measure separately comprehension of mate¬
rial read, chronological judgment, judgment of evidence,
evaluation of the importance of facts, and the under-

326 IMPROVEMENT IN SCHOOL SUBJECTS

standing of cause and effect relationships. Other tests
stress historical information. The Van Wagenen Ameri¬
can History Scales measure information, thought and
character judgment.

h. Foreign Languages. — There are now many good
tests for the measurement of ability in French, Spanish,
Latin and German. The Columbia Research Bureau
Tests in French, Spanish and German are good examples
of modern work in this field.

i. Science . — The large field covered under this general
heading leads to many diverse tests. There are general
science tests, such as the Ruch-Popenoe General Science
Test containing items from botany, chemistry, physics,
zoology, astronomy, geology and so forth. Then there are
tests in the separate sciences, such as the Powers General
Chemistry Test, the Ruch-Cossmann Biology Test and
the Iowa Physics Test. Naturally in this wide field there
is room for a great many tests.

j. Miscellaneous Subjects. — In addition to the more or
less standard subjects which we have surveyed above,
there remain such subjects as music, hygiene, cooking,
sewing, shop work and in the high schools the vocational
subjects, such as typewriting, stenography and the like.
There are educational achievement tests or scales in prac¬
tically all of these subjects. In some the measurement
is rather meager, but in others the tests are well-estab¬
lished. In all the fields the possibility of measurement
has been demonstrated.

More elaborate accounts of achievement tests will be
found in the works of Kelley (2), Monroe (3), Symonds
(5), and Ruch and Stoddard (4).

Tests and School Objectives. — Our survey of the edu¬
cational tests and scales shows very definitely that any¬
thing which the school sets out to teach can be measured
by suitable tests. Training, schooling, education or what¬
ever else we call it means, from the psychological stand-

EDUCATIONAL RATINGS

327

point, the modification in some way or other of the
reaction habits of the child. These reaction habits lend
themselves to measurement and our educational tests are
the measures of the different kinds of reactions which
the school endeavors to bring about in the child.

Frequently the aim of education is expressed in lofty
and general terms, such as better citizenship, a higher
morality, a Christian life and so on. So long as these
terms remain undefined and vague, there is considerable
agreement among people as to the desirability of these
objectives. But, as soon as we attempt to be more spe¬
cific as to just what is good citizenship, as to just what is
the Christian life, disagreement at once begins. We can¬
not set out vaguely to make children good citizens. We
have to be concrete and definite, and set out to teach them
so much reading and writing and history and geography
and habits of health and definite ways of reacting to
this, that, and the other thing in their environment. Im¬
mediately we do this, we arrive at tangible reactions
which are open to psychological measurement. Let the
educator set out definitely to modify the reactions of a
child in a particular way, and the psychologist will be
able to measure the amount of change or modification
that is taking place.

Interpretation of Educational Ratings. — In all of the

educational achievement tests discussed above there re¬
sults from the testing some kind of score or numerical
value. These scores vary greatly from test to test and
in order to be able to interpret them, they must be trans¬
muted into some standard rating.

One of the best ways of interpreting any particular
score is to compare it with the distribution of a large and
random sampling of other children’s scores. Since most,
if not all, distributions of educational scores tend towards
a normal distribution, the usual way is to compare the
individual score with the standard deviation of the com-

328 IMPROVEMENT IN SCHOOL SUBJECTS

parable group. In this way a score is expressed as so
many times the standard deviation of the standard group.
Standard deviations are calculated from the mean and
therefore we have plus and minus quantities. These are
difficult to handle. Because of this, modifications of the
direct standard deviation units are more customarily em¬
ployed. The best known and most useful modification is
the T-Scale devised by McCall. A T-unit is one-tenth of
the standard deviation of a random sampling of twelve-
year-old children. T-units run from zero to one hundred,
and fifty is the mean score for twelve-year-olds. The dis¬
tribution of any age group can then be interpreted in
terms of the standard twelve-year-old group. It is also
possible to calculate T-scores for every age group or
indeed for any group of children. Tables of T-scores are
frequently given for standard educational tests.

Another method of interpreting educational test scores
is to convert them into educational ages. Just as an
intelligence test score is converted into a mental age
(M.A.), so an educational test score is converted into an
educational age (E.A.). The E.A. is obtained by an age
standardization for those ages for which the test is most
useful. The meaning of the E.A. is simple and easily
grasped. An E.A. of 10 on a test means that the child in
question is achieving what ten-year-olds ordinarily
achieve; an E.A. of 12 means he is up to average twelve-
year-old performance and so on. The scheme is simple
and useful but not as accurate as any of the standard
deviation methods. The drawback of the E.A. method
is that the E.A. becomes less and less reliable as we pro¬
ceed to the higher ages. In these higher ages an E.A.
becomes the average score of a much more selected group
than the E.A. of the lower ages. Whereas an E.A. of 11
or 12 derived from an extensive standardization of ele¬
mentary school pupils may be said to represent fairly the
achievement of twelve-year-olds in general, an E.A. of

EDUCATIONAL RATINGS

329

15 or 16 obtained from school children cannot in any
sense be considered the average achievement of fifteen- or
sixteen-year-olds in general. These fifteen- or sixteen-
year-olds are a very selected group. The E.A. should
not, therefore, be used in the high school, and teachers
should be aware of the fictitious nature of E.A.’s in the
upper ages when they make use of them to interpret very
high achievement scores on any test.

If we use the E.A. for any educational achievement
test, we have the Educational Age on that particular
test. Sometimes we are making use of several tests in
different subjects, and so we obtain several E.A.’s. To
distinguish between these it is quite customary to give
them specific names, and so we hear of Arithmetic Age,
Reading Age, Geography Age, and so on, and sometimes
these are contracted into symbols such as A.A., R.A.,
and the like. They are all, however, educational ages of
some sort or other.

Following the analogy of the Mental Age and Intelli¬
gence Quotient, we have for educational tests the Educa¬
tional Quotient accompanying the Educational Age. The
E.Q. is the E.A. divided by the C.A., just as the I.Q. is
the M.A. divided by the C.A. The meaning of the E.Q.
is, therefore, similar to that of the I.Q., keeping in mind
the different sort of thing we are measuring. Thus an
E.Q. of 100 is normal or average for the age in question;
an E.Q. of 125 is above average and an E.Q. of 80 is
below. The E.Q. of 80 tells us that for some reason or
other the child in question is doing quite inferior work for
his age. The reason may be lack of intelligence, faulty
grade placement, faulty teaching and so forth. The E.Q.
cannot reveal the reason. The E.Q. of 125 tells us that
the child is doing much better educational work than is
usually accomplished by children of his age. This may
be due to superior intelligence, superior schooling, excep¬
tional opportunity and so forth. We must remember that

330 IMPROVEMENT IN SCHOOL SUBJECTS

the E.Q. is in terms of the average educational accom¬
plishment of children of different chronological ages. Its
reference point is always the accomplishment of the av¬
erage child under present-day instruction, with average
opportunities. It does not refer to what might be ac¬
complished under ideal conditions.

All educational tests have furthermore another type
of rating which can be used in addition to either of the
above methods. This rating is a grade rating of some
type or other. Such ratings are based upon grade stand¬
ardizations, in contrast to the age standardizations dis¬
cussed above. The grade ratings tell us how any par¬
ticular child or class compares with the average for grades
in general. The grade rating is important particularly from
the supervisor’s or superintendent’s point of view. Its
significance, however, can only be determined by taking
into consideration such factors as the intelligence of the
children making up any particular grade, and the chron¬
ological age of the children to determine whether they
are under- or over-age for the grade in question. We
shall see how these factors are important in our discus¬
sion of the uses of educational tests.

Summary

1. This chapter deals with the measurements of the modifica¬
tions which the school is trying to make in its children.

2. An educational test score represents a point on a learning
curve for some child in some particular function.

3. Repeated tests allow us to plot learning curves for particu¬
lar subjects.

4. Progress in educational subjects is not necessarily steady
and continuous.

5. The standard grade or age norms for a test may be con¬
sidered as composite learning curves for children in general,
assuming such standards to be based upon the same kind of
selection of children at each age or grade.

SUMMARY

331

6. Educational tests may be considered according to (1) their
general purpose and (2) the subject matter measured.

7. Purposes may be to measure speed or power or quality.

8. Tests considered from the aspect of subject matter may be
further divided into: general or survey tests covering sev¬
eral fields of knowledge; and specific tests dealing with one
field only.

9. Examples of these general and special tests have been given
and it is seen that almost every subject of school instruction
is now covered by tests. This applies to both the elementary
and the high school. Some fields have been well covered
and others very meagerly.

10. It is concluded that any specific objective which the school
may set up is measurable by means of achievement tests.

11. Scores obtained on educational tests must be transmuted
in order to be interpreted.

12. Two common ways of transmuting scores are the standard
deviation method and the educational age method.

13. There are limitations to the usefulness of Educational Age
and Educational Quotient.

14. All educational tests give us the opportunity of interpreting
the scores in terms of grade norms.

332 IMPROVEMENT IN SCHOOL SUBJECTS

Review

True-False Statements

As a review mark the following statements true or false:

1. Achievement tests should determine the ideal type of sub¬
ject matter to be taught in schools.

2. Achievement test scores are generally more reliable than
teachers’ marks.

3. Repeated educational tests of the same child give us a learn¬
ing curve for that child.

4. No one achievement test will measure all the various aspects
of any school subject.

5. Most achievement tests are merely speed tests.

6. Which arithmetic test we choose will depend upon what
specific aspect of arithmetic we desire to measure.

7. Power tests are those which emphasize how quickly a child
can do the test.

8. The educational quotient is derived in the same way as is
the intelligence quotient.

9. Complete this sentence:

The E.Q. is the quotient resulting from dividing the-by

the-.

Advised Reading

Gates, A. I.: The Improvement of Reading, New York, 1927.

References

1. Hollingworth, L. S., and Cobb, M. V.: Twenty-Seventh

Yearbook of the National Society for the Study of Educa¬
tion, Part II, 1928, Chapter 1.

2. Kelley, T. L.: Interpretation of Educational Measurements,

Yonkers, 1927.

3. Monroe, W. S., et ah: Educational Tests and Measurements,

Boston, 1924.

4. Ruch, G. M., and Stoddard, G. D.: Tests and Measurements

in High School Instruction, Yonkers, 1927.

5. Symonds, P. M.: Measurement in Secondary Education, New

York, 1927.

LIST OF TESTS

333

List of Tests Referred to in this Chapter

Composite Educational Tests

Stanford Achievement Test: World Book Co., Yonkers.

Pintner Educational Survey Test: College Book Co., Colum¬
bus, Ohio.

Iowa High School Content Examination: Bureau of Educa¬
tional Service, University of Iowa.

Arithmetic and Mathematics

Courtis Standard Research Tests, Series B.: Courtis, Detroit,
Michigan.

Woody Arithmetic Scale: T. C. Bureau of Publications.

Woody-McCall Mixed Fundamentals: T. C. Bureau of Pub¬
lications.

Buckingham Scale for Problems: Pub. Sch. Publishing Co.

Hotz Algebra Scales: T. C. Bureau of Publications.

Columbia Research Bureau Geometry Test: World Book Co.,
Yonkers.

Reading

Thorndike-McCall Reading Scales: T. C. Bureau of Publica¬
tions.

Burgess Picture Supplement Scale: Russell Sage Foundation.

Monroe Silent Reading Tests: Pub. Sch. Publishing Co.

Gray Oral Reading Test: Pub. Sch. Publishing Co.

Thorndike Test of Word Knowledge: T. C. Bureau of Pub¬
lications.

Gates Silent Reading Tests: T. C. Bureau of Publications.

Handwriting

Thorndike Handwriting Scale: T. C. Bureau of Publications.

Ayres Handwriting Scale: Russell Sage Foundation.

Spelling

Morrison-McCall Spelling Scale: World Book Co., Yonkers.

Iowa Spelling Scales: Pub. Sch. Publishing Co.

Briggs’ Sixteen Spelling Scales: T. C. Bureau of Publications.

Composition

Thorndike Extension of Hillegas Scale: T. C. Bureau of
Publications.

Van Wagenen English Composition Scales: World Book Co.,
Yonkers.

334 IMPROVEMENT IN SCHOOL SUBJECTS'

Language

Wilson Language Error Test: World Book Co., Yonkers.

Charters Diagnostic Language Test: Pub. Sch. Publishing
Co.

Kirby Grammar Test: University of Iowa.

Literature

Abbott-Trabue Exercises in Judging Poetry: T. C. Bureau
of Publications.

Van Wagenen Reading Scales in English Literature: Pub.
Sch. Publishing Co.

Geography

Gregory-Spencer Geography Test: University of Cincinnati.

Posey-Van Wagenen Geography Scales: Pub. Sch. Publish-
ing Co.

History

Barr Diagnostic Tests in American History: Pub. Sch. Pub¬
lishing Co.

Van Wagenen American History Scales: T. C. Bureau of
Publications.

Foreign Languages

Columbia Research Tests: World Book Co., Yonkers.

Science

Ruch-Popenoe General Science Test: World Book Co., Yon¬
kers.

Powers Test for General Chemistry: World Book Co., Yon¬
kers.

Ruch-Cossmann Biology Test: World Book Co., Yonkers.

Iowa Physics Test: Pub. Sch. Publishing Co.

CHAPTER XV

THE USES OF EDUCATIONAL TESTS

Mental Ability and Educational Achievement. — The

intelligence test attempts to answer the question, “How
much can this child achieve?’’ The educational test
gives an answer to the question, “How much has he
achieved?” The two instruments are measuring a child’s
reactions from different viewpoints. Sometimes indeed
the very same type of reaction is used in the two types
of tests, as when reading reactions and arithmetical reac¬
tions are included in intelligence tests. But usually the
reactions called for in intelligence tests are more general.
They are gathered from a wider field. Whereas the reac¬
tions called for in educational tests are more specific.
They are gathered from a narrower field. They are
limited very specifically to what the school is concerned
with teaching. The intelligence test is essentially prog¬
nostic in function. It looks towards the future and pre¬
dicts the type of achievement of which the child should
be capable. It is this predictive aspect of intelligence
tests that makes them so valuable. They indicate the
amount of education a child should be capable of achiev¬
ing. If a child has a high I.Q. his educational achieve¬
ment should be high. If he has a low LQ. we cannot
expect his educational achievement to be up to the aver¬
age for his age. An inherently bright boy should do bet¬
ter than one of merely normal ability, and it is no credit
to the bright boy to achieve the average educational ac¬
complishment for his age. To men of genius we look for
works of genius. From men of feeble capacities we do
not expect works of genius nor even average products.

335

336 THE USES OF EDUCATIONAL TESTS

The A.Q. Technique. — In this sense, therefore, all
educational testing must be interpreted in the light of
the intelligence of the children tested. To evaluate cor¬
rectly the educational achievement of a pupil, we must
have an accurate measure of his intelligence. In the
actual measurement of pupils, several techniques have
been developed to accomplish this. The easiest to under¬
stand and one frequently used is the A.Q. technique.
This is built upon the I.Q. and the E.Q. A child of a
given I.Q. is expected to attain a similar E.Q. This ratio
of E.Q. to I.Q. is called the A.Q. or Achievement Quo-

^ „ E.Q. E.A.

tient. The A.Q. is thus ——- or _ -

I.Q. M.A.

If the E.Q.’s and the I.Q.’s are very similar we shall
have A.Q.’s clustering around 100, and this means that
educational attainment is keeping step with intellectual
ability. If the E.Q. is appreciably below the I.Q., then
our A.Q.’s will be small or appreciably below 100. This
means that all the intelligence of the child is not being
used in his educational work. For some reason or other
he is not using all his intellectual capacity. The reason
for this, of course, will not be given by our measurements.
We must look for the reason elsewhere. A very common
one is faulty grade placement of children of high I.Q.
They are frequently in grades doing relatively simple
work and hence they do not have the opportunity to
accomplish more on their educational tests. Other rea¬
sons may be laziness, lack of proper motivation, illness
and so on. If the E.Q. is appreciably above the I.Q. then
we obtain A.Q.’s above 100. These indicate children who
are accomplishing more than is ordinarily accomplished
by children of their I.Q. level. We are reminded here of
what was pointed out in the previous chapter, namely,
that E.A.’s and E.Q.’s are based upon normal or average
accomplishment and not upon maximum accomplishment.
Hence it is perfectly possible for a child at any given

LIMITATIONS OF THESE TECHNIQUES 337

I.Q. level to achieve more than is ordinarily accomplished.
This is brought about frequently by pushing some chil¬
dren faster than usual, by stimulating them to work by
placing them in a grade where the mental age level is
higher than their own and trying to keep them up to
standard. This condition is frequently found among
docile, industrious, dull children. They advance with
their class because they are of the right chronological
age and because their docility and industry impresses
the teacher. They are constantly given extra attention
and training by the teacher because they are the slowest
in the class. All of this tends to raise their E.A. above
that of children of like M.A. and hence we get A.Q.’s
above 100. This illustrates very well the fact that our
educational test standards are norms of average accom¬
plishment and not standards of maximum accomplish¬
ment under ideal conditions. The average A.Q. of a
class will be a measure of the efficiency of the class and
similarly with the average of a whole school or school
system.

The F Score. — The standard deviation methods of
rating educational tests can also be used to evaluate
educational achievement in terms of mental ability. The
difference between the sigma position of the pupil on the
educational and the mental tests is such a rating. In
McCalFs T score technique we have the F or efficiency
score which is the achievement T score plus 50, minus the
intelligence T score. The plus 50 is merely included so
that the average or normal shall be 50, as in the T scale.

Limitations of These Techniques. — In addition to
these A.Q. and T and F techniques, there are several
others, but these will suffice as the most commonly used
methods. All of these techniques must be used cautiously.
They all depend upon a very thorough standardization
of the educational and intelligence tests used. They all
assume that the standardization groups of the two sets

338 THE USES OF EDUCATIONAL TESTS

of tests, educational and intelligence, are similar if not
identical. Obviously, if this is not the case, it will be
impossible to interpret the A.Q.’s or F’s in the usual way.
If the intelligence test has been standardized by means
of a poor group of children and the educational test by
means of an average group, then when we come to use
these standards on an average group of children, our

I.Q.’s will be increased above what they really should be,
and our E.Q.’s will be according to expectation. Not
knowing the faulty standardization of the intelligence
test, there will be nothing to show us the distortion of
the I.Q.’s. Using, then, these distorted or elevated I.Q.’s,
we will obtain depressed A.Q.’s. These we may interpret
as due to poor teaching or what not, instead of to faulty
standardization of the measures used. Ideally these tech¬
niques should not be used except with intelligence and
educational batteries standardized on the same group of
children.

General Uses of Educational Tests. — We can classify
some of the common uses of educational tests as follows:

1. Survey and Inventory Purposes.

2. Efficiency of Instruction.

3. Sectioning and Promoting.

4. Diagnosis of Individual Peculiarities.

5. Motivation.

6. Research.

1. Survey and Inventory Purposes. — No school survey
nowadays is complete without the use of objective tests
and scales for the measurement of the school product.
All the paraphernalia of modern education, buildings,
playgrounds, superintendents, teachers, books and equip¬
ment, are after all solely useful in so far as they are modi¬
fying the child’s reactions in definite desired ways. The
tests and scales are measures of such modifications, and
hence they help to give an answer as to the degree in

USES OF EDUCATIONAL TESTS

339

which the school equipment and staff are justifying them¬
selves in the product produced. The results of such tests
may be used to compare city with city, or the results in
any one city may be compared with the norms of a well
standardized test to see if the city in question is reaching
the standards of the country at large. In making such
comparisons there is obvious need for taking into account
the intelligence of the children, and also the chronological
age of the child where grade norms are used.

The tests used in such surveys need not be long and
detailed. If class averages are merely to be made use of,
the composite brief survey test is to be selected because
of its economy in operation, even though it may not
be very reliable as far as individual scores are concerned.
Sometimes a particular subject may be measured in de¬
tail and then of course a special test for that subject
should be used.

The use of a test for comparing city with city and with
the standard for the country at large is shown in Table
9, which is taken from a survey of the schools of Tampa,

TABLE 9

Achievement in Silent Reading in Tampa, Florida, Schools
for White Children Compared with Other Places

May, 1925

Thorndike-McCall Test for Understanding of Sentences

City

Grade

4 Senior

6 Senior

7 Senior

8 Senior

Tampa, Florida

39.7

50.0

56.0

61.0

Baltimore, Maryland

43.3

53.1

58.1

61.2

33 Wisconsin Cities

40.9

52.6

55.3

58.0

Paterson, N. J.

35.5

49.0

51.7

53.5

St. Paul, Minn.

41.8

53.5

58.0

62.5

Louisville, Ky.

39.1

51.7

59.8

60.7

Hackensack, N. J.

41.4

51.9

56.2

Standard

41.8

53.7

58.3

59.6

340 THE USES OF EDUCATIONAL TESTS

Florida (Strayer, 2). The table shows the average scores
made by pupils in various grades in different cities. From
this we see how Tampa compares with other cities, and
by comparing the scores of Tampa with the standard
scores on the bottom line, we note that Tampa in general
falls a little behind the country at large.

In another survey (Port Arthur, Texas) standard tests
are used to answer the following four questions:

“1. How well are the pupils in the schools of Port Arthur doing
in each of the subjects for the grade they are in?

“2. How well are they doing for their ages?

“3. Are they achieving up to their ability to achieve?

“4. What are some of the apparent difficulties and short-com¬
ings ?”

2. Efficiency of Instruction. — Educational tests are in¬
valuable for supplementing and checking the judgments
of supervisors and superintendents with reference to the
efficiency of their teachers. To depend solely on the
impressionistic method of dropping in to a classroom to
judge the efficiency of a teacher means to be at the mercy
of countless chance incidents, of numerous extraneous
factors which warp our judgments. The total personality
or some special mannerism of a teacher may prejudice the
supervisor in one way or another. Hence the value of
objective tests in addition to these personal judgments.

In the use of tests in this connection, it is obvious at
once that intelligence tests must be used in order to eval¬
uate the work of the teacher. A lower educational score
in a class made up of inferior children may mean much
better instruction than a higher score in another class of
superior children. The difference between the average
educational achievement score of two classes cannot
directly measure the efficiency of instruction. We must
interpret this in the light of the intelligence, the chrono¬
logical age and the previous instruction of the children.

USES OF EDUCATIONAL TESTS

341

When this is done the educational achievement scores will
give credit where credit is due. Many a good teacher
has been blamed for poor teaching because the poor qual¬
ity of the intelligence of her children has not been recog¬
nized. Many an indifferent teacher has been commended
for good work simply because she has had a class of su¬
perior children who have made progress in spite of her.

3. Sectioning and Promoting. — In large schools where
there are several sections of the same grade, educational
tests along with intelligence tests, can be of great service
in determining the children that should be placed in each
section. The best instruction can be given to a group
which is homogeneous mentally and educationally. A
homogeneous intelligence group means a group of chil¬
dren of similar M.A.’s and I.Q.’s. Such a group is grow¬
ing mentally at the same rate. They will be able to keep
pace with each other in learning and if they all start from
the same starting point in any subject, i.e. are homo¬
geneous educationally, we have an ideal group for mass
instruction. The proper use of tests will aid us to ap¬
proach such an ideal.

Fast-moving and slow-moving sections can be more
accurately determined by means of a judicious use of
educational and intelligence tests. Some schools adopt
the so-called three-track plan of fast, average and slow
classes. There is no special virtue in two, three, four or
any number of tracks. This may be determined by the
number of pupils at any one grade level. The important
thing is to have the sections as homogeneous, mentally
and educationally, as is possible.

Educational tests may further help in determining
promotion from one grade to another or from one section
to another. They are merely to be considered a help in
this matter, for they merely give a measure of a small
sample of the work of the class. In the more individual¬
ized instruction of the freer type of school, educational

342 THE USES OF EDUCATIONAL TESTS

tests may be used to check up on the work of the indi¬
vidual, to determine his progress, to decide whether he
has mastered a certain unit of subject matter and should
be allowed to go on to the next. Furthermore, in all
schools standard educational tests are of the greatest
help in the grade placement of new pupils. The probable
grade levels of such new pupils may be very quickly
determined.

The use of educational tests in sectioning, promoting
and placing pupils gradually merges into their more gen¬
eral use in educational and vocational guidance. In this
connection an educational achievement score is but one
small item of the many that must be assembled in at¬
tempting to give sound educational or vocational advice.

4. Diagnosis of Individual Peculiarities. — Progress in
any field means overcoming the difficulties that confront
the individual. What is easy to some, is a stumbling-
block to others. To help the individual, the teacher
must know what are his particular difficulties. Educa¬
tional tests help to discover the particular weaknesses of
each child.

Almost any of the tests will give some kind of a diag¬
nosis of the individual case, but obviously the survey
type of test is too short to be of any practical use. In
general the more specific and detailed a test is, the more
useful will it be for diagnostic purposes.

The best example of achievement tests constructed for
diagnostic purposes is to be found in the Gates Reading
Tests (1). There are four types of tests:

Type A — Reading to Appreciate the General Signifi¬
cance of a Paragraph.

Type B — Reading to Predict the Outcome of Given
Events.

Type C — Reading to Understand Precise Directions.

Type D — Reading to Note Details.

USES OF EDUCATIONAL TESTS

343

If we wish to know something of a child’s reading
ability, we must give all four tests. Gates gives the fol¬
lowing example of the scores of four pupils on the four
types of tests:

Tests

Pupil

100

The figures are the percentages of accuracy. Gates makes
the following comments about these four pupils.

“ Pupil 1 is a thoroughly accurate reader. In all four
tests he made not a single error in comprehension. Pupil
2, of about the same age, grade, and intelligence, is ex¬
ceedingly inaccurate in his grasp of the material. In
Tests A, B, and D, he errs on three or four exercises in
every ten and in Test C (Directions) he misses seven out
of ten. Whatever else may be discovered about these two
pupils, the fact will remain that Pupil 1 is accurate in all
types of reading and therefore not in need of cautions
and drills in accuracy; whereas Pupil 2 is inaccurate in all
types and is in need of help in this respect. Pupil 3 is
fairly accurate in reading save in that type which re¬
quires very precise and well-integrated understanding,
as in Test C. Pupil 4 is very accurate in getting details,
either rigidly related ones as in Test C or more inde¬
pendent ones as in Test D, but he is inaccurate in the
more general interpretation of a paragraph as a whole,
as required in Tests A and B. Pupils 3 and 4 illustrate
the value of using the team of tests. Had only Test D
been used, both would have been diagnosed as “ per¬
fectly accurate readers,” whereas, in fact, Pupil 3 is very
inaccurate in exacting reading (Test C) and Pupil 4 is
very inaccurate in reading to predict something beyond
the substance given (Test B) and in reading for the gen-

344 THE USES OF EDUCATIONAL TESTS

eral impression (Test A). On the basis of one test, it is
unwise to make a statement concerning a pupil’s general
accuracy. The team of four gives much greater insight.”

5. Motivation. — Educational test scores may act as a
splendid method for motivating children’s work. Pupils
may be allowed to keep a record of repeated standard
tests in any given subject, so that they may measure their
own progress. If they put these on a chart and draw a
graph, they are virtually constructing their own learning
curves. As we have seen in our general discussion of the
laws of learning, work without knowledge of results is
never as good as work with knowledge of results. Suc¬
cess is satisfying and a record of progress will therefore
help to strengthen the desirable bonds and eliminate the
undesirable ones. Nothing succeeds like success. One
success spurs us on to a second. This kind of motivation
is on general principles to be preferred to the motivation
caused by rivalry with others. Although the latter has
its merits, yet rivalry with oneself, trying to beat one’s
own record, is more desirable.

Scales such as the Handwriting Scale are often dis¬
played in the classrooms and children are encouraged to
rate their own handwriting and strive to reach a higher
level. The average score of the whole class on any test
may be used to foster group loyalty and each individual
be made to feel his share in helping or hindering the gen¬
eral progress of the group.

So far in educational work the use of tests for motiva¬
tion has not received much attention. Undoubtedly
their value in this direction is great and will be increas¬
ingly recognized. Each teacher must use good judgment
in such use. To make the child keep too many records
at one and the same time would lead to confusion. To
continue a record for too long a period may lead to lack
of interest. To strive to reach an unattainable goal would
not be good motivation. The use of tests for purposes

SUMMARY

345

of motivation will only be justified so long as they really
continue to function as such.

6. Research. — To attempt any record of the uses of
tests in educational research would resolve itself into
the impossible task of recording the major portion of
educational research during the past fifteen or twenty
years. The old-fashioned dependence upon the opinion
of authority in matters educational is rapidly giving way
to the exact methods of quantitative measurement.

The relative merits of various methods of teaching
must be investigated by means of tests. The problem of
the amount of transfer of training from any one subject
to any other can only be satisfactorily handled with the
help of tests. In the numerous experiments requiring
experimental and control groups, tests are necessary to
equate such groups. And so on through the long and
varied list of modern educational research we note every¬
where the value of educational tests.

Summary

1. The interpretation of educational tests must always be in
terms of the intelligence of the pupil.

2. The A.Q. technique is a method whereby the I.Q. and the
E.Q. of the pupil are used to determine his general efficiency.

3. Another such method is the use of T scores on educational
and intelligence tests in order to arrive at F or efficiency
scores.

4. All such techniques assume a similar standardization group
from which the different norms have been derived.

5. Six general uses of educational tests have been discussed.

6. General surveys of school systems employ the survey type
of educational test to obtain a general measure of the educa¬
tional product of the system.

7. Efficiency of instruction is best measured by a combination
of intelligence and educational tests, in order to evaluate
the work of the teacher in terms of the intelligence of his
pupils.

346 THE USES OF EDUCATIONAL TESTS

8. Educational tests can be used to divide a grade into more
homogeneous sections, to help in grade placement of new
pupils, to help in deciding on doubtful cases for promotion.

9. Individual difficulties and peculiarities may be discovered
by appropriate educational tests. This is called the diagnos¬
tic use of the tests.

10. Records of tests may be used to motivate a pupil’s work.
Rivalry with self is a better motive than rivalry with others.
All pupils, good and poor alike, can make progress.

11. For all kinds of accurate educational research the educational
test is indispensable.

REFERENCES

347

Review

True-False Statements

As a review mark the following statements true or false:

1. Achievement tests should be substituted for the usual school
examination.

2. A high A.Q. means that the child is working too hard and he
should be forced to rest.

3. A teacher should be held accountable for the low A.Q. of any
child in her class.

4. For survey purposes a relatively short test may be sufficient.

5. Achievement tests are of great help in the proper placing of
children in high school classes.

6. The Thorndike-McCall Reading Test would be a good test
for diagnosing the specific reading difficulties of any child.

7. An A.Q. above 100 means that a child is working beyond his
intellectual capacity.

8. Use of the A.Q. technique is best restricted to tests standard¬
ized upon the same population.

9. The best tests for diagnostic purposes are those which are
detailed and .specific.

Advised Readings

McCall, W. A.: How to Measure in Education, New York, 1922,
Chapters I to VI.

Strayer, G. D. (Director): Report of the Survey of the Schools
of Port Arthur, Texas. Bureau of Publications, Teachers
College, Columbia University, 1926, Chapters V and VI.

References

1. Gates, A. I.: The Improvement of Reading, New York, 1927,

Chapter VIII.

2. Strayer, G. D. (Director): Report of the Survey of the Schools

of Tampa, Florida. Bureau of Publications, Teachers

College, Columbia University, 1926, Chapter VIII.

CHAPTER XVI

NEW-TYPE EXAMINATIONS

Classroom Examinations. — One of the functions of a
teacher is to score or grade the work of the pupils in the
particular subject they are being taught. The teacher dur¬
ing the course of the year attempts to measure the specific
modifications he has been trying to bring about in his
pupils. From the standpoint of educational psychology
the teacher’s examinations are measures of the learning
progress of the pupils. These classroom examinations
differ from standard educational tests, because they are
made up by the teacher himself with reference to the
specific things he has been teaching, because there are no
general standards or norms and thus no outside criterion
of the worth or value of an answer or paper. Neverthe¬
less, these classroom examinations are important and,
indeed, necessary for our present system of promotions
and credits. They are also useful to the teacher himself
in order to find out how effectively his teaching is being
grasped by the pupils. It is obvious that the more reliable
such examinations can be made, the more efficiently will
they serve all these purposes.

The Essay Type of Examination. — Until recent times
the most commonly used type of examination was the
essay type. This type of examination expected the pu¬
pils to write an answer in essay form to some question
beginning, “Discuss,” “Describe,” “Criticize,” “Tell
about,” “Compare,” “Contrast,” “Evaluate,” or some¬
times more directly beginning with “What,” “Why,”
“How,” “Who,” or “Where.” An examination may
consist of one such question or a great many, depending

348

RISE OF THE NEW-TYPE

349

upon the length of the answers expected. In the scoring
of such papers the examiner generally gives his judgment
upon the question as a whole, assigning a mark of some
sort to denote this judgment. If there are several ques¬
tions the total of these marks represents the total mark
for the paper. The standards for scoring are purely
subjective and rarely does a teacher decide beforehand
just what are the specific requirements for a perfect
answer. Usually the examiner depends upon the total
impression made upon him by an answer and assigns a
mark accordingly. In grading papers for a large class,
say a college class of 100 or more, the difficulty of keeping
any standard constant in one’s mind is very great, and
the monotony of the whole procedure is well known.
Hence this essay type of examination is difficult to score,
it is monotonous and wearisome to grade and it repre¬
sents a burden which should not be borne by the teacher.

The Essay Type is Unreliable. — In addition to what
we have said in the last paragraph about the undesirability
of the essay type of examination, there is additional
criticism that it is unreliable. The scoring of such exam¬
ination papers is unfair to the student. There cannot be
much significance in a mark if competent teachers of
English assign marks all the way from 60 to 98 to the
same paper, as has been found to be the case. Again in
scoring a history paper, 115 competent teachers assigned
marks all the way from 70 to 100, and to another paper
gave marks all the way from 45 to 90. In scoring one
answer to a question in geography 91 teachers assigned
marks from 2 to 20, the maximum score allowed being
20. Such reports as these created dissatisfaction with
the essay type of examination and, thus, long before the
actual coming of the new-type examination, the need for
something better was felt.

The Rise of the New-Type Examination. — The greatest
stimulus to the new-type examination was the rapid

350

NEW-TYPE EXAMINATIONS

spread of group intelligence and standard educational
tests. Undoubtedly here and there teachers had made
occasional use of similar devices in an unscientific ten¬
tative fashion.* But the methods now used were first
developed in connection with group intelligence and edu¬
cational tests.

The Types of Objective Examinations. — There are
many definite types of objective questions and a brief
survey of the more common types will be profitable.

The True-False Type . — Here a statement is presented
and the pupil must indicate in some manner whether the
statement is true or false.

1. Indicate by encircling T or F:

T F The normal pulse rate is about 70.

2. Underline true or false:

Any chord passing through the center of a circle is called a

diameter. True False

3. Write “true” or “false” or T or F before the statement :

-. The attack on Fort Sumter helped the Northern

people, for it united them as they had not been united before.

The statements set before the student may range all
the way from mere information to such as require judg¬
ment and thought. Statements may be constructed
which require considerable critical ability on the part of
the student, if he is to answer them correctly. The prep¬
aration of difficult, thought-provoking statements is by
no means easy. The examiner must guard carefully
against ambiguity. The scoring is either the number
right or else the number right minus the number wrong.
The latter penalizes random guessing. Students should
be told whether or not to guess those questions they are
uncertain about. Perhaps the best results are obtained
by telling pupils not to guess.

* The writer remembers that one of his high school teachers made use
of the one word answer type of question in testing for factual informa¬
tion in history, geography, and literature before educational or intelli¬
gence tests had been heard of.

TYPES OF OBJECTIVE EXAMINATIONS 351

The Multiple Response Type .— Here a statement is
made with two or more answers or completions, and the
pupil is called upon to indicate the best or right answer.
Usually at least three choices are offered and the best
tests offer four or five choices. A good examination
should contain at least four choices so that the influence
of chance correct answers by guessing will be very slight.

Underline the correct or best answer:

1. The most important product of Chile is: (1) gold; (2) nitrates;
(3) cattle; (4) wheat.

2. The architectural style of the great cathedral at Amiens is:
(1) Corinthian; (2) Gothic; (3) Ionic; (4) Doric; (5) Roman.

3. The Romantic movement was interested in: (1) the spread
of religion; (2) the life and ideas of the middle ages; (3) the
problems of science; (4) the maintaining of the existing order
of things; (5) a return to the Roman Catholic church.

The Completion Type. — Here one or more words in a
sentence are omitted and the pupil is asked to fill in
words so as to make the best or truest statement. Ex¬
amples are as follows:

1. -is the capital of France.

2. Robert E. Lee, of-, was-of the--army.

3. Boyle’s Law states that when a-is subjected to com¬
pression and kept at a constant temperature the-is

-proportional to the pressure.

The possibilities of the completion type of question are
very great. The questions may range from the very
simplest to extremely hard ones. The objection to this
form of examination is the difficulty of scoring. The
scoring is less objective than that of many of the other
types. Many degrees of goodness may be represented in
the answers given, and the scorer must call upon his sub¬
jective judgment in marking these answers.

The Matching Type. — Here two lists are presented to
the pupils and the problem is to match an item of the

352

NEW-TYPE EXAMINATIONS

first list with the appropriate item of the second. For
some types of information this is a very good exercise.

Place the number of the author opposite the appro¬
priate book:

Authors

1. Burns

2. Tennyson

3. Byron

4. Goldsmith

5. Keats

6. Thackeray

7. George Eliot

8. Scott

9. Dickens
10. Macaulay

Books

Eve of St. Agnes
Mill on the Floss
Oliver Twist
Idylls of the King
Deserted Village
Cotter’s Saturday Night
Kenilworth
Prisoner of Chillon
Lays of Ancient Rome
Vanity Fair

Such lists may be made more difficult by not having
the same number of items in each column. It is best to
make the first list a little longer than the second.

The Analogy Type. — Here the student is presented
with two words in a certain relationship and he must then
find the appropriate word that will fit a third word in the
same relationship, or in other words he must complete
the analogy. Examples are:

1. Automobile is to carriage as motorcycle is to (1) horse, (2)
walking, (3) buggy, (4) bicycle, (5) train.

2. Circle is to square as sphere is to (1) circumference, (2) cube,
(3) round, (4) corners, (5) ball.

3. Anger is to fighting as fear is to (1) rage, (2) attack, (3) ela¬
tion, (4) flight, (5) pain.

Preparation and Use. — The samples above are merely
the most common types of the new objective examina¬
tion. There are other types and teachers may adapt any
type to suit their convenience. It will be seen further
that the construction of proper objective examinations is
much more difficult than the construction of the old-
fashioned essay-type examination. We cannot do better

PREPARATION AND USE

353

than quote here the suggestions of Paterson (1) with ref¬
erence to the preparation and use of objective questions:

“1. Questions covering every phase of the course should be
utilized to insure wide sampling of pupil knowledge.

“2. An excess number of questions should be prepared to allow
ample opportunity for the selection of the best questions
for the examination proper.

“3. Ambiguous questions both with respect to meaning and
possible answer should be rejected.

“4. The apparent difficulty of a question should not be the
basis for either accepting or rejecting a proposed question.

“5. Acceptable questions should include an equal number of
easy, hard, and moderately difficult questions.

“6. The first half dozen or so questions should be so easy that
practically all can answer them, thus serving as a ‘ shock
absorber.’

“7. Each acceptable question should be an independent unit
in the examination.

“8. Each acceptable question should be short.

“9. The examination should include a very large number of
questions.

“10. Each form or type of question should be segregated, the
examination consisting of as many parts as there are types
of questions.

“11. Within each part of the examination the questions should
be arranged according to topical sequence in the course.

“12. The examination itself should be preceded by suitable
general directions.

“13. Specific directions should be given for each segregated
group of questions.

“ 14. There should be a random arrangement of true-false ques¬
tions, with approximately an equal number of true and
false statements.

“ 15. The correct answers among the alternative answers in the
single-choice and in the plural-choice questions should be
placed according to chance.

“16. A uniform method of marking the papers, together with
the use of a colored pencil in scoring, should be used.

354

NEW-TYPE EXAMINATIONS

“ 17. Scoring formulae should not be used except possibly for the
true-false type of questions, when a right-minus-wrong
scoring formula may be used.

“18. ‘Weighting’ of questions according to difficulty or import¬
ance is rendered unnecessary in new-type examinations.

“19. Total scores should be computed for the examination
papers, distributed on a graph or table, and then a key for
converting total scores into letter grades derived.

“20. The examination should be mimeographed or printed, and
both used and unused copies should be kept under lock
and key in order to avoid the possibility of coaching.

“21. The prevention of coaching should also be accomplished
by using duplicate forms for classes in the same subject
taking the examination at different hours or on different
days and by changing the examination questions from
semester to semester.

“ 22. A large file of questions should be developed for each course,
so that a reservoir of from 1500 to 2000 objective questions
would be available from which examinations in endless
variety could be quickly assembled and used as occasion
demands. The ideal plan is to determine the diagnostic
significance of each question, thus developing a large list
of valid questions to be used in the preparation of examina¬
tions.”

Advantages of New-Type Examination. — The great
advantage of this type of examination is the reliability
of the scores, because of the objective system of marking.
Reliabilities can be calculated in the usual way by the
coefficient of correlation. Furthermore, the marks of
such examinations lend themselves more readily than do
the marks of the old-type examination to adequate sta¬
tistical treatment. We may, if we wish, compute T
scores for our class and compare the progress of the class
or the pupil from time to time with a great deal of relia¬
bility. The distributions of marks will tell us whether
our examinations are adequately testing the class or
whether they are too easy or too hard. In short, the

SUMMARY

355

marks of adequate new-type examinations give us a reli¬
able measure of the learning progress of our pupils.

There is no need to discard the old essay-type of exam¬
ination entirely. It can still be used with profit, but it
should be used sparingly, and the teacher should be
aware of the difficulties in marking which it presents.

Summaky

1. Class-room examinations are the teacher’s measure of the
particular modifications that he is trying to bring about.

2. The old essay-type examination depends too much upon the
subjective impression created in the examiner.

3. This type of examination has been shown to be very unreliable.

4. The rise of the objective examination is due to the develop¬
ment of techniques in connection with group intelligence and
educational tests.

5. Useful types of the objective examination are True-False,
Multiple-Response, Completion, Matching and Analogy.

6. In the preparation and use of these examinations many de¬
tails must be watched.

7 . The great advantage of the new-type examination is its relia¬
bility as a scientific measure of the progress of the pupil.

356

NEW-TYPE EXAMINATIONS

Review

True-False Statements

As a review mark the following statements true or false:

1. Teachers should not be allowed to use the essay-type of
examination any more.

2. The best type of objective examination is the true-false type.

3. A new-type examination should contain easy, medium and
difficult questions.

4. The marks given by different teachers to the same essay-
type of examination paper have been found to vary enor¬
mously.

5. One advantage of the new-type examination is its greater re¬
liability.

Advised Readings

Paterson, D. G.: Preparation and Use of New Type Examina¬
tions , Yonkers, 1925.

Russell, C.: Classroom Tests, Boston, 1926.

References

1. Paterson, D. G.: Preparation and Use of New Type Examina¬
tions, Yonkers, 1925, 64-66.

CHAPTER XVII

CONCLUSION

We have now come to the end of this brief and ele¬
mentary survey of educational psychology. Starting
with the thesis that the school child or rather the learn¬
ing individual is the center of interest for our study, we
found it necessary to make an artificial division into the
original tendencies of the individual and the modifica¬
tions made by the school. Any such division is merely
a matter of convenience. It helps us to bring order into
our thinking. We never find original tendencies and
modifications separated or clearly differentiated in any
individual. Modifications begin at birth and are con¬
tinually taking place. Yet at the same time every mod¬
ification is influenced by the original tendencies of the
child. There is, thus, a continual inter-action, which
makes any child at any given moment a very complex
and difficult subject to sum up. Hence the attempt on
the part of psychologists to attack this complex situation
from two points of view; from the point of view of the
original tendencies possessed by the child and from the
point of view of the modifications built upon these orig¬
inal tendencies.

The difficulty of keeping to this division has already
been apparent to the student in reading this book. The
conditioning of instinctive tendencies might well have
been taken up in Part II as an aspect of modification,
because it is a change or modification of such tendencies.
We have kept this discussion in Part I to show that there
are original tendencies very early in life and that they
are very modifiable. Again, much of the chapter on the

357

358

CONCLUSION

measurement of non-intellectual traits might easily
have been placed in Part II, as showing the measurement
of non-intellectual traits conditioned by the environ¬
ment. There can be no clear-cut division. It is a matter
of convenience as to where we make the division.

All living, all experience means modification of the
experiencing organism. Within certain limits human
beings are very modifiable. Each human being again
possesses his own amount of modifiability. Some take on
changes very readily. Others are very refractory to
change. Helping to cause these differences in modifia¬
bility are the differences in original tendencies possessed
by different individuals. Hence the measurement prob¬
lem for the psychologist is two-fold. On the one hand he
attempts from the measurement of one set of reactions to
infer the original tendencies possessed by the individual.
And on the other hand he attempts by the measurement
of other reactions to infer the amount of modification that
has taken place. There is nothing inherently different
in the reactions themselves. What inferences as to mod¬
ifications or original tendencies we may draw from the
test-reactions will depend upon the group being tested.
Thus we frequently see similar test material used in intel¬
ligence and in educational tests. Whether it is good for
the measurement of original tendencies or of modifica¬
tion will depend upon the children being compared. For
all educational measurement is a comparison between
children. Our norms or standards are always derived
from what some group of children has done.

All life, all experience, we have said, leads to modifica¬
tion. In civilized society, the school is the most important
agency set up by man to make certain definite modifica¬
tions. The school works intensively upon the raw mate¬
rial of human nature for many years, in order to achieve
certain modifications. At times the cynic feels inclined
to say that the modifications achieved by the school are

CONCLUSION

359

mighty few, and so he unconsciously pays a tribute to
the power of original nature. We also pay homage to the
force of these original tendencies when we say, “human
nature cannot be changed,” meaning thereby that all
modifications made are relatively slight and superficial.
On the other hand no one would dream of abandoning
the school in civilized society, and the optimists among
us look to it very largely to change and reform society.
If the school is not doing all it should, their answer is
more schooling.

These conflicting modes of thought place before the
psychologist a two-fold task — the measurement of orig¬
inal tendencies and the measurement of their modifica¬
tion. The school has always been aware of this problem.
From the very beginnings of organized education, we
note the tendency to examine, in order to find out what
modifications the school has made. And in the educa¬
tional world of to-day, the school examination plays a
large and important part. The other aspect of the situa¬
tion — the measurement of original tendencies — has
never entered very clearly into the minds of educators.
Although conscious of the differing abilities of children,
they have not taken them into consideration very seri¬
ously until the present time. The tests of the educa¬
tional psychologist are attempting to meet this need.

The problem of the modifiable individual is not only a
problem of testing, it is also a problem of investigating
how modifications actually take place. And hence a
large part of our subject deals with the measurement of
modifications as they are taking place under varied con¬
ditions. From such experimentation we gain some idea
as to how children learn and as to the most favorable
methods of learning. Only the general results are treated
in a text such as this. The specific application of these
results must be worked out for every department of
study. This is being done in some cases, and so we speak

360

CONCLUSION

of the psychology of arithmetic, the psychology of reading
and so forth. As a matter of fact, research in educa¬
tional psychology is so recent, that merely the broad
outlines have been established at the present time. The
teacher must take these general rules and suggestions
and apply them specifically to the problems of the class¬
room as he meets them in his daily work.

GENERAL BIBLIOGRAPHY

Book, W. F.: The Intelligence of High School Seniors, New York,

1922.

Buckingham, B. R.: Research for Teachers, New York, 1926.
Carr, H. A.: Psychology, New York, 1925.

Colvin, S. S.: The Learning Process, New York, 1914.
Freeman, F. N.: Mental Tests, Boston, 1926.

Galton, F.: Hereditary Genius, New York, 1880.

Gates, A. I.: Psychology for Students of Education, New York,

1923.

Gault, R. H.: Social Psychology, New York, 1923.

Gesell, A.: The Mental Growth of the Pre-School Child, New
York, 1925.

Hartshorne, H., and May, M. A.: Studies in the Nature of
Character, I. Studies in Deceit, New York, 1928.
Hollingworth, L. S.: The Psychology of Subnormal Children,
New York, 1920.

Hollingworth, L. S.: Gifted Children, New York, 1926.
Jordan, A. M.: Educational Psychology, New York, 1928.
Koehler, W.: The Mentality of Apes, New York, 1925.
McCall, W. A.: How to Measure in Education, New York, 1922.
McDougall, W.: An Introduction to Social Psychology, London,
1922.

Meumann, E.: The Psychology of Learning, New York, 1913.
Morgan, J. J. B.: The Psychology of the Unadjusted School
Child, New York, 1924.

Norsworthy, N., and Whitley, M. T.: The Psychology of Child¬
hood, New York, 1918.

Perrin, F. A. C., and Klein, D. B.: Psychology , New York,
1926.

Peterson, J.: Early Conceptions and Tests of Intelligence, Yon¬
kers, 1925.

Pillsbury, W. B.: Education as the Psychologist Sees It, New
York, 1925.

361

362

GENERAL BIBLIOGRAPHY

Pintner, R.: Intelligence Testing, New York, 1923.
Poffenberger, A. T.: Applied Psychology, New York, 1927.
Pyle, W. H.: The Psychology of Learning, Baltimore, 1928.
Robinson, E. S.: Practical Psychology, New York, 1926.

Ruch, G. M., and Stoddard, G. D.: Tests and Measurements in
High School Instruction, Yonkers, 1927.

Russell, C. R.: Classroom Tests, Boston, 1926.

Sandiford, P.: Educational Psychology, New York, 1928.
Spearman, C.: The Abilities of Man, New York, 1927.

Starch, D.: Educational Psychology, New York, 1927.

Symonds, P. M.: Measurement in Secondary Education, New
York, 1927.

Terman, L. M.: The Intelligence of School Children, Boston,
1919.

Terman, L. M., et al.: Genetic Studies of Genius, Volume I,
Stanford University Press, 1925.

Thomson, G. H.: Instinct, Intelligence and Character, New York,
1925.

Thorndike, E. L.: Educational Psychology, Three Volumes,
New York, Teachers College, Columbia University, 1913.
Thorndike, E. L.: The Psychology of Arithmetic, New York,

1922.

Thorndike, E. L., et al.: The Psychology of Algebra, New York,

1923.

Thorndike, E. L., et al.: The Measurement of Intelligence, New
York, Teachers College, Columbia University, 1927.
Thorndike, E. L., et al.: Adult Learning, New York, 1928.
Trabue, M. R.: Measuring Results in Education, New York,

1924.

Van Wagenen, M. J.: Educational Diagnosis, New York, 1926.
Watson, G. B.: Experimentation and Measurement in Religious
Education, New York, Association Press, 1927.

Watson, J. B.: Behavior, New York, 1914.

Watson, J. B.: Psychology from the Standpoint of a Behaviorist,
Philadelphia, 1924.

Woodrow, H.: Brightness and Dullness in Children, Philadel¬
phia, 1919.

Woodworth, R. S.: Psychology, New York, 1921.

APPENDIX I

THE ASSOCIATION TEST

Do not read or glance at what follows until you have
taken the test described in Chapter III, page 56. If you
read this before taking the test, you may as well not take
the test, as your results will be perfectly useless either
to yourself or to anyone else.

If you have carried out the instructions as laid down in
the text, you are now ready to score your paper. Below
is the list of 18 stimulus words followed by two response
words. If you have responded to the stimulus word by
either of the two response words, put a check mark on
your papers. Check only the common responses given
below:

Stimulus Word Common Responses

1. Color

Red 61

Blue 26

2. Furniture

Chair 80

Table 8

3. Flower

Rose 61

Violet 10

4. Letter of Alphabet

A 76

B 8

5. Metal

Iron 46

Gold 29

6. Historic Personage

Washington 50

Napoleon 14

7. Part of Speech

Noun 46

Verb 36

8. Geometrical Figure

Triangle 43

Square 23

9. Verb

Run 27

Go 24

10. Tool

Hammer 46

Saw 14

11. Article of Food

Bread 52

Meat 7

12. Part of Body

Arm 35

Head 23

13. Day of Week

Monday 46

Sunday 20

14. Room in House

Parlor 28

Dining 18

15. Animal

Dog 37

Horse 27

16. Book

Psychology 12

Bible 6

17. Girl’s Name

Mary 20

Helen 9

18. Country

America 27

U. S. 20

363

364

APPENDIX I

The numbers following the response words show the
percentage of students in university classes, that re¬
sponded with the words in question. These percentages
are simply put there as a matter of information.

Now count up the number of common responses which
you obtained in the experiment. We shall call that your
score. Let us now compare your score with scores ob¬
tained by other students who have taken this test. The
writer has given this test to many groups of students
studying educational psychology. Here are the results.
Qi means the lower quartile and shows the score below
which 25 per cent of the class fell; m is the median score —
the score made by the middle person in the class, half of the
scores lie above and the other half below this median score;
Q 3 is the upper quartile and 25 per cent of the class lie
above this score; the range shows the lowest and highest
scores obtained by any member of the class; n means the
number of students in the class who took the test.

Class

Median

Range

1 .

4-17

119

4-17

5-17

262

4-18

6-17

130

4-17

131

6-16

6-18

In general the middle 50 per cent of students in a class
are likely to have scores between 9 and 14. If your score
lies between 9 and 14, then you have reacted in the most
normal or ordinary way for students of educational psy¬
chology in general to react. If your score lies above 14,
anywhere from 15 to 18, then you have responded with
more common responses than is usual. You belong among
a group of 25 per cent of students who give these higher
scores. As your score tends to reach 18, you become

APPENDIX I

365

more and more unlike the mass of students in your asso¬
ciations; you have too many common responses. If your
score lies below 9, anywhere from 4 to 9, then you be¬
long to the 25 per cent of students who give few common
responses. If your score should be below 4, then you have
fewer common responses than any of the 872 students I
have so far tested.

This is just a little experiment to give you an idea of
how association tests work. Do not be alarmed if you
have given too many or too few common reactions. I am
comparing your scores with results obtained by normal
university students. There is here no question of in¬
vestigating abnormality or insanity. Just what the
significance of the number of common responses is, it is
rather hard to say. Perhaps the more sociable we are,
the more we think alike. The less sociable, the less likely
are we to think like our fellows and the more likely are
we to develop thoughts and associations somewhat dif¬
ferent from our fellows. One of my students tried to
follow up this idea and obtained a rating of “sociability”
from the students themselves and from other students.
When she correlated these ratings with the number of
common associations on this test she found a positive
correlation of .45 with the self ratings and of .33 with the
ratings by others.

Just why we react with a certain word, it is frequently
very difficult to say. If you have a class discussion on
the results of this test, you will be surprised at the differ¬
ent kinds of reasons that the students will give for the
specific reactions they have made. Some of the reactions
can be explained by frequency, that is, they are associa¬
tions between words or ideas that are very frequently
used, such as “furniture—chair,” or “ geometrical figure—
triangle.” Because they are the first of a series probably
accounts for ‘Tetter—A,” and “day of week—Monday.”
Note in the latter connection that Monday is generally

366

APPENDIX I

the most frequent response regardless of the day on
which the test is given. “Book—psychology” indicates
the general nature of the environment under which the
test was taken, but remember that the day of the week
did not influence the response. It is interesting to note
further that Napoleon is thought of as an historic person¬
age more frequently than Lincoln, that many girls re¬
spond to “girl’s name” by “Mary” or “Helen” and
very rarely do they give their own name. Every now
and again some student will give one or more responses
of an uncommon type that are directly connected with
some striking event that has just occurred to him. See
if you can find samples of such reactions in the class, but
be critical about accepting all of the explanations offered.
I have known students to say that they responded to
“historic personage” by “Napoleon,” because they had
just been studying or reading about Napoleon. I doubt
such an explanation, in view of the fact that in general
about 14 per cent of students give such a reply.

This test can be given to children in the grades. You
can try it with classes of school children, if you are a
teacher. It is surprising how early in life these common
associations seem to be formed. Here are the commonest
responses given by 236 older children, age thirteen and up,

and by 119 younger

children below age

thirteen:

Stimulus

Older Children

Younger Children

1. Color

Red

2. Furniture

Chair

3. Flower

Rose

4. Letter of Alphabet

5. Metal

Gold

6. Historic Personage

Washington

7. Part of Speech

Noun

8. Geometrical Figure

Triangle

Square

9. Verb

Run

10. Tool

Hammer

11. Food

Bread

APPENDIX I

367

Stimulus

Older Children

Younger Child

12. Part of Body-

Arm

Head

13. Day of Week

Monday

14. Hoorn in a House

Kitchen

Bedroom

15. Animal

Horse

16. Book

History

17. Girl’s Name

Mary

18. Country

America

A Town

APPENDIX II

JUDGMENT OF INTELLIGENCE FROM

PHOTOGRAPHS

The ranking according to the intelligence tests of the
twelve children is given below. The children were tested
by the Yerkes-Bridges Point Scale and the intelligence
rating is the C.M.A. or Coefficient of Mental Ability,
which is not identical to the I.Q. or Intelligence Quo¬
tient. C.A. means Chronological Age and M.A. means
Mental Age.

Rank

Child

C.A.

M.A.

C.M.A.

Description

5.7

1.93

Very bright

7.2

1.64

Very bright

12.

1.21

Bright

10.8

1.09

Normal

6.8

.97

Normal

12.5

.90

Normal

.88

Slow

9.7

.81

Slow

10.

.80

Slow

9.8

.72

Very dull

8.6

.72

Very dull

7.6

.47

Feebleminded

369

370

APPENDIX II

The instructor should work out a rank correlation
with the class. He should work one example on the
blackboard and have the class follow step by step as
explained in the text. Here is a sample of how the work
should appear on the blackboard and on each student’s
sheet.

Child

My Rank

Test Rank

Difference

D 2

f P

= i _ 6 ^d 2

n(n 2 ~ 1)

1128

= 1 -

1716

100

= 1 — .66

= + .34

188

1128

The following coefficients of correlation were reported
by the writer for different groups of individuals who have
taken this test.

APPENDIX II

371

Distribution of Coefficients of Correlation

Physi¬

cians

Psychol¬

ogists

Miscel¬

laneous

Students

Teachers

-08

-10

-06

-14

-07

-21

-19

-08

-32

-24

-01

-51

-30

-02

-09

-05

-63

-07

-26

-28

-29

n .

Av. r .

+ .05

+ .18

+ .05

+ .03

+ .09

Av. rank r. .

+-ii

+ .31

+ .06

+ .03

+ .17

Median r for all cases = + .10.

A full description of the original experiment will be
found in the following: Pintner, R., “Intelligence as Es¬
timated from Photographs,” Psych. Rev., Vol. 25, No. 4,
July, 1918, pp. 286-296.

372

APPENDIX II

Below is given another distribution of correlation co¬
efficients obtained from a large class of graduate stu¬
dents in education who made their judgments from a
screen projection of the same pictures.

Distribution op Correlation Coefficients

Coefficients

No. of Cases

Plus 63 to plus 87

Plus 38 to plus 62

Plus 13 to plus 37

Minus 12 to plus 12

Minus 37 to minus 13

Minus 62 to minus 38

Total

Median Coefficient = + .14

Sv.

• —t \ i i, i l

INDEX OF SUBJECTS

Abbott-Trabue poetry tests, 325,
334

Accuracy, 245-246
Achievement, educational, 308-
334

Achievement quotient, 288, 336-
338

Alcohol, 303-304
Analogy tests, 352
Arithmetic, learning of, 224-227,
231-232; tests of, 319-320, 333
Army alpha test, 111-113, 135
Association tests, 54-62, 363-367
Ayres handwriting scale, 323, 333

Conflict, 40

Constancy of I.Q., 125-127
Courtis standard tests, 316, 319,
333

Cousins, resemblance of, 169
Cramming, 242

Deaf, 156-157
Deceit, tests of, 78-81
Delinquency, 152-156
Downey will temperament test,
83-84

Dreams, 53

Drugs, effect of, 303-305

Barr American history tests,
325, 334

Binet scale, 54, 94-100, 160
Blind, 156

Brigg’s spelling scales, 324, 333
Buckingham arithmetic problem
scale, 320, 333

Burgess picture supplement
scale, 321, 333

Capacities, 6-7
Character, 70-78, 168
Charters diagnostic language
tests, 317, 325, 334
Classification by intelligence
tests, 136-143
Coffee, 303-304
College entrance, 144-146
Columbia Research Bureau for¬
eign language tests, 326, 334;
geometry test, 320, 333
Community of ideas test, 56-59,
363-367

Completion tests, 351
Complex, 48-49
Conditioning, 28-36

Educational age, 328
Educational quotient, 329
Educational tests, 308-334; uses
of, 335-346

Emotions, 6-7, 18-20, 27-32;
and learning, 232-237; tests
of, 76-77

English, tests of, 324-325, 333-
334

End spurt, 218
Errors, 244-245
Examinations, 348-356
Extroversion, 41

F scores, 337

Family history studies, 171-175
Fatigue, 287-300
Foreign languages, tests of, 326,
334

Forgetting, 254-261
Formal discipline, 266-268, 278
Freudianism, 49-51

Gates reading tests, 323, 333,
342-344

Geography, tests of, 325, 334
373

374

INDEX OF SUBJECTS

Gray oral reading tests, 322, 333
Gregory-Spencer geography
tests, 325, 334

Group tests of intelligence, 100-
114

Guidance, 143-144

Handwriting scales, 323, 333
Herring revision of Binet Scale,
100

History tests, 325-326, 334
Honesty, tests of, 78-81
Hotz algebra scales, 320, 333

Imagery, 248-249
Improvement, knowledge of,
243-244

Inheritance, 165-179
Initial spurt, 218
Instincts, 6, 16-18
Intelligence, 89-161; inheritance
of, 165-179; judgments of, 90-
93, 369-372; tests of, 94-116
Intelligence quotient, 99
Interests, tests of, 81-83
Introversion, 42
Iowa high school content exam¬
ination, 319, 333
Iowa physics test, 326, 334
Iowa spelling scales, 324, 333

Kent-Rosanoff test, 55, 59
Kirby grammar tests, 325, 334
Kuhlmann revision of Binet
Scale, 114

Learning, age and, 246-248; effi¬
ciency of, 224-250; laws of,
185-194; permanence of, 254-
261

Learning curves, 197-221
Learning periods, distribution of,
239-243; length of, 239-243

Manipulation, 20-21
Matching tests, 351-352
Memory devices, 233-236
Mental discipline, 266-268, 278

Merrill-Palmer tests, 115
Mnemonics, 233-236
Models, value of, 230
Monroe silent reading tests, 322,
333

Moral traits, 78-81
Morrison-McCall spelling scales,
324, 333

Motivation, 214-217, 248, 298-
299, 344

Multiple response tests, 351
Musical capacity tests, 115-116

National intelligence test, 105-
106, 130, 135

Negroes, intelligence of, 158
Nervous breakdown, 296
New-type examinations, 348-
356

Occupation and intelligence,
149-152

Otis self-administering test, 111,
135

Performance tests, 100-101
Personality, 67
Physiological limit, 218-220
Pintner-Cunningham mental
test, 103-105, 135
Pintner educational survey test,
318, 333

Pintner non-language test, 108-

110, 130, 135, 156-157, 160
Pintner-Paterson Performance

Scale, 100

Pintner rapid survey test, 110—

111, 135

Posey-Van Wagenen geography
scales, 325, 334

Powers chemistry test, 326, 334
Psychoanalysis, 51-52
Psychology, field of, 3-12

Racial differences in intelligence,
157-161

Rating scales, 73-75
Rationalization, 45-47

INDEX OF SUBJECTS

375

Readiness, 191

Reading, learning of, 227-228,
232; tests of, 321-323, 333,
339

Reflexes, 6

Ruch-Cossmann biology test,
326, 334

Ruch-Popenoe science test, 326,
334

Science tests, 326, 334
Seashore musical capacity tests,
116, 152
Sleep, 300-301

Social status and intelligence,
176-178

Special classes, 140-143
Specific capacities, tests of, 11 fl-
lie, 152
Speed, 245-246
Spelling tests, 324, 333
Stanford achievement test, 310-
312, 314-315, 318, 333
Stanford revision of Binet Scale,
94-98, 125-128, 169, 170
Starch spelling test, 312
Study, how to, 249-250
Sublimation, 43-45
Suggestibility, 77
Survey of intelligence, 146

T scale and T scores, 328
Temperament, tests of, 83-84
Terman group test, 105-108, 135
Tests, character, 71-85; intelli¬
gence, 93-115; educational,
309-334

Thorndike handwriting scale,
317, 323, 333

Thorndike-Hillegas composition
scale, 324, 333

Thorndike intelligence examina¬
tion, 108, 135

Thorndike-McCall reading
scales, 316, 321, 333, 339

Thorndike word knowledge test,
322, 333

Tobacco, 304-305

Transfer of training, 263-284

True-false tests, 350

Twins, resemblance of, 169-170

Unconditioning, 30-32

Van Wagenen composition
scales, 324, 333

Van Wagenen history scales,
326, 334

Van Wagenen reading scales,
325, 334

Ventilation, 301-303

Vocalization, 21-23

Whole and part learning, 237-239

Wilson language error test, 325,
334

Woodworth personal data sheet,
76

Woody arithmetic scales, 316,
320, 333

Woody-McCall mixed funda¬
mentals, 320, 333

Yerkes-Bridges point scale, 369

LIBRARY

INDEX OF NAMES

Angell, 272
Arai, 291, 292
Ash, 291

Bennett, 272
Bergstrom, 290
Bettman, 290
Bolton, 291, 295
Book, 118, 145, 147, 164, 257,
262, 361
Briggs, 276
Buckingham, 361
Burgerstein, 290
Burt, 154, 164

Caldwell, 286
Carr, 361
Chapman, 213
Clothier, 151, 153, 164
Cobb, 313, 332
Colvin, 361
Coover, 272
Courtis, 286

Dawson, 295
Day, 157

Dearborn, 271, 273
Dickson, 164
Downey, 88
Dugdale, 181

East, 181

Ebbinghaus, 254, 255, 295
Ebert, 271, 273

Feleky, 19
Foster, 212
Fracker, 272
Freeman, 134, 361
Freud, 49, 50, 51, 53
Friedrich, 295
Fusfeld, 157

Galton, 166, 167, 171, 181, 361
Gates, 66, 227, 228, 252, 257,
262, 323, 332, 343, 347, 361
Gault, 361

Gesell, 22, 23, 39, 72, 88, 361
Gilbert, 272
Goddard, 174, 175, 181
Graves, 127, 134

Haggerty, 176, 181
Hart, 46, 66

Hartshorne, 78, 80, 85, 88, 168,
361

Healy, 154, 155, 164
Heck, 295
Henmon, 291
Hertzberg, 231, 253
Hewins, 276

Hollingworth, H. L., 303, 304,
307, 361

Hollingworth, L. S., 313, 332,361
Holmes, 290
Hull, 307
Hurlock, 248, 253

Irwin, 117, 141, 164

James, 271, 273
Jones, H. E., 260, 262
Jones, M. C., 31, 39
Jordan, 361
Judd, 247, 272

Kafemann, 291
Kelley, 326
Kent, 55, 66
King, 295
Kitson, 252
Klein, 196, 361
Kline, 272
Knight, 298, 307

378

INDEX OF NAMES

Koehler, 361
Kornhauser, 252
Kuhlmann, 134

Laser, 295
Laslett, 307
Lindley, 291

Mais, 286

Marks, 117, 141, 164
Mathews, 76, 88
Maxfield, 214

May, 78, 80, 85, 88, 168, 361
McCall, 114, 328, 337, 347, 361
McDougall, 361
McGeoch, 256, 262
Meumann, 271, 273, 361
Monroe, 326, 332
Morgan, 66, 361
Morrison, 262
Muscio, 299, 300
Myers, 244, 253

Nash, 176, 181
Noble, 139
Norsworthy, 361

Oehrn, 290
Otis, M., 77

Paterson, 135, 353, 356
Paulu, 216
Pavlov, 33, 34
Penney, 286
Perkins, 276
Perrin, 196, 361
Peterson, 361
Pillsbury, 361

Pintner, 58, 66, 92, 114, 134, 135,
139, 157, 161, 164, 362, 371
Poffenberger, 164, 297, 362
Pyle, 241, 253, 362

Raup, 187, 196
Rayner, 39
Remmers, 298, 307
Ritchie, 291
Ritter, 295

Robinson, 292, 307, 362
Rosanoff, 55, 66

Ruch, 326, 332, 362
Rugg, 276
Russell, 356, 362

Sandiford, 176, 181, 362
Scott, 151, 153, 164
Seashore, 116, 135
Sikorski, 295
Sleight, 272
Spearman, 362
Specht, 291

Starch, 217, 240, 276, 286, 291,
312, 362

Stoddard, 326, 332, 362
Strayer, 340, 347
Stutsman, 135
Swift, 276

Symonds, 36, 39, 278, 286, 326,
332, 362

Taylor, 257, 262
Terman, 73, 82, 88, 95, 117, 118,
135, 143, 164, 174, 176, 181,
362

Thomson, 362

Thorndike, 18, 39, 119, 120, 123,
135, 223, 224, 225, 227, 242,
243, 247, 252, 253, 255, 262,
272, 277, 278, 279, 286, 289,
290, 291, 295, 302, 362
Trabue, 362

Van Wagenen, 362
Vogt, 290

Watson, G. B., 85, 88, 362
Watson, J. B., 27, 28, 39, 362
Watts, 45, 66
Webb, 272
Wells, 66
Wentworth, 170
Weygandt, 290
Whipple, 252, 291
Whitely, 256, 262
Whitley, 361
Winch, 272, 276
Woodrow, 215, 362
Woodworth, 18, 27, 39, 76, 272,
362

'.’Xnflv wuumho *

■

t*« 7 S7

I i ¥+'.*

!W P

« hi eru m

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Educational psychology : an introductory text"

See other formats