Skip to main content

Full text of "ERIC ED010498: EXPERIMENTATION WITH COMPUTER-ASSISTED INSTRUCTION IN TECHNICAL EDUCATION. SEMI-ANNUAL PROGRESS REPORT."

See other formats


REPORT RESUMES 

£0 010 198 08 

EXPERIMENTATION WITH COMPUTER-ASSISTED INSTRUCTION IN 
TECHNICAL EDUCATION. SEKI-AJNUAL PROGRESS REPORT. 

BY- HITZEL, HAROLD E. BRANDON, GEORGE L. 

PENNSYLVANIA STATE UNTV., UNIVERSITY PARK 

REPORT NUMBER BR-5-0035 PUB DATE 31 DEC 66 

REPORT NUMBER ERD-I99 
CONTRACT OEC-5-85-D74 

EDRS PRICE MF-S0.16 KC-S5.00 125P. 

DESCRIPTORS- ^COMPUTER ASSISTED INSTRUCTION, ^PROGRAMED 
INSTRUCTION, 4PR0GRAMED MATERIALS, COMPUTER ORIENTED 
PROGRAMS, COMPARATIVE ANALYSIS, *VCCATIOMAL EDUCATION, 
INSTRUCTIONAL MATERIALS, INFORMATION DISSEMINATION, 
INSTRUCTIONAL DESIGN, ^EDUCATIONAL STRATEGIES, UNIVERSITY 
PARK, PENNSYLVANIA 

THE THIRD 6-MOMTH PERIOD OF OPERATION OF A 
COMPUTER-ASSISTED INSTRUCTION (CAI) EFFORT IN TECHNICAL 
EDUCATION WAS COVERED IN THIS REPORT. THE OBJECTIVES OF THE 
TOTAL PROGRAM (A 4-YEAR EFFORT) WERE (1) TO COMPARE CAI WITH 
OTHER EDUCATIONAL STRATEGIES, (2) TO PREPARE INSTRUCTIONAL 
MATERIALS, (3) TO TRAIN PERSONNEL, Al\!0 {4, TO DISSEMINATE 
RESULTS OF RESEARCH. FIVE BARRIERS *0 YtfE DEVELOPMENT OF CAI 
WERE DISCUSSED— (1) DELAY OF PROGRAM DEVELOPMENT WAITING FOR 
IMPROVED HARDWARE AND VICE VERSA, (2) LACK OF EXPERIENCE AND 
METHODOLOGY FOR CONSTRUCTION OF ACHIEVEMENT MEASURES, (3) 
EXCESSIVE TIME REQUIRED TO WRITE A CAI PROGRAM t (4) LACK OF 
KNOWLEDGE Of THE APPROPRIATE BALANCE BETWEEN CAI AND TEACHER 
INSTRUCTION, AND (5) RESTRICTION OF EXCHANGE Or CAI PROGRAMS 
DUE TO LACK OF COMPATIBILITY Or COMPUTERS. IT WAS CONCLUDED 
THAT IT WAS INAPPROPRIATE AT THE TIKE OF REPORTING TO 
MAINTAIN THAT CAI DOES OR DOES NOT PROVIDE MORE EFFICIENT 
METHODS OF TEACHING. (AU 



i 

i' 

t 

> 



o 






5* Oo5s 
EP.P- Iff 



1 

I 

1 

3 

I 



o© 

ON 

Sf- 



Ui 



I 

I 

I 

I 

I 

I 

i 

1 



i 

I 




ASSISTED IHSmUCm LABORATORY 



l COLLEGE OF EDUCATION • CHAMBERS BUILDING 

1 



1HE PENNSYLVANIA . „ 
STATE UNIVERSITY ^NiVERsITY rfliSC, PA. 



EXPER t MENTA? t ON WITH 
COMPUTER -ASSISTED INSTRUCTION IN 
TECHNICAL EDUCATION 



&& DS?A2TMS*,T C? f2«!2 T», ETJSSfiCfJ «£> ^'HLF^.RS 

OTfcd&swrca ra r®e?ted frtmlife 

Fg&ft Ci C'-r-^n >t r <* & cjm&ss 

Cs ut v [/ Kpr*xl tfn& x .effw si bscsism 
f&ftsn Gf ^Vr/s 



SEM I -ANNUAL PROCRESS REPORT 
DECEMBER <§!» ISS6 




THE PENNSYLVANIA STATE UNIVERSITY 



COMPUTER ASSISTED INSTRUCTION LABORATORY 
UNIVERSITY PARK 3 PENNSYLVANIA 



Semi-Annual Progress Report 
EXPERIMENTATION WITH COMPUTER-ASSISTED INSTRUCTION 
IN TECHNICAL EDUCATION 
Project No. 5-35-074 



PRINCIPAL INVESTIGATORS 



Harold Eo Mitzel 



George L. Brandon 



ASSOCIATE INVESTIGATORS 



Marilyn A. Adams 
Terry A. Bahn 



Helen L « K. Farr 
David A. Gilman 
Keith A. Hall 



Harriett A. Hogan 
Robert V. Igo, Jr. 
Donald W. Johnson 
Betta H. Kriner 



Kenneth H. Wodtke 



CONTRIBUTING AUTHORS AND RESuUCHERS 



Karl Borman 
Bobby R. Brown 
Patricia Fredericks 



Clara Gargula 
Nancy Harvil chuck 
Harold Sands 



John Tardibuono 



December 31, 1966 




TABLE OF CONTENTS 



TABLE OF CONTENTS 
LIST OF TABLES . 
LIST OF FIGURES . 



Page No . 
♦ 

. i 

• • • 

. m 

. iv 



CHAPTER 

I. INTRODUCTION ] 

Physical Facilities and Equipment 2 

Summary of Coursewriter Author Language 3 

Development of CAI Course Materials in 

Technical Education ...... 9 

Five Major Barriers to the Development of 

Computer- Ass is ted Instruction . , 13 



Student Performance Summaries for CAI Courses 21 

Some Comments Concerning Efficiency in the Preparation 
of Materials for Computer-Assisted Instruction 27 



II. ENGINEERING SCIENCE . . . 43 

III. TECHNICAL MATHEMATICS 55 

IV. COMMUNICATION SKILLS . 61 

V. RESEARCH REPORTS 



Cueing and Feedback in Computer-Assisted Instruction ... 75 

Relative Effectiveness of Three Modes of Presentation 
through Computer-Assisted Instruction , . . , . . 79 



i 



TABLE OF CONTENTS 



Page No. 

RESEARCH REPORTS (continued) 

# 

Application of a Modified Gagne'" Type Model to Computer 
Assisted Testing and Instructional Branching ...... 83 

An Instrument for the Measurement of Expressed 

Attitude toward Computer-Assisted Instruction ...... 95 

APPENDIX A 

Students within Sequence Numbers, All Separate 108 

Sequence Numbers within Students, All Separate 1C9 

Students Separate, Sequence Numbers Pooled . , 110 

Sequence Numbers Separate, Students Pooled ....... Ill 

Statistical Summary Listing .......... 112 

APPENDIX B 

Computer-Assisted Instruction for Technical Education . . 115 
APPENDIX C 

Student Attitude toward Computer-Assisted Instruction . . 121 
Spelling Achievement Test 127 







w 



LIST OF TA3LES 



TABLE TITLE Page No. 



1.1 Summary of Course Materials Developed* Tested and 

Revised as of December 31, 1966 10 

4, 1 Biographical Information on $s in Field Trial of 

Remedial Spelling Course ................ 67 

4.2 Scores on 37- Item Spelling Test for 25 Students in 

Field Trial , . . . . 68 

4.3 Rank Ordering of Absolute Gains with Number of Course 

Segments and Equipment Malfunctions < . . . . 69 

4.4 Attitude toward CAI by Gain and Sex .......... 70 

4.5 Tabulated Responses for Attitude Questionnaire Item . . 71 

5,1 Listing, by Item Number, of Negative, Positive, and 

Neutral Items ....... . . 98 

,2 Description of Study Group as to Courses Taken, 

Number and Percentage of Forms Sent Out, and 
Number and Percentage of Forms Returned . , 100 

5.3 Analysis of Variance Summary Table of Reliability » . . 101 



*6,4 Analysis of Variance Summary Table of 
Reliability by Strata ....... 



102 



FIGURE 



LIST OF FIGURES 



Page No, 

1.1 Sample Page from a Programming Workbook ........ 34 

1.2 Flowchart of Programming Options Available for Author 

Using Macro Techniques ...... ..... 38 

1.3 Flowchart of Steps Followed by Author Using a Macro 

p r°9ram . . . 39 

2.1 Flowchart for Example 1 - Introduction to Physics 

Program * . 47 

2.2 Flowchart for Example 2 - Atomic Energy Program .... 50 

2.3 Flowchart for Example 3 - Atomic Energy Program .... 53 

4.1 Flow Diagram of Spelling Program 62 



5.1 Hypothetical Model of Knowledge Hierarchy in which 
Lower Level Learning Sets Mediate Positive Transfer 

to Higher Level Sets . . * ....... . ... . . . . 85 

5.2 Partial Hierarchy of Learning Sets for the Criterion 
Performance of Converting a Number from One Base to 

Its Equivalent in Another Base 90 



CHAPTER I 
INTRODUCTION 



This report spans the third six months period. (Duly 1, 1966 
to December 31* 1966) of operation of a computer-assisted 
Instruction (CAI) effort In technical education and Is designed 
to show Penn State University stewardship of Its own resources 
and the federal funds awarded to It under the provisions of 
Section 4(c) of the Vocational Education Act of 1963. 

Briefly, the objectives of the original proposal were as 
follows : 

1. To evaluate the articulation of computer- 
assisted Instruction with other educational 
strategies and, by means of careful experi- 
mentation, determine optimum ways of pre- 
senting core courses in technical educa- 
tion curricula. 

2. To prepare curriculum materials for computer 
presentation with emphasis on the instruction 
of post-high school students In technical 
mathematics o engineering science, and 
communication skills. 

3. To train an Interdict pi inary group of 
individuals to prepare course materials and 
to do research on 'computer application in 
technical education. 

4. To disseminate the Information and evi- 
dence concerning the innovation of CAI 
and Its application to occupational educa- 
tion. 

Progress has been made toward all of these objectives 
and the evidence Is detailed In the following report. The 
first chapter deals with the physical facilities provided by 
the University, the equipment configuration in operation 
during the past six months, and & digest of the Saursawrlt&r 



language now being used by investigators in the Laboratory, 
Chapters II through IV describe the course development activities 
in technical education subjects Brief reports of research 
findings are presented ;n Chapter V. 

The project has been programed for an additional 30 months 
In which further attempts will be made to establish reliable 
knowledge concerning the application of the computer-assisted 
instruction technology to the field of occupational education. 

Physica l Facilities and Equipment 

• For the P ast year, the project has been housed in recently 
remodeled quarters located in 201 Chambers Building, at The 
Pennsylvania State University, University Park, Pennsylvania. 

The new quarters provide ample office space for approximately 
ten professional staff members, ten graduate research, assistants , 
and ten technicians. In addition to the office space for staff 
members, the Laboratory contains four air-conditioned, sound- 
proofed terminal rooms, two 8' 4" x 6' and two 6' 2” x 6‘, each 
containing one CAI student terminal, audio-visual components, 
and a printing desk calculator. 

The Laboratory is operating by means of dial access telephone 
to an IBM 1410 computer system which is housed in the Computation 
Center on the Penn State campus. In addition to the four terminal 
which are located in the CAI Laboratory on the campus, two 
terminals are now operating at the Williamsport Area Community 



3 



College, Williamsport, Pennsylvania, and two are located in 
Ivyside Building at Penn State's Commonwealth C»m P M S Altoona, 
Pennsylvania. CAI course materials are teleprocessed to students 
at these remote locations from the 1410 computer on the main 
campus. Students at the Williamsport Area Community College 
and at the Altoona Campus have been receiving computer-assisted 
instruction and have been participating in experiments since 
August, 1966. 

An IBM Magnetic Tape "Selectric" Typewriter and a 105" 

Card Reader have been acquired during the period to accelerate 
the preparation and computer input of CAI course materials. 

Summary of Co ursewri ter Author Languag e 

The courses listed in Table 1.1 have been written in the 
CAI author language known as Coursewriter . A complete description 
of Coursewriter is beyond the scope of this report; however, 
a summary of the functions of each of the operation codes in 
the language is provided below. This list covers most of the 
basic operations in the language. However, an author wishing 
to prepare CAI programs should study the more detailed Cou rsewriter 
manual (Coursewri ter 1965). In sdditlsn, Gilman and Harvilchuck 
(1967) have developed a set of training materials for providing 
new authors with instruction in the use of the Coursewriter 
language. These materials- will be reproduced as a separate 
technical report of the Penn State CAI Laboratory In the near 
future. 




Summary of Coursewriter Operation Codes 

Primary ^ 

" Computer types text and waits for the student to siqnal 
completion. Commonly used to display^ readinq assiqn- 
ment to a student. y 

Same as rd, but does not update restart address. 

.9Ji ~ Computer types text and waits for student to type a 

response. Commonly used to display questions or problems 
to a student. 

$ an * e ss qu, but does not update restart address. 



Major 



Cj3 - Correct answer to be stored in memory for comparison 
with student’s answer. 

cb^ - Similar to ca -- used to identify a set of alternate 
correct answers-- the subsequent action is to be the 
same regardless of which answer in the set is matched 
by the student's response. 

wa - Wrong answer to be stored for comparison with student's 
answer. 

wb - Similar to wa - used to identify a set of alternate 
wrong answers - the subsequent action is to be the 
same regardless of which answer in the set is matched 
by the student's response. 

aji - Anticipated answer - similar to ca and wa, but not 
followed by implicit branch. 

ub - Similar to aa - used to identify e. set of alternate 
anticipated answers - the subsequent action is to be 
the same regardless of which answer is matched by the 
student's response. 

un - Texc co be typed if the student's answer is not one 
of the specified correct or wrong answers. 

nx - Instructs the computer to execute the i nstruction(s ) 
immediately following the nx. »he purpose of the nx 
is to change a minor operation code (e-g., ty or fn) 
to a major ope rati or* code. 



2LL * Time limit -- computer ignores anything typed by the 
student after the specified time has lapsed. 



Minor 



ty_ - Computer types text and continues without waiting for 
any response from the student. 

b_r - Branch -- alters the sequence of execution 
Unconditional: branch is always taken 

Conditional: branch is taken only if a particular 

condition is satisfied. 

ad - Adds (algebraically) a number or the contents of a 
counter to a counter. Commonly used for accumulating 
a student’s errors or response times. 

11 “ Load -- clears a counter and adds a number (or the con- 
tents of a counter) into a counter. Load may also be 
used to set a switch to 0 or 1. 

dv - Divide contents of a counter by a constant or by the 

n cimhor ^ *1 p nf am 

MiliM w I r * i u WVUll VW V ■+ 

m£ - Multiply contents of a counter by a constant or by the 
-'umber in a counter. 

fpl - Display a slide. 

jjjO- Seek and position a slide, but does not display it. 
tpl - Play a tape recorded message. 

tj)0- Seek and position tape recorded message but does not 
play it. 

tpr - Record a tape recorded message. 

fn - Computer executes the specified function. Functions 

are special series of Instructions (written in a machine 
language subprogram) so that the computet can do 
processing which cannot be done by using only the 
Coursewriter operation codes. 

fin - slide//n - The displ ay slide function is used to present 
a slide; n represents tFFlTumber of the slide to be 
displayed. 



! ; 



6 

fn 



sli de//nx - The seek and pos i ti on slide function will 
seek and position slide n, but will not show the slide 
until a display slide function occurs in the program. 



fn - tape//n - The play tape function will play tape recording 
number n. 



n 

L 



r < 

u 






fn - tape//nx - The seek and pos i ti on tape function causes tape 
recording number n to be positioned. The recording will 
not play until a tape play function occurs in the 
program. 

fn •• dc// - The display c-counter function is used to display 
the contents of a c-counter to the student. 

fn - dx// - The display x-counter function is used to display 
the contents of an x counter to a student. 

fn - wait// - The wart function allows the author to delay the 
program before execution. 

fn - kw // - The key word function allows the author to specify 
one or more key words which must be matched in the 
student’s answer. 



P! 



tL 

m 



m 



fn - kwo/7 - The key words ordered function is similar to the 
key word function. However, the key word ordered 
function also requires that the matched key words in the 
student's response are entered in a specified order. 



fn - 



kwi// - The key words 

words that have been entered 
function finds a word in the 
the ca or wa, the function is 



initial f uncti on 
in the ca 



searches for the 
or wa. If the 
student's response not in 
termi nated. 



fn - 



ordered_ a nd initial 
"s tudentf 1 
s 



kw//io - The key words 
searches for key words in the 
function insists that the student' 
certain order and also checks to insure 
no unmatched key words in the student's response 



function 

s response. This 
response be in a 
that there are 



&£j 



Iff?* 

n 

y 



p 

<L'3 

E3 



fn - 



fn - 



lim - The 1 i mi ts function allows the 
mathematical limits within which the 
response will be acceptable. 



author to specify 
student's numerical 



paO// - The parti al answer zero function allows 
to disregard extraneous discrepancies between a 
response and the text of a ca, wa, or aa. This 
is used to process answers which are misspelled 
partially correct. 



an 



author 



student's 
function 
or 



0 



m 

0 






^ 1 * 
y i 



u 

O : 

ERIC 



p , 

I M 



H- 

h 




7 



i r 3n d/ / - The pseudo random i nteger function allows 
authors to specify that a pseudo random inteqer be placed 
in a c-counter. 

ic// - The initial characters function allows authors to 
specify that only a certain designated number of initial 
characters in the student’s response are to be compared 
with a subsequent ca or wa. The function also allows the 
author to specify that characters in certain positions 
of the student's response are irrelevant and are to be 
considered matched. 

ed// - The edi t function allows an author to edit the 
student s response by replacing or deleting characters. 

sb// and rb // - The save and branch function (sb) allows 
the author to insert in one place within a course a 
certain sequence of material (subroutine) which can be 
brancned to repeatedly thus limiting the necessity for 
programming the same material at repeated places within 
the course. The return branch (rb) function returns 
the student to a point in the course as directed by an 
address Indicated in the text of the sb function* 



9 



PRECEDING PAGE BLANK- NOT FILMED 

Development of CAI Course Materials in Techni cal Educatl on 

Summary of Course Materials 
Developed, Tested, and Revised 

One objective of the Penn State CAI Laboratory is to 
produce useful educational products in the form of CAI programs 
for post-high school technical education programs. The majority 
of CAI programs developed in the Laboratory are experimental 
in the sense that they have been designed to explore teaching 
strategies, and to develop the unique instructional capabilities 
of a high-speed electronic computer. Although the CAI courses 
are being developed in small segments for experimental purposes, 
many of these segments appear to have practical utility as instruc- 
tional materials. Preliminary evaluations of several courses 
have been completed and the data are encouraging with regard 
to the degree of student learning resulting from the courses. 

Course segments are being developed in three major areas: 
Engineering Science, Technical Mathematics, and Communication 
Skills. Table 1.1 represents an accumulative summary of all 
CAI course segments developed on the project as of December 31, 
1966. The table indicates the extent to which each course 
segment utilizes audio-visual communication, static displays, 
the number of students who have taken the course to date, the 
total number of system hours of student instruction, the length 
of each course segment estimated by the average time taken per 
student to complete instruction, and a column indicating whether 
the course segment has been revised following the examination 
of student performance on the course. 



10 



Table 1.1 

Summary of Course Materials 
Developed, Tested, and Revised 
as of December 31, 1966 



Course Segment 


1 No. of Slides 


No. of Tape j 

Messages 


No. of Static 
Displays 


No. of Students 


Q> 

_ to 
c= 

01 ^ 
-POP 

U) o c 
>> 0) 

</> C E 

O CD 
r— Q} 

$ !</, 

O 

f— 

HRS/MINS 


Q) 

E 

1 « |»» 
Olh- 

> 

C -P 

c 

• OJ 
P C5 
tfl 3 
LJ -P 
CO 

HRS/MINS 


Revised following 
evaluation? 


Communication Skills: 
Spelling Introduction 


2 


3 




36 


9 


0 


i 

0 


15 


yes 


Vocabulary 


9 


12 


-- 


36 


15 


0 


0 


25 


yes 


Diagnostic Test 




37 




25 


25 


0 


1 


0 


yes 


Plurals 


15 


— 


— 


8 


6 


0 


0 


45 


yes 


Suffixes 


-- 


10 


— 


5 


<*« 

O 


45 


0 


45 


yes 


Words with "e" or H y" 


1 


1 


— 


7 


2 


20 


0 


20 


yes 


Syllables 


— 


15 


— 


12 


6 


0 


0 


30 


yes 


Words with "i" or 'e“ 


— 


10 


— 


2 


0 


40 


0 


20 


yes 


Compounds 








6 


3 


0 


0 


30 


yes 


Discrimination 


4 


29 


— 


1 


0 


35 


0 


35 


yes 


Homonyms 


— 


10 


— 


17 


11 


14 


0 


1 

."N — 

0/ | 


yes 


Demons 


— 


29 


— 


19 


11 


5 


0 


35 


yes 


Proof readi ng 


— 


wm 


1 


25 


12 


30 


0 


30 


— 


Posttest 


— 


36 


— 


25 


25 


0 


1 


0 


— 


State Capitals - Paired- 
Associate Learning 

Keyword Function 








10 


25 


40 


2 


34 


yes 


Partial Answer Processing 








8 


22 


30 


2 


48 


yes 



Course Segment 


cn 

Cl 

•© 

CO 

o 

• 

o 


No. of Tape 
Messages 


No. of Static 
Displays 


(A 

4-> 

C 

03 

3 

CO 

4- 

o 

• 

o 

z 


O 

e£ 

Q) 3 
P O+J 

wUc 
>> <» 
v» C £ 
o 

/— Q) 

ns <u </> 
*» E 
o *<- 

h- ! — 

HRS/MINS 


01 

£ 

• *r* 

0)h- 

> 

<C -P 

c 

• 0) 

-o 

1A 3 
OJ 

CO 

HRS/MINS 


Revised following 
evaluation? 


Techni ea 1 Ma thema ti ts : 




















Converting Number Systems 




— 


4 


87 


187 


30 


2 


10 


yes 


Metric System 


4 






4 


4 


0 


1 


0 


yes 


Significant Figures 


17 


13 




5 


9 


0 


1 


45 


yes 


Intro, to Prob. Solv. 


mum 


— 


— 


3 


3 


0 


1 


0 


yes 


Kinematics & Calculus 


9 


11 


«• u 


6 


4 


0 


0 


40 


yes 


Vector Analysis 


17 


13 


w w 




0 


0 


2 


0 


— 


Trigonometry 


31 


20 


20 


i 

7 | 


i 

i Q* 

! 7 


0 


3 


30 


yes 


Printing Calculator 


3 


3 


4 


— 


0 


0 


0 


0 




Significant Figures 1 


— 


— 


7 


14 


22 


30 


1 


30 


yes 


SigfiglOO 






7 


29 


40 


50 


1 


25 


yes 


SigfiglOl 


— 


— 


7 


36 


48 


0 


1 


20 


yes 


Sigfig200 


-- 


— 


7 


11 


15 


0 


1 


15 


yes 


Exponential Rules and 




















Logan thms 


13 


(MW 


13 








! 2 


0 


— 


Engineering Science 




















Intro, to Physics 


10 


8 




2 


1 


45 


0 


52 


yes 


Working with Units 


9 


2 




55 


88 


6 


i 


29 


yes 


(flbr 


9 


2 


ww 


2 


2 


30 


l 


15 


yes 


(flcr 


9 


2 


— 


26 


29 


45 


0 


68 


yes 


{■Fisk 


9 


2 


-- 


20 


18 


0 


0 


52 


yes 


(medl 3 versions) 


9 


2 


16 


115 


115 


0 


1 


0 


yes 



*Represents student instruction time to cover one segment only 




12 



Course Segment 


* 7 

e/i 

O 

T5 

00 

4- 

O 

o ! 


No. of Tape 
Messages 


o 

to cn 

4- a. 
O C/7 

♦i— 

• Q 
O 

z 


No. of Students 


r " 

pc 

Total System 
s: Time on Course 
== Segment 


E 

* *r~ 

CT>*— 

> 

<C *-> 

* c 

o 0 

+-> -a 

CO 3 
LU 

</) 

HRS/MINS 


Revised following 
evaluation? 


Scientific Notation 


25 


— 




4 


4 0 


1 0 


yes 


Atom 


20 


19 


— 


14 


12 30 


1 0 


yes 


Basic Magnetism 


12 


7 


— 


2 


2 30 


1 15 


yes 


Optics* Part I 


28 


0 


— 


— 


*» *• m m 


1 30 


-- 


Optics, Part II 


31 


0 


12 




— 


1 30 


— 


Electrostatics 


22 


13 


»- 


— 


— 


2 15 


-- 


Micrometer & Vernier Caliper 


14 


— 


14 






2 0 


— 


Heat, Part I 


29 










2 0 


— 


Heat, Part II 


15 










1 30 


— 


Differential Knowledge Results 


» 

» 














Area of Mechanics 


50 










1 0 




Science Misconceptions 


30 ; 










2 0 


-- 






m 



& 

tL 



to 



ea 



m 



ti'TA 

fe 



P 

B 



m 



o 

ERIC 



13 



Five Major Barriers to the Development 
of Computer Assisted Instruction 



Harold E. Mitzel 



There are at least five plausible* more or less accurate 
definitions of computer-assisted instruction: (1) computer 

terminal as an adjunct calculating device and laboratory instru- 
ment used in a typical classroom, usually where mathematics or 
physics is taught; (2) computer used as a record keeper and 
retriever of student biographical and performance data; (3) 
computer simulation of real-life problem-solving, i.e., medical 
diagnosis, equipment trouble shooting, etc.; (4) computer as 
a recitation-receiving and evaluation device or "homework monitor" 
and (5) computer as a pre-programed control device utilizing 
multiple displays which tutor the learner in subject-matter 
content. For purposes of these remarks, I plan to restrict 
myself to the tutorial definition. 

The first major barrier to the development of CAI is the 
hardware-software gap. In a sense, this problem resembles 
the chicken-egg controversy. Some experts maintain that the 
electronic and mechanical devices, together with their care- 
fully tested operating systems, must be made available before 




1 



Remarks prepared for American Management Association 
Meeting, Americana Hotel, New York City, August 12, 1966. 












large scale content programs can be prepared on a meaningful 
basis. These specialists emphasize the necessity for compati- 
bility between various systems in order that carefully constructed 
programs of instruction may receive widespread dissemination 
and testing. They claim that the equipment is the horse that 
must precede the proverbial cart. 

An equally cogent argument is made by the group which 
believes a large number of courses should be developed first, 
and then the representatives of equipment manufacturers could 
examine these content materials and arrive at specifications 
for needed display and response gear. Both groups are correct 
from their own points of view, but the long-term development 
of CAI will probably be harmed by too rigid adherence to either 
extreme position. 

A second major barrier to the development of CAI is the 
lack of experience and methodological know-how for constructing 
appropriate criterion measures useful in evaluating CAI courses. 
The ubiquitous, end-of-term achievement examination will just 
not do for measuring the yield from a sophisticated, individu- 
alized CAI program. The difficulty is that we have, >n 
traditional instruction, become accustomed to offering a 
standard bill of fare to the learner, and we have based marks 
and “Brownie points" upon each learner’s scale position rela- 
tive to all other learners who were exposed to the same lec- 
tures, readings, stale jokes and class discussions. CAI, 




at its best, should offer a distinctly individualized course 
of instruction in which gaps in the learner's Knowledge are 
filled in during the course by means of diagnostic and reme- 
dial sequence steps. Thus, it seems to be theoretical ly 
appropriate to ask the typical CAI learner to achieve mastery 
of the content as long as we allow him a reasonable amour.t of 
time. By the same reasoning, the learner with limited aptitude 
may be unable to finish the course, even though he is given an 
unrestricted time limit. The major variable for CAI programs 
under these postulates is a ti me £core or a number-of- attempts 
score rather than an achievement score. For these reasons, 

CAI and conventional instruction may be fundamentally not 
comparable . Certainly, the typical end-of-term examination 
on which half of the learners miss half of the questions suggests, 
under CAI conditions, that there was something wrong with either 
the examination or the program which preceded it. If CAI 
lives up to its potential, it should be unnecessary to provide 
a special off-line immediate examination for any learner. If 
the examining is appropriately done at intervals throughout 
the program, then every learner should have achieved mastery 
of the content up to the limits of capacity. 

A third barrier to CAI development is the inordinate amount 
of time required of the subject-matter author in preparing a 
course of instruction for CAI presentation. In a few subjects 
involving drill materials, it is possible to form a series of 
words, symbols, or algorithms into a standard sequence. These 



n 




16 

sequences * called "macros," /e handy devices for building up 
a backlog of CA1 displays in limited areas of instruction . But, 
most curriculum material is focused on objectives of instruction 
that go beyond simple associ ati cnal learning. In our work 
with college level material at Penn State, we clearly recognize 
the "iceberg" analogy in CAI programs. When one student goes 
through a CAI programed sequence, typically his printout of 
the interaction between himself and the computer represents 
only about 12 to 20 per cent of the total stored program on 
that topic. In general, the more sophisticated the program 
in te^ms of alternatives and remedial opportunities, the more 
material is beneath the surface which the typical learner 
may never encounter. 

Simpler, easier-to-use programing languages will help 
to alleviate this problem to some extent. But perhaps the 
most headway will be made by setting aside a small portion 
of a computer executive system which will enable an author 
to make "on-line" input of a block of material at the termi- 
nal, and to test that material immediately without going through 
a laborious compiling process. This capability is becoming 
available in some of the recently developed CAI systems. We 
are currently exploring other strategies for getting a maximum 
of finished product from authors for a minimum expenditure 
of their time. Some appear to be promising, but are as yet 
untried. 



L aJ 



I \ 



R 

5 s 



n 



P’n 




F? 

i i 



n 



0 



n 



u 



i 

LsJ 

n 

LI 




i , 



r~ 



* 



t 1 

' — i 



r\ 

i ■ 

5. ! 




I s' 






*- 

r 

$ 

i 



k 



ERIC 



Li 



r j 

H 



i * 



La. 

Cj 



@ 



ri 

»' ; 



E 4 
* 1 



r> 

*i -> 

t. t 



lUJ 



n 



F? 



i ‘ 

fF » 

iu 



£7jJ 






17 



A fourth harrier to the development of CAI is our Uck 
of know ledge concerning the appropriate '‘mix'* between computer- 
madieteti instruction and tcacner-irediated components of instruc- 
tion. In a sense the teaching machine movement, in its headlong 
rush to solve all the problems of education, failed to come 
to nrios with the problem of the "mix." Should the learner 
spend 50 per cent of his time with the teacher and 50 per 
cent on the computer program? Can tin's ratio be 90-10 or 
10-90 and still be effective? How do the various conceptions 
of CAI use influence the cost benefit analysis? 

It seems entirely likely that the optimum recipe for 
intermingling CAI into existing forms of instruction will 
be different for different content, different learners and 
different situations, but the fundamental guidelines i.aye 
not yet been derived either from empirical research or experience. 

At Penn State during the recent spring term, we tried 
a field trial in three college courses in which CAI was 



designed to substitute for two of three regularly scheduled 
lectures per week. All students met with the instructor during 
the week’s first period, and the CAI students scheduled themselves 

V. 

on the computer terminals for approximately two and one-half 
hours on the average during the remainder of the week. Other 
students attended the lecture-c 1 ass discussion in the usual 
way. In a modern mathematics course, this arrangement worked 
well and the students functioned satisfactorily in the program. 

In two courses where heavy emphasis was placed on marks and 






hfliflaffamiflaa 



1 o 

i u 




on examination performance as the key factor io achieting 
mei'ks, the students felt restricted by the pace of the CAI 
program. Many wanted to see all of the material to be 
covered in the examination so that they could organize a 
self-study program tailored to the emphasized examination. 

These observations impressed upon us the fact that com- 
puter teaching programs are not as flexible for the student 
as existing textbooks and lectures. College students have 
certain expectations of the way in which courses are conducted, 
and they have rather firmly fixed stylized ways of responding 
to instructional "systems." 

A fifth barrier to the orderly development of CAI is the 
lack of compatibility between computer systems which tends to 
retard the free exchange of programs of instruction developed 
in different laboratories and curriculum centers. In my 
opinion, not one of the three or four major operating systems 
in use today offers much advantage over the others; yet as far 
as I know, no two programs of different origin will "run" 
on the same computer. In our own Laboratory, we cannot operate 
courses which "ran" last year because we elected to "improve" 
the system, as it were. 

A dimension of the compatibility problem is found in the 
controversy between single-purpose and multi-purpose computers 
for the CAI application. One leading manufacturer has recently 
announced an experimental class-sized configuration that will 



L 

n 

Lj 

n 
I > 

f i 



n 
\ > 
u 



ii 



iJ 



□ 

r-i 

u 



LJ 

R 

U 

f*l 






EJ 



u 



n 

L-j 

n 

y 



f* - 



1 r* 
I ^ 



r 



Sj 




n 



u 



fr-3 



tnske it pos c ^ bis to operas r P thirty-two student ststiuns 
simultaneously . This is a single purpose system that is not 
st this time compatible with any other known operational 
system, hence, teaching materials for the new system must be 
constructed frorr the beginning. Other manufacturers of hard- 
ware are apparently going to devise their own CAI operating 
systems in order to enhance the marketability of their 
equipment. 



What mechanisms should be devised In order to Insure 
compatibility of expensively produced programs on a variety 
of systems t The spectre of the teaching machine development, 
where there were at one time more different brands of machines 
than good programs, still hangs over us and we ought to 
demonstrate that we can learn from past mistakes, both our 
own and these of others. 



a 

n 

LJ 



n 



u 



n 



n 

L_J 



n 



n 

t . 

Li 



21 



PRECEDING PAGE BCANK-NOTFICMED 

Student. t_dr for ri an eg S umroa ^*1 e s 
for CAI Courses 

Terry A. Sahn 

This report on student records is divided into four 
secti ons : 

1. Uses of student records 

2. Requirements of good student records 

3. How our student records meet these requirements 

4. Heans by which our student records are obtained 

Uses of Student Records 

At this stage of CAI development at Penn State, the primary 
use of student records is for obtaining data from experimental 
studies. However, in the future student records will be used 
for examining progress made by students and for evaluating 
course segments, ecg., if many students have a high error rate 
on a segment, perhaps that segment is too difficult and should 
be subdivided. 

Bequl rements of Good Student Records 

(1) Student records should be complete; that is to say, 
they should contain all pertinent information including course 
name, student number, restart address of segment, response, 
response latency, response category, and the condition of all 
author-manipulated counters at the time the segment was executed. 
Other things which are useful, though not quite so pertinent. 



t 



oo 



tm U 



ara date, time, response character count, number- <?f trials of 
student on that segment, and a count of the unanticipated responses 

(2) Student records must be accurate. This is very 
important in all uses of such records. Stringent checks must 
be made to maintain high accuracy* especially on the more 
pertinent information. Unavoidable errors, e.g., those occurring 
at logging time in the Dean, should be flagged to indicate 

that they are in error so that CAI investigators may take the 
eonditiori into account. 

(3/ A method of obtaining a statistical summary must be 
available in order to facilitate the discovery of trends and 
deviations in performance and to facilitate segment evaluation 
and study of student progress. 



(4) Another important requirement is economy in terms 
of money and time. While monetary economy is a self-evident 
requirement, time economy may need some explanation. A one 
week turn-around would be desirable between the performance 
of the last learner in a study group and the availability of 
a student record summary of all learners in the group. 

(5) A final requirement of student records is flexibility. 
Student records should have the ability to adapt to investigator 
needs in relation to grouping order, pooling, and the suppression 
of information extraneous to author's purposes. 



1 

The control and monitoring system for IBM 1410 CAI 
system. 





Student Records Program at Penn State 

How well does the student records program at Penn State 

meet the requirements of instruction? 

Completeness , The student records are quite complete, 

including all of the items of information mentioned on the 

previous page (see APPENDIX A-5, p. 112). 

Accuracy . The student records are reasonably accurate 

including some flagging of log errors. However, there is 

a definite need to check the Dean in order to correct log 
* 

errors in the response categories. 

Statistical Summary . The statistical summary of student 
records is complete and flexible enough for most author uses. 

The summary includes student number, sequence number, attempts, 
mean latency, standard deviation of the latency, and frequency 
of correct response. The author has a choice of one of four 
methods (see APPENDIX A, p. 107) of statistical summary: 

1. Students within sequence numbers, all separate (p. 108) 

2. Sequence numbers within students, all separate (p. 109) 

3. Students separate, sequence numbers pooled (p. 110) 

4. Sequence numbers separate., students pooled (p. Ill) 
Since frequency of correct response will always be 1 

when both students and sequence numbers are taken separately, 
it is not listed in subroutines 1) and 2). 

If students or questions are pooled, the author can specify 
an error criterion. If the error rate exceeds this criterion. 




24 



the entry is flagged with the symbol *T* in the student record 
printout. This procedure helps the author to identify high or 
low error rate frames and students. 

Econo « . Student records for a typical course segment 
of 75 frames taken by 75 students costs approximately $50 
to $75 in computer time. This figure includes a statistical 
summary. 

Time - Student records for a typical course segment taken 
on-line by 10-30 students can be gathered in about one to 
three weeks from the request date. Unfortunately, this turn- 
around time is too long for instruction where more immediate 
feedback to teacher and learner is desirable. We are presently 
seeking ways to shorten this turn-around time. 

Flexibility. The student records program has sacrificed 

I 

some flexibility in order to increase efficiency and economy. 
However, they seem to be sufficient for present needs and 
can be adapted to future needs as the occasion arises. 

Methods Used for Obtaining Student Records Listing 

Two different machines and three different programs are 
used to obtain the student records listing. Program I, used 
on an IBM 1401, edits a 339 character log record to a 195 
character record. Program II is the IBM sort/merge program 
operated on an IBM 7074. This program sorts records according 
to course name, student, sequence number, time and anything 
else an author may desire. Finally, Trogram III lists the 
student records according to requested course, students and 




25 



sequence numbers (restart addresses) and punches cards container? 
wOur^c name , student number* sequence number, date, latency, 
response character count, the XI and X2 counters, response 
category, unknown response count, clock arid trials. Program 
III operates on the IBM 1401 and allows the printing of certain 
information to be suppressed if so desired. All three of 
these programs have facilities for input/output error detec- 
tion and Programs I and III have facilities for correcting 
such errors. 

Sie.ti s ti cal Sum ma ry 

The statistical summary is obtained by means of a program 
with four subroutines written in D.A.F.T. (Dual Autocoder- 
Fortran Translator) which operates on the IBM 7074. This 
program uses the cards punched by Program III above for 
obtaining its data. 



2 ? 



PRECEDES PAGE 6CKNK- ROT FICMEO 

Some Comments Concerning Efficiency in the Preparation 
of Materials for Computer- Ass is ted Instruction 

David Alan Gilman 

Several authors (Dick, 1965; Sllberman and Coulson, 1962; 
Gentile, 1966} have assessed the difficulty and amount of time 
required for the preparation of instructional materials for 
computer-assisted instruction. Time estimates vary due to the 
fact that different standards of comparison are used. Sllberman 
and Coulson (1962) estimate that preparation of a two-hour 
lesson may take several months. Gentile (1966) finds agreement 
among professional programers that only a few good frames can 
be written in a day. The difficulty of programing a lesson 
(in a subject matter familiar to a teacher) is compounded 
with the difficulty of writing the program in a form acceptable 
to a computer. Gentile feels that this combination is too 
much to ask of a course author. 

In the experience of this author, under optimum conditions 
an experienced author-programer can hope for no better than a 
1/40 ratio of instructional time tc preparation time. Comparison 
of this ratio to that of other experienced authors indicate 
that this ratio is considerably better than theirs, with 
typical figures being 1/70 or 1/100. 

There exist several points of view concerning the best 
method for preparing programmed material for computer-assisted 




instruction. These points of view are evident in the philoso- 
phies of the various languages available for programming mate- 
rials Tor different computer systems. 

The level of strategy required In a course may dictate the 

% 

type of programming that is to be done. At the simplest level of 
computer language are languages such as PLATO 1 (Bitzer, Braunfeld, 
and Lichtenberger, 1962) which require only that the author 
enter his text and rules for evaluating his answers. The 
computer has been programmed to accept the text and the author 
does not need to learn the computer language * but rather learns 
a few procedural rules f or preparation of course materials. 

The Coursewriter language (Maher, 1964) is an example of a 
second level of difficulty. Coursewri ter requires that the 
author program his pattern of instruction in a language which 
might be described as relatively simple, but the use of the 
language becomes complex as the author's strategy becomes 
more and more Involved. The highest level of programming 
technique is required by the author who writes his own computer 
program in machine language to carry out instructional strategies. 
This allows the use of the full capability of the computer,, 
but necessitates a high level of competence in computer program- 
ming as well as in instructional strategy. 



1 

Acronymn for Programmed Logic for Automatic Teaching 
Operati ons . 




The Coursewri ter language, developed by IBM scientists, 
has been written in such a way that the course author can 
use practically any strategy he desires provided he is 
skilled enough to program it. The use of such a language 
requires that the author be familiar with the language a»d 
know a number of programming tricks in order that he can 
accomplish the desired objectives. Some attempt has been 
made to make the Coursewriter language compatible with 
natural language, but if an author is to use a rather complex 
instructional strategy, he will need to know how to program 
this strategy. The Planit 1 language (Feingold and Frye, 1366), 
put together at Systems Development Corporation, represents 
an attempt to use the computer to assist the course author 
by informing the author at each step in. the programming as 
co what type of statement he should program next. Thus if 
the next statement in the program should be a question, the 
computer will type "sq" to the author, which Indicates that 
the author should now specify the question. A message of 
"sa" typed to the author indicates that the author should 
now specify the anticipated answer. In essence, Planit 
provides guidance to the author in programing his course. 

While the Coursewriter approach represents a more flexible 
approach and therefore an approach which can be more fully 



1 

Acronym for Programming Language for Interactive 
Teaching. “ 



utilized to present complex learning situations, much can 
be said for the greater efficiency in the method employed 
in Planjt. The question which must be answered in the choice 
of a language for a computer assisted instruction system is* 
“How complex are the instructional strategies going to be 
for a given curriculum? 11 

The complexity of instructional strategies ranges from 
using the computer as a tool to present paired associates 
learning materials to using the computer to teach concepts 
in a manner akin to the way a classroom teacher or tutor might 
present them. A language limited to the preparation of mate- 
rials for paired associates learning would not be particularly 
valuable, since there are many devices such as the Gerbrands 
memory drum which can efficiently accomplish the same objec- 
tives at considerably reduced costs and with much less effort 
required from th experimenter. 

At the other extreme, it is not an efficient use of a 
teacher's time to code complex computer programs. Even if 
programing were as easy as writing a book, It would be 
questionable to expect a large number of classroom teachers 
to have the time to write good books. Bugelski (1964) predicts 
that program writing will ultimately be left to professional 
programers who will work with subject matter consultants. 

Several alternatives to these two extremes are available 
and can readily be put to use. Each of these involve the 
use of a fairly simple language, such as eouri,ewrltar and 



the writing of questions, anticipated responses, and appropriate 
feedback for each frame by an author familiar with the subject- 
matter. 

The essence of the solution to the problem of the great 
length of time and great amount of effort required for the 
preparation of materials for computer-assisted instruction 
lies in the ability of the language to accomplish complex 
strategy without the accompanying burdens of complex program 
coding. 

Some alternatives to having an author program course 
material are the use of (1) author-course programmer teams, 

(2) a programming workbook, (3) an automatic author prompting 
device, and (4) macro programs or standard sequences of 
program options into which a variety of different text mate- 
rials can be fitted. Each of these have been tried with some 
degree of success by the Penn State CAI Laboratory and each 
are being considered for further development. 

Each method herein described is capable of providing 
fairly sophisticated strategies of instruction for a course 
author. 

Author-Coder Teams . The use of author-coder teams has 
been the most evident and probably the most widely used method 
of increasing programming efficiency which we have employed 
at Penn State. This method is fairly simple. The author 
first plans his instructional strategy and brings the material 
to a coder. It is not necessary that the coder be skilled 



32 



in languages other than Coursewriter . since Coursewriter should 
be able to accomplish most strategies planned by the author. 

The coder can approach the problem in any of two ways. First, 
he can instruct the author as to how to program the material 
and give him step-by-step directions for preparing the course, 
or he can do the program coding himself. 

The question which arises at this point is the amount 
of skill and training required of the coder. The CAI Laboratory 
at Florida State University, for example, employs clerical- 
technical personnel as consultant-programmers. The author 
prepares as much of the material as he wishes and then turns 
the material over to a technician who finishes the programming. 
It is fairly easy to train a technician to become a skillful 
coder. However, care must be taken to insure that the author 
has control over the programming of his course and that when 
the technician utilizes special programming techniques that 
these techniques do not change the author's intended teaching- 



strategies . 

A second point of view is that the coder should also be 
a skilled instructor or psychologist :o that he can evaluate 
the author's strategy and make suggestions as to how the course 
could be best taught. The difficulty in this approach is that 
it would probably be rather difficult to find such a skilled 

person who would be willing to confine his activities to course 
coding. 



E"j ksj nrananEicin Era n 



33 




A third approach, utilized we understand by the computer- 
assisted instruction project of the Westinghouse Corporation, 
is using a skilled professional programmer to prepare the 
author's program in a computer language such as Autocoder 
or SP$_. The greatest difficulty in this approach is that - 
the author, because of his lack of understanding of the 
computer language, becomes far removed from the course prepa- 
ration and is unable to contribute course improvements. 

In summary, the use of an author-coder team can improve 
the efficiency of course preparation. However, there exists 
a danger that the author may lose control of course strategy 
and may also lose some flexibility If he does not understand 
what the coder can do with the author language. 

Programming Workbook . This method of improving efficiency 
involves the preparation of ? standard instructional frame or a 
series of instructional frames on a form in such a way that 
the author needs only to complete a few blanks in order to 
accomplish the programming task. 

The difficulty in using this method <s that the author is 
limited to one (or a few) instructional strategies in any 
course. Figure 1.1 shows a typical series of Coursewri ter 
statements arranged in a programming workbook. The author 
needs only to prepare the appropriate instructional messages, 
questions, anticipated responses, and feedback, and write 
them in the blank spaces. The course can then be punched on 
cards vn the regular manner by a technician. 



ro 




(author specifies label name) 


ty 




{author specifies instructional 






message) 


Id 


0/7 S3 




qu 




(author specifies question) 


nx 






ad 


x0//x2 




fn 




(author specifies type of processing) 


ca 




- (author specifies correct answer) 


fn 


sb/ // stl //aO///sayright 


(feedback for correct response) 


Id 


l//s3 




ad 


l//cl 




br 


ac- /7s3//l 


(students who respond correctly are 






branched) 


uni 




(feedback for responses which are in- 






correct ) 


un2 






qu 




(author specifies remedial question) 


nx 






ad 


X0//X2 




fn 




(author specifies type of processing) 


ca_ 




(author specifies correct answer) 


fn 


sb///stl//aO///thatsfcetter 


(feedback for correct response) 


Id 


1//C2 


- 


Id 


1//S3 




br 


ac //s3//1 


(students who respond correctly are 


UR 




branched) 


un 






br 


ac 


(students who reach this point 



in the program do not understand 
and are branched to more remedial 
work) 



Fig. 1.1 Sample page from a programming workbook. 






L 



a ; 

Li 



i . 
$ % 
> I 
Lj 



r"j 

h ; 

u 



13 



n 

* _ * 

Li 



I i 



n 

o 




0 



a 



pi 

y 



35 

Programming workbooks provide a simple and direct way 
of preparing course materials. However, the courses prepared 
by this meth-oo may become dull and repetitious for the student 
since the same type of questions are presented In each frame. 

Automatic Author Prompting D evice . IBM has recently 
marketed a device called the Magnetic Tape “Selectric" Type- 
writer. This machine allows a sequence to be stored on a 

magnetic tape and a typist can manually insert changes, 

additions or deletions in the program. A programming work- 
book could be stored on the tape and utilized in such a way 

that a set of options could be provided the author at every 

step and the author could provide the text of these statements 
by simply typing them. The program and the author's state- 
ments are then stored on tape. Errors can be corrected by 
erasing the tape in the area where the error occurred and 
by retyping the correction. The revised course is then 
stored on the tape. 

It should be possible to connect the Magnetic Tape 
"Selectric 5 * Typewriter to a card punch in such a way that 
cards containing the course material can be punched for 
computer input automatically after the last revision is made. 

The disadvantages of this method of preparation are similar 
to those of the programming workbook. In fact this method 
is, essentially, a mechanized programming workbook. In 
addition some authors may experience difficulty in learning 
to operate the equipment. 



si *■**-] Wi*l 



36 



% 




A Macro Program , A macro program is a sequence of pre- 
programmed options Into which text material may be fitted. 

Some of the inadequacies of the programming workbook approach 
can be overcome if one utilizes a prepared program to accomplish 
any of several strategies. In using a macro program such as 
the one developed at Penn State, the author merely types a one 
digit number (or a combination of one digit numbers) to indicate 
the type of strategy he wishes to use for a particular question 
frame. After typing the number or numbers, the author merely 
needs to specify the feedback which goes with each of several 



responses . 

The options contained in a macro program designed and 
in operation at Penn State and their corresponding code numbers 
are indicated below: 

1. No feedback 

2. Feedback of "wrong" if the response is incorrect 

3. Feedback of why the student's response is incorrect 

4. Feedback of "correct" if the student's response is 
correct 

5. Feedback of why the student's response is correct 

6. Feedback containing a statement of the correct response 

7. Feedback of why the correct response is correct 

8. Repeat of question if response is incorrect 

9. Repeat of all questions missed in program until all 
frames are answered correctly 



The author needs only to program a series of integers to 
indicate his strategy. An author wishing to have feedback of 




"wrong*' when the student's response was incorrect, a feedback 
of why the student's response was incorrect, and repeat of 
the question if the response is incorrect would simply program 
238, The order of the integers is unimportant. Any combinations 
of alternatives 2 through 9 is possible, out choice 1 is not 
compatible with the remainder of tpe choices. Appropriate 
feedback for each of the responses must also be listed by the 
author. This is all the programming the author needs to d). 

Other improvements are possible. The author may choose 
to use feedback indicating a correct answer from a series of 
potentially reinforcing comments such as "right," "correct," 

"very good," or "hot dog." These comments may be chosen randomly 
from a prepared list by use of a subroutine. 

The original macro programs may be prepared by any pro- 
grammer skilled in the use of Coursewrf ter . Since the macro 
itself is written in Coursewriter , some authors may use the 
macro while other authors may continue to do their own pro- 
gramming in Coursewri ter . 

Figure 1.2 indicates a flow chart containing the options 
available to an author who uses the macro described in the 
previous paragraphs. Figure 1.3 indicates the programming 
steps used by the course author. Such a programming arrangement 
enables an author who f nothing of the programming language 
to prepare courses which utilize rather complex strategy. 
Improvements cr new strategies may be programmed by providing 
changes in the macro. 





Fig. 1.2. Pircurawlng options available In each frame for authors using macro techniques. 



■ ge L 1 1 - - m 






. o 

iniic 



BESS ca o oca era era era era era ca 






Multiple 

Choice 

Frame 



Constructed 

Answer 

Frame 



39 




Yes 



Fig. 1.3. Steps followed by author using a macro program. 







Such a plan is ideal for the programming of multiple 
choice programs in the intrinsic programming style recommended 
by Crowder (1962), Also, through the use of t*>e Coursewri ter 
edit function, constructed responses can be made equivalent 
to numbers. These numbers may then be used to represent the 
response alternatives. Thus, the first anticipated response 
may be made equivalent to a 1, the second anticipated response 
may be made equivalent to a 2 , etc. By using this technique, 
the macro may be used for constructed response programs as 
well as for multiple choice programs. 

Cone! usion 

There are several methods for improving the efficiency 
in the preparation of instructional materials for computer- 
assisted instruction. Among the methods discussed were the 
author-coder team, the use of programming workbooks, an 
automatic author prompting device, and the macro program. 

It is probably asking too much of an author to expect him 
to prepare instructional materials and also program them for 
computer presentation. 

Author-coder teams represent increased efficiency. The 
coder may serve as a consultant or may actually do the program- 
ming for the author. The use of programming workbooks is 
another possibility, although preparation of materials by 
this method limits versatility of instructional strategy. 

The same difficulty is true of an automatic author prompting 



device such as a Magnetic Tape "Selectric" Typewriter. This 
device may also be rather difficult for authors to use. 

The use of a macro program represents a highly efficient 
method or course preparation. By using a macro program, it 
is possible to have a natural language program without sacrific 
the versatility of tne programming language. Manv strategy 
variations may be utilized simply with the use of a macro 
program. Eventually, Cfll programming should be no more 

different than preparing a well -thought through lesson for 
the classroom. 

It should be. emphasized that each of these methods is 
relatively inexpensive when compared with the on-line costs 
for direct author input of material in an interpretative mode. 
These methods may represent considerable savings when Increased 
efficiency in the use of the author's time is considered. 







42 



References 



Blt zer, D. L. , Braunfeld , P. G., and Lichtenberger , W. W. 
Plato II: A multiple-student, computer-controlled 

automatic teaching device. In J. E. Coulson (Ed ), 
Programmed l earning and_ computer-based instruction. 
New York: wtley and bon^, 1962. Pp. ZW^W. 



Bugelski , B. R. 
teaching . 



.Ihe. p sychology of learning applied io 
Indian apolis: BoFbs -Merrill Co., 195T. 



Crowder, N. A. 
Programmed 



The rationale of intrinsic programming 
Instruction , April 1962, 1, 3-6. 



Dick, Walter. The development 
based instruction. Amer. 
41-54. 



and current status of computer- 
educ . Res . J . , 1965, 2, 



Fein gold, S. L. and Frye, C. H. A user's guide to Planit. 
Programming Language for Interactive Teaching. Systems 

1966 l0Pment Corporatlon Tech * Memo . TM-3055/000/00, 



Gentile, J. Ronald. The first generation of computer- 

assisted instructional systems: an evaluative review. 

Audio-Visual Communications Review , ir press. Spring 1967 

Maher, Anna. Computer-based instruction (CBI): Introduction 

to the IBM project. IBM Research Report RC 1114, 1964. 



Silberman, H. F. and Coulson, J . £. Automated teaching. 

In H. Borko (Ed.), Computer application in the 
P |hlVTQral s ciences . New Jersey: Prentice-Hall, 1962. 



CHAPTER I! 

ENGINEERING SCIENCE 
David Alan Gilman and Robert Xgo 

The engineering science course is being developed in 
such a way that it will serve a dual purpose. First, and 
most important, the course segments are being programmed to 
serve as material for research studies. Second, the course 

segments are being combined to form a comprehensive curriculum 
in engineering science. 

Primary emphasis has been given to the development of 
materials suitable for the teaching of the basic skills for 
the engineering technician. Among these are the use of sig- 
nificant figures, measurement, working with units, the metric 
system, and the use of scientific notation. Several versions 
of course segments in these areas have been prepared and are 
available for the use of experimenters. Also, materials have 
been prepared in the areas of optics, heat and thermodynamics, 
atomic structure, electrostatics, and electronics. The 
Engineering Science course segments now total approximately 
23 1/2 hours of instruction. 

One area currently under development is a simulated physics 
laboratory suitable for technical education students. The goal 
of the simulated laboratory is to take advantage of the highly 
motivating aspects of “guided discovery 1 * learning and inquiry 
metnods. Experiences are planned which will enable students 



44 



to demonstrate their knowledge of the various elements which 
go inco the solution of a problem and to guery the computer 



for those items of information which they do not possess. 

During the past six months, two studies have been con- 
du cted. Each of the studies involved teaching phys i cs to 
technical education students. The first study was a comparison 
of three branching techniques in teaching a programed sequence. 

The second attempted to compare the effectiveness of differing 
the verbal content of programs for students of high and low 
verbal ability. 

Articles describing the computer-assisted instruction 

project in technical education at Penn State appeared in the 

September 1966 issue cf Tech nical Education News and the 

■ ™ ■■ ■ 

November 1966 issue of School Shop (see Appendix B, p.115). 

An article describing a study in engineering science has been 
accepted for publication by the Journal of Educational Research . 

The three examples following are instructional sequences 
taken from course segments in engineering science. These sequences 
illustrate several different programing strategies, as shown 
in the accompanying flowchart for each example. An understanding 
of the role of each of the mnemonic codes or operands can be 
obtained by referring to Chapter 1, page 4, of this report. 



Example 1 

Segment: Introduction to Physics 

% 

Institution: Penn State 

Author: D. Gilman 



45 



The program first checks to determine whether or not the 
student has given the words “matter" and “energy" in his answer. 
If the answer does not contain both words, it is tested for 
the presence cf either word. If "matter" or "energy" is found 
in uhe answer, the student is told he is partly correct and 
directed to type a complete answer. The u£, which tells the 
student the answer, is typed only if the student's response 
contains neither "matter" nor "energy." 

This relatively simple coding method gives the student 
who is on the right track the opportunity to improve his answer. 
There are three different treatments provided for the completely 
correct, partly correct, and completely wrong responses. 

The program is written so that the students' responses 



will be typed 


in red and the computer messages 


will be typed 


in black. 






LABEL OP 


TEXT 




qu 


What is physics the study of? 


(PreA)* 


nx 






fn 


ed//#// // // s 




fn 


kw//2//c2 




ca 


, matter , energy 




ty 


Correct . 





♦Programed so that automatic ribbon shift from black to 
red occurs and student's response will be typed in red. 




4 



45 



LABEL 



wrong 



OP TEXT 

nx 

br wrong//c2//l//l 

t y You sre partly correct. Type the complet# 

answer. (PreA) 

t «• a 

ur xiu 

un Your answer is not correct. Physics is the 

study of matter and energy. Type the correct 

answer to the question. (PreA) 

rd 



X 







4 ? 



L 



Li 



0 



0 

‘ i 




es 

is 







Fig, 2.1. Flowchart for Example 1 (Introduction to Physics) 



■2 

f7 
■ « 



o • 
ERJC 



Continue 

course 





Example 2 



Segment: Atomic Energy 

Institution: Penn State 

Author: D. Gilman 

The author pregrams the computer to generate a random 
two-digit number by the function "irand," and then uses that 
number to pose a problem to the student. The "lim" function 
is used to check the student's answer. By using the "lim" 
function it is possible to store the contents of the counter 
containing the random number as the correct answer. He is 
looped through the same sequence, getting a newly generated 
number each time, until he is correct on one. 



LABEL OP 1 

a-80 rd 

fn 
br 
ty 

fn 

qu 

nx 



TEXT 

irand//c2//c2 

a-b0//c2//e//0 

How many protons would there be in an atom 

having ,, 

dc//c2 

,, electrons? 



1 

Refer to Chapter 1, page 4, for an explanation of «nemo»1c 
codes and operands characteristic of Coursewrit r language. 




r 



n 








i * 

u^ 




fn 


ed//$//s 


t * 

i ■ 




fn 


1 1m//l//c3 


G 




ca 


1//99 




fc r 


a-80-l//c2//ne//c3 


0 

□ 




ty 


Right. They are equal 


l 

CO 

0 

1 


Ult 


The* should be equal. 


D 




br 


a-80 




a-84 


rd 





4y 



J 



i 

ij 



8 



i 



imc 





o 



i r 




as o ita oa d lea ezs era d n ej e o in 



Segment: Atomic Energy 

Institution: Penn State 

Author: D. Gilman 



Example 3 



51 




The author wants to accumulate the student : s response 
data in this question and type out his total response time 
to the student. The author first clears counter xl . He then 
inserts a blank rd before the question so the counter will 

not be re-cleared if the student takes more than one session 
to complete the question. 

Each time the student answers the question, his response 
time (from when he received the PROCEED light until he entered 
the EOB signal) is added to xl . His answer is then processed. 
The student is told whether he is correct or wrong and how 
many seconds he has spent on the question. 



LABEL OP* TEXT 
rd 



fpl 35 

Id G//xl 

rd 




qu Examine the slide. What is the atomic number 
of sodium? 



1 

Refer to Chapter 1, page 4, for an explanation of mnemonic 
codes and operands characteristic of Coursewriter language. 





T 



LABEL 



okay 



OP TEXT 

nx 

ad x0//xl 

f n 1 i m 

ca 28//23 

br okay 

nx 

fn ed//#// // //s//- 

fn kw//l 

ca ,twentythree 

ty Correct. It took you ,, 

fn dc//xl//-2 

ty 99 seconds to answer the question correctly, 

wa go 

ty 23 

un Wrong. You have now spent ,, 

fn dc//xl//-2 

ty 99 seconds trying to answer this question. 




PRECEDING PAGE BLANK* NOT FILMED 



55 



CHAPTER III 
TECHNICAL MATHEMATICS 

David Alan Gilman 
and 

Nancy Ann Harvilchuck 

Computer-assisted instruction in technical mathematics 
has utilized for its foundation the principles emphasized in 
numerous technical mathematics texts in current use. These 
texts, together with the guidelines suggested in Curri cul a 
for Six Technologies , by Arnold and Schili (1965), have been 
used extensively in the planning, preparation, and evaluation 
of course material. 

During the last six months, the major activities of the 
technical mathematics program have been the preparation of 
course material and the planning of research studies. 

A total of 14 mathematics course segments, approximately 
twenty hours of student instruction time, have been developed 
at Penn State to date. One recently developed course considers 
logarithms and exponents. The student first reviews exponential 
notation and then is introduced to the concept of logarithms. 
After some practice in the manipulation "of logarithms, the 
student begins to solve simple problems. He then utilizes 
his knowledge to work problems of a practical nature. Since 
the program uses static displays, the student may refer to 
previous examples at any time. 

A sample of the logarithm program follows: 



Sample of Logarithm Program 
OP 1 TEXT 

qu 7. Here's a tricky one. 

4 4 

a i a » 

Solve this using the division law for exponents 
nx 

fn kw//l 

wa $l$one 

ty , Your answer is correct but what is it in exponential 

form? (a ? ) 
nx 

fn kw//l 

ca $aO$a°$AO$A° 

(The student has been tolcl to type ad (azero) 

. 0 . 
for a .) 

ty Correct. Whenever you divide one number by itself 

you get one. So by definition a 0 = 1 . 

4 4 4-4' 0 

un a v a = a = a Type aO. 

The program tries to help the student solve examples as 
illustrated below: 

q u ^6. Example: To express 537 in standard notation 

1 . First write 5.37 

2. To obtain 537, we must multiply 5.37 by ? 

Refer to Chapter 1, page 4. for an explanation of mnemonic 
codes and operands characteristic of Coursewriter language. 





57 



nx 

fri kw//l 

ca $hundred$100 $L00 

ty Good. 

un 5.37 x = 537. 

ad l//cl 

un v ou must multiply by 100, Type 100. 

ad ' l//cl 

qu 17. 100 can be expressed as 10 squared or 10 to 

what power? 
nx" 

fn kw//l 

ca $ 2 $two$econd 

ty Good. The final result is 4.37^ x 10 in standard 

notation . 

fi 1 2 

un 10° = 1, 10 = 10, 10 = 100, etc. Try again. 

What power is needed? 
ad l//cl 

un The answer is 2. Type 2. 

ad l//c 1 

t 

After some practice with standard notation, the student begins 
solving simple problems. 

qu 33. Log to the base 9 of 3 can be written as 

log 3 * n. Find n. 




fn kwo//3 

ca $1 $/ $2 

ty Good 

un This is tricky. 9° = 3. What's n? 

^ * 

un Hint: n is less than 1, but greater than 0. Try again, 

un n = 1/2. Type 1/2. 



By the end of the program, the student is ready for complex 
practical problems as illustrated beiow. 



qu 



The weight w in pounds which will crush a solid 
cylindrical cast iron column is given by the formula 



w = 98,520 



d 3.55 



where d is the diameter in inches and 1 is the length 
in feet. What weight will crush a cast iron column 
8 ft. long and 5.2 inches in diameter? (3.55 and 1.5 
are exponents) 



nx 

fn 

ca 

ty 

un 

un 

un 



kw//l 

$1,004,000 $1 ,004,000 $1 ,004.000 .$1004000. 

Correct. Now for the last problem. 

Wrong. Recheck your work and try again. 

Wrong. Log 2 = log 98,920 + 3.55 log d -1.5 log 1. 
Log 2 = 6.00173 . Find w. 




59 



Research studies being planned involve comparisons of 
various media for instruction and drill in mathematics and 
the use of several varieties of feedback to teach mathematics. 

Two studies are being planned which will compare the efficiency 
of teaching a student by an instructional sequence :!iich the 
student selects, with an instructional sequence selected by 
the course author. Considerable emphasis is also being given 
to the appropriate amount of verbal content that a program 
should contain in order for it to teach most effectively. 

Perhaps the most significant aspect of computer-assisted 
instruction is its branching and decision-making capability. 

The computer has the potential to branch on the basis of any 
of several predictors (i.e., error rate, response latency, 
mental age, attitude scores, and other variables). Our. research 
will attempt to identify predictors which correlate differentially 
with performance in different instructional treatments. From 
this information it will be possible to develop branching 
decisions based on empirically validated predictors. Cut-off 
scores can be established so that students are branched to 
instructional treatments which optimize learning. 



References 



Schill, Willi am John and Arnold, Joseph Paul. Curricula content 
for six technologies . Illinois: University of Illinois! 

1965. 



6 ) 



. . CHAPTER IV 

CCMMUNICATiOIT SKILLS 
Harriett A. Hogan and Helen L. K. Farr 

Summary of Activities 

During the third six-month period of the project, the 
final remedial spelling segments, .pf the communication skills 
programs were completed, put on the CAI system, revised, and 
then submitted to a field trial. A total of 14 spelling segments 
was available for use. 

The purposes of the spelling programs are to evaluate 
the spelling competencies of students preparing to be technicians 

v 

and to provide remedial instruction in nine areas of spelling 
as needed by each individual student. Major emphasis in each 
spelling program is directed toward the utilization of the 
decision-making capacity of the computer to individualize Instruc- 
tion and toward the optimum use of the audio-visual equipment 
associated with the computer terminals. 

In Figure 4.1 a flow diagram of the complete spelling 
program is shown c 

The program includes an orientation to the IBM Selectric 
typewriter, tape recorder and photographic slide compute, out- 
puts, a word study unit, the diagnostic test, nine .emedial 
segments, a proofreading exercise and a spelling achievement 



test. 






*Njtje Instructional Areas: 

1. sppl - plurals 

2. spsuffix - prefixes snd suffixes 

3. spe - final e words 

4. spi-e - words containing ie 
comoi nation 

5. spsyl - studying words by syllables 

6. spcomp - contractions and compound 
words 

7. spdlscr - similar words 

8. sphom - homonyms 
spdemon - demon words 



1 ™ 

Posttest 

"spptest" 




t 

Stop 




Fig. 4.1* Flow diagram of CAI spelling program. 




The word study unit emphasizes the Importance of system- 
atic word study In the Improvement of spelling. Students are 
encouraged to look at each word, pronounce It and then write 
It 7- instruction In dividing words Into syllables Is Included. 

Word study also Includes examining words for "trouble spots" 
such as silent letters, difficult vowel combinations, and un- 
phcnetlc sounds. 

The diagnostic test Is used to evaluate each student's 
spelling performance In nine problem areas. Five Items in 
each of the nine areas are Included In the 37 word diagnostic 
test. T&fs is accomplished by isoine of the same words 

for more than one purpose. For example. In the word perceive, 
if the first syllable is spelled incorrectly. It is tallied 
as a prefix-suffix error; If the ei combination Is reversed, 

V 

an ei-ie error is tallied. If a student makes two or more 
errors of one type he Is branched to the appropriate remedial 
program. Unanticipated responses In either syllable are tallied 
as errors which will branch the student to the remedial program. 

The objectives of the proofreading unit are to emphasize 
the importance of proofreading written work and to Improve 
the spelling of words In context. A segment of a technical 
report such as might be prepared by a technician is displayed 
to the student and he identifies and corrects misspelled words. 

The achievement test, like the diagnostic test, is made 
up of words containing nine types of spelling problems (Appendix C 
p.127). Difficulty of the two tests Is equated so that change 
scores from the diagnostic test to the achievement test can be 
obtained. 




Words used throughout the spelling program are selected 
from lists of most commonly used words and lists of technical 
wordso The selection Gf words and the context in which they 
are used are intended to interest the technical student and 
to illustrate to him how spelling may be an occupational skill 
important to him in his work as well as in general usage. 

In the fall, the two staff members working on communica- 
tion skills visited the Williamsport Area Community College 
(WACC) and the A'toona campus of The Pennsylvania State Uni- 
versity (APSU) to: (1) observe students in two-year technical 

programs in English and speech classes; (2) discuss strategies 
of CAI with instructors; and (3) lay the groundwork for a field 
trial of the completed spelling segments. 

Students at, WACC and APSU were asked by local staff members 
to participate in the field trial , with the understanding that 
participants would be paid a nominal sum when they completed 
the tasks involved. 

Because of the reading involved in the CAI spelling pro- 
gram, it was important to have an indication of the students' 
reading ability. No standardized reading or English achieve- 
ment scores were available for the WACC students; therefore, 
arrangements were made to administer to them the reading com- 
prehension and English placement tests constructed by The 
Pennsylvania State University Office of Student Affairs Research 
These.* tests constitute part of the tost battery administered 
to students applying for admission to The Pennsylvania State 
University (PSU), and consequently, the scores obtained by the 




APSU students on the tests were available from the Office o< 

Student Affairs Research * 

A 40-item questionnaire (Brown, 1966 and Wodtke. 1965) 
designed to elicit student attitude toward CAI was expanded 
to cover 44 items. This questionnaire was administered to 
all participants after they had completed the CAI spelling 
course (Appendix C, p> 121). 

The Field Trial 

A field trial was designed to obtain information about 
the general "takeabi li ty" of the spelling program segments, 
and to provide bases for editing and revising the program. 

The study was conducted at WACC and APSU from October 31, 1966, 
to December 9, 1966. 

Twenty-five students at WACC and 16 students at APSU 
participated in the field trial. Because of equipment mal- 
functions and proctor errors, data from 12 Subjects (Ss ) at 
WACC and 4 Ss at APSU were incomplete and are not included In 
this. All Ss were enrolled in two-year programs of study at 
their schools. The 25 participants were majoring in the following 
courses at their schools: 1C in technical subjects (e.g., 

electronics, drafting, construction, machine and electrical 
technology) and 15 in business courses (e.g., general business, 
accounting, business management, retailing, agricultural business) 
The numberof students enrolled in each area is shown in Table 
4.1. The ages of the Ss ranged from 18 to 27 years; 14 Ss 
were men and 11 were women. 



66 



Preliminary Findings and Apparent Trends 

The analysis of the data gathered f^om the study has not 
yet been completed. Nevertheless, several trends have emerged 
from the preliminary analyses, and it is clear that the informa- 
tion will be helpful in editing and revising the program. 

The following preliminary findings are already evident: 

(1) The range of all scores for the APSU Ss was narrower than 
for the WACC Ss. (2) Gains were recorded by two-thirds of 
the APSU Ss; all but one of the WACC Ss made gains. More of 
the Ss at WACC made larger gain scores than the Ss at APSU. 

The mean gains were greater at WACC than at APSU. hen students 
generally made greeter gains than women students except that 
sex differences were confounded with school as shown in Table 
4.2. (3) As shown in Table 4.3, there appears to be a positive 

correlation between achievement gain scores and the number of 
course segments covered by the student. There is no apparent 
relationship between the number of equipment malfunctions en- 
countered by the student and the gain. (4) Generally, the scores 
on the attitude questionnaire appear unrelated to the amount 
of gain revealed by the gain scores as shown in Table 4.4*. 

Women Ss had a higher mean score (positive) on the attitude 
questionnaire than men. 



1 

A complete correlational analysis was unavailable for this 
report; however, .the final analysis will be completed in the near 
future . 





67 



Table 4.1 

Biographical Information on Ss in Field Trial 
of Remedial Spelling Course 



3 



I 

3 

3 

3 

3 





APSU 


WACC 


Hen 


3 


11 


Women 


9 


2 


TOTAL 


12 


13 



Course Majors: 




• 


Business 


General Business 




2 


Accounting 




1 


«. Business Management 




2 


Retailing 


9 




Agricultural Business 


1 




Technical 


Electronics 




3 


Drafting 


1 


1 


Construction 




2 


Machine 




1 


Electrical Technology 


1 


1 . 


TOTAL 


12 


U 



3 

3 








Table 4.2 



Scores on 37-Item Spelling Test 
for 25 Students in Field trial 

(Maximum possible score 50^) 







APSU 








WACC 






Sex 


Pre- 

test 


Post- 

test 


Gain 

Scores 


Terminal 

Time 


Sex 


Pre- 

test 


Post- 

test 


Gain 

Scores 


Terminal 
T 1 me 


u 

M 

M 


26 

32 

26 


34 

39 

33 


8 

7 

7 


5:00 

5:48 

5:42 


W 

.M 

M 


20 

29 

22 


43 

47 

34 


23 

18 

12 


8:23 

5:35 

10:09 


W 

W 

M 


45 

47 

35 


48 

50 

33 


3 

3 

3 


2:25 

2:53 

7:01 


M 

M 

M 


28 

25 

21 


37 

34 

28 


9 

9 

7 


6:41 

8:20 

7:34 


W 

H 

W 


46 

35 

45 


48 

37 

45 


2 

2 

0 


2:25 
6:45 
1 : 56 


W 

H 

i*i 


31 

38 

33 


37 

43 

36 


6 

5 

5 


5:18 

3:05 

4:16 


w 

w 


42 

43 


42 

41 


0 

-2 


2:26 

3:23 


N 

M 


34 

39 


38 

42 


4 

3 


5:58 

2:31 


w 


41 


39 


-2 


3:24 


M 


40 


41 


1 


3:06 












M 


39 


39 


0 


2:32 


Mean 


38.58 


41.08 


2.58 


4:50 




30.69 


38.53 


7.84 


5:38 



!L§.* Maximum score of 50 on this 37-word diagnostic spell 
type ^of error ( see^p? Ilf"** S6Veral WOrdS d1agnose more th#n on * 





69 



Table 4.3 

Rank Ordering of Absolute Gains with Number of 
Course Segments and Equipment Malfunctions 





APSU 






WACC 




Gain 

Scores 


Number of 
Segments 
Studied 


Weighting 
of Equip. 
Malfunction 
(#30)* 


Gain 

Scores 


Number of 
Segments 
Studied 


Weighting 
of Equip. 
Malfunction 
(#30)* 


8 


9 


3 


23 


14 


4 


7 


9 


3 


18 


' 9 


3 


7 


8. 


5 


12 


11 


5 


3 


8 


2 


9 


9 


2 


3 


5 


3 


9 


9 


4 


3 


5 


2 


7 


9 


2 


2 


8 


1 


6 


8 


3 


2 


5 


3 


5 


7 


4 


0 


5 


5 


5 


7 


3 


0 


5 


3 


4 


7 


4 


-2 


6 


3 


3 


6 


3 


-2 


5 


2 


1 


6 


3 






• 


0 


5 


5 



♦Item 30 on Attitude Questionnaire reads: 

While on Computer Assisted Instruction I encountered mechanical 
malfunctions. 



(Weighting) (1) 



( 2 ) 



(3) 



(4) 



(5) 



Very Often Often Occasionally Seldom Very Seldom 



9 



Attitude Toward CAI by Gain and Sex 



APSU WACC / 



Positive 

Attitude 

Scores 


Gain 

Scores 


Sex 


Positi 

Attitude 

Scores 


Gain 

Scores 


/ 

/ 

/ 


180 


3 


W 


168 


23 


w 


177 


2 


M 


166 


18 


M 


170 


7 


M 


165 


12 


M 


169 


0 


W 


163 


9 


H 


165 


0 


K 


161 


4 


K 


163 


-2 


W 


160 


9 


M 


161 


-2 


■A 


160 


6 


W 


157 


3 


W 


160 


0 


K 


154 


7 


M 


159 / 


3 


M 


152 


2 


W 


157 / 


2 


H 


144 


8 


w 


151 / 


7 


M 


140 


3 


K 


147' 


5 


H 


- 






124 


1 


M 


Kean 161 * 


2.58 




fjean/157 


7.61 




Hen I * 

Homen 3f = 


154.6 

163.1 




/ - 

Men X = 

Women X s 


155.7 

164.0 





/ 

/ 



Discussion 



4 

/ 



It must be remejpbered that coiolete statistical analyses 
have not yet beery^ompleted on the data discussed In this report, 
and the finding^ reported above should be regarded as preliminary 



until all of the planned statistical analyses have been completed. 




Table 4,5 



Tabulated Responses for 
Attitude Questionnaire Item 



Item 44 : 



How long do you feel you could work efficiently in computer- 
assisted instruction at one sitting: (circle one) 



_ t More than 

1/2 hour 1 hour l-i/2 hours 2 hours 2 .hours 



APSU 

WACC 




TOTAL 0 7 



6 11 1 



However, the preliminary exami.,siion of the data does suggest 
answers to some of the questions that this field trial was 
planned to explore. 



1. The narrow range of scores for APSU Ss may be 
a consequence of two facts: 

(a) The APSU Ss were a sample from a popu- 
lation previously screened by meeting 
university admissions requirements. 

(b) Of the 16 APSU Ss , 13 were enrolled in 
the same course of study, thus reducing 
variability in spelling achievement to 
some extent. 

2. The program, even without needed editing and re- 
vision, appears to be a way for students to learn 
spelling skills. As might be expected, the pro- 
gram was especially helpful to those Ss who were 
poor spellers as Indicated by their lew pretest 
performance. This is Inferred from the comparison 
between those whose performances showed gains 
(i.e., 20 Ss) and those whose performances showed 
no gain (i.e., 3 Ss) or a loss (i.e., 2 Ss). 












U P 






72 



3. The greater gains shown by men as contrasted 
with women students is in keeping witn usual 
expectations for enrollees in technical courses. 
These findings are of course confounded with 
the schools which the students* were attending 
and further field trials will be undertaken with 
student samples from both sexes in both locations. 
The women had higher scores on the pretest; con- 
sequently, they had less opportunity to make 
gains, and considerably less need for a course 

of instruction in remedial spelling. Furthermore, 
the women in the field trial had higher mean 
scores on the PSU reading comprehension test 
(* = 12.5) and on the English placement test 
(X = 50.1) in contrast to men whose mean scores 
on the same two tests were 11.5, and 34.6, re- 
specti vely, 

4. Unless Ss experience a minimum of equipment mal- 
function when they are taking a CAI course, they 
tend to regard CAI as an interference in their 
learning, rather than as a help. Of the 40 Ss 
who originally enrolled to participate in this 
study, the data from 15 could not be included 

in the analysis because the Ss encountered extensive 
machine failure and proctor error. The mean 
attitude score of these excluded students was 
less positive (X = 151.8) than that of Ss v/hose 
data were retained (X * 159). 

5. Confidence In the validity of the pretest has 

been increased by the positive relationship between 
the number of spelling course segments studied 
and the positive gain scores. Ss were routed 
into remedial course segments according to their 
performances on the pretest. 



Implications 

The implications made by the preliminary analysis of data 
from this field trial are promising. Depending on the com- 
pletion of the data analysis, it appears that further research 
on the use of a CAI spelling program with poor spellers in 
two-year technical courses is indeed merited. 



i 





73 



Efforts to eliminate malfunctions of equipment should 
be of primary concern to researchers. There is some evidence 
that this may Involve the more careful and thorough training 
of terminal proctors assisting in a study, as well as the more 
rigorous and systematic testing and retesting of the equipment 
used, and of the computer program as it is executed at the 
terminal . 

Since the program has demonstrated that remedial spelling 
courses can be taught by CAI , it would seem that a new field 

trial should be initiated with a revised and edited version 
of the course. 



74 



References 



B * A instrument for the measurement of expressed 

toward computer-assisted instruction. In 
S r*!l± nu ! 1 ? r ?sr e fs Report Experiwentatinn with 

^r?Scct r no SS 5'ny 7i ^ trUCt1 °!r r^ J ^^^ ca1 E^ti on , 

al?! December If; m6 rePar W E< fi7Ii ® r 'H 

Wodtke, K. H., Mitzel, H. E., and Brown, B. R. Some preliminary 

the _ react1ons of students to computer-assisted * 
Association*. ilf bT P - ~ — IbUSSB. Psychological 



z &m m a 



wm 



75 



CHAPTER V 
RESEARCH REPORTS 

Cueing and Feedback in Computer-Assisted Instruction 
Keith A. Hall, Marilyn Adams, and* John Tardibuono 

It is not appropriate at this point of development to main- 
tain that the CAI can or does provide more efficient methods 
of teaching. Rather it Is more appropriate to scrutinize the 
-particular characteristics of a given system for improvement 
of learning. Many of these characteristics or variables can- 
not be judged in terms of previous experimentation because 
they do not exist in other learning experiments or situations. 
Further, studies conducted in laboratory situations cannot 
be readily transferred to an educational environment. This 
study attempts to remove the psychological learning experiment* 
from the artificial world of animal laboratories and nonsense 
syllables and to place it in educational context. 

The CAI system at Penn State University has two Coursewriter 
functions (among others) for providing feedback to students 
regarding the correctness of their response. Each of the functions 
gives the author flexibility In selecting the type of feedback 
to provide to the student. At this point though, there is 
no information available to guide the author In making a choice. 

The two functions under Investigation are keyword (kw) and 
partial answer zero (paO). (See Chapter I for an Interpretation 
of mnemonic codes.) The keyword function can be used to match a 
student's response against a stored correct response in a block 






-2~r 







76 

consisting usually of a complete word or several complete words. 
A decision is made by the system on the basis of a complete 
word matching or not matching. The paO function causes the 
system to natch a student's response against the stored correct 
response a single character at a time or in groups of characters 
at a time depending on the author's decision. In each case 
feedback can be given to the student based upon what was matched 
and what was not matched. 

The simplest form of feedback with kw is for the system 
to type at the student terminal the complete response which 
he should have made* The simplest form of feedback with paO 
is for the system to match the response made by the student 
character by character against the stored correct response 
and type back at the student terminal the characters which 
matched correctly, dashes for lowercase characters which did 
no* match, and underscores for uppercase characters which di d 
not match. 

A typical student-system interaction using kw with 
immediate feedback of the correct response might look like 
this. The system types "Colorado" and the student is to type 
the capital of that state. If the student types "Boulder," 
the system will respond "Denver." Using the paO function the 
interaction would differ. If the student typed "Boulder" the 
system would respond "De-r-r," In this case the system searched. 

4 

the student's response "Boulder" and found three characters 
which are part of the correct response -- d, e, and r. The 



O : - 

ERJC 



0 



o 



K - 
r 



i 



m 

Imwur ** s 

^ v \*/ 



\l 




77 



system rearranged these into the proper order and typed 
them for the student in their proper location. 

Programs have been written for investigating the effec- 
tiveness of these two kinds of feedback. A paired-associate 
learning task is employed using fifty pairs which the student 
must learn* The fifty state names of the United States 
are presented as the stimulus items and the student will 
learn to respond with the names of the capitals. The items 
are presented one at a time to the student at the terminal 
typewriter in random order. If the student responds correctly 
on his first trial to that stimulus, it drops from the program. 
The program recycles until each student has responded correctly 
on his first trial to each of the stimulus items. After he 
had "acquired" each of the fifty pairs,he again goes through 
the list as a posttest. Each student takes a retention test 
two weeks after the initial treatment. 

Each of the experimental programs contains the following 
features : 

1. A list of warm-up items consisting of five countries 
presented as stimuli and their capitals as response 
items. 

2c A typing test which reports a student's time and 
accuracy In typing an alphabetic sentence. 

3. A progress report to the student after each trial 
through the list of states consisting of 

(a) total number of responses 

(b) total number of stimuli presented 
{c) total number of items acquired 










78 



(d) total response latency 
> (e) current clock reading 

| 4. An automatic five-minute break approximately half 

% way through the task. 

%■ 

| 5. An automat i c connecti on to a system administered 

r student opinion survey regarding CAI. 

i, A preliminary investigation was conducted during the Fall 

| Tt nn 1966 , with ten subjects on each program. The results, 

f though not conclusive, have encouraged the authors to continue 

F; collecting data with a large sample of students. The two expert 

mental treatments are now being administered to students from 

t the Williamsport Area Community College. 

C 



! 

1 



! 



9 

ERIC 









T 







03 GO C3 CT3 C~i CO 



79 



Re 1 at 1 ve Ef fecti veness of three Modes of Presentati on 
Through Computer-Ass? sted Instruct! on 

Donald W. Johnson 



This study was undertaken to provide some a'.swer to the 
computer-student interface problem. Often when writing material 
for CAI , an author is confronted with the problem of deciding 
which mode of presentation he should use. His choices cur- 
rently are among the typewriter output, audio tape messages, 

2 x 2-inch photographic slides, or in some cases static displays. 
(Static displays are usually in the form of papers, booklets, 
or three-dimensional models.) In some instances these choices 
are determined by the subject matter being presented, (i.e a , 
when presenting stimulus material to test spelling ability, an 
audio message is the logical choice). However, the presentation 
mode in many instances is not so well defined. 

The purpose of the study is to determine the relative 
effects of three modes of presentation, audio tape, typewriter 
output, and static displays on total time required for subjects 
to complete the course and on competence as determined by a 
posttest score. 

Method 

Course material chosen for the study was a physics sequence 
on “Working With Units." It is a basic physics series designed 
for vocational-technical students who have finished high school 
and have a limited background in mathematics and physics. 




j 












\ 7 >- : 



80 

The sequence originally contained material presented through 
typeout with nine slides supplementing the printed verbal material* 
Questions were included throughout the -sequence to provide for 
student interaction. Written verbal typeout contained the basic 
instruction and could be considered the core of the short course. 

This core instructional material was transcribed word-for- 
word on audio-tape vo provide the audio mode and was placed in 
a booklet to provide the static display mode. The original 
material was labeled the type mode. 

All subjects in all modes viewed the same slides and answered 
the same Questions: only the instructional material was altered 
through a change of presentation mode. A fourth, g-ouo, identi- 
fied as the control group, took the final examination only and 
received no instruction. Subjects were placed in groups by 
random assignment. 

.he subjects were approximately ninety juniors and seniors 
.n the College of Education enrolled in an introductory instruc- 
tional media course. They did not have a mathematics or physics 
background. 



In the type mode, the subject signed on 
course , the computer typed out instructions then typejka^question , 
the subject typed the answer to the question, a^Tide was shown, 
illustrative material about the slide was tyjfed, etc., until 
the lesson was completed. The subject was immediately tested 
on his knowledge at the terminal by a computer administered 
test. Total instructional time and test score was recorded. 



Li 




It 

o 



/ ; ’ i ; l -i' '< „ . .. ''l ;! .■ 1 .M 






J ‘ 



; ' '' / jv? v 








Audio Mode . Subjects who received instruction/through 
the audio mode heard the same material that the/subjects in 
the previous mode read from the typecut. Subjects could re- 
peat each message as often as desired bpt could not read or 
by-pass any material. When finished; they took the same com- 
puter administered test. 

Display Hp. de. Subjects' in the display mode ha-d .a book- 
let in their hands containing all of the messages heard pr 
read by the above g^ups. The typeout instructed them to refer 
to and read th^^proper page of instruction. Slides were shown 
in the sequence by the computer and questions asked 

at the/proper time. Answers to the questions were typed by 
re subject. The same computer administered test was taken 
by this group. 

CPntrdl Group . The control group only took the test at 
thd computer terminal. 

We have just completed the instruction of the subjects 
and the collection of the data; it has not yet been analyzed. 






83 



PRECEDING PAGE BLANK* NOT FiLMED 

Appl i cat 1 on of Modi f i ed Gagne Type Model to 
Computer-Assisted Altitude Testing and 
Inst ructi onal Branching 

Harold Sands 

Although twentieth century concern with individual differ- 
ences has come to the fore in such modern developments as mental 
testing, counseling arid guidance, multi -tract curricula, ability 
grouping, ungraded classrooms, perhaps one of the most direct 
attacks cn the problem vis-a-vis instruction has come from the 
recent arrival on the educational scene of computer-assisted in- 
struction, The computer has been heralded for extending the 
decision-making capabilities of educators far beyond potentials 
which have existed in the past; however, it is doubtful if we have 
even begun to utilize these capabilities. Perhaps the most glar- 
ing examples of this are in methods used for aptitude testing and 
branching. It is an old adage to . . start with a student 
where, he is . . , " In instructional practice, use of the com- 

puter has brought about few changes which really extend the ad- 
vantages accrued beyond traditional aptitude and achievement 
testing. In instructional branching, where one might expect even 
greater changes, progress has not been much better. 

Although such variables as latency and error rate are 
currently being explored, the best current method for branching 
is based on programmers* intuitive judgment . If a student makes 
an error in responding, he is branched to other stimulus material 
because in the judgment of the programmer, all students who have 





84 



madfc that error need that particular change in their instruction. 
Such approaches, while they are certainly an attempt In the di- 
rection of individualizing instruction, have also been accomplished 
through scrambled-book programs, and can hardly be said to be 
making full use of the decision-making capability which educa- 
tors purportedly will soon have in computer-assisted instruction. 

The recently initiated research described in this summary is 
an attempt to further enhance the individualizing capability of 
computer-assisted instruction by extending the testing and branch- 
ing capabilities traditionally available. 

Theoretical Framework 

Gagne (1962) maintains that the acquisition of knowledge 

proceeds in a hierarchal fashion from lower level learning sets 

to higher level sets until mastery of the criterion is achieved. 

Learning sets are conceived as specific behaviors which mediate 

positive transfer to the next superordinate set in the hierarchy. 

theoretically, one cannot achieve mastery of the next higher 

learning set unless he has mastered all sets subordinate to it. 

Such hierarchy might be organized into blocks of learning sets 
as shown in Fig. 5.1. 

Theoretically, for example, a student cannot master Level X 
unless he has mastery of Sets A, B, and C of Level II and of their 
subsets. In several studies, Gagne and his associates have been 
able to demonstrate the existence of such hierarchies in the 
teaching of mathematics. In one study (Gagne'', Mayor, Garstens and 



85 



Paradise; 1962), a hlerachy for the addition of integers was 
identified. After the administration of a program, the acqui- 
sition of learning sets at successively higher stages of the 
hierarchy was found to be dependent upon prior mastery of sub- 
ordinate learning sets. Predictions of instances in accord 
with predictions ranged from 97 to 100 per cent.. Confirmatory 
results have been obtained in a study by Gagne and Paradise 
(1961) which was concerned with the class of tasks of solving 
linear algebraic equations, and by Gagne and Basler (1962) 

r 

in which the task was that of specifying sets, intersections of 
sets, and separations of sets, using points, lines and curves. 



LC^tL I 



LEVEL II 



LEVEL III 




Fig. 5.1 Hypothetical model of knowledge hierarchy in which 
lower level learning sets mediate positive transfer to higher 
level sets. 



86 



Implications of Gaone's Idea for 
Computer-Assisted Aptitude Testing 

If this construct is correct * then it seems to be par- 
ticularly relevant to computer-assisted testing. If learning 
is really hierarchical, and if tests are developed to measure 
performance of each set in the hierarchy, then tests could 
be administered in a manner similar to the Stanford-Binet . 
in which the highest set in the hierarcy for which the individual 
had mastery was identified. If the test and hierarchy are 
valid, it would not be necessary to test lower or higher sets 
for which the individual had mastery, since presumably, he 
must have mastery of all subordinate sets and cannot have mastery 
of superordinate sets. It should be noted that such a test 
requires the availability of test items for possibly hundreds 
of learning sets, an interpretation of student responses, and 
a decision-making capability by the tester to either test 
further or to prescribe instruction. Such a test must be admin- 
istered individually; therefore, it requires a live tester 
or an adequate substitute, perhaps a computer. The latter 
is considerably more practical and is probably more accurate. 

A student who is to learn mathematics could be seated at a 

terminal and tested in order to identify his highest level 

of learning set achievement. Instructional sequences or "tracts" 

from which he could derive benefit could be made available 

by a teacher; or if decision-making criteria had been programmed 

into the system, instructional assignments could be made by 

the program of the computer. 



' 4v~‘n- 



Implications of Gagne's Idea for Branching 

The testing functions discussed thus far described a system 
in which a student is tested and differentially placed into 
a teaching sequence which represents his highest level of learning 
set achievement. Once placed into a sequence according to 
past achievement, individual differences in the learning of 
new material must also be accounted for. Whereas pre-testing 
may allow for individual differences in past achievement, branching 
allows for individual differences in learning abilities. Branch- 
ing can, for convenience, be thought of as consisting of two 
functions. First, a student is tested and his response is evalu- 
ated. Secondly, the student is presented with stimulus material 
which, on the basis of the test response, is optimal for him 
at that time. Although other criteria are involved in maximizing 
the advantage of such a student-tutor relationship, one of 
.the requirements is that there be a unit of identifiable behavior 
to be tested and to which the student is to be branched. In the 
jargon of programmed instruction these have sometimes been re- 
ferred to as "steps." In branching, it seems that learning sets 
could serve as the logical units of behavior for testing as a 
student progresses through a program and for branching intended 
to allow for individual differences in learning abilities. If a 
student cannot master a particular learning set, there would, 
theoretically be little advantage in advancing him to the next 
higher learning set. In fact, subordinate learning sets might be 
tested until there was some assurance that the student has mastery 
of all requisite sets. 





The Current Research 



The purpose of this research Is to Investigate Gagne's 
notion that higher mental processes are structured by learning 
sets which are hierarchically ordered, and the full Implications 
of this Idea for testing as well as branching in a computer - 
assisted Instructional system. Actually, several modifications 
have been made in the type of hierarchy Identified and In methods 
for arriving at it. The study has been divided Into three phases 



1. The development of methods for analyzing subject 
matter and Identifying hierarchies of learnlnq 
sets. 

2. The development of methods for testing and 
identifying a student's level In a hierarchy 
of learning sets. 

3. The development of computer-assisted instruc- 
tional materials which Incorporate the appli- 
cation of theoretical findings to a more highly 
developed system of individualized Instruction 
through branching and other teaching strategies 
based on the methods of diagnostic testing de- 
veloped and using sets In hierarchies as the 
primary unit of learned behavior. 



The first phase which is reported here Is concerned with the 
identification of a subject matter hierarchy. A hierarchy has 
been identified and Is currently In the process of being validated. 
The task selected was a simple one, to convert a number In one 
base to its equivalent in another base. The remainder of this 
paper is devoted to a description of how the hierarchy was de- 
veloped and to a discussion of how such a hierarchy can be vali- 
dated. 



• * 






89 



The Hierarchy 

The procedure for analyzing the task (converting a number 
from one base to its equivalent in another base) and identifying 
a hypothesized hierarchy was slightly modified from that of 
Gagne's ; however , the hierarchy whi ch results is somewhat di ffer- 
ent , having two rather than one dimension, and consisting of con- 
ceptual levels, each of which contains the total components of 
criterion behavior, but at different conceptual levels. To 
arrive at this type of hierarchy, the questions were asked of 
the final criterion, "What does a person do when he performs 
this task?" "Can what he does be described by dividing the final 
task into two subsets?" If not two, then three and so on. These 
latter sub-sets become Level II learning sets, and each of these 
was in turn divided into Level III sets, with the procedure con- 
tinuing until five levels were identified. In all, the model 
consists of approximately 30 learning sets. The first three 
levels of this hierarchy are shown in Figure 5.2, and will be 
used as a more specific example of how the hierarchy was formed. 
Starting with Level I, the criterion performance, it was decided 
that what a person does in making the conversion could be described 
as consisting of two subordinate sets. These are the two sets 
shown as Level II. Had the final Level I performance not been 
divisible by two, then successive divisions of three, four and 
so on would have been tried until the least number of sets were 
arrived at which were greater than one, but which completely de- 
scribed the criterion performance. Each of the two sets at 
Level II were subdivided following the same procedure. 





Ill 



II. 



I. 



1. Convert e number 
Li. one base to 
• equivalent 
;n another base 



Converts a number 
to base . fc .en 



T 




Simplifies an 
expansion 



i 

> 

i 

| 

I 

I 

l 

I 

? 



T 



{ 

i 

t 



Converts . a base 
ten number to 
another base 



Divides a place 
ten number up by 
place values of 
another base 



T 

l 

i 

V 



4. Rearranges the 
quotients of a 
converted number 
into correct 
sequence 



Sequences * 
Learning Sets 



Fig. 5.2. Partial hierarchy of learning sets for the 
criterion performance of converting a number from one base to 
its equivalent In another base. 



L, 

o 



Tsssasar 






91 



r*~* 



list* 



m 

rz 

vZ ; 



I 



& 



I 



m 



hs 



Learning sets 1 1 1 - 1 and III-2, according to the theory 
of hierarchal learning, mediate positive transfer to set II~K 
Set II-l and 1 1 - 2 mediate transfer to 1-1. If any of the 
subordinate learning sets are not mastered, then theoreti rally , 
learning of the higher level set cannot occur. For example, 
if an individual has not mastered levels ITI-3 and III -4 , he 
should not be able to learn set II-2. 

A further examination of Figure 5.2 will reveal that as 
one proceeds from the top set at each level downward, the sequence 
of learning sets is the same sequence in which the criterion 
behavior must be performed. This is another modification of 
the way in which Gagne organises his hierarchies; however, 

5in~e integration of subordinate sets plays such an important 
role in his construct, it was felt that by organizing the sets 
sequentially, the meaning of "integration” or t.he "putting 
together" of subordinate sets was made clearer. This assumes, 
of course, that integration consists primarily of selecting 
or recalling relevant learning sets and of performing them 
in a sequence appropriate to the given task. In cases where 
the sequence is arbitrary, the learning sets in question are 
joined by boxes which are contiguous to each other, but none 
occurred in the first three levels. 

For purposes of this study, only the first four levels 
which consist of 15 learning sets, are being used. 



i! 

L 

r 



n 



o 

ERIC 



92 



Validation of the Hierarchy 

The hierarchy is in the process of being validated, and 
data should be available for the next reporting period. Gagne's 
premise has implications for validity. Sets in the hierarchy 
provide fairly specific behaviors against which test items can be 
compared. Secondly, if the test is accurately measuring the 
highest level of achievement in the hierarchy, then it should 
be an easy matter to demonstrate that individuals who lack 
mastery of a Tower set, lack mastery of higher sets* Finally, 
the validity of such tests might be demonstrable on the basis of 
showing a relationship between the measured level of learning 
set attainment and conventional achievement scores taken at the 
end of a training period. 




References 



Gagne, R. M. The acquisition of knowledge. Psychol. Rev., 

1962, 69, 355-365. 

Gagne, R. M. and Bassler, 0. Study of retention of some topics 
of elementary nonmetric geometric. J. Educ. Psychol., 

1963, 54, 123-131. * 

/ 

Gagne, R. M. , Mayor, d. R. , Garstens, H. L., and Paradise, N. D. 
Factors 'in acquiring knowledge of a mathematical task. 
Psyche! , Monogr ., 1962, 76_ (7, Whole No. 526). 

Gagne, R. M. and Paradise, N. E= Abilities and learninq sets in 
knowledoe acquisition. Psychol. Monoqr., 1961. 75~ 

(14, Whole No. 518). — 



PRECEDING PAGE BLANK- NOT FILMED 



95 



An Instrument for the Measurement of Expressed 
Atti tude Toward Computer Assisted Instruct! on 1 

Bobby R. Brown 

It is generally agreed that student attitude and motivation 
are important variables in the traditional learning situation. 

In tiie lecture method the student, it is often said, should 
have a positive attitude toward the teacher and the subject 
matter content as well, if optimal learning is to take place 
(Bugelski, 1964). However, in programmed instruction, and 
especially automated instruction, such as computer-assisted 
instruction (CAI), there will be less direct contact between 
teacher and student. It seems important therefore to assess 
the effects on learning of student attitude toward this kind 
of instruction. The results of some previous investigations 
seem to indicate that student attitude toward automated instruc- 
tion may not be of much consequence relative to the amount 
learned (Eigen and Feldhusen, 1964; Wodtke, Mitzel, and Brown, 
1965; Wodtke, 1965). It may well be that the effects of student 
attitude only show up over an extended period of time and that 
these studies were too short-termed to give a true indication 
of the effects of student attitude. One study which did report 
significant changes in attitude was interestingly enough a 
study of eight months* duration (Campbell and Chapman, 1965). 



1 

Developed while the author was on an IBM fellowship at 
The Pennsylvania State University, wt*th helpful advice from 
Professor Robert Lathrop*. 




96 



The studies cited above, however, have one thing In common: 
they all measured attitude with experimenter-constructed tests 
of either unknown or unreported reliability. If anything defini- 
tive Is to be said about student attitude and programmed Instruc- 
tion, some effort must be put into developing better measuring 
instruments. This is a report of one effort to develop a more 
satisfactory measure of student attitude toward CAI, and a de- 
scription of the Instrument in its present state of development. 

General Description and Statement of 
intended Application of the instrument 

The attitude instrument consists of forty Items, each 
of which Is a statement that could be made about CAI or about 
oneself 1r> relation to CAI, and each of which is to be responded 
to by marking one of five choices on a Likert-type scale. The 
response device was designed to be used for research purposes 
only and was planned specifically to be used in research in 
the CAI Laboratory at Penn State. 

The inventory was designed and preliminary data have been 
analyzed with the plan to use the method of summated ratings 
to obtain one score of student attitude (Edwards, 1957). The 
user should refrain from making statements on the basis of 
responses to individual items in the test, and be aware that 
the reliability of individual items will at best be quite low. 
Also, because of the difficulty encountered in interpreting a 
summated rating score wich reference to a neutral point or neutral 
score, it becomes difficult to assign individuals to favorable or 




unfavorable catego-ies. This difficulty is not encountered, 
however, if one's research interest calls for measuring change 
in attitude or for correlation measures. 

Sources of the I terns 

The items in the inventory (see Appendix C, p .121) came 
from three sources: 

1, Four items were taken from a student reaction 
inventory previously developed by Dr. Wodtke 

at The Pennsylvania State University CAI Laboratory. 
These four items were reworded to make them 
compatible with the five-point Likert-scale 
used in the present instrument. 

2. Sixteen items were based on comments previously 
written by students who had been engaged in 



3< Twenty items were based on the author's ob- 
servation of students' reactions while they 
were engaged in CAI and on students' comments 
to the author during informal conversation 
after they had been engaged in CAI. 



Wording of the Items 

In order to add some protection against haphazard marking 
of the items and to minimize the possible effects of response 
set, some of the items were stated negatively (i.e., were 
statements of negative attitude toward CAI), and some were 
stated positively. Two statements were incomplete until the 
student marked his responses, and his responses determined 
whether the statements were negative or positive. One of 
the incomplete or neutral statements was: "Concerning the 

course material I took by CAI, my feeling toward the material 



before I came to CAI was: Very Favorable , Favorable, Indifferent 

Unfavorable, Very Unfavorable." 

Unanimous agreement among six judges was obtained on the 
classification of items as negative or positive. The break- 
down of the items as negative, positive, or neutral is given 
in Table 5.1. 

Table 5.1 . 

Listing, by Item Number, of Negative, 

Positive, and Neutral Items 



Type Statement Item Number Total 



Negati ve 


1, 


3, 4 


> 6, 


8, 


10, 


12, 


13, 


16, 


17, 




19 


, 22, 


23, 


24, 


26, 


29, 


30, 


32, 


, 34 


• 


35 


, 36, 


40 

* 1 * 














Positi ve 


2, 


5, 7, 


, 8, 


11, 


14, 


15, 


18, 


20, 


, 21 




25 


, 31, 


33, 


37, 


38, 


39 









Neutral or 

Incomplete 27, 28 



In order to make the students' choice of response more 
compatible with the wording of the items, four different 
labelings of the scale were used in the test. These are as 
follows : 

Strongly disagree. Disagree, Uncertain, Agree, Strongly 
agree 

All the time. Most of the time. Some of the time* Only 
occasionally. Never 

Quite Often, Often, Occasionally, Seldom, Very seldom 

Very favorable. Favorable, Indifferent, Unfavorable, 
Very unfavorable. 






99 



Subjects 

The subjects used for the pilot administration reported 
here were taken from an available group of students who had 
received some instruction by CAI sometime during the three 
previous terms at Penn State. Approximately twenty-five of 
the students were from the local high school, thirty-one from 
an introductory educational psychology class, and twelve Penn 
State students from an audiology class who were receiving 
instruction on CAI as part of a course they were taking for 
credit . 

The students received their instruction in one of four 
courses and two students received instruction on two courses. 

It will be observed that the available group differed on at 
least the following dimensions: age, amount of time on CAI, 

educational level, whether or not they were taking the course 
for credit, and length of the interval since they were on 
CAI. From this group sixty-eight students were available. 

The breakdown of student by course is given in Table 5.2. 

Two students had both modern mathematics and physics but are 
counted only in physics. 

Administration 

The CAI attitude inventory was mailed to each of the 
students along with a letter and a self-addressed, stamped 
envelope. Thirty-four forms were returned. The number and 
percentage of returns is given by class in Table 5.2. Six forms 



were returned too late to be Included In this analysis, and 
eftjht forms were returned unopened because the students Had 
moved. Adjustment for these 14 untallied forms yields a total 
return of 67 per cent. 

Table 5.2 

Description of Study Group as to Courses Taken, 

Number and Percentage of Forms Sent Out, 
and Number and Percentage of Forms Returned 



Course 


No. sent 
out 


% of total 
sent out 


No. 

returned 


% of total 
returned 


% returned 
by class 


Acphys. 


23 


33.8 


17 


50 


73.9 


Audlol . 


12 


17.6 


1 


2.9 


8.3 


Hod Math 1 


18 


26.5 


11 


32.4 


61.1 


RSMH 


15 


22.2 


5 


14.7 


33.3 


Totals 


68 


100.02 


34 


100.0 


50.0% 



Analysis of Data from Pilot Administration 

Rel i abi 1 i ty . The 34 tests returned were scored and inter- 
item correlations were obtained. A principle-components factor 
analysis and varimax rotation of the factor loading were performed 
in order to determine whether there appeared to be strata of 
items within the Instrument. Six factors were extracted in 
the factor analysis and their Eigen values are 12.4, 2.8, 2.7, 

2.3, 2.2, and 2.1. This Indicates that the first factor is 
accounting for approximately 31 percent of the total variance 
in the form and the remaining factors something less than 



101 



7 per cent of the variance for each factor. The internal consis- 
tency reliability of items was computed. This coefficient 
was computed by the analysis of variance method. The coefficient 
obtained by this method is identical to the coefficient obtained 
by Kuder-Richardson Formula 20 (Dick, 1S65; Rabinowitz and 
Eikeland, 1964). This coefficient was .885. The analysis of 
variance summary table for this estimate of reliability is pre- 
sented in Table 5.3. 



Table 5.3 

Analysis of Variance Summary Table of Reliability 



Source 



- D.F. 



Sum of Squares 



Mean Squares 



Subjects 



33. 



238.36 



7.2231 



Items 



39. 



1072.58 



Residual 



1287. 



1067.87 



0.8297 



R = 1 - 0.830/ 7.223 



0.885 




Mean intraitem correlation = 0.162 
Non-strati fied reliability = 0.885 



In addition, internal consistency coefficients were computed 
using the factor loadings from the factor analysis to construct 
strata of items. As set forth by Rabinowitz and Eikeland (1964), 




this method will give a better estimate of the reliability 
of a test if there are clusters or strata of items in the test. 
The coefficients given by this analysis are .859 if strata 
are considered random and .890 if strata are considered fixed. 
The analysis of variance summary table for these estimates 
of reliability are presented in Table 5.4, 

Table 5.4 

Analysis of Variance Summary Table 
of Reliability by Strata 



Source D.F. Sum of Squares Mean Squares 



Subjects 


33. 


238.36 


7.2231 


Strata 


6. 


170.75 




It/Strat 


33. 


901.83 




S X S 


198. 


202.17 


1.0211 


Re si d^al 


1089, 


865.70 


0.7950 


Total 


1359. 








Rel 1 abi 1 ity 


Estimates 




R = 1 • 


• ( 1.021/ 7.223 


) = 0.859 


Strata Random 


R R 1 ■ 


• ( 0.795/ 7.223 


) = 0.890 


Strata Fixed 



.It should be observed that very little change in the 
coefficients resulted when the stratified analysis was performed. 
This result considered in connection with obtained Eigen values 
of the factors on which the strata were based suggests two 
possibilities: (1) the Inventory may consist of one group 




of relatively homogeneous items, or (2) the fact that there 
was a considerable interval between the students* experience 
with CAI and the administration of the test may have resulted 
in their expressing a generalized attitude toward CAI which 
would prevent any meaningful factors showing up in the analysis. 
Regardless of which of these possibilities has occurred, the 
factors of items become rather meaningless and no attempt will 
be made to deal further with the factors until additional data 
are obtained. 

In addition to the usual caution to be exercised in inter- 
preting reliability coefficients, the nature of the available 
group and the method of administration must be kept in mind. 

It is quite possible that attitude toward CAI was a factor 
related to whether or not students returned the form. The 
reliability reported here might not have held up if data for 
all students had been available. 

Validity There is no statistical evidence of the validity 
of this attitude inventory at the present time. Evidence of 
content validity is presented in this section. 

The device appears to be measuring student attitude toward 
CAI.. The items consist of possible statements of attitude 
toward aspects of CAI or comparative statements about CAI and 
traditional instruction. 

The strongest argument presently available as to the valid- 
ity of this form is that the items can be considered to have 
a quasi -sampl ing or logical validity. The items were constructed 



104 



from actual observad or written expressions of students' stated 
attitudes about CAI. Because it was felt that an expression 
of attitude toward CAI would involve little emotional or ego 
involvement, the items were written to get at “expressed 
attitude" and no effort was made to get an indirect measure 
of attitude. Inasmuch as the inventory was designed to measure 
expressed attitude and because the items were constructed from 
instances of actually expressed attitudes, it is felt that 
evidence of more substance than mere face validity has been 
presented fo. the scale. 








References 



Bugelski , B, R. The p sycho 1 ogy of learning app 1 i ed to teaching. 
New York: Bobbs-Merri 1 1 Company, Inc., 1964. ~ 

f 

Campbell, Vincent N. and Chapman, Madalynne A. Research in 
degree of student control over programmed instruction: 
long-term cumulative effects on problem-solving and 
transfer. American Institutes for Research/Palo Alto. 
AIP-E20-12/65-TR. 

Dick, Walter. Comp 1 ete A . 0 . V . reliability . Computer Program. 
Office of Examination Services, The Pennsylvania State 
University, March, 1965. 

Edwards, Allen L. Techniques of attitude scale construction. 

New York: Appleton-Century-Crofts , Inc., 1957"! 

Eigen, Lewis D. and Feldhusen, John F. Interrelationships 
among attitude, achievement, reading, intelligence, and 
transfer variables in programmed instruction. In 
J. P. DeCecco (Ed.), Educa tional Technology, New York: 

Holt, Rinehart, and Winston, 1964, pp. 376-386. 

Rabinowitz, William and Eikeland, Hans-Magne. Pedagogisk 

Forskning , Nordisk Tidsskrift for Pedagogikk. SAERTRYKK, 

Fra Argang, 1964, Uni versi tetsf orl aget . 

Wodtke, Kenneth H , Relationships among attitude, achievement, 
and aptitude measures and performance in computer-assisted 
instruction. In Semi-Annual Progress Report, Experi mentation 
ft 1 th Computer-Assi sted Ins t ruct i on i n Technical Education , 
Project No. 5-85-074, prepared by Harold E. Mitzel et al., 
December 31 , 1965 . 

Wodtke, K. H„ „ Mitzel, H. E., and Brown, B. R, Some preliminary 
results on the reactions of students to computer-assisted 
instruction. Proceedings of the American Psychological 
Associ ati on , September, 1965. 



PRECEDING PAGE BLANK* NOI FILMED 



APPENDIX A 



Students within Sequence Numbers, All Separate 
Sequence Numbers within Students, All Seoarate 
Students Separate, Sequence Numbers Pooled 
Sequence Numbers Separate, Students Pooled 
Statistical Summary Listing 



Appendix A.l 



Students Within Sequence Numbers, All Separate 



COURSE NAME - SigfiglGO AUTHOR - Tracy Logan S/20/66 - 11/12/66 



Student No„ 


Sequence No. 


Attempts 


Mean Latency 


Sd of Latency 


586 


AA-0010-018 


2 


79.53 


71.057 


589 


AA-0010-018 


2 


86.30 


26.121 


590 


AA-0010-018 


2 


83.95 


79.309 


595 


AA-0010-018 


1 


65.03 


0.000 


615 


AA-0010-018 


2 


49.28 


17.748 


618 


AA-0010-018 


1 


13.33 


0.000 


628 


AA-0010-018 


2 


38.43 


31.190 


633 


AA-0010-018 


1 


150.00 


0.000 


634 


AA-0010-018 


2 


65.24 


19.184 


637 


AA-0010-018 


2 


47.72 


48.317 


641 


AA-0010-018 


1 


76.20 


0.000 


b47 


AA-0010-018 


3 


75.32 


48.671 


662 


AA-0010-018 


1 


137.75 


0.000 


618 


AA-0010-018 


1 


88.86 


0.000 


801 


AA-0010-018 


2 


89.16 


57.177 


586 


AA-0020-020 


1 


20.23 


0.000 


589 


AA-0020-020 


3 


40.95 


12.604 


590 


AA-0020-020 


3 


46.88 


14.043 


595 


AA-0020-020 


1 


28.70 


0.000 


615 


AA-0020-020 


1 


24.10 


0.000 


618 


AA-0060-020 


2 


25.98 


1.803 


628 


AA-0020-020 


3 


26.19 


15.128 


633 


AA-0020-020 


3 


41.18 


17.450 


634 


AA-0020-020 


4 


44.24 


15.942 


637 


AA-0020-020 


1 


79.27 


0.000 


641 


AA-0020-020 


1 


51.90 


0.000 


647 


AA-0020-020 


1 


89.60 


0.000 


662 


AA-0020-020 


1 


26.87 


0.000 


801 


AA-0020-020 


2 


27.97 


11.646 


586 


AA-0040-020 


1 


63.72 


0.000 


589 


AA-0040-020 


1 


96.85 


' 0.000 


590 


AA-0040-020 


1 


75.58 


0.000 


595 


AA-0040-020 


1 


71.14 


0.000 


615 


AA-0040-020 


1 


50.16 


0.000 


618 


AA-0040-020 


1 


21.24 


0.000 


628 


AA-0040-020 


1 


69.68 


0.000 


633 


AA-0040-020 


1 


108.34 


0.000 


634 


AA-0040-020 


1 


14.15 


0.000 


637 


AA-0040-020 


1 


159.33 


0.000 


641 


AA-0040-020 


1 


133.68 


0.000 


647 


AA-0040-020 


1 


18.34 


0.000 


662 


AA-0040-020 


1 


21.17 


0.000 


801 


AA-0040-020 


1 


38.38 


OcOOO 


586 


AA-0050-030 


1 


118.38 


0.000 


589 


AA-0050-030 


1 


46.82 


0.000 



Appendix A 2 

Sequence Numbers Within Students, All Separate 
COURSE NAME - SigfigiOO AUTHOR - Tracy Logan 9/20/66 - 11/12/66 



Student No, 


Sequence No. 


Attempts 


Mean Latency 


Sd of Latency 


586 


AA-00 10-018 


2 


79.53 


71,057 


586 


AA-0020-020 


1 


20.23 


0.000 


586 


AA-0040-020 


1 


63.72 


0.000 


586 


AA-0050-030 


1 


118.38 


0.000 


586 


AA-0060-020 


1 


80.80 


0,000 


586 


AA-00 80-0 10 


1 


16.95 


0.000 


586 


AA-0090-020 


1 


24.73 


0,000 


586 


AA-0 100-020 


1 


16.90 


0.000 


586 


AA-0 110-020 


1 


35.42 


0.000 


586 


AA-0 120-020 


1 


11.35 


0.000 


586 


AA-0130-020 


1 


14.92 


0.000 


586 


AA-0 140-020 


1 


71.99 


0.000 


586 


AA-0150-020 


1 


18.68 


0.000 


586 


AA-0 160-020 


2 


35.36 


2.326 


586 


AA-0160-229 


1 


16.13 


0.000 


586 


AA-0180-010 


1 


220*28 


0.000 


586 


AA-0180-030 


1 


20.75 


0.000 


586 


AA-0200-020 


1 


39.65 


0.000 


586 


AA-0210-020 


1 


13.13 


0.000 


586 


AA-02 30-010 


2 


34,34 


0.375 


586 


AA-0240-020 


1 


26.62 


0.000 


586 


AA-0250-090 


2 


19.54 


10.062 


586 


AA-0 300-0 20 


1 


16.89 


0.000 


586 


AA-0 310-0 20 


1 


12.40 


0.000 


586 


AA-03 30-040 


1 


28.64 


0,000 


586 


AA-0 340-020 


1 


11.87 


0.000 


586 


AA-0 3 50-020 


1 


8.29 


0.000 


586 


AA-0 360-020 


1 


18.60 


0.000 


586 


AA-0410-100 


1 


5.30 


0.000 


586 


AA-0420-020 


1 


20.21 


0.000 


586 


AA-0430-040 


1 


25.46 


0.000 


586 


AA-0440-020 


1 


13.92 


0.000 


586 


AA-0450-010 


1 


29.27 


0.000 


586 


AA-0460-020 


1 


11.60 


0.000 


586 


AA-04 70-020 


2 


37.60 


12.127 


586 


AA-0470-030 


1 


40.82 


0.000 


586 


AA-0510-020 


1 


29.06 


0.000 


586 


AA-0530-020 


1 


7.80 


0.000 


586 


AA-0530-100 


1 


88.85 


0.000 


586 


AA-0531-200 


1 


9.88 


0.000 


586 


AA-0531-250 


1 


11.55 


0.000 


586 


AA-0 5 3 1-340 


1 


56,72 


0.000 


586 


AA-0531-370 


1 


3.40 


0.000 


586 


AA-0550-018 


2 


32.16 


10.479 


586 


AA-0560-020 


1 


66.12 


0.000 


586 


AA-0610-010 


2 


59.62 


19.375 


586 


AA-0 6 10 -019 


1 


21.28 


0.000 


589 


AA-00 10 -018 


2 


86.30 


26.121 


589 


AA-0020-020 


3 


40.95 


12.604 


589 


AA-0040-020 


1 


96.85 


0.000 


589 


AA-0050-030 


1 


46.82 


0.000 


589 


AA-0060-020 


1 


19.88 


0.000 


589 


AA-0080-010 


1 


24,38 


0.000 





no 



Appendix A. 3 

Students Separate, Sequence Numbers Pooled 




r : 



H 



o 



l 



COURSE NAME 


- SigfiglOO 


AUTHOR - Tracy Logan 9/20/66 - 11/12/66 


Student No. 


Attempts 


Mean Latency 


Sd of Latency 


Frequency 


586 


54 


35.64 


5.110 


52 


589 


67 


31.35 


3.765 


65 


590 


58 


37.61 


5.467 


56 


595 


49 


28.79 


5.125 


48 


609 


13 


21.08 


3.985 


12 


615 


45 


21.93 


2.645 


43 


618 


60 


23.42 


3.768 


58 


628 


47 


37.59 


7.767 


45 


633 


41 


60.04 


8.741 


40 


634 


56 


37.82 


4.797 


54 


637 


69 


60.55 


11.296 


67 


641 


22 


42.76 


7.753 


21 


647 


56 


46.22 


8.350 


53 


660 


32 


29.19 


3.520 


31 



662 63 



27.82 



2.903 



62 




Appendix A. 4 111 





Sequence Numbers Separate, 


Students Pooled 




COURSE NAME - 


SigfiglOO 


AUTHOR - Tracy Logan 9/20/66 


- 11/12/66 


Sequence No. 


Attempts 


Mean Latencv 
♦ 


Sd of Latency 


Frequency 


AA-0010-018 


25 


73.45 


9.213 


24 


AA-0020-020 


9 


37.39 


5.967 


9 


AA-0060-020 


1 


24.70 


0.000 


1 


AA-0020-020 


17 


41.76 


6.225 


17 


AA-0040-020 


14 


67.27 


12.436 


14 


AA-0050-030 


13 


88.31 


15.945. 


13 


AA-0060-020 


5 


31.02 


14.193 


5 


AA-0080-010 


23 


27.29 


3.456 


23 


AA-0090-020 


14 


26.06 


4.563 


14 


AA-0060-020 


8 


47.27 


7.598 


8 


AA-0100-020 


14 


27.19 


9.157 


14 


AA-0110-020 


16 


74.85 


44.695 


16 


M-0120-020 


20 


18.03 


2.624 


20 


AA-0130-020 


14 


12.52 


1.368 


14 


AA-0140-020 


14 


56.34 


9.364 


14 


AA-0150-020 


14 


50.80 


23.753 


14 


AA-0160-020 


18 


45.18 


6.908 


17 


AA-0160-229 


14 


26.67 


5.751 


14 


AA-0180-010 


14 


157.76 


26.482 


14 


AA-Q130-03L 


21 


19.99 


3.133 


21 


AA-0200-020 


14 


46.17 


5'. 880 


14 


AA-0210-020 


13 


7 . 24 


3.469 


13 


AA-0460-020 


1 


18.02 


0.000 


1 


AA-0410-280 


1 


114.81 


0.000 


1 


AA-0210-020 


2 


11.91 


0.297 


2 


AA-0230-010 


21 


29.24 


4.397 


20 


AA-0 240-020 


26 


34.66 


9.533 


26 







112 





vO 


pv 


O' 


rH 


CM 


O' 


O' 


m 


vO 


CM 


>n 


00 


CM 


O' 


a> 




m 


rH 


m 




CM 


CO 


CM 


CO 


rH 


CM 


m 


CO 


O 




iH 


9 ^ 


• 


• 


• 


• 


• 


• 


• 


• 


• 


• 


• 


• 


• 


• 


• 


c o 


rH 


VO 


vO 


O' 


vO 


O' 


tn 


vO 


00 


CO 


rH 




00 


‘O 


CM 


1 o 


i th 


o 


CO 


o 


m 




rH 


o 


o 


o 


<P 


CM 


o 


CM 




11 *H 


00 


o 


00 


o 




m 


rH 


CO 


<r 


rH 


rH 


.H 


CM 


00 


<p 


n ° 


o 


o 


o 


rH 


o 


o 


CM 


CM 


CM 


CO 


CO 


<P 


vi- 


rH 


O 




pv 




cn 


o\ 


00 


00 


o 


O' 


O' 


O 


O 


vO 


vo 


o 


O' 




rH 


rH 


rH 


rH 


rH 


rH 


CM 


rH 


rH 


CM 


CM 


rH 


rH 


CM 


rH 


11 CM 


* 






* 


* 


* 




* 


* 


* 


-K 


& 




* 


* 


Un 


o 




o 




o 




O 


o 




o 




O 




O 


o 


H CM 
































n ^ 
































U rH 




























• 


• 


X5 
































1-1 


o 


CM 










00 


CM 


rH 


o 


rH 


pv 


CM 






pci 


rH 


rH 












rH 




rH 












VO 


VO 


vO 


vO 


vO 


vO 


vO 


VO 


vO 


vO 


VD 


VD 


vO 


vO 


vO 


0) 


VO 


VO 


vO 


vO 


vO 


vO 


vD 


VO 


vO 


vO 


VO 


vD 


vO 


vD 


vO 


— 


— 


' — 


\ 


— 


— 


— 






— 


— 


\ 


— 


\ 


\ 


I «u 


o\ 


O' 


00 


00 


00 


00 




vO 


VO 


CO 


co 


pv 


Pv 


m 


m 


cd 


CM 


CM 


CM 


CM 


CM 


CM 


o 


o 


o 


o 


o 


rH 


r— * 


o 


o 


Q 


\ 


— , 


— , 


— , 


— , 


— 


— . 


\ 


— 


— 


\ 




— 




\ 




o\ 


o% 




o\ 


o\ 


ov 


o 


o 


O 


o 


o 


o 


O 


o 


o 














rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


* 
































o 


f^ 


00 


pv 


CO 


CO 


pv 


CO 


CO 


CO 


VD 


CO 


00 


Pv 


O 


o 


c 


pv 


CM 


pv 


00 


o 


00 


O 


00 


r^ 


00 


CO 


<r 


CO 


O 


oo 


0) 


• 


• 


• 


• 


• 


• 


« 


• 


• 


• 


• 


• 


e 


• 


• 


4J 


CJ\ 


o\ 


<P 


r^ 


o 


pv 


m 


rH 


VO 


CO 


CO 


o 


VD 


o 


00 


cd 


CM 


CM 


o 


vO 


<r 


CM 


vO 


VO 


CO 


00 


rH 


VD 


rH 


m 


pv 




rH 




rH 




rH 


















rH 






00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


• 


rH 


rH 


rH 


H 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


o 


o 


O 


O 


o 


O 


o 


O 


O 


o 


O 


O 


O 


O 


O 


O 


53 


1 


1 


I 




1 


1 


( 


1 


1 


1 


1 


1 


1 


1 


1 




o 


o 


o 


o 


o 


o 


o 


o 


o 


o 


o 


o 


O 


o 


o 


• 


rH 


cH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


CT 


o 


O 


O 


o 


O 


o 


O 


O 


o 


O 


O 


O 


O 


O 


O 


0) 


o 


o 


O 


o 


O 


o 


O 


O 


o 


O 


o 


O 


o 


O 


O 


CO 


i 


1 

3 


!. 

3 


i 


i 


i 


i 


i 


i 


i ■ 


i 


i 


i 


i 


i 




CM 


a 


§ 










a 




CO 








a 


s 






V 


O 


& 








o 












o 


rj> 




vO 


VO rH 


Ov 


OV rH 


o 


O H 


in rH 


m rH 


in rH 


00 


00 rH 


00 


00 


CO rH 


<P rH 




CO 


00 tA> 


00 


oo m 


o\ m 


o\ m 


o\ m 


rH UO 


rH m 


tH tH 


rH in 


CM 


CM rH 


co m 


co m 


CO 


m cm 


m cm 


m 


m cm 


m <p 


m cm 


m cm 


vD tH 


VO CM 


VO rH 


VO CM 


VO 


VO m 


VO CM 


vO CM 




O VO 


O VO 


o m 


O vO 


o co 


O vO 


O vO 


O P^ 


O VO 


O O' 


O VO 


O CO 


O vO 


O VO 


o oo 




CO • 


CO • 


CO • 


CO • 


CO • 


CO • 


co • 


CO • 


CO • 


CO • 


CO • 


CO VO 


CO • 


CO • 


CO • 




CM 


CM 


CM 


CM 


00 


CM 


CM 


CM 


CM 


CM 


CM 


CM 


CM 


CM 


CM 




00 


00 


CO 


00 


pv 


00 


00 


00 


00 


CO 


00 


00 


00 


00 


00 




o 


o 


o 


o 


o 


O 


o 


o 


o 


o 


o 


O 


o 


o 


o 


0) 


o 


o 


o 


o 


o 


O 


o 


o 


o 


o 


o 


O 


o 


o 


o 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


rH 


CO 


00 


00 


oo 


op 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


u 


*H 


*H 


•H 


•n» 


•H 


•H 


•H 


•H 


•H 


•H 


*H 


•H 


♦H 




*H 


0 


4-1 


4-1 


14-1 


14-1 


4-4 


4-4 


14-1 


4-j 


4-4 


4-1 


4-1 


4-1 


4-1 


4-1 


4-4 


o 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


o 


*H 


*H 


•H 


*H 


•H 


•H 


•H 


*H 


•H 




•H 


*H 


*H 


*H 


H 




CO 


CO 


CO 


CO 


CO 


CO 


00 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 




rH 


CM 


rH 


CM 


tH 


CM 


rH 


rH 


CM 


rH 


CM 


rH 


CM 


rH 


** 








(continued on next page) 





113 



n 



i / 



o 

o 

rH 

u 



. 



CM 

p 



rH 

X 



<D 

03 

Q 



00 

CO 

• 

00 

LO 

m 

o 

o\ 



X! 



vO 

rH 

• 

vO 
O 
c n 
co 
vO 
rH 



CO 

H 

• 

<■ 

<J\ 

CO 

vO 

rH 



CO 

o 

rH 

o 

CM 

rH 



rH 

CM 

• 

CM 

o 

rH 

CM 



* 

O 



m 



m 

co 

m 

o 

rH 

CM 



* 



CO 

rH 

• 

vD 

CO 

vO 

O 

rH 

CM 






CM 


oo 


CM 


CM 


m 


CO 


rH 


CM 


m 


rH 


* 


\ 


• 


• 


• 


co 


IT. 


r^ 


<r 


rH 


rH 


O 


rH 


CO 


CM 


vO 


vO 


r^ 


o 


CM 


CO 


m 


m 


rH 


rH 


00 


o\ 


o\ 




(Tv 


rH 


rH 


rH 


rH 


rH 


X 


o 


* 


CO 




o 


o 




O 


o 



O 

in 

* 

vO 

CM 

CO 

rH 

C\ 

rH 



•K 



00 

O 

• 

CM 

CM 

rH 

o> 

rH 



X! 



y 

§ 

4J 

cd 

P 



o 

25 



co 



CO 

rH 



vO 

vO 

un 

"o 

o 

rH 



\0 



m 



oo 

rH 

o 

i 

o 

rH 

_o 

o 



vO 

vO 

O 

rH 

O 

rH 



00 

00 

• 

rH 

oo 



oo 

rH 

o 

I 

o 

rH 

o 

o 



vO 

vO 

o 

rH 

O 

rH 



m 

m 

• 

CO 

rH 



00 

rH 

o 

I 

O 

rH 

o 

O 

I 



CO 



vO 

vO 

(T> 

rH 

o 

rH 



vO 

vO 

CM 

rH 

O 

rH 



vO 

vO 

CM 

rH 

O 

rH 



rH 

rH 



vO 

vO 

CM 

rH 

O 

rH 



CM 



vO 

vO 

rH 

rH 

O 

rH 



CM 



vO 

vO 

O 

CM 

O 

rH 






vO 

vO 

X 

o 

CM 

o 

rH 



00 



vO 

v0 

C\ 

CM 

o\ 



vO 

vO 

00 

CM 

o^ 



vO 

vO 

00 

CM 

<7^ 



o 


CFi 


O 


vO 


m 


ov 


CM 


rH 




CO 




m 


• 


• 


• 


• 


• 


• 


vO 










Ok 




m 


CM 




fO 


CM 




rH 






rH 


tH 



CO 


CO 


m 


CO 






CM 


<r 


CO 


00 


• 


• 


• 


• 


o 


00 


o 


rH 


CM 


• 




CM 


vO 


CO 


CM 



00 

rH 

O 

I 

o 

rH 

O 

o 



00 

rH 

O 

I 

o 

rH 

O 

o 



00 

rH 

O 

I 

o 

rH 

o 

O 



00 

rH 

o 

I 

o 

rH 

o 

O 



555 



00 

rH 

O 

I 

O 

rH 

o 

O 

I 



co 

rH 

o 

I 

o 

rH 

O 

o 



00 

rH 

o 

1 

o 

rH 

o 

o 



3 5 5 



o 

CM 

o 

I 

o 

CM 

o 

o 

I 



o 

CM 

O 

I 

o 

CM 

Q 

O 



O 

CM 

O 

I 

o 

CM 

O 

O 



5 3 5 



o 

CM 

o 

I 

o 

CM 

o 

0 

1 



CM 



vO 

vO 

00 

CM 

G\ 




a 

u 



CM 

5 



r-» oo 


oo 


rH 


CM 


rH 


m 


m 


m 


vO 


m 


vO CM 


vO CM 


>o CM 


vO 


CM 


O vO 


O VO 


O VO 


O vO 


cn • 


CO • 


CO • 


CO 


• 


CM 


CM 


CM 




CM 


00 


00 


00 




00 



m 

m 



2 

o 



rH 


rH 


rH rH 


vO 


a* 


<j\ 


<7\ 


S 


O 


m 


o m 


00 


00 CO 


00 


00 


G 


00 


CM 


00 CM 


m 


m u 


m 


m 




o 


m 


O VO 


o o 


o a) 


o 


o 


o 


CO 


• 


CO • 


CO tH 


co u 


co 


co 


rH 




CM 


CM 




a) 


s 








00 


oo 


rH 


S 


co 




rH 





O 


O 


o 


o 


o 


o 


O 


O 


O 


O 


o 


o 


o 


o 


<D 


o 

rH 


O 

rH 


o 

rH 


o 

rH 


o 

rH 


o 

rH 


O 

rH 


o 

rH 


O 

rH 


O 

rH 


o 

rH 


O 

rH 


O 

rH 


O 


CO 


CO 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


J-l 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


P 


4-4 




4-4 




4-t 


MM 


4-4 


4-4 


4-t 


m 


4-4 


4-4 


4-4 


4-4 


o 


00 - 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


o 


*H 


•H 


•H 


*H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 


•H 




CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


H 


CM 


rH 


CM 


rH 


rH 


CM 


CO 


rH 


rH 


CM 


rH 


rH 


CM 


CO 



m 

y 

& 



u 



n 

y 

a 

y 



e* r 



O 

ERIC 





PRECEDING PAGE MISSING 



APPENDIX 



0 — 



Page No. 



H5 



New, faster teaching methods supplement, not supplant, the teacher 

Computer-Assisted Instruction 
for Technical Education 



By David A. Gilman 

Co m put er- Assisted Instruction Laboratory , Pennsylvania State University , University Pork 



Reprinted from November 1966 SCHOOL SHOP, Ann Arbor, Michigan 

WITH PERMISSION 



■^VOCATIONAL EDUCATION faces 

f many problems, some of which are 
due to the changing character of our 
society. In certain areas of technology, 
the rapidity of technical and scientific 
advances has become sc great that it is 
literally impossible to create curricu- 
lums for vocational-technical education 
that are net obsolete by the time they are 
taught in the schools. The extensive 
training required for employment in 
many technical occupations is imposing 
a great burden on existing programs in 
vocational and technical education. The 
severe shortage of curriculum experts 
find teaching personnel has been a tra- 
ditional problem throughout the national 
scene. 

A research and development project 
now under way at Pennsylvania State 
University is attempting to solve these 
and other problems in vocational-techni- 
cal education. The purpose of the proj- 
ect (supported in part by the Bureau of 
Research, USOE) is to prepare course 
material in the three core areas — tech- 
nical mathematics, communications 
skills, and engineering science — suitable 
for youth and adults in the first two 
years of post-high-school technical ed- 
ucation. These materials are presented 
to students at a typewriter terminal with 
an audiovisual display unit controlled 
by a computer. 

What is Comput' r-Asststed instruction? 

Computer-assn ed instruction (CAI) 
is, in reality, instruction prepared by 
a human teacher for presentation under 



computer control. Experienced teachers 
prepare materials for a teaching pro- 
gram. The student receives instruction 
by means of slides, tape messages, and 
typed information. Then, questions and 
problems are typed and the student re- 
sponds by typing an individualized an- 
swer. The student’s responses to ques- 
tions determine hov the instruction will 
progress. Students who do not fully 



understand the material are branched 
to remedial instruction, and thus avoid 
repetition of old material. 

Computer-assisted instruction en- 
at les each student to have a device 
which provides private specialized tu- 
toring. CAI has the potential to accele- 
rate the learning process by avoiding 
needless repetition of drill after the 
student has mastered it. It can also be 



This diagram shows 
schematically the 
computer center and 
its service areas. The 
instructor goes to the 
computation center, 
types up a new pro- 
gram, or revises an 
old one, and it is 
ready to be used by 
students in the several 
schools which are 
hooked up to the 
center. The computer 
described in this arti- 
cle can accommo- 
date eight remote 
teaching terminals, 
any five of which may 
operate at same time. 



5JVS6NT TERMINALS 

WILLIAMSPORT AREA COMMUNITY COLLEGE. 

UJ'tliamsporh, Po, 




STUDENT TERMINALS 
Altoona c&rtot 
Altoona, Pa. 



FIGURE 2 



Schematic Drawing cf Equipment 
Configuration * 




PRECEDING PAGE BLANK- NOT FILMED 





PROGRAMING COURSES 

The segment of the program on 
the left was taken from a program 
in trigonometry. The segment 'ieals 
with the law of sines and the law of 
cosines. Knowledge of these laws is 
necessary for the analysis of force 
vectors, for the determination of 
velocity components, for finding the 
components of an alternating cur- 
rent, and for other tasks required in 
technical occupations. 

used to diagnose a student’s problems 
and to provide appropriate remedial 
measures. 

CAI differs from programed texts in 
that the student’s responses are evalu- 
ated against anticipated answers stored 
in the memory of the computer. Since 
the correct answers are stored in the 
computer, the student cannot look up 
the answer and he cannot proceed in the 
program until he demonstrates that he 
understands the material. 

Some programed texts contain 
branching programs. These include mul- 
tiple-choice questions and instructions 
to turn to a certain page. CAI, as pres- 
ently developed at Penn State, does not 
present alternatives to the student, but 
rather requires that the student con- 
struct his own response to each ques- 
tion. The computer compares the stu- 
dent’s response with a number of stored 
possible answers. Correct answers are 
accepted, the student is provided with 
encouragement, and is presented with 
new instruction. When the student gives 
an incorrect answer, the computer pre- 
sents a diagnostic comment and 
branches the student to remedial in- 
struction. 

The computer can be programed to 
present material on the basis of the 
student’s response history or on the ba- 
sis of other relevant information, such 
as response time, scores on achievement 
tests, error rate, or amount of time spent 
on instruction. Combinations of these 
criterions may also be used. 

CAI offers several potential improve- 
ments over the so-called teaching ma- 
chines. Most teaching machines are 
merely mechanisms for the presentation 
of programed texts. In most cases, these 
devices are used to eliminate page turn- 
ing or to eliminate the possibility that 
the student might omit some of the in- 
struction. The instruction which these 
machines presents is little different from 
that presented in a programed text. 
CAI, however, provides greater flexi- 
bility in the presentation of material, 
utilizes audiovisual techniques, evalu- 

c»tca student luipuiuo, keeps detailed 










i i i ■ -4nr> i m 9 m m § 

Students at the various teaching terminals progress at their own speeds, receiving special, reme- 
dial aid when necessary, or acceleration and enrichment if the situation calls for it. This student 
is receiving what Penn State researchers consider "tailored" instruction. 



records of the student’s progress, and in 
a sense, tailors the instructioi 1 sequence 
to each student. 

Advantages for Vocational Students 

There are a nu nber of reasons why 
CAI may be advantageous for voca- 
w.onal students. Potentially, it permits 
an efficient use of expensively and high- 
ly trained teachei s. An instructor may 

f J J* - *!.'■ h ! *ht !, -• 

: *= [Lt '= j u!' - M i':l ^!udct^^ at I 
j! I 1 '" - i » r.u t! % |t j 1 it 

through a program prtpartd b) an cx 
penenied Ititih^i. A substantial iiiiitdM 

in lb.- *! ., !•'! t ? • j h. r rati ., »» 
i 1 ‘ f « '* ' f ■ I.?., j ; ? t ‘.ai 

Wurh for the tear. her. 

Computer assisted instruction makes 
p' -sit It an at< tit r at t ii and individual 

Lalion of instruction which has often 
been desired in educational theory, but 
rarely achieved in practice. The com- 
puter reacts to features of the students 
performance, presents appropriate re- 
medial instruction when a student is 
not succeeding, and presents accelerated 
material when a student is insufficiently 
challenged. 

Another factor which assists the stu- 
dent is the gaming interaction between 
the student and the machine. Here, the 
role of the machine is that of an oppo- 
nent with which the student interacts, 
just as he interacts with problems in 
laboratories and real situations. 

An instructor may update a curricu- 
lum which has been programed for the 
computer by merely typing the program 
revisions at the typewriter terminal. 



Often, a complete revision process may 
be accomplished in one day. 

It is not necessary that the instructor 
be skilled in computer programing. A 
simple programing language is used 
which most persons can learn after a 
few hours of instruction. The most im- 
portant factor in preparing the course 
material is that the student must be able 
to comprehend it. The effectiveness of 
the program depends on the eSwtive- 
ness of the teacher who prepares it. It 
is imperative that the instructor use a 
carefully planned teaching strategy. 

In fact, analysis of the data obtained 
from students who have received in- 



struction may be used to show the in- 
structor ways to improve his program. 
The teacher has an opportunity to ob- 
serve the strengths and weaknesses of 
his presentation. 

The computer we are using is a:- IBM 
1410. It can accommodate eight remote 
teaching terminals, any five of which 
may operate simultaneously. Costs for 
operating the system include computer 
rental, terminal rental, and telephone 
transmission charges. Present costs are 
high. However, larger systems accom- 
modating upwards of 40 remote termi- 
nals are possible if a larger computer is 
used. It has been estimated that the 
cost of operating such a system could 
be less than one dollar per hour per 
student. 

How CAI Will Serve Education 

Since the terminals and audiovisual 
i nits are available commercially, it is 
possible for any institution to use 
courses programed for computer-assist- 
ed instruction. The courses may be used 
for supplemental on the job training, 
adult retraining programs, and for pre- 
senting current techniques and informa- 
tion to fmpkjfs, as >>t!l as fox instruc- 
tion of students in technical institutes. 

It is not anticipated that all courses, 
<>r that am parinuiar fit 

# 1 i J - f * * f * i - , * i ’ - f , - { 1 

r J , 1 . • ' > ’ }[«» 

^ i ^ . - . i . * . 



*n occupational education. 



Pictured is a typical computer-controlled typewriter terminal and audiovisual unit to which the 
student goes for instruction. Costs for operating the system include computer rental, terminal 
rental, and telephone transmission charges. 





119 



APPENDIX C 



C.l Student Attitude toward Computer-Assisted 

Instruction 



C.2 Spelling Achievement Test 



PRECEDING PAGE BLANK- NOT FILMED appendix c 

STUDENT ATTITUDE TOWARD COMPUTER ASSISTED INSTRUCTION 

COMMUNICATION SKILLS 

This is upt a test of information; therefore, there is no one "right" answer to a 
quest, on. We are interested in your opinion on each of tlia statements below. Your 
opinions will be strictly confidential. Do not hesitate to put down exactly how you 
feel about each item. We are seeking information, not compliments; please be frank. 

NAME: DATE 



NAME OF COURSE 

^ !■ m ■ 



CIRCLE THE RESPONSE THAT /OST NEARLY REPRESENTS YOUR REACTION TO EACH OF THE STATEMENTS 
BELOW: 

1. While taking Computer Assisted Instruction I felt challenged to do my best work. 



Strongly Disagree Uncertain Agree Strongly 

Disagree Agree 

2. The material presented to me by Computer Assisted Instruction caused me to feel that 
no one really cared whether I learned or not. 



Strongly Disagree 

Disagree 



Uncertain 



Agree 



Strongly 

Agree 



3. 



The method by which I was told whether I had given 
monotonous . 






wrong answer became 



Strongly Disagree Uncertain Agree 

Disagree 

4. I was concerned that I might not be understanding the material. 



Strongly 

Agree 



Strongly Disagree Uncertain Agree 

Disagree 



Strongly 

Agree 



5. I was not concerned when I missed a question because no one was watching me anyway. 



Strongly Disagree Uncertain Agree Strongly 

Disagree Agree 

6. While taking Conqmter Assisted Instruction I felt isolated and alone. 



All the 
time 



Most of 
the time 



Some of Only Never 

the time occasionally 




While taking Computer Assisted Instruction I felt as if someone were engaged in 
conversation with me. 



All the 
time 



Most 'of 
the time 



Some of 
the time 



Only 

occasionally 



Never 



The responses to my answers seemed appropriate. 



All rne 



+1-1 

t .Hi 

time 



Most of 
the time 



Some of 
the time 



Only 

occasionally 



Never 



I felt uncertain as to my performance in the programmed course relative to the 
performance of other. 



All the 
time 



Most of 
the time 



Some of 
the time 



Only 

occasionally 



Never 



I found myself just trying to get through the material rather than trying to learn. 



All the 
time 



Most of 
the time 



Some of 
the time 



Only 

occasionally 



Never 



I knew whether my answer was correct or not before I was told. 



Quite often Often Occasionally 

I guessed at the answers to questions . 



Seldom 



Very seldom 



Quite often 



Often 



Occasionally 



Seldom 



Very seldom 



In a situation where I am trying to learn something, it is important to me to know 
where I stand relative to others. 



Strongly Disagree Uncertain Agree 

Disagree 

I was encouraged by the responses given to ray answers of questions 



Strongly 

Agree 



Strongly 

Disagree 



Disagree 



Uncertain 



Agree 



Strongly 

Agree 



As a result of having studied some material by Computer Assisted Instruction, I am 
interested in trying to find out more about the subject matter. 



Strongly 

Disagree 



Disagree 



Uncertain 



Agree 



Strongly 

Agree 



In view of the time allowed for learning, I felt too much material was presented 




All the 
time 



Most of 
tha tima 



Some of Only Never 

the time occasionally 



I was more involved in running the machine than in tinders tanding the material. 



• * 

All the Most of Some of Only 

time the time the time occasionally 

I felt I could work at my own pace with Computer Assisted Instruction. 



• • 

Strongly Disagree Uncertain Agree 

Disagree 

Computer Assisted Instruction makes the learning too mechanical. 

: : : : 

Strongly Disagree Uncertain Agree 

Disagree 



Never 



Strongly 

Agree 



Strongly 

Agree 



I felt as if I had a private tutor while on Computer Assisted Instruction. 



Strongly Disagree Uncertain Agree 

Disagree 

I was aware of efforts to suit the material specifically to me. 

: : : : 
Strongly Disagree Uncertain Agree 

Disagree 



Strongly 

Agree 



Strongly 

Agree 



I found it difficult to concentrate on the course material because of the hardware. 



All the 
time 



Most of Some of Only 

the time the time occasionally 



The Computer Assisted Instruction situation made me feel quite tense. 



Never 



■ • • 
Strongly Disagree Uncertain Agree 

Disagree 



Strongly 

Agree 



Questions ere asked which I felt were not relevant to the material presented. 



All ie 
time 



Most of 
the time 



Some of Only 

the time occasionally 



Never 





L 



25. Computer Assisted Instruction is an inefficient use of the student's time. 



r 

L 



Strongly 

Disagree 



n 26. 

I ' 

U 



Disagree Uncertain Agree 

I put in answers knowing they were wrong in order to get information from the machine. 



Strongly 

Agree 



0 



Quite often 



Often 



Occasionally 



Seldom 



Very Seldom 



27. 



f!nnr*<=>mi no f-ho Miirco rr.ot-ov! ol T ♦ „ a...* .... j t _ ... . . . . 

o w ~‘ a - — *■*■»»•*■ * wjr v/uuijiULc i ais l6u instruction, my feeling 

toward the material before I came to Computer Assisted Instruction was: 



Very 

favorable 



Favorable 



Indifferent 



Unfavorable 



Very 

unfavorable 



28, 






y 



Concerning the course material I took by Computer Assisted Instruction, my feeling 
toward the material after I have been on Computer Assisted Instruction is: 



Very 

favorable 



Favorab le 



Indifferent 



Unfavorable 



Very 

unfavorable 



29. 



I was given answers but still did not understand the questions. 

: : : : 
Very often often Occasionally Seldom 



Very Seldom 



30. While on Computer Assisted Instruction I encountered mechanical malfunctions 



31. 



Very often Often Occasionally Seldom 

Computer Assisted Instruction made it possible for me to learn quickly, 



Very Seldom 



Strongly 

Disagree 



Disagree 



Uncertain 



Agree 



Strongly 



Agree 




S 32. I felt frustrated by the Computer Assisted Instruction situation. 



■ 

■ 



Strongly 

Disagree 



Strongly 

Agree 



1 1 



33. 



Disagree Uncertain Agree 

ihe responses to my answers seemed to take into account the difficulty of the question. 



I 



Strongly 

Disagree 



Uncertain 



34. 



Disagree 

I could have learned more if I hadn't felt pushed, 



Agree 



Strongiy 

Agree 



I 



l 



Strongly 

Disagree 



Disagree 



Uncertain 



Agree 



Strongly 

Agree 



jERJC " 



i 









r 



i 

r 



■ 

l 



i 






» 



i 




35. The Computer Assisted Instruction approach is inflexible. 



p 37. 

Li 



Strongly 

Disagree 



Disagree 



Uncertain 



Agree 



Strongly 

Agree 



36. Even otherwise interesting material would be boring when presented by Compter 
Assisted Instruction. 



Strongly 
Dis igree 



Disagree 



Uncertain 



Agree 



Strongly 



In view of the effort I put into it, 
Computer Assisted Instruction. 



ngree 

I was satisfied with what I learned while taking 



J 



o 



39. 



hO. 



Strongly 

Disagree 



Disagree 



Uncertain 



Agree 



Strongly 

Agree 



38. In view of the amount I learnea, I would say Computer Assisted Instruction is 
superior to traditional instruction. 



Strongly 

Disagree 



Disagree 



Uncertain 



Agree 



Strongly 

Agree 



With a course such as I took by Computer Assisted Instruction, I would prefer 
Computer Assisted Instruction to traditional instruction. 



Strongly 

Disagree 



Disagree 



Uncertain 



Agree 



Strongly 

Agree 



I am not in favor of Computer Assisted Instruction because it is just another step 
toward de-personalized instruction. 



Strongly 

Disagree 



Disagree 

41. Jfy typing training and experience has been: 



Uncertain 



Agree 



Strongly 

Agree 



Very 

Extensive 



Extensive 



Some 



Little 



Very 

Little 



42. Typing experience is necessary in order to perform easily on the machine. 



Strongly 

Disagree 



Disagree Uncertain Agree 

43. I think Computer Assisted Instruction would be best for learning: 



Strongly 

Agree 



Spelling 



Punctuation 



Grammar 



Report 

Writing 



Vocabulary 



< 

'i 



r. 



r=s*"» 



'Ll 








<T: 

it i 








U 44. 


How long do you feel you could work efficiently in 


computer guided instruction at 


r: 
1 ^ 


one sitting? 


(circle one) 




I ; 


Half hour 


1 hour 1-1/2 hours 2 hours 


More than 2 hours 


C 






(Approx, how many hours 



THIS SPACE IS PROVIDED FOR ANY COMMENTS YOU CARE TO MAKE ABOUT COMPUTER ASSISTED INSTRUCTION. 



- l - 



1 

a 



i 

i 

i 

■ 

H 

a 

■ 

i 

i 

1 10/31/66 
Form 2b 

Communication Skills 

i 



Appendix C.2 
Achievement Test 

Thirty-seven words are pronounced by tape, used in sen- 
tences, and pronounced again. The student then types the 
word. At the conclusion of the test, the score is typed out 
with an indication of the kinds of errors made. There is a 
possibility of fifty errors in the following 37 words: 



1. 


decei tful 


19. 


bountiful 


2. 


echoes 


20. 


piece 


3. 


dyeing 


21 . 


can ' t 


4. 


laboratory 


22. 


access 


5. 


two-thirds 


23. 


complement 


6. 


quiet 


24. 


matnemati cs 


7. 


principle 


25. 


theses 


8. 


calendar 


26. 


intercede 


9. 


armies 


27. 


angri ly 


10. 


prescribe 


28, 


vein 


11. 


achieving 


29. 


they * 1 1 


12. 


derivative ^ 


30. 


personal 


13. 


self-improvement 


31. 


invisible 


14. 


formerly 


32. 


thieves 


15. 


capital 


33. 


relies 


16. 


pri vi lege 


34. 


temperature 


17. 


knives. 


35. 


thirty-one 


18. 


circumference 


36. 


instance 






37. 


grammar