Skip to main content

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 *' 


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





v> k —-— * 

F i) i 10 In C 

M, U. 

'R. YAL Kjg 




^ tA 

//'\\ s 


\ * ( 3 1 » 

1- » 

/} /' 

f»|! 1 

in on 

i \j L*. i 



V \ *V 

0 i\._ 

.. y-" 











v . */' 



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 

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 



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 



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. 



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 - ~~~ ? 




I. The Field of Educational Psychology . 3 


II. The Original Tendencies .... 15 

III. The Functioning of Original Tendencies 40 

IV. The Measurement of Non-Intellectual 


V. The Measurement of Intellectual 


VI. The Uses and Results of Intelligence 

Testing .136 

VII. The Inheritance of Mental Traits . 165 


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 





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 


F C' - 



B. M. J. 

■Cp/CA L 


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 


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

23. Distribution of Gifted and Unselected 


24. Frequency Curves for High School Seniors 118 

25. Composite Curve for Ninth Grade . .119 





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 


32. Occupational Intelligence Ratings 

33. Occupational Standards on Intelligence 


34. The I.Q.’s of Repeated Offenders 

35. Comparison of Deaf and Hearing Children 

36. Intelligence of Foreign-born Men in the 


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 





























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 


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 


75. Progress in Spelling. 

76. Learning Curves for Addition of Fractions 

77. Improvement in Educational Achievement 























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 


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 . 














]- J r : i-'Q cy re 


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. 



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 



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 

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. 


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 



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 


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 

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 


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 

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- 


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. 


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 






6 . 

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

Tests of 

Continuous Modification. Fatigue. 

Measurement of Modifications in Educa- school sub- 





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 

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 

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

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





B. M. J. 





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 




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. 




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 



reaching, handling 

answering in some way 


pick up receiver 


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 

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 

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


1. Sensitivities 

2. Attentiveness 

3. Bodily control 

4. Food, habitation 

5. Fear, fighting 

6. Social instincts, gregarious¬ 

7 . Manipulation 

8. Expression of emotions 

9. Learning, remembering 


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. 

Manipulation, exploration 




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 

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

<f \ 


( ' ’> 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.) 



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 



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 

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 



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 

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. 



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 

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- 




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 



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 

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 


M anipu- 













with par¬ 
ents and 



proval of 



Toys, use¬ 
ful things 







to other 

proval of 











proval of 











proval of 





















tions of 
all sorts 







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 



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- 



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. 



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. — 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. 



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 


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 

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¬ 


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. 



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.” 


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

S. 2. Rat-■> 

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

S. 2.-■> 


S. 3. Rabbit-> 

S. 4. Fur Coat-> 

No Transfer. 

S. 5. Cotton Wool-> 

S. 6. Mask-■> 


S. 2.-» 

R. 1. Crying (original tend¬ 

R. 2. Manipulation (original 

R. 1. Repeated seven times 

R. 1. Original tendency condi¬ 

R. 1. 

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¬ 


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


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 



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 



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 



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. 


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. 



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

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 




True-False Statements 

As a review mark the following statements true or false: 

1. Manipulation can be called an instinctive tendency to 

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¬ 

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. 



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¬ 

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. 


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 

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. 



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 




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 


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 

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 



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 


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 



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 


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 



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. 


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- 



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 

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 


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, 



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 


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¬ 

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 


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 



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 

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¬ 

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 

(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 


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 


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 

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 



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.) 


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 



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 







Time in Seconds 








Stimulus Word 

Reaction Word 

Time in Seconds 

7. Ask 



8. Wool 



9. Spiteful 



10. Lake 



11. Sick 



12. Ink 



13. Swim 

Can Swim 


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 

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 



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 


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. 


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

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 

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. 



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. 



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 

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 

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 

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 

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. 



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

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 

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


( ) 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. 


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

of work rules his wife and children with an iron 

( ) 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 


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. 


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, 

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. 




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 




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: 


big stick 
rough and ready 

physically weak when young 

writer of books 


very informal 
typical blusterer 

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


writer of interesting letters 

man of letters 

high idealism 

an introvert 


died broken-hearted 


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



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 

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- 



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¬ 

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- 


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 



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- 



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 

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 



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


* pH 



















• -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 

' ^ 

« S « 

P 02 3d 
02 r O 












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 








CD biD O 
02 d 

® Cl 


i y=i 

«+h cd 


02 02 

9? 3 £ & £ 

£ «g 

* s 







CO 03 

4J dj 



a . < 

02 4i> 03 

S'S $ 


d be r „ 


d 2 ■ r/ 

H °S 


>5 2 



=4* I 

02 P 
02 o3 

• rH 



* i*"i 

to p- 02 

8 -j# M 

d to 

2 02 

.5 -a 

O! 44 






1 > 








£ A 

Tj *4-3 

a> s 

3 £ 

C2 to 








< 4-1 









, ri 

44 d 02 
*h d d bJO 

£ £ £ $ 

d gg 3 

- 02 cd 









f -1 












o H 03 

C 4 






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 
























b£) d 

C4 .2 

02 *43 


>; d 


02 445 

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



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. 



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 

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 



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 



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 



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. 



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 



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- 



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 



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. 



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¬ 

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. 



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. 


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. 



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. 




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 

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 

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

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

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) - 



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 

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. 


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. 




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 



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) 



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 


“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 

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 


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 

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 


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. 



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¬ 

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 


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¬ 

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. 



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. 




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


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. 



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 

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 


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¬ 


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.) 




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- 


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 


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. 


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 



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) 


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





( 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 






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.) 


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 

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 



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- 


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. 



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 — 












(S) 10 

(T) 5 

(U) 14 

(V) 15 



















(A) 8 

(B) 11 




* ... 







(F) 28 

(G) 30 

(H) 29 


(K) 26 









(L) 10 









5 — 














(M) 2 









63 — 

(D) 64 


(F) 47 

(G) 55 











(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 






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. 


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 



2. ®©©©©®@(D® 

5. O O O A/o 




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 



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 



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- 


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. 



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

the Merrill Palmer School.) 


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- 


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. 



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. . . 

Register of Voice. . 

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- 




.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.) 


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.) 


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- 


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 






















, r r 




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- 


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 

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¬ 

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 


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.) 


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- 


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 

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 


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 

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 



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 

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 

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: 










































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

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 

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- 


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 


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 


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 

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- 



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. 


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¬ 

4. Intelligence tests may be divided into group and individual 

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. 


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 

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 

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 

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. 




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 

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¬ 

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 

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. 


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


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 


-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: 


2 .- 


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. 


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, 


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

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



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, 


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, 


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, 

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

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




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 

























12-11 18-11 14-11 















5B2 1 











5B3 2 













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 



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: — 





















Total 93 


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 

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- 



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 

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. 


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 

























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 



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. 



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- 



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 







A or B 







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- 



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- 



90 100 


UO 120 130 140 150 160 170 

10 25 


75 90 95 






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 



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 

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 


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 

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. 



D - 



C + 










































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 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 


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 



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. 



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 

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 



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.) 



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. 


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

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 



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 

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 


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. 



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 











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.) 



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. 



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). 


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. 



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 

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 

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




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. 



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. 


1. Book, W. F. : The Intelligence of High School Seniors, New 

York, 1927. 

2. Burt, C.: The Young Delinquent, New York, 1925, Chapter 


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 

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. 



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- 



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 

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 



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- 


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 



with increase in accuracy and adequacy of the trait 

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, 


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 



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 


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¬ 

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 


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 


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 




4 . mo. 




<L 1837. 

In] B 












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- 


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. 


Fathers of 







Public Service 









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 



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 



















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 

This general hierarchy of occupations with reference to 
the intelligence of the men in them can be seen in Figure 


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 


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 



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 

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. 



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. 



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. 


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. 




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 




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. 



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 



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 

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 



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. 



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 



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 



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 



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. 



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. 


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 

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 

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. 




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 

5. Ordinarily individuals do not differ very much in readiness 
to respond. 

6. Reactions followed by satisfaction tend to be repeated and 

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¬ 

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 

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-. 



Advised Reading 

Perrin, F. A. C. and Klein, D. B.: Psychology , New York, 1926, 
Chapter V. 


1. Raup, R. B.: Complacency , New York, 1925. 



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. 




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- 


Time in Seconds for Writing the Alphabet Backwards: 

Thirty Trials 

Trial No. 


Trial No. 






























































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 



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 



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 

Figure 42. — Practice Curve as in figure 40; average of three 


Figure 43. — Practice Curve as in figure 40; average of five 




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 


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 


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 



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- 


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¬ 

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. 



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 


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 


Figure 49. — Practice Curve. Writing the alphabet backwards. 

Subject F. 



Figure 50. — Composite Learning Curve. Writing the alphabet 


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- 



ever, shows steady improvement from the beginning of 
the practice to the end. 


Time in Seconds for Writing the Alphabet Backwards: 
Medians and Quartiles for Each Trial for 88 Students 



































































































































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 


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). 







2 8 

£ g 






0123456 7 89 10 


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. 














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. 









= 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 








5 6 





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 












g 1000 


1 2 3 4 5 6 7 8 9 10 


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 



) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 


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 

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 


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 

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.) 




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. 



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- 



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.) 


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 



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- 



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 



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. 


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 



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. 




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 

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 


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¬ 

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. 



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. 

(c) we are satisfied with our present level. 

(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. 



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, 



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 

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- 



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¬ 

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 



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 

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- 



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 

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. — 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. 



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- 


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. 




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 



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: 









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 



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: 

















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. 



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 



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 


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 



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. 



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, 



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 



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. 



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 


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- 



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 

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 , 



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 

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 



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. 






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 



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 



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. 



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 

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 

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. 




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¬ 

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. 



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¬ 

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. 

(c) never go together. 

26. Check the best response: 

Learning is a process of: 

(a) passive absorption. 

(b) eliminating original tendencies to react. 

(c) forming new connections. 

(d) storing up knowledge. 

Advised Readings 

Whipple, G. M.: How to Study Effectively, Second Edi¬ 
tion, Public School Publishing Co., Bloomington, Ill., 

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, 


1. Gates, A. I.: The Improvement of Reading, New York, 


2. Gates, A. I.: New Methods in Primary Reading, Teachers 

College, Columbia University, New York, 1928. 



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. 



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 — 






*3 60 




% 40 




20 — 












0 ! 2 6 14 20 

Interval between learning and relearning: in days. 


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 




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 

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 


Figure 67. — A curve of forgetting for meaningful material. 
(From McGeoch and Whitely, Journal of Educational Psychology , 

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- 



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 

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. 


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 

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 



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 

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 


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 


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. 




1. Disuse weakens. Forgetting begins as soon as we stop 

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. 



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 

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 

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 

Advised Readings 

Thorndike, E. L.: Educational Psychology: Briefer Course , 
Teachers College, Columbia University, New York, 1914, 
Chapter 17. 


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. 



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 




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 


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¬ 

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 



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 



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 



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 



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.” 



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 

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. 


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¬ 

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 


Laboratory Experiments in Transfer of Training 




Transfer to 





Memory pas- 

Other memory 

Slight or 




Ebert and 


Non-sense syl- 

Numbers, let- 

22% resid- 


ters, words, 

ual gain 

poetry, etc. 

acc. to 



TABLE 4— Continued 

Author Date 


Transfer to 




Memory of 

Similar stimuli 

16% resid- 

sound in- 




3% resid- 





Poetry, prose, 

Dates, sylla- 

3-4% resid- 


bles, etc. 






10% resid- 




Gilbert and 

Reaction to 






touch, color, 

No controls 

Thorndike and 


Areas, weights, 

Similar stim- 

52% down 

lines, letters, 


to zero 



Marking words 

Other words 

75% as much 
as in train- 





Nouns, verbs 


of e. s. 

Less than 






Other colors 


tion of shades 
of red 


Coover and 

Sound intensi- 


10% gain 








Theory of re- 

Target under 




ment when 
depth is 




Other mazes 


Coover and 


Very little 



Sorting Cards 


or doubtful 


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 



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- 


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 




School Experiments in Transfer of Training 




Transfer to 





Mental mul- 

Adding, divid- 

29% resid- 


ing, etc. 






Slight or 










Correct usage 


Foreign Languages: 



Latin and 


Latin most 







Size of English 

3-4% in 


favor of 

Latin stu¬ 




Spelling, defini- 


tions, etc. 




Foreign lan- 

Writing and 



reading tests 

(no transfer 
for memory 






Biological tests 

34% resid¬ 


5% resid- 







32% resid- 




7% resid- 

rical tests 


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. 


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- 



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 


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 



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 

The Mechanism of Transfer. — How does transfer 
take place? We have suggested this in all that has gone 


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 



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 



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. 


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 

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 



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 

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¬ 




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 

10. There may be transfer of attitudes as well as transfer of 

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 

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. 



17. To promote a professor, distinguished for research, to the 
position of dean indicates a belief in a great amount of 

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. 

(c) associative shifting. 

(d) concentration. 

Advised Readings 

Starch, D.: Educational Psychology, New York, 1927, Chapter 

Thorndike, E. L.: Educational Psychology: Briefer Course, 
Teachers College, Columbia University, New York, 1914, 
Chapter 18. 


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. 



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 



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- 


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- 



Summary op Laboratory Fatigue Experiments 

Experimenter Date 









ing Ger¬ 






















for digits 



tion, areas 




1 hour 


None; per¬ 
haps gain 

4 hours 



36 min. 



or none 

1000 trials 


Slight in¬ 
crease in 

2 hours 

10 adults 

Very slight 
fatigue ef¬ 
fects. In¬ 
in nonsense 

90 min. 

3 adults 

erable. One 
fourth bet¬ 
ter after 

90 min. 


able. 24% 
better after 

75 min. 


9% better 
after rest 

3 to 8 





TABLE 6 —Continued 

Experimenter Date 

























3 place X 

3 place 






tion 2 place 
X 2 place 



tion 4 place 
X 4 place 

Starch & Ash 










1 hour 

3 adults 


1 hour 


erable de¬ 

90 min. 8 


2% better 
after rest 

1 hour 




even after 

10 min. 

2 hours 



4 to 12 


more time 
than after 

\y 2 to 2 


effect of 


2 hours 




12 hours 


Increase in 
equals 75% 
of initial 

2 hours 


4.3% loss 

10 hours 


Slight drop 
in number 
rise in er¬ 


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 


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 


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 

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- 




Results of School Fatigue Experiments 







Writing from dictation 

Errors increase slightly 



Memory for digits 

No loss in efficiency 



Addition and mul- 

Work improved (prac- 





Addition, multiplica¬ 
tion, dictation 

Very slight decrease 



Memory for digits. 

No difference. No 

Addition, multiplica- 

real difference. 

tion. Completion ex- 

Decrease in effi- 


ciency (?) 



Marking letters. Dic¬ 

No difference. 

Dictation of unrelated 

Slight decrease 



Adding, multiplying, 

No difference between 

cancelling, memorizing, 

late and early work 



Addition, multiplica¬ 
tion, judgment 

No difference 




Increase in amount; 
decrease in accuracy 




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 


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 



1 Good 






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. 


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 



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. 



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- 


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. 



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 


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 

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- 



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, 



Comparing the Effect of Alcohol and Coffee on Various 

Tests (after Hollingworth) 














+ 8 

+ 10 

+ 19 








— 7 

— 13 

— 14 


+ 3.8 


— 6 

— 10 

— 20 


— 4.0 

Color Naming 

— 2 

— 7 

— 12 


+ 4.4 


— 5 

— 12 

— 23 


+ 6.2 



— 15 

— 16 


+ 2.5 











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 

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 



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. 


1. There is no real difference between mental and physical 

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 

4. Laboratory experiments in fatigue show us that an individ¬ 
ual may work several hours at a difficult task without serious 
loss of efficiency. 


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 




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 

5. The curve of satisfyingness always follows the curve of 

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. 


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, 




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 




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 

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 


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 

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. 


Figure 73. — Improvement in Reading and Arithmetic for one 


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- 


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¬ 

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 



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 


a « 
2 § 

O e 
■+* ps 







• rH 





















'S3 ° 
. -S 3 00 
CO o ^ 


O £ 

*C -S 

a os 

S3 >; 
02 ^ 
. 02 


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 


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 

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. 



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 


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 



“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 


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 



are not meant to give detailed information about specific 

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 

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 


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 

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. 



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 


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 — 

Reading has so many aspects that no one test can 
measure all of them. The most thorough-going attempt 



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 


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- 



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 

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- 


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- 



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- 


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 



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 


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. 


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. 



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. 



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 


Advised Reading 

Gates, A. I.: The Improvement of Reading, New York, 1927. 


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 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, 

Woody Arithmetic Scale: T. C. Bureau of Publications. 

Woody-McCall Mixed Fundamentals: T. C. Bureau of Pub¬ 

Buckingham Scale for Problems: Pub. Sch. Publishing Co. 

Hotz Algebra Scales: T. C. Bureau of Publications. 

Columbia Research Bureau Geometry Test: World Book Co., 


Thorndike-McCall Reading Scales: T. C. Bureau of Publica¬ 

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¬ 

Gates Silent Reading Tests: T. C. Bureau of Publications. 


Thorndike Handwriting Scale: T. C. Bureau of Publications. 

Ayres Handwriting Scale: Russell Sage Foundation. 


Morrison-McCall Spelling Scale: World Book Co., Yonkers. 

Iowa Spelling Scales: Pub. Sch. Publishing Co. 

Briggs’ Sixteen Spelling Scales: T. C. Bureau of Publications. 


Thorndike Extension of Hillegas Scale: T. C. Bureau of 

Van Wagenen English Composition Scales: World Book Co., 



Wilson Language Error Test: World Book Co., Yonkers. 

Charters Diagnostic Language Test: Pub. Sch. Publishing 

Kirby Grammar Test: University of Iowa. 


Abbott-Trabue Exercises in Judging Poetry: T. C. Bureau 
of Publications. 

Van Wagenen Reading Scales in English Literature: Pub. 
Sch. Publishing Co. 


Gregory-Spencer Geography Test: University of Cincinnati. 

Posey-Van Wagenen Geography Scales: Pub. Sch. Publish- 
ing Co. 


Barr Diagnostic Tests in American History: Pub. Sch. Pub¬ 
lishing Co. 

Van Wagenen American History Scales: T. C. Bureau of 

Foreign Languages 

Columbia Research Tests: World Book Co., Yonkers. 


Ruch-Popenoe General Science Test: World Book Co., Yon¬ 

Powers Test for General Chemistry: World Book Co., Yon¬ 

Ruch-Cossmann Biology Test: World Book Co., Yonkers. 

Iowa Physics Test: Pub. Sch. Publishing Co. 



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. 



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 


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 

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 


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 

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 



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, 


Achievement in Silent Reading in Tampa, Florida, Schools 
for White Children Compared with Other Places 

May, 1925 

Thorndike-McCall Test for Understanding of Sentences 



4 Senior 

6 Senior 

7 Senior 

8 Senior 

Tampa, Florida 





Baltimore, Maryland 





33 Wisconsin Cities 





Paterson, N. J. 





St. Paul, Minn. 





Louisville, Ky. 





Hackensack, N. J. 










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. 



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 


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 

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 

Type C — Reading to Understand Precise Directions. 

Type D — Reading to Note Details. 



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: 



























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- 


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 



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. 


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 

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 


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. 




True-False Statements 

As a review mark the following statements true or false: 

1. Achievement tests should be substituted for the usual school 

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. 


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. 



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 




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 



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. 


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 



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: 


1. Burns 

2. Tennyson 

3. Byron 

4. Goldsmith 

5. Keats 

6. Thackeray 

7. George Eliot 

8. Scott 

9. Dickens 
10. Macaulay 


Eve of St. Agnes 
Mill on the Floss 
Oliver Twist 
Idylls of the King 
Deserted Village 
Cotter’s Saturday Night 
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 



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 

“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 

“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. 



“ 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¬ 

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 



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. 


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. 




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¬ 

5. One advantage of the new-type examination is its greater re¬ 

Advised Readings 

Paterson, D. G.: Preparation and Use of New Type Examina¬ 
tions , Yonkers, 1925. 

Russell, C.: Classroom Tests, Boston, 1926. 


1. Paterson, D. G.: Preparation and Use of New Type Examina¬ 
tions, Yonkers, 1925, 64-66. 



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 




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 



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 



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. 


Book, W. F.: The Intelligence of High School Seniors, New York, 


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, 


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, 

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, 

Peterson, J.: Early Conceptions and Tests of Intelligence, Yon¬ 
kers, 1925. 

Pillsbury, W. B.: Education as the Psychologist Sees It, New 
York, 1925. 




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, 

Terman, L. M., et al.: Genetic Studies of Genius, Volume I, 
Stanford University Press, 1925. 

Thomson, G. H.: Instinct, Intelligence and Character, New York, 

Thorndike, E. L.: Educational Psychology, Three Volumes, 
New York, Teachers College, Columbia University, 1913. 
Thorndike, E. L.: The Psychology of Arithmetic, New York, 


Thorndike, E. L., et al.: The Psychology of Algebra, New York, 


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, 


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. 



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 

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 




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. 







1 . 
















































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 



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 



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 



Older Children 

Younger Children 

1. Color 



2. Furniture 



3. Flower 



4. Letter of Alphabet 



5. Metal 



6. Historic Personage 



7. Part of Speech 



8. Geometrical Figure 



9. Verb 



10. Tool 



11. Food 






Older Children 

Younger Child 

12. Part of Body- 



13. Day of Week 



14. Hoorn in a House 



15. Animal 



16. Book 



17. Girl’s Name 



18. Country 


A Town 




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. 












Very bright 






Very bright 
















































Very dull 






Very dull 










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 


My Rank 

Test Rank 


D 2 




















f P 

= i _ 6 ^d 2 






n(n 2 ~ 1) 












= 1 - 












= 1 — .66 






= + .34 














The following coefficients of correlation were reported 
by the writer for different groups of individuals who have 
taken this test. 



Distribution of Coefficients of Correlation 







































































n . 






Av. r . 

+ .05 

+ .18 

+ .05 

+ .03 

+ .09 

Av. rank r. . 


+ .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. 



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 


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 




Median Coefficient = + .14 


• —t \ i i, i l 


Abbott-Trabue poetry tests, 325, 

Accuracy, 245-246 
Achievement, educational, 308- 

Achievement quotient, 288, 336- 

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, 

Cousins, resemblance of, 169 
Cramming, 242 

Deaf, 156-157 
Deceit, tests of, 78-81 
Delinquency, 152-156 
Downey will temperament test, 

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, 

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- 

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, 

Forgetting, 254-261 
Formal discipline, 266-268, 278 
Freudianism, 49-51 

Gates reading tests, 323, 333, 

Geography, tests of, 325, 334 



Gray oral reading tests, 322, 333 
Gregory-Spencer geography 
tests, 325, 334 

Group tests of intelligence, 100- 

Guidance, 143-144 

Handwriting scales, 323, 333 
Herring revision of Binet Scale, 

History tests, 325-326, 334 
Honesty, tests of, 78-81 
Hotz algebra scales, 320, 333 

Imagery, 248-249 
Improvement, knowledge of, 

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- 

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, 

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- 

Occupation and intelligence, 

Otis self-administering test, 111, 

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, 

Rating scales, 73-75 
Rationalization, 45-47 



Readiness, 191 

Reading, learning of, 227-228, 
232; tests of, 321-323, 333, 

Reflexes, 6 

Ruch-Cossmann biology test, 
326, 334 

Ruch-Popenoe science test, 326, 

Science tests, 326, 334 
Seashore musical capacity tests, 
116, 152 
Sleep, 300-301 

Social status and intelligence, 

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, 

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, 

Woodworth personal data sheet, 

Woody arithmetic scales, 316, 
320, 333 

Woody-McCall mixed funda¬ 
mentals, 320, 333 

Yerkes-Bridges point scale, 369 



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, 

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 



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, 

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, 

'.’Xnflv wuumho * 


t*« 7 S7 

I i ¥+'.* 

!W P 

« hi eru m