 

                  

I..
. 0.0.300” . .1
....br. .0... .... 00
0.1.0.0003: J...
. . .

 

 

00:0. 0... . . 0)....
.010- .llAuld-ﬂ...“ 000.. n

300"!“ .ﬁrIR-‘f... I0 0.
. .

 

..
.P...’ 0..
l I. .
.w

l
. r

       

      

                

..0r I . I00...
. ’f’li" !
, .1 .I giill’lgsPaI ..I
,0 “00.004 exv‘n‘é
. $0... .1 ...-I.)
0 . .o
.0000
.I I
001‘)
00.00....
-.. .0I
0.0..
. 0.0

0
I 'k
HM

 

...00 Q ‘
=00 _. .. . ‘

     

 

 

   

 

 

           

      

         
         

. 0
. .00.. .0 ..
. - ... . ... .0 .0 ...:
i > 0 . 0.0,.
00.00..-. .0. ...0.
. . \‘b‘n...’.).‘0 .
0 . . 0....) .00: 0.0.
. 09.030-020.000).
..0..I. ..I 00.. .0... .
.0 I ... ..0...0 .. 00.00-... .03.. . ..
. I .0 v 0 . . . - . .. . .0 .
. .000 .. - . I ..0 30.0-0UI 100.0 .0. . .00 .
0 u. . 0 I 0 -0...00 000.u.0. 0. 9.0.0.00005 H0 ... ..0 0.0109020300103001? ....
0 . . 0 ..I0 0100. .... ..000001 . ..0 .le..0...0.t.~.005.00.001.150} .0. .0090 0..)[0000‘ ll
. I 0 I 00 0 0 .. ... I; .0 Q . .0. u" . .001 .IV'./. .01 0:0...01..I.:.\I;o .. i 0. 0|
0 -. . .00....0 I.. 0. .0..- 0.. .. ..00.. 0 ., .0. 13.00.51: 00000000300900.“ 0.00000000000P100.HH.
IIII l l I -\~..,AIL‘O- E‘. Unit. .
7\ 0. l ' \ - - 0 \Ibil.l 00-0.. '0. . In”...
- 0 . . . .0. . . ...-.0! .i. -.....
I I I . 0 0 -. ... I 0 0 0.0.0.0... ...I
- -I I.-I .00.. .. ..- ....000 .
0. . 0 0000.. .PIII. 0.00-0 . ...
.- -. . I . ..-.00...
. 2 .III .0- 0| ..
. .. . ..n. ...“. .. .0.0
. .0. .00... .
. ..I .00 I
. P 0. 00-.. .
. . I 0. :3-
. .. 00.00..
I I .
.0 0....-. v
I .....I.. .UKPx
0.. .... ,
.. 0......

0| 000
00.00 L. 20.00.
I I I I 0

.00 0

 

.HI . .....
. I 00 0-
0.0 y \
- 0 0 .
. - 0.. .
. . 00
0 0 ..0 u .
I - . -000 . 0.
0 . 0 . .000 ...
. . I- 0 0.. ..
. 0 0 I ‘II‘
. 0.0 . 0 II.
. 0 0 I . I
. 0 .0 0 I
—I I . . I 0.0.0..~.
w 0 0 a). .
00 . .
w I o 10.. I
...I I0. 0 . I- 000
. I .0 ... ..- “0
. 0 I I . II - .00
. . 0 0 I
V . I 0 0 0 - 4 .0
. I0. 000
v. 00 . 0 '00 . 0 I 0 l.
m 0 . 0 0.- I v. I0I) . . 0
v. 0 0 ‘I II I 0' - 0
0I . ,. I . . . 0.-
-00 . .0 00 ... ..
s. . 0-00Ml 0 I -0P000) 00
. I 0 I 0- 0’ 0 000. 0.
WI- - .. I 00. I0 00- 0.00I I
n 0 ..I I 00 . . - 0.. - 03.0..
I I 0 . .0 l 0 .- . 00.0. I .
. . I I 0. 0000. 000 . .
u 0.00 0 .. 0 . 0 .00. .0.|I.0._'. LII I . I. . I 0W0. .
. 0.00. 00 . 0 .00 '0. f 0.0. .0000. 0.
a I .0 ' 0 . LI! 0 0 I0 0 . . 0.0‘ 0 0 0000. I‘. .000 0.0 0000 001.0..0
. I 0, - 0 00 0|: . 70-000. 0.0000300000 0... 00.. 10.00.0000
. . 000 ...It..0.0r.00000 0.0000000!!! . 30.000.000’00000900
I.. ...........0.0.00.-0.0.I.0)0.0.00.000| 00000 00lvl‘00000ll0000II-y0'0000000
. . . . .0000. .0I0v000.000003o00. 0500.. 0003000000301
.. .1 00000000000010

 

 

NIVERSITY LIB

lllllllllllilllllllllllllllll ill'lllll

31293 0088

lllllolllllllll

 

 

 

 

This is to certify that the

dissertation entitled

Minimalist Design Principles Applied to

the Problem of Computer Anxiety

presented by

Christopher Branko Reznich

has been accepted towards fulﬁllment
of the requirements for

Doctor of W degree in Educational]. Systems
Development

 

 

Date \/L1‘ZU:’ /"T); /??.3

MSU is an Afﬁrmative Action/Equal Opportunity Institution 0— 12771

 

 

 

 

 

F‘ «
LIBRARY
Michigan State
University

n J

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

______________—7
DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
cmmmuq

MINIMALIST DESIGN PRINCIPLES APPLIED TO
THE PROBLEM OF COMPUTER ANXIETY

BY

Christopher Branko Reznich

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Counseling, Educational Psychology
and Special Education

1993

ABSTRACT

MINIMALIST DESIGN PRINCIPLES APPLIED
TO THE PROBLEM OF COMPUTER ANXIETY

By
Christopher Branko Reznich

Computer anxiety is one possible barrier to acquisition of computer
skills. To test whether an instructional intervention could decrease
computer anxiety while training subjects in basic word processing skills,
instruction developed according to the Minimalist Design principles of
John Carroll and was compared with a commercially available
instructional unit.

The design of the experiment was a repeated measures control
group with random assignment of subjects. TWO instructional sessions
were held, separated by one month. Subjects were pre- and posttested
during each instructional session with respect to computer anxiety and
desire for further instruction in computer skills. Subjects worked on two
similar word processing tasks, a practice and a transfer task, during the
ﬁrst instructional session, and on another similar word processing task
during the second instructional session. Subjects' times to complete the
word processing tasks were recorded, and their performances on the
word processing tasks were scored according to a performance checklist.
For the experimental group, there was an statistically signiﬁcant
decrease in computer anxiety immediately following the first instructional
treatment. For the control group, there was also a decrease in computer
anxiety, but which occurred over the one month time experimental tirne
frame. Both experimental and control treatments were equally effective

 

in training
signiﬁcant
further corl
lt we
according *

in computr

 

 

 

in training subjects to perform the word processing tasks. There were no
signiﬁcant diﬂ‘erences between the groups with respect to desire for
further computer training as a result of the treatments.

It was concluded that word processing instruction designed
according to Minimalist Principles can produce an immediate reduction

in computer anxiety among more computer anxious subjects.

 

Parents. P

This dissertation is dedicated to my wife, Pornpimol June; to my

parents. Paul and Eleanore: and to my brother, Brian.

 

 

Inc‘
mdrsupp
Henry, Ph
go as well
Kalaian. :
analysis.
alFuUer
Scale.

li
SUﬁprI
atadern
“ﬁnan1
[kVekn

Educat

ACKNOWLEDGMENTS

I would like to thank the members of the dissertation committee for
their support and advice: Ann Austin, Ph.D., Joe Byers, Ph.D., Rebecca
Henry, Ph.D., and Stephen Yelon. Ph.D. (Dissertation Director). Thanks
go as well to Pat Mullan, Ph.D., Betsy Becker, Ph.D., Ms. Hripsime
Kalaian, and Richard Houang, Ph.D. for their guidance on the data
analysis, and to George Marcoulides, Ph.D., of California State University
at Fullerton, for his advice and permission to use the Computer Anxiety
Scale.

I would also like to express my gratitude to the many teachers,
staff persons and colleagues who have helped me throughout my
academic career at Michigan State University, but most especially to
William Anderson, Ph.D., Ofﬁce of Medical Education Research and
Development, and Stephen Yelon. Ph.D., Department of Counseling,
Educational Psychology, and Special Education.

r:

 

 

 

LIST OF T;
CHAPTER

The
1an

 

Meat
Inter
lnsu

Deﬁ
Res:

TABLE OF CONTENTS

LIST OF TABLES ............................................................................ viii
CHAPTER 1: INTRODUCTION ......................................................... l
The Broader Context: Why Don't Some People Perform? .......... 1
Identiﬁcation of the Problem: Computer Anxiety ..................... 2
Purpose of the Study .............................................................. 4
Measurement of Computer Anxiety ........................................ 4
Interventions to Decrease Computer Anxiety .......................... 5
Instructional Design Approach .............................................. 7
The Research Idea ................................................................. 9
Deﬁnitions ............................................................................. 9
Research Questions ............................................................... 1 l
Hypotheses ............................................................................ 12
Implications of the Study ....................................................... 13
Summary and Order of Presentation ...................................... 14
CHAPTER 2: REVIEW OF THE LITERATURE ................................... 16
Introduction .......................................................................... 16
Anxiety .................................................................................. 16
Computer Anxiety and Its Measurement ................................ 18
Human Performance Considerations ...................................... 20
Instructional Design Considerations: Keller's ARCS Model
and Carroll's Minimalist Principles ............................... 22
Summary ............................................................................... 32
CHAPTER 3: METHODOLOGY ......................................................... 34
Introduction ........................................................................... 34
Design of the Study ................................................................ 34
Design of the Instructional 'n'eatment .................................... 35
Subjects and Subject Recruitment ......................................... 39
Instrumentation ..................................................................... 40
Experimental Procedures ....................................................... 41
Data Handling ....................................................................... 44
Data Scoring .......................................................................... 45
Data Analysis ......................................................................... 45
Power Analysis ....................................................................... 47
Summary ............................................................................... 50

vi

 

 

CHAPTER

lntr
Sigr
Lim
Dirt
Cor.

APPEKDH

CHAPTER 4: RESULTS .................................................................... 52

Introduction ........................................................................... 52
Description of the Sample ...................................................... 52
Summary of the Descriptive Statisch .................................... 69
Hypothesis Testing ................................................................. 70

Computer Anxiety ......................................................... 72

Word Processing Task Performance .............................. 8 1

Desire to Learn More Computer Skills .......................... 94
Summary of the Hypothesis Testing 108
Controlling for Pretest Group Differences:

Analysis of Covariance ................................................. 1 12
Summary of the ANCOVA Analysis ....................................... l 18
Exploring Gender Differences ................................................ l 19

CHAPTER 5: DISCUSSION ............................................................. 122
Introduction .......................................................................... 122
Signiﬁcance of the Study Results ........................................... 122
Limitations of the Study ......................................................... 128
Directions for Future Study .................................................... 130
Conclusion ............................................................................. 133

APPENDICES .................................................................................. 134
Appendix A: Control TYeatment Instructional Materials ......... 134
Appendix B: Experimental Treatment

Instructional Materials ...................................... 1 53
Appendix C: Computer Anxiety Scale .................................... 164
Appendix D: Background Information Form (Phase I) ............ 166
Appendix E: Background Information Form (Phase II) ........... 168
Appendix F: Feedback Form ................................................. 169
Appendix G: Data Collection Schedule .................................. 170
Appendix H: Consent to Participant in a Research Study ....... 171
Appendix 1: Machine Scoring Form ...................................... 172
Appendix J: Directions for Students (Phase I) ....................... 174
Appendix K: Time Sheet(Phase I) ........................................... 175
Appendix L: Word Processing Task 1 .................................... 176
Appendix M: Word Processing Task 2 .................................... 177
Appendix N: Scheme for Naming WordPerfect
Document Files ................................................. 178
Appendix 0: Word Processing Task (Phase II) ........................ 179
Appendix P: Directions for Students (Phase 2) ...................... 180
Appendix Q: Time Sheet (Phase II) ......................................... 181
Appendix R: Word Processing Task 1 Checklist ..................... 182
Appendix S: Word Processing Task 2 Checklist"; ................ 183
Appendix T: Word Processing Task Checklist (Phase II) ......... 184
LIST OF REFERENCES .................................................................... 185

vii

Table 1:

Table 2:
Table 3:

Table 4:

Table 5:
Table 6:

Table 7:
Table 8:

Table 9:

Table 10;
Table l 1:
Table 12:
Table 13;
Table 14;

 

Table 15;
Table 16: )
Table 17;
Table 18:
Télble 19;
Table 20:
Table 21;
Table 22:
Table 23:
Table 24:

Table 1:

Table 2:
Table 3:

Table 4:

Table 5:
Table 6:

Table 7:
Table 8:

Table 9:

Table 10:
Table 1 1:
Table 12:
Table 13:
Table 14:
Table l 5:
Table 16:
Table l 7:
Table 18:
Table 19:
Table 20:
Table 2 1:
Table 22:
Table 23:

Table 24:

LIST OF TABLES

Means and Standard Deviations of

Subjects' Age and GPA .................................................... 53
Counts and Percentages of Gender of Subjects ............... 54
AN OVA Table: Comparison of Phase I CAS Pretest

Scores Based Upon Gender ............................................. 54
Numbers of Semester/Terms of Subjects' Previous

Computer Study by Gender ............................................ 55
Counts and Percentages of Academic Majors of Subjects. 56
ANOVA Table: Comparison of Phase I CAS Pretest

Scores Based Upon Recoded Academic Majors ................. 57
Counts and Percentages of Class Year of Subjects ........... 58
Number of Semesters/Terms of Subjects'

Previous Computer Study ................................................ 59
Counts and Percentages of Subjects' Current Word
Processing Skills ............................................................. 60
Counts and Percentages of Subjects' Typing Skills ....... 61
Subject Attendance at Phase I Only and Phases I and II.. 62
Subjects' Mean Computer Anxiety Scale Scores: Phase I.. 62
Subjects' Mean Computer Anxiety Scale Scores: Phase II. 63

Means and Standard Deviations of Subj ects’

Desired Time to Learn Word Processing ........................... 64
Means and Standard Deviations of Subj ects'

Interest in Iemg Computer Skills ............................... 65
Means and Standard Deviations of Subjects'

Desired Time to Learn Computer Skills ............................ 66
Means and Standard Deviations of Subjects' Word

Processing Task Performance Measured by Checklist ...... 67
Means and Standard Deviations of Subjects'

Word Processing Task Time in Minutes ........................... 68
ANOVA Table for Computer Anxiety Scores

Repeated Measures, Control Group ................................. 74
Computer Anxiety Score Mean Comparisons,

Control Group ................................................................. 74
AN OVA Table for Computer Anxiety Scores

Repeated Measures, Experimental Group ........................ 75
Computer Anxiety Score Mean Comparisons,

Experimental Group ........................................................ 76
CAS Mean Score Comparison, Experimental Group,

Upper 50% ....................................................................... 77
CAS Mean Score Comparisons, Control Group.

Upper 50 ......................................................................... 78

viii

Table 25'.
Table 26:
Table 27:
Table 28:
Table 29:
Table 30: I
Table 31:
Table 32:
Table 33
Table 34:
Table 35;
Table 36;
Table 37;
Table 38:
Table 39;
Table 40;
Table 41:
Table 42.
Table 43
Table 44
Table 4 5
Table 46
Table 47
Table 48

 

Table 25:
Table 26:
Table 27:
Table 28:
Table 29:
Table 30:
Table 3 1:
Table 32:

Table 33

Table 34:
Table 35:
Table 36:
Table 37:
Table 38:
Table 39:
Table 40:
Table 4 1:
Table 42:
Table 43:
Table 44:
Table 45:
Table 46:
Table 47:
Table 48:

CAS Mean Score Comparisons, Experimental Group,

Lower 50% ..................................................................... 78
CAS Mean Score Comparisons, Control Group,

Lower 50% ..................................................................... 79
Phase I Pretest CAS Scores: Phase I Only and

Phases I and II ................................................................ 79
AN OVA Table Comparing CAS 1 Score,

Group by Attendance ...................................................... 80
AN OVA Table: Word Processing Task One Performance

as Measured by Checklist ................................................ 82
AN OVA Table: Word Processing Task TWO Performance

as Measured by Checklist ................................................ 83
ANOVA Table: Word Processing Task Three Performance

as Measured by Checklist ................................................ 83
Percentages of Subjects Scoring "9" on the Performance
Checklist ......................................................................... 84
ANOVA Table for Word Processing Scores

Repeated Measures, Experimental Group ........................ 85
Word Processing Performance Mean Comparisons.
Experimental Group ........................................................ 85
ANOVA Table for Word Processing Scores

Repeated Measures, Control Group .................................. 86
Word Processing Performance Mean Comparisons,

Control Group ................................................................. 86
Word Processing Task Performance Comparisons,
Experimental Group, Upper 50% CAS ............................. 87
Word Processing Task Performance Comparisons,
Experimental Group, Lower 50% CAS .............................. 87
Word Processing Task Performance Comparisons,

Control Group, Upper 50% CAS ...................................... 88
Word Processing Task Performance Comparisons,

Control Group, Lower 50% CAS ...................................... 88
ANOVA Table: Word Processing Task One

Time in Minutes .............................................................. 89
ANOVA Table: Word Processing Task No

Time in Minutes .............................................................. 89
AN OVA Table: Word Processing Task Three

Time in Minutes .............................................................. 90
AN OVA Table for Word Processing Time

Repeated Measures, Control Group ................................. 9 1
Word Processing Time Mean Comparisons,

Control Group ................................................................. 9 1
ANOVA Table for Word Processing Time

Repeated Measures, Experimental Group ........................ 92
Word Processing Time Mean Comparisons,

Experimental Group ....................................................... 92
Word Processing Time Comparisons.

Experimental Group, Upper 50% CAS ............................. 93

Table 49:
Table 50:
Table 51:
Table 52:
Table 53:

 

Table 54:
Table 55: {
Table 56:
Table 57:
Table 58;
Table 59;
Table 60;
Table 61;
Table 62;
Table 63;
Table 64;
Table 65:
Table 66:
Table 67;
Table 68:
Table 69;
Table 70:
Table 71:

Table 72:

Table 49:
Table 50:
Table 5 1:
Table 52:
Table 53:
Table 54:
Table 55:
Table 56:
Table 57:
Table 58:
Table 59:
Table 60:
Table 6 1:
Table 62:
Table 63:
Table 64:
Table 65:
Table 66:
Table 67:
Table 68:
Table 69:
Table 70:
Table 7 1:
Table 72:

Word Processing Time Comparisons,

Experimental Group, Lower 50% CAS .............................. 93
Word Processing Time Comparisons,
Control Group, Upper 50% CAS ...................................... 94
Word Processing Time Comparisons,
Control Group. Lower 50% CAS ....................................... 94

ANOVA Table: Desired Time to Learn Word Processing,

Phase I Pretest ................................................................ 96
ANOVA Table: Desired Time to Learn Word Processing,

Phase I Posttest .............................................................. 96
ANOVA Table: Desired Time to Learn Word Processing,

Phase II Pretest ............................................................... 97
ANOVA Table: Desired Time to Learn Word Processing,

Phase II Posttest .............................................................. 97
AN OVA Table: Interest in Learning Computer Skills,

Phase I Pretest ................................................................ 98
AN OVA Table: Interest in Learning Computer Skills.

Phase I Posttest ............................................................... 98
AN OVA Table: Interest in Learning Computer Skills,

Phase II Pretest ............................................................... 99
AN OVA Table: Interest in Learning Computer Skills.

Phase II Posttest .............................................................. 99
ANOVA Table: Desired Time to Learn Computer Skills.

Phase I Pretest ................................................................ 100
ANOVA Table: Desired Time to Learn Computer Skills,

Phase I Posttest ............................................................... 100
ANOVA Table: Desired Time to Learn Computer Skills,

Phase II Pretest ............................................................... 101
ANOVA Table: Desired Time to Learn Computer Skills,

Phase II Posttest .............................................................. 101
AN OVA Table: Desired Time to Learn Word Processing
Repeated Measures. Control Group ................................. 102
Desired Time to Learn Word Processing

Mean Comparisons, Control Group ................................ 102
ANOVA Table: Desired Time to Learn Word Processing
Repeated Measures, Experimental Group ....................... 103
Desired Time to Learn Word Processing

Mean Comparisons, Experimental Group ........................ 103
ANOVA Table: Interest in Learning Computer Skills,
Repeated Measures, Control Group ................................. 104
Interest in Learning Computer Skills

Mean Comparisons, Control Group ................................. 104
ANOVA Table: Interest in Learning Computer Skills, '
Repeated Measures, Experimental Group ........................ 105
Interest in Learning Computer Skills

Mean Comparisons, Experimental Group ........................ 105
ANOVA Table: Desired Time to Learn Computer Skills
Repeated Measures, Control Group ................................. 106

Table 73:
Table 74:
Table 75:
Table 76:
Table 77:

Table 78:
Table 79:
Table 80:
Table 81:
Table 82:
Table 83:
Table 84:
Table 85:

Table 86:
Table 87;

 

Table 7 3:
Table 74:
Table 7 5:
Table 76:
Table 77 :

Table 78:
Table 79:
Table 80:
Table 81:
Table 82:
Table 83:
Table 84:
Table 85:

Table 86:
Table 87:

Desired Time to Learn Computer Skills

Mean Comparisons, Control Group ................................. 106
ANOVA Table: Desired Time to Learn Computer Skills
Repeated Measures, Experimental Group ........................ 107
Desired Time to Learn Computer Skills

Mean Comparisons, Experimental Group ........................ 107
Beta-Coefﬁcient for the Regression YCAS 2 over XCAS 1.
Experimental Group ....................................................... 1 14
Beta-Coefficient for the Regression YCAS 2 over XCAS 1.
Control Group ................................................................. 1 14
ANCOVA Table for Adjusting CAS 2 Means ...................... 1 15
Adjusted Mean Diﬁ'erences: CAS 2 ................................... 1 15
ANCOVA Table for Adjusting CAS 3 Means ...................... 1 16
Adjusted Mean Differences: CAS 3 ................................... 1 16
ANCOVA Table for Adjusting CAS 4 Means ...................... 1 17
Adjusted Mean Differences: CAS 4 ................................... 1 17

ANOVA Table: Experimental vs. Control Group, CAS 2.... 1 18
Means. Standard Deviations, Standard Errors.

Experimental and Control Group, CAS 2 ......................... 1 18
Two Factor ANOVA Table for Gender and Group: CAS 1 .. 120
Two Factor ANOVA Table for Gender and Group: CAS 2.. 120

xi

 

The Brgac

Peor
reasons. .
underlyin,
for perforr

and obsta

 

have no e1
nonperfon
for DEFICIT
emironmf
4) employ:
Causes of
While a la
Coaching
beneﬁts 01'
Mag
ProbIEmS

CHAPTER ONE
INTRODUCTION

WWW

People choose not to perform a required task for a variety of
reasons. Mager and Pipe (1970) note that lack of skill is but one reason
underlying performance deﬁciencies. Other reasons include punishment
for performance instead of reward, lack of consequences for performance.
and obstacles to performance. In these latter instances, training will
have no effect due to the fact that the real conditions causing
nonperformance are not addressed. Rossett (1992) suggests four reasons
for performance problems: 1) employees' lack knowledge or skill, 2) the
environment is in the way, 3) there are no, few, or proper incentives, and
4) employees are unmotivated. Rossett further suggests that different
causes of performance problems will necessitate different interventions.
While a lack of skill or knowledge might require training, job aids, or
coaching, lack of motivation may require informing workers of the
beneﬁts of performance, or promoting early success at a task.

Mager and Pipe, and Rossett remind us that not all performance
problems can be solved by training alone. Awareness of the real causes
of performance problems must inform any attempt at improving human
performance. This approach underlies the following study, which
focuses on one case of the wider problem of people who do not want to

learn how to use a new technology because of motivational diﬂiculties:

the case

because t

Idgng'ﬁggi

Mar
"Compute
individua
possibility
Rosen an:
C0mputer
than com
mew h:
Comptlter
New as

The
for StUder
centmy. I
[1988), 81“.
for 0011601

2
the case of people who do not want to learn how to use computers

because they are "computer anxious."

W

Many people exhibit fear at the thought of using computers.
"Computer anxiety" is deﬁned as "the fear or apprehension felt by
individuals when they use computers, or when they consider the
possibility of computer utilization" (Simonson, et al., 1987 , p. 238) .
Rosen and Maguire (1990), in a meta-analysis of 8 1 research reports on
computer anxiety stated that, on average, 25% of all people "feel less
than completely comfortable with computers" (pg. 180). Computer
anxiety has been shown to be a signiﬁcant barrier to learning to use
computers (Marcoulides, 1988). Research has identiﬁed computer
anxiety as a signiﬁcant problem of human performance.

The ability to use a personal computer is becoming a "basic skill"
for students, workers and professionals in the closing decade of the 20th
century. According to a report by the omce of Technology Assessment
(1988), an overwhelming body of evidence suggests that new technologies
for collecting. storing, manipulating, and communicating information are
revolutionizing the structure and performance of the national economy.
As of 1989, there were 45 million computers in use, a ﬁgure which is
growing exponentially (White, 1991). David L. House, Senior Vice
President of Intel, predicted that 22 million computers would be sold in
1991 (White, 1991). Skill in the use of a computer is becoming a

prerequisite for success in the workplace, and there is a great need for

3
instruction in the use of computers that enables people to take

advantage of these important tools. Given the overwhelming importance
of acquiring competence in the use of computers, combined with the
prevalence of computer anxiety, there is a need for interventions
designed speciﬁcally for individuals who are computer anxious but who,
nevertheless, must acquire competence in the use of computers.

One can deﬁne the problem of computer anxiety as a performance
problem according to Rossett's taxonomy as having aspects of skill
deﬁciency and motivational deﬁciency. Given Marcoulides' observation
that computer anxiety is a barrier to learning to use computers, we can
assume that many computer anxious individuals need some form of
training in order to acquire competence in using computers: even
strongly motivated learners will need some form of training materials,
instruction, manuals. or job aids, in order to obtain the procedural
knowledge required to use a computer.

The computer anxious learner not only lacks necessary procedural
knowledge, but faces a motivational barrier brought about by his or her
fears. Thus, training, in whatever form, will not be enough. It will also be
necessary to address the motivational problem besetting computer
anxious individuals. Rossett (1992) suggests four possible interventions
addressing lack of motivation: 1) informing workers of the beneﬁts,
impact, and value of performance, 2) linking performance to work
challenges, 3) using role models, and 4) promoting early successes to

instill conﬁdence.

was

The

 

word proc
motivatio:
students r
computer
processin
more abor
purpose 0
any, are s

in 0
1“5111111181“
anxiety, p
relevant p
PTObIEm a

IEVIEW of I.

W

 

Test
Raub's Att

11981); Ma
Educauon;
(1981): am
“9891- Th.

Wily

The purpose of this study is to determine whether instruction in
word processing that is designed to address both the skill deﬁciency and
motivation deﬁciency of computer anxious university undergraduate
students will have a more positive effect upon 1) the levels of subjects’
computer anxiety. 2) the ability of students to learn basic word
processing skills, and 3) the desire of computer anxious students to learn
more about computers, than would a control treatment. It is also the
purpose of this study to determine whether or not treatment effects. if
any, are sustained over time.

In order to introduce the research idea. methodology and
instruments. it is necessary to introduce the measurement of computer
anxiety, possible interventions for decreasing computer anxiety, and
relevant principles of instructional design. These areas, along with the
problem area of computer anxiety, will be explored more thoroughly in a
review of the research literature in the next chapter.

WW

Tests have been developed to measure computer anxiety, including
Raub's Attitudes Toward Computers scale and Computer Usage Checklist
(1981): Maurer's Computer Anxiety Index (1983), a revised version of the
Educational Innovation Survey developed by Rohner and Simonson
(1981): and the Computer Anxiety Scale developed by Marcoulides
(1989). These instruments all use some form of paper and pencil self

report us,

Cronbac

Interventi

111W

 

overcome
Howard,
11987) de
clinical rr
desensiti
gr OUps.
Which in.
on relax;
811113.
Re
nmum,
Segregat
Special e
to realili
Studena
of instm
CaIllbre
mCIUdrn‘
of Stu der

suggest 1

5
report using a Likert-scale format. Reliability coefﬁcients as measured by

Cronbach's alpha are generally .80 and above (Cambre and Cook, 1985).

WWW

Investigators have tested several methods of helping people
overcome their computer anxiety (Bloom, 1985: Cambre and Cook, 1987 ;
Howard, 1984: Howard and Thomas, 1987; Weil, 1987). Weil, et. al..
(1987) developed an approach to computer anxiety reduction based upon
clinical methods including individualized relaxation training. systematic
desensitization, cognitive-behavioral thought-stopping and support
groups. Bloom (1985) suggests an "anxiety management" approach
which includes education about stress responses, skill building focusing
on relaxation and altering distracting thoughts, and practice of coping
skills.

Researchers have asked if computer anxiety can be diminished
with instruction in computer use. Howard and Thomas (1987)
segregated computer anxious students into separate classes and, with no
special effort to design the course to combat computer anxiety, were able
to realize reductions in computer anxiety scores of the treatment group
students. Howard stressed the need for research to determine what kinds
of instructional approaches are most likely to reduce computer anxiety.
Cambre and Cook (1987) found that a non-threatening environment
including no tests nor grades helped to reduce computer anxiety scores
of students in a one-week introductory course on microcomputers. They

suggest that computer anxiety can be remediated through positive

6
computing experiences. Unfortunately. Cambre and Cook do not specify

the exact nature of their instruction.

A study by Howard (1984) provides further support of the effect
instruction can have on computer anxiety. Working with two groups of
relatively low computer anxious students, Howard compared the eﬁ'ect of
two instructional treatments on computer anxiety scores. The control
treatment was a lecture on BASIC programming language, and the
experimental treatment was a demonstration and hands-on practice
session on DOS. He found that, while there was a difference in the post-
instruction computer anxiety scores for the two groups. they were due
more to an increase in the computer anxiety scores for the control group.
He attributed this increase primarily to the content and delivery of the
instruction: rather than being a true neutral control treatment, it had the
eﬁ'ect of increasing anxiety scores. This gives support, negative though it
may be, to the argument that instruction can have an effect on students'
computer anxiety.

One problem in the research on the effect of instruction upon
computer anxiety is the lack of adequate description of the instnictional
treatments used by the researchers. This experiment will seek to
describe fully the exact nature of the instructional intervention. thus

addressing a weakness of some past studies.

v—a
[3

cor
a s
or
int
ma
lea

do

Oh.
of
cor
ac.

WC

the
thl
Mi

(ll)
Ta‘
ad

21C

WW

Signiﬁcant work has been done by John Carroll to develop a set of
instructional design principles that directly address the domain of using
computers (Carroll. 1984a, 1984b, 1990). The principles are the result of
a series of qualitative studies of adults learning to use computers with
typical commercial computer manuals, which included the use of
introductory training materials. Carroll (1990) found that most such
materials actually obstructed learning because they interfered will the
learners natural leaming processes. The principles that Carroll
developed. called "Minimalist Principles," are intended to minin'rize the
extent to which reference materials used as instructional materials
obstruct learning and refocus the design of training materials on the goal
of supporting leamer- directed activity and accomplishment. This aim is
consistent with current theories of cognitive psychology in which learners
actively construct cognitive representations, or mental models. of the
world around them (Resnick, 1989). Minimalist documentation includes
mechanisms by which learners can access immediately the information
they need to attain a self-directed goal, thereby emciently constructing
their own mental models of how computers function. Support for
Minimalist Principles is also provided by cognitive load theory (Chandler
and Sweller, 1 99 1) which suggests that eﬁ'ective instructional material
directs cognitive resources toward activities that are relevant to leaniing
rather than toward preliminaries to learning. Minimalist documentation
adheres to this principle by "slashing the verbiage" so that learners may
access required and only required information, rather than slogging
through pages of unnecessary reading. One of the intended outcomes of

instn
at tas
sugge

motiv

word
mater
needs
needs
learn.
with j
meet
illogie
Studr

may

8
instruction designed according to Minimalist Principles is early success

at tasks involving the use of computers, which is one way Rossett (1992)
suggests for addressing the motivational needs of learners suffering from
motivational deﬁciencies.

The sheer volume of reference material that accompanies common
word processing software is daunting. Furthermore. instructional
materials intended to get users started are developed to address the
needs of a mass market for computer software, rather than the speciﬁc
needs of any individual user or group of users. Users have goals and I
learning styles which software documentors cannot imagine nor address
with precision. If commercial computer instructional materials cannot
meet learners' needs effectively, as suggested by Carroll (1984a), it is
illogical to suggest that they can meet the needs of computer anxious
students. Given that instruction can be manipulated as a variable, it
may be fruitful to attempt to design instruction according to Minimalist
Principles and to test whether or not such instruction has an effect on
computer anxiety, on computer anxious students' desire to learn
computer skills, and on computer anxious students' ability to use a
computer to complete a task. The Minimalist Principles form a potential
framework for developing computer instruction which can then be tested
with computer anxious learners to see if such instruction is eﬁ'ective. A
thorough discussion of the Minimalist Principles is contained in the
literature review in the next chapter

desig
of co:
instr
and 1'
corn}
USlIl:
madi
com]
are r
scho

corn]

C7
m
T")

L

flhﬂlﬂaﬁamdnldma

.r‘
’v

The proposed study is intended to test whether instruction
designed according to Carroll's Minimalist Principles can meet the needs
of computer anxious learners. If it can be shown 1) that a brief
instructional treatment can have an effect on reducing computer anxiety
and increasing the desire to learn how to use computers, and 2) if
computer anxious students can actually learn useful computer skills
using the instructional treatment, then the proposed study will have
made a useful contribution to addressing the problem of training
computer anxious learners. Computer anxious education majors who
are required to take at least one course in computer applications in
school settings, or computer anxious adults who must undergo
compulsory job upgrading related to computers, are examples of learners

who might beneﬁt from thoughtful applications of Carroll's principles.

Deﬁnitions

computer anxiety: "the fear or apprehension felt by individuals
when they used computers. or when they considered the
possibility of computer utilization" (Simonson, et al., 1987,
p.238L

computer skill: ability to use computers to perform tasks. such as

using word processing to write a term paper.

10
desire to learn: the expressed desire for instruction or continued

learning in a domain. The desire to learn be expressed either
verbally through self-report, through written assessment
such as completion of a motivational inventory or through

actions such as further study.

Minimalist Principles: a set of instructional design principles
developed by Carroll and his colleagues that are intended to
guide the development of instruction in the use of
computers. They include using real tasks, slashing the
verbiage, facilitating instant access to desired content, and
coordinating the system with the instruction.

word processing task: any task which requires the use of a

computer and word processing software.

practice task: a task undertaken by an individual in the process of
learning a skill.

transfer task: a task which closely resembles a practice task in
form, but which varies from the practice task in content.

 

 

11
We

1 . Would a g'oup of computer anxious students that is taught
basic word-processing skills using instruction designed
according to Minimalist Principles show a decrease in levels of
computer anxiety as measured by the Computer Anxiety Scale
than would a control group receiving commonly available

commercial instruction?

2. Would a group of computer anxious students that is taught
basic word-processing skills using instruction designed
according to Minimalist Principles perform better on a word
processing task more than would a control group receiving
commonly available commercial instruction?

3. Would a group of computer anxious students that is taught
basic word-processing skills using instruction designed
according to Minimalist Principles express more desire to learn
how to use computers than would a control group receiving
commonly available commercial instruction, as measured by
direct questions about their desire for further computer skills
instruction?

4. Would a group of computer anxious students that is taught
basic word-processing skills using instruction designed
according to Minimalist Principles display more desire to learn

how to use computers than would a control group receiving

12
commonly available commercial instruction. as measured by

attendance at a follow-up instructional session?

5. If there are any diﬁerences between the control g'oup and the
experimental group with respect to computer anxiety scores or
expressed desire to learn more computer skills, will these

differences be evident one-month after the initial study?

Hypotheses

l . Tine experimental group will score lower on a measure of
computer anxiety, the Computer Anxiety Scale, than will an
equivalent control group after initial traimng in the use of word

processing.

2 The experimental group will score lower on a measure of
computer anxiety, the Computer Anxiety Scale, than will an
equivalent control group when measured during a word
processing follow-up session attended one month after initial

training.

3. The experimental group will score higher on performance of a
word processing task than will an equivalent control group.
Diﬁ‘erences in performance scores will be stable after a follow-

up treatment.

13
4. The experimental group will indicate a greater desire to learn

more computer skills as measured by 1) subjects' estimates of
the number of hours they would be willing to devote to further
word processing skills study. 2) subjects' estimates of their
interest in learning other computer programs, and 3) subjects'
estimates of the number of hours they would be willing to
devote to developing their computer skills, than will an
equivalent control group, when measured immediately after
instruction.

5. The experimental group will indicate a greater desire to learn
more computer skills as measured by 1) subjects' estimates of
the number of hours they would be willing to devote to further
word processing skills study. 2) subjects‘ estimates of their
interest in learning other programs. 3) subjects' estimates of the
number of hours they would be willing to devote to developing
their computer skills, and 4) subjects' attendance at a second
computer session, than will an equivalent control group, when

measured during a one-month follow-up computer session.

lmnhsationsottheﬁtudy

If the hypotheses are upheld, this will be evidence for a powerful
instructional and motivational effect upon computer anxious learners.

Given the stakes for individuals who remain unskilled in the use of

sh
fre

th.

w)

sh

10

DE

Es

Pu

ins

C01

W0

14
computers in terms of lower productivity and employability. it is

important to seek such solutions.

In a wider context, the present study may have implications for
future technologies that arouse anxiety in potential learners. If it can be
shown that Minimalist Principles provide a useful instructional
framework for addressing the problems of computer anxious learners,
they may also be useful for addressing the problems of future learners
who will feel anxious about as yet undeveloped technologies.

Another implication of the study is in the wider area of learner
motivation and its influence on performance. The study may be able to
shed light on the question of how important learner motivation really is

to performance of a task.

W

This chapter set the broad problem area of why some people don't
perform, noting that there are many reasons for nonperformance.
Essential concepts of computer anxiety and its measurement, and
Minimalist Principles of instructional design were introduced. The
purpose of the study, research idea, deﬁnitions. research questions.
hypotheses and implications of the study were also introduced.

This study seeks to compare the efﬁcacy of computer skills
instruction designed according to Minimalist Principles with that of
commercially available materials in training learners to acquire basic
word processing skills, in reducing computer anxiety. and in promoting

15
learner motivation expressed as the desire to learn how to use

computers.

Chapter 2 will present a review of relevant literature, chapter 3 will
present the methods used in the study, chapter 4 will present the study
results, and chapter 5 will present a discussion of those results and
suggestions for further study.

we]
adc
an:
effe
ab<
prc

W0

I?

Ct.

deS

CHAPTER TWO

REVIEW OF THE LITERATURE

introduction

The purpose of this study was to determine whether instruction in
word processing that is designed according to Minimalist Principles to
address both the skill deficiency and motivation deﬁciency of computer
anxious university undergraduate students would have a more positive
effect upon 1) the desire of computer anxious students to learn more
about computers, 2) the ability of students to learn basic word
processing skills, and 3) the levels of subjects' computer anxiety, than
would a control treatment of commercially available materials. The
research idea, questions and hypotheses were developed upon review of
the literature on anxiety, computer anxiety and its measurement, human
performance considerations, and Minimalist Principles. A review of that
literature is presented in this chapter.

Angela:

Anxiety has been described as a 20th century construct (Cambre,
et. al., 1985). Its early history has been traced to ﬁend, who identiﬁed
"angst" as a clinical syndrome in 1895 (Spielberger, 1966). May (1950)
describes anxiety as a characteristic symptom of modern times; its

16

 

f0
re
pr
re
013
th.
of

Ch;
ﬂu
Sta

COT

sys
be)
ran
resI
iRte

17
contributors include pressure for social change produced by rapid

scientiﬁc and technological advances. Epstein (1972) describes anxiety as
a concept composed of three basic sources, each having a unique feeling
state associated with it. The ﬁrst source is primary overstimulation. which
involves frantic feelings of being overwhelmed with stimulation to the .
limits of tolerance. The second source is cognitive incongruity, which is
associated with situations involving a discrepancy between an
individuals cognitive plan and reality, and the resulting failure to
formulate a plan for coping with the discrepancy. The ﬁnal source is
response unavailability, a condition that occurs when the object
producing the arousal is unknown, when there is a waiting period
required before a response can be made, there is a conﬂict between
opposing response tendencies, or when the required response is not in
the individual's capability. Cattell and Scheier (1961) identiﬁed two types
of anxiety which they referred to as "trait" and "state" anxiety. TYait
anxiety is considered a relatively permanent and stable personality
characteristic. while state anxiety is a transitory condition which can
ﬂuctuate over time and treatment. Spielberger (1966) reﬁned the trait-
state anxiety distinction. He deﬁned anxiety states as "subjective.
consciously perceived feelings of apprehension and tension, accompanied
by or associated with activation or arousal of the autonomous nervous
system" (p. 16). He deﬁned anxiety traits as motives or acquired
behavioral dispositions that predispose an individual to perceive a wide
range of objectively nondangerous circumstances as threatening, and to
respond to these with anxiety state reactions disproportionate in
intensity to the magnitude of the objective danger. Spielberger (I 970)
developed the State-Trait Anxiety Inventory (STAI) as a measure of these

18
two dimensions of anxiety. Subsequent research on the STAI has

become the basis for development of measures of computer anxiety
(Cambre and Cook, 1985).

The fact that state anxiety can, as Cattell and Scheier note,
ﬂuctuate over time and treatment, opens the possibility that an
appropriately designed instructional treatment can be developed which
reduces a person's computer anxiety while training an individual to
acquire computer skills, if it can be shown that computer anxiety is
primarily a "state" rather than a "trait."

WWW

Researchers have attempted to deﬁne computer anxiety (Maurer
and Simonson, 1984: Raub, 1981: Simonson, et. al., 1987). Simonson,
et. al. (1987) deﬁned computer anxiety as "the fear or apprehension felt
by individuals when they used computers, or when they considered the
possibility of computer utilization" (p. 238). Thus, computer anxiety
arises in response to a speciﬁc action or thought related to computers, or
a "state” in Cattell and Scheier's view. Raub (198 1) deﬁned computer
anxiety as "the complex emotional reactions that are evoked in
individuals who interpret computers as personally threatening. This
deﬁnition describes an anxiety state, in contrast to an anxiety trait" (p. 9,
underlines in the original). Maurer and Simonson ( 1984) reported that a
computer anxious person would exhibit avoidance of computers and the
areas in which they were located; excessive caution when using

computers; negative remarks toward computers and using computers;

19
and attempts to shorten the times when computers were being used.

This view of computer amdety reinforces the state rather than the trait
characteristic.

Several researchers (Raub, 198 l; Marcoulides, 1985; Simonson
and Rohner, 1981; Simonson, et. al., 1987) have designed and tested
instruments intended to measure computer anxiety. Raub (198 I) has
developed an "Attitudes Toward Computers" scale, a 25-item scale which,
after factor analysis, identiﬁed one factor Raub labeled "computer
anxiety. " Maurer's "Computer Anxiety Index" (1984) is a revised version
of an instrument called the "Educational Innovation Survey" developed
by Simonson and Rohner (1981). The original instrument was thought to
reﬂect "intent to use computer" more accurately than computer anxiety,
and Maurer sought to overcome the limitations of the original survey.
The resulting instrument was a 26-item survey with a reliability
coefﬁcient (Cronbach's alpha) of .94. The Computer Anxiety Scale
developed by Marcoulides (1985) is a 20-item inventory developed as a
measure of perceptions by students of their anxiety in different situations
related to computers. Exploratory and conﬁrmatory factor analysis
revealed two factors, one Marcoulides labeled a general computer anxiety
factor, the other labeled an equipment anxiety factor. Variables related
to the general factor or concerned with anxiety arising ﬁom direct
experiences with the computer, the role of computers in society, and
ﬁom the impact of the computer on individuals at work. The equipment
anxiety factor reﬂected anxiety over operation of a personal computer,
looking at computer printers and printouts, and watching someone use a

computer. The internal consistency coefﬁcient alpha for the test was .97.

20
The Raub instrument was intended to investigate and measure the

attitudes associated with computer anxiety, and thus was not intended
as a measure of computer anxiety itself. The Maurer instrument is a
commercially available product which costs approximately .50 per
administration. An instrument that is reliable, measures the
phenomenon of interest, and requires little or no ﬁnancial outlay is the
most desirable. The Computer Anxiety Scale developed by Marcoulides
met all the criteria. Permission was sought from and granted by Dr.
Marcoulides to use his instrument in this study.

Humanﬁerfonnancegonsideraﬁoas

It was previously noted that people do not perform for a variety of
reasons, and that skill deﬁciency is only one reason for performance
deﬁcits. Mager and Pipe (1970) have provided a useful framework for
analyzing the nature and causes of performance problems. They suggest
that performance problems can be caused by skill deﬁciency,
punishment for desired performance, reward for non-performance,
desired performance making no diﬁ‘erence, and the existence of obstacles
to desired performance. Within Mager and Pipe's framework, computer
anxiety may constitute an obstacle to performance insofar as computer
anxiety is an interior state which prevents the learner from attempting to
acquire computer skills. The negative feelings that accompany computer
anxiety can also be thought of as a negative consequence of desired
performance. A computer anxious person, when faced with the need to
acquire computer skills, will feel uncomfortable and will seek to avoid the

conditic
suggestI
andzuTJ
interfercl
man
perform '
approach

F (i
considerl
intervenI
problemI
Suggests
looking )
deﬁcﬁs,
€0ng on
Procedu;

T1]
the Prob I
that motl
indm‘duél

addressh

 

21
conditions that provoke the uncomfortable feelings. Mager and Pipe

suggest removing obstacles to performance when they are encountered,
and arranging for positive consequences when negative consequences
interfere with performance. While Mager and Pipe illustrate their ideas
with numerous examples, the issue of anxiety interfering with
performance is not directly addressed, and we need to look elsewhere for
approaches to dealing with computer anidety.

Following Cattell and Scheier (1961), computer anxiety can be
considered a transitory condition that may be alleviated via instructional
intervention. Rossett's approach to analyzing human performance
problems ( 1992) includes the category of employee motivation. She
suggests that human performance professionals, while traditionally
looking to phenomena external to the employee to explain performance
deﬁcits, should also consider the employee's internal state, and what is
going on within the employee as he or she contemplates a new system or
procedure.

The approach which underlies the present study, is to assume that
the problem of computer anxiety leads to learner motivation deﬁcits, and
that motivational deﬁcits which occur within computer anxious
individuals interfere with their desire to acquire computer skills. By
addressing the motivational needs of learners, one may be able to oﬁ‘er a
learning experience that will help the computer anxious learner to "set
aside" their anxiety sufﬁciently for computer skills learning to occur. An
analogy could be made to a person suffering from a common cold. While
medical science does not currently have much to offer by way of curing a
cold, it offer a pharmacopoeia of medicines aimed at symptomatic relief.

A worker can avail him or herself of this pharmacopoeia and feel well

enough
Assumi

then by

effects c

 

 

22
enough to continue working. The approach taken here is similar.

Assuming that a motivational deﬁcit is symptomatic of computer anxiety.
then by addressing the "symptom" one may be able to overcome the
effects of the "disease."

15.1 .on. L‘_191,on0_‘.__o_r'sﬁ‘ six 510012.10, «410-:

MW

Fortunately, the ﬁeld of instructional theory and design provides a
number of approaches to dealing with motivational issues, and may
provide a potential arsenal of treatments to address the motivational
needs of computer anxious learners.

Keller (1983a, 1983b, 1987) has developed a model of motivational
requirements which may be addressed in order for people to be motivated
to learn. His ARCS model synthesizes the work of numerous
psychological theorists into a framework for integrating strategies for
increasing learner motivation. According to the ARCS model, there are
four general motivational requirements: attention, relevance, conﬁdence,
and satisfaction. Keller brieﬂy discusses the aim of each requirement,
outlines requirement subcategories, and suggests strategies for meeting
each requirement:

Attention. The motivational concern is getting and sustaining
learners' attention. Subcategories of attention include perceptual
arousal: creating curiosity by using novel approaches and injecting
personal or emotional material: W: stimulating an attitude
of inquiry by asking questions, creating paradoxes, nurturing thinking

challen.

present

 

unexpet

R;
needs.j
stateme
present
persona
positive
analogie

Q

achiever

 

 

23
challenges: and mm: sustaining interest by variations in

presentation style, concrete analogies, human interest examples and
unexpected events.

m. The motivational concern is responding to perceived
needs. Subcategories include Wm: providing examples or
statements of the utility of the instruction, presenting or having students
present the goals of instruction: Wishing: providing learners with
personal achievement opportunities, leadership responsibilities, and
positive role models: and mm: providing concrete examples and
analogies related to the leamers' work.

Wm. The motivational concern is promoting

achievement and success, especially early on. Subcategories include
WHEELS: building a positive expectation for success by
establishing trust and by explaining the evaluative criteria; succcss
W: increasing belief in personal competence by providing
many varied and challenging experiences which increase learning
success: and 12W: tying learner success to their own efforts
by providing feedback on performance.

W. The motivational concern is creating

satisfaction so that there will be continued motivation to learn.
Subcategories include W: providing meaningful
opportunities for learners to use their new knowledge or skills by
providing problems, simulations, or work samples: 2951mm
ccnscgucnccs providing reinforcement of learner successes by using
verbal praise. real or symbolic rewards, and student presentation of their
work. and scum: assisting students in anchoring positive feelings about
their accomplishments (i.e., when comparing their work with others',

they rer
requirei

consistti

aCCOIIl-i}

areas oi

 

and self
attemptI
one mo<

T:
plan for
Particul
Student

f0Cusin .

 

 

24
they remain satisﬁed with their own work) by making performance

requirements consistent with stated expectations and providing
consistent measurement standards for all leamers' tasks and
accomplishments.

The ARCS model is based upon supporting studies from many
areas of research on human motivation, including reinforcement theory
and self-eﬁicacy theory. It is something of an "omnibus" model - it
attempts to synthesize a wide variety of approaches and perspectives into
one model. Its practical utility has been supported through ﬁeld-testing.

The ARCS model is intended as a foundation upon which one can
plan for meeting the motivational needs of any group of students. It is
particularly attractive in that it addresses the motivational needs of
students throughout the educational process, rather than narrowly
focusing upon how to gain students' interest.

By addressing the motivational concerns raised in the ARCS
model, one may provide enough "symptomatic relief" from their anxiety
states to allow computer anxious learners to acquire computer skills.
Ideas within the ARCS model can be applied to the challenge of designing
instruction for computer anxious learners. The attention concern can be
addressed by including analogies that relate to leamers' previous
experience. The relevance concern of perceived need can be addressed
by having learners work on tasks meaningful to them. The conﬁdence-
building concern of promoting achievement and success can be
addressed by providing opportunities for successful task completion.
The satisfaction- generation concern can be addressed by including

meaningful tasks on which leamers can work.

'I
framew
meets t
many it
anxious
skills.

\\
that is i
for guid
0f knov.
provide.
in the a

C
to gUidel
Practica
Series 0.
that inv

to 163m
Calls the

 

idem-f“
attajmn
betwen
they req
attaining
being. L
”Hangm-

a SenSe (

 

 

25
The ARCS model of motivation in instruction provides an omnibus

framework for approaching the challenge of making instruction which
meets the motivation needs of learners. Its application can result in
many ideas for designing instruction which may alleviate computer
anxious leamers' fears sufficiently so that they can acquire computer
skills.

While the ARCS model is a framework for designing instruction
that is motivating to learners, it is "content-free" in that it is not intended
for guiding the development of instruction in any particular body or type
of knowledge. The work of John Carroll (1984a, 1984b, 1990, 1992)
provides a set of principles intended explicitly for developing instruction
in the area of computer skills.

Carroll has suggested a set of what he calls "Minimalist Principles"
to guide instructional developers facing the task of creating instruction in
practical skills of computer use. The principles were developed out of a
series of qualitative research studies conducted by Carroll and associates
that investigated the behavior of adult computer learners while they tried
to learn how to use their machines. He begins by describing what he
calls the "training problem." Learners of computer skills begin by
identifying appropriate goals such as printing a document, the means for
attaining the goals (printer, keyboard, monitor), and drawing connections
between goals and means. Training materials provide assistance, but
they require analysis by learners as to which parts are relevant to
attaining their goals, and which parts could be ignored for the time
being. Learners can end up learning things and doing things that are
unanticipated or unintended by the trainmg designers. This can lead to
a sense of W on the part of the learner - Epstein's

notion I.
effort tc
order tt’
syrnptol
by new:

A
Carroll
overwh.
intende
interest
instmer
plunge
them 91.

8
t0 Carrt
goals ar
They 5k
[1984b,

that intt

"jg
at

 

26
notion of primary overstimulation is relevant in this context. Thus, in the

effort to divine out of a mass of information what they need to learn in
order to accomplish personal goals, learners can begin to exhibit
symptoms of anxiety as deﬁned by Epstein, a sense of being overwhelmed
by new information and technology.

Another facet of the training problem is called W.
Carroll notes that training designs often respond to the problem of
overwhelming learners by using overviews in the front of their manuals
intended to help orient learners to the system. Oﬁen learners aren't
interested in reading, they are interested in aging. Rather than obey the
instruction, "Don't do anything until you‘ve read this ﬁrst," learners will
plunge ahead at the ﬁrst mention of a procedure that they feel will move
them closer to a personal goal, frequently with troubling eﬁ‘ects.

Skipping is another aspect of the instructional problem. According
to Carroll, "People come to the learning task with a personal agenda of
goals and concerns that can structure their use of training materials.
They skip crucial material if it doesn't address their present concerns"
(1984b, p. 125). Thus, learners will browse ahead until they ﬁnd a t0pic
that interests them, ignoring prerequisites.

Other aspects of the training problem include:

"reasoning instead of reading" - drawing defective conclusions
based upon incomplete or incorrect understandings.

"ignoring the screen" - "keeping one's nose in the book" while
attempting procedures, rather than attempting a step, then looking

how lea

comput

goals f0}

 

Accol-dh
probIEm
StratEgn

embodie

I’Eaj life
hater if

27
at the screen to determine whether it has been effectively carried

out.

"trouble recovering from errors" - Typically, instruction assumes
that learners will not make errors. Learners are expected to carry
out procedures exactly as they are written. The reality is that
learners do make errors. The problem is that most instruction

makes no adequate allowances for this fact.

"wanting to do real wor " - people learn to use word processing to
create documents, immediately: they are not interested in rote

learning of procedures.

As a result of these studies, Carroll described his observations of
how learners actually learned. He observed that people learned to use
computers by exploring them. by trial and error. by setting personal
goals for themselves, and by actively manipulating the machine.
According to Carroll (1 990), the appropriate orientation to the training
problem is directing and supporting the natural learning styles and
strategies of the learners. This is the essence of Minimalist Design and is
embodied in the following principles:

W: Learners will learn better if they work on

real life tasks. For example, secretaries will learn to use computers
better if they begin by learning to create common ofﬁce/ organization

documents such as business letters.

compu
instruc
materizi
become
to start

introdu

E
elabora
in mem
pose ar
inStruc
appropi

I30

Order a g
maten’a

needed.

they ijl
fTOm thf

 

28
W: Learners will try to do things with the

computer as soon as they feel they can: they will not “wait for the
instruction." Most computer manuals provide far too much introductory
material. Learners will get ahead of the manual, make mistakes, and
become frustrated. Effective instructional materials will allow students
to start as quickly as they want to by providing minimum amount of
introductory material.

WW: When learners actively generate

elaborations of learned material, this makes the material more accessible
in memory. Instructional materials can explicitly encourage learners to
pose and then investigate questions that interest them. A key role of
instruction is to guide learners to pose productive questions and to adopt
appropriate methods to investigate them

W: Learners tasks typically will not follow the
order as prescribed in a computer manual, therefore the instructional

materials need to allow the learner to access needed information as

needed.

WW: Successful training Systems
incorporate linkages between the instructional materials and the action

on the screen. It is not enough to expect people to follow directions alone:
they must be able to coordinate directions with the feedback they receive
from the system. .

f

a
E

approa.

[Carroll

1.6811161
compuf

must it

instruc

useful i

If"!

training
motivat
desktop
example

 

design 0

actimy .

29
WW: Conventional
instructional materials designed according to traditional systems

approaches assume error-less performance on the part of the learner
(Carroll, 1 990). According to Carroll, this is an untenable position.
Learners, especially learners of difficult new technologies such as
computers, can be expected to make mistakes. Effective instruction
must include plans for this contingency. Carroll suggests an
instructional module that directly addresses the most common and

useful error recovery strategies.

WW: One of the most important aspects of
training design is understanding the user's prior knowledge and

motivation and then ﬁnding ways to exploit it. The typewriter and
desktop metaphors have been used successfully in interface design, for

example.

W: The learning situation is full of meaning and
experiences. the chance for discovery, and the possibility of achievement.
The ﬁne details of a real situation such as the contents of a memo to be
typed, or an individual's preferences and choices in a learning situation,
can guide the development of instruction.

The key to the Minimalist Principles is minimizing the extent to
which instructional materials obstruct learning and of refocusing the
design of training materials on the goal of supporting learner-directed
activity and accomplishment. According to Carroll (1990):

30
Presenting real tasks that learners already understand and

are motivated to work on, helping them to get started rapidly
on these tasks, allowing them to rely on their own reasoning
and improvising, reducing the instructional verbiage they
must passively read, organizing material to support skipping
around and to facilitate the coordination of attention
between the training and the system, and addressing
important user errors can produce better training material
than the current state of the art. (p. 183)

There are many ways in which the Minimalist Principles dovetail
with Keller's ARCS model of motivation in instruction. Carroll's notion of
"presenting real tasks that learners already understand and are
motivated to work on" is directly related to Keller's motivational concerns
of relevance and satisfaction generation. The Minimalist Principle of
"getting started fast" can be very conﬁdence-building by providing early
opportunities for success, again addressing one of Keller' s motivation
concems. Keller's notion of personal control is supported by Carroll’s
principle of reading in any order: learners can control their access to
required information at any point in the learning experience. It is
apparent that instruction designed according to Minimalist Principles will
also in large part address the motivational concerns outlined by Keller.

Carroll (1990) reports that work has been done which conﬁrms the
efficacy of the Minimalist Principles. Olfman (Olfman, 1987 ; Olfman and
Bostrom, 1988: Olfman and Bostrom in Carroll, 1990) designed a
manual for teaching spreadsheets according to minimalist principles and
found that students who were trained with the "minimalist manual"

31
tended to use the spreadsheet longer than those in the control group.

Participants evaluated the minimalist manual as better facilitating their
learning. Participants in a study by Oatley, Meldrum, and Draper (in
Carroll, 1990) performed twice as efﬁciently in both leaming and testing
in skills in glossary preparation with a word processor compared with a
control group using a commercially developed systems-style manual.
Warner (l 987) at Michigan State University developed minimalist
instruction in the Microsoft Disk Operating system and examined brevity
and coordination of the system and training. He found that both factors
enhanced achievement and learner attitudes and decreased learning
time. Although he found no difference with respect to computer anxiety
scores for the experimental and control groups, this could attributable to
the fact that he used an instrument intended to measure general state
anxiety (Spielberger's State Anxiety Inventory) rather than an instrument
designed speciﬁcally for measuring computer anxiety.

mndence is accumulating supporting the use of Minimalist
Principles in designing computer skills instruction. The proposed study
seeks to contribute to this literature by applying Minimalist Principles to
the problem of computer anxiety.

32
Simmer:

This chapter presented a review of the literature on anxiety.
computer anxiety and its measurement, human performance
considerations, and the Minimalist Principles. The literature on anxiety
identiﬁed two types, "trait" and "state" anxiety. Trait anxiety is a relatively
enduring personality characteristic, while state anxiety can ﬂuctuate over
time and treatment. The literature on computer anxiety and its
measurement deﬁned computer anxiety as fear or apprehension when an
individual uses a computer or thinks about using a computer. Many
researchers consider computer aanety as a state rather than a trait. and
thus possible amenable to treatment. Several instruments used to
measure computer anxiety have been developed: the present study
adopted Marcoulides' Computer Anxiety Scale.

Several reasons for why individuals don't perform were noted in the
literature on human performance: skill deﬁciency, punishment for
desired performance, reward for nonperforrnance, desired performance
making no difference, and the existence of obstacles to desired
performance. It was suggested that computer anxiety could be conceived
of as an obstacle to performance in this framework, and the negative
feelings associated with computer anxiety operating as punishment for
desire performance. Rossett suggested the need to consider lack of
employee motivation as a phenomenon explaining some performance
deﬁcits. The approach taken by this study was that motivational deﬁcits
can be expected to occur within the computer anxious individual which
interfere with that individual's ability and desire to acquire computer
skills. It was suggested that, by dealing with the motivational deﬁcit

33
"symptom" of computer anxiety, one might be able to mitigate its

deleterious effects. Keller's ARCS model suggested four motivational
requirements — attention, relevance, conﬁdence, and success - which
need to be attended to in order to design instruction that is motivating to
learners. Carroll's Minimdist Principles such as "presenting real tasks,"
"getting started fast." and "reading in any order" were shown to dovetail
with Keller's motivational requirements. Research on application of
Carroll's Minimalist Principles provide evidence for their efﬁcacy in
training individuals to use computers. This study is intended to
contribute to that effort.

CHAPTER THREE

METHODOLOGY

Introduction

This chapter presents the methodology used in the study.
Separate sections present the study design, design of the instructional
treatment, instrumentation, description of the subjects and subject
recruitment, experimental procedures, data handling, data scoring, data

analysis, and a power analysis.

Desimattheﬁtudx

Campbell and Stanley (1963) suggest that randomization of
experimental and control groups is the most adequate assurance of lack
of initial biases, controlling for the effects of testing, history, maturation,
regression, and selection by virtue of assuring group equality at the time
of assignment. The study was a pre/posttest control group design with
respect to computer anxiety and motivation measures, and a posttest
only control group design with respect to performance at a word
processing task . There was random assignment of control and
experimental groups. One extension suggested by Campbell and Stanley
is to include "testing for effects extended in time" (p. 31) to see if there
are any enduring treatment effects. The experiment was performed two

34

35
times with the same subjects. Phase I and Phase II, the second

administration separated from the ﬁrst by one month.
The pre/posttest design with random assignment of subjects can
be depicted graphically in the following manner:

01 X1 02 03 X1 04

On X2 02 03 X2 04

The posttest only design with random assigiment of subjects can
be depicted gaphically in the following manner:

X1 02 X1 04
R
X2 02 X2 04
W

The study is trial of a set of instructional materials. Thus a
thorough discussion of the treatment is warranted.

Both the experimental and the control treatments are instructional
materials designed to teach basic skills in the use of the WordPerfect©
word processing software, version 5. l (Appendices A and B).

The experimental instruction was developed to incorporate ﬁve of
Carroll's Minimalist Principles: slash the verbiage, reading in any order.

36
coordinating system and training, getting started fast, exploiting prior

knowledge, and supporting error recognition and recovery. It was felt that
these principles were under the control of the instructor and were thus
amenable to experimental marnipulation. Training on real tasks was
incorporated into the word processing task set for both the experimental
and the control goups. Economy of design of the study dictated that an
identical word processing task be used to gauge the effectiveness of the
instructional materials.

The experimental treatment was intended as a self-instructional
module. It would be similar to the "let's get started" types of sections
accompanying commercial word processing packages or of an
introductory tutorial in a course on basic word processing skills. An
expanded version of this instruction has been used as tutorial material
for a progam of basic computer skills instruction for academic
physicians participating in the Primary Care Faculty Development
Fellowship Progam at Michigan State University. It was decided to
investigate a variation of the "let's get started" format because it is
assumed that many learners of computer packages get their instruction
in this manner.

The control materials are the "Getting Started" section of the
WordPerfect reference manual (1989), pages 3 - 19 (Appendix A).

The following discussion compares the experimental materials with
the control materials with respect to the Minimalist Principles.

W: The experimental materials contain 10 pages
of text, while the control materials contain 17 pages of text. Several of the

experimental material pages contain very little text. For example, the

37
section on "deleting text" has four lines of text, one line composed of the

word "or." The section in the control materials which contains
information on deleting text contains 2 1 lines of text, and the key
information regarding deleting text is irnbedded within text explaining
other "Keys to Know." Overall, the experimental materials contain 1309
words in 357 lines of text, while the control materials contain 4,279
words in 519 lines. The principle of slashing the verbiage is manifest
throughout the experimental material by the relative brevity of the
explanations.

W: The experimental material incorporates a

set of index tabs that allow the user relative "random access" to the
desired section. The use of the index tabs is a variation of a technique
used by Carroll (1990) in which a set of tabbed instructional cards
replaced the more common instructional manual. The tabs should allow
the user to access a desired topic of instruction with a minimum of
searching. The control material provides a traditional table of contents.

We: The experimental material

includes several examples of screen information that would appear upon
correct application of the instructions. An example is on the ﬁrst page.
which contains a mock-up of a blank screen in WordPerfect ready for text
entry. This is the screen the learner should see if they have correctly
followed the procedure to start up the progam. Another example is in
the section on "Exiting WordPerfect," where the screen prompts that
appear during the process of exiting the progam are represented in the
material:

38
2. This message will appear at the bottom of the screen:

Save document? Ies (No)

The purpose of such prompts is to allow the learner to coordinate
their act of reading through the instruction with the action of the system
itself.

W: The experimental instruction dispenses with
customary introductory "read this ﬁrst" types of material. The learner is

instructed in the use of the word processing system beginning with the
very ﬁrst page - how to start up the word processing software.
Explanatory material not directly related to use of the system is kept to a
minimum or dispensed with entirely.

WW: References are made to the typewriter
and the process of typing as a means of relating text entry to readily.

available learner concepts. This is evident in the section "I‘yping Text.
Moving Around" in which the metaphor of typing is exploited to highlight
the differences between the act of typing, which requires a carriage
return after every line, and the act of text entry and the concept of "text
wrapping," which is explained without reference to the term itself.

W: A section on "Undoing"

is included in the experimental materials to allow the learner quick

reference to two methods of error recovery, using the "'cancel" command

39
and the "escape" key. The control materials include only the "cancel"

command, and irnbed instructions for using the command within a larger
section on "Keys to Know."

W

The coordinator of the Human Subjects Pool of the Department of
Psychology at Michigan State University was contacted for recruitment of
study subjects. The Human Subjects Pool is comprised of undergaduate
introductory psychology students who, in order to fulﬁll course
requirements, agee to participate in psychological research. Students in
the pool may choose the experiments in which they participate. Subject
recruitment procedures allow the researcher to specify desired subject
attributes. For this experiment, it was speciﬁed that subjects should
meet the following criteria: 1) no previous experience using WordPerfect
for IBM and compatible computers, and 2) consider themselves afraid of
computers. Standard departmental sign-up procedures for recruiting
subjects from the pool were followed which involved creating and posting
sign-up sheets with the title of the experiment, the time comrrnitrnent
required, the desired subject attributes, and the location and time to
report for a brief survey. Subjects had a choice of three different times at
which the could report for the screening test. A total of 200 students
sigred up for the study, of which 173 reported for the initial
administration of the Computer Anxiety Survey.

Instanmentation

Computer anxiety was measured using Marcoulides' Computer
Anxiety Scale (Appendix C), a 20-item questionnaire using a 5-point
Likert scale for each question. The internal consistency of the test as
measured by Cronbach's alpha is .97 . The Computer Anxiety Scale was
adrrrirnistered immediately prior to the treatment and immediately after
the treatment, during both Phase I and Phase II of the study.

In order to gather demogaphic data and to determine
pretreatment desire to learn computer skills, the Backgound
Information form (Appendix D) was administered to all subjects. The form
includes questions about subj ects' class standing. age. sex. GPA.
previous computer experience, previous word processing experience.
typing ability, academic major, and three questions regarding subjects'
desire to learn computer skills. The full Backgound Information
questionnaire was administered to all Phase I subjects. An instrument
with only the three questions pertaining to subject desire to learn
computer skills was administered during Phase II (Appendix E).

In order to gather data on the desire to learn computer skills after
experimental and control treatments, a Feedback questionnaire
(Appendix F) containing the same three questions about subject desire to
learn computer skills which appeared in the Background Information
form was adnninistered after treatments in both Phase I and Phase II of
the study. The Feedback questionnaire also contained two questions that
are intended to obtain qualitative data regarding subjects' perceptions of
the instruction.

41
Experimentalmcedurss

Upon acceptance of the doctoral dissertation committee and
successful review by the University Committee on Research Involving
Human Subjects, the researcher conducted four pilot tests of the
experiment, each test with one subject. The results of the pilot test were
used to reﬁne the experimental procedures.

Prior to subject recruitment, a schedule of computer laboratory
times was arranged with the Microlab Services Coordinator (Appendix G).
The schedule was arranged so that there would be a total of four weeks
of data collection divided into two phases of two weeks duration each.
Data collection extended over an eight-week period, each week of data
collection followed by a week without data collection. During each week
of data collection, there were ﬁve data collection periods scheduled.
Recalling that the experiment was performed in two phases, a total of ten
goups, ﬁve experimental and ﬁve control, were studied. The numbers of
afternoon/ evening sessions versus morning sessions were balanced so
that there was an even distribution of experimental and control goups
over the different time slots.

After subjects signed up for the study, the Computer Anxiety
Survey was administered on three separate occasions, at which times
informed consent to participate in the study was be obtained (Appendix
H). Those students who signed up and attended one of the CAS
administrations were allowed to participate in Phase I and Phase II of the
study.

Two initial goups of n=86 subjects each were formed via random
assignment. This was accomplished by writing the name of each subject

on

CD

83'

3P
wit

de:

sul

ati

C01

“'6.

cm

42
on a slip of paper, putting all the slips of paper into a large manila

envelope, mixing the slips of paper 5 mirnutes to scramble tlnem, then
withdrawing the slips of paper and assigning subjects alternatively to
each goup.

Subjects were contacted by phone to arrange the time at which
they would participate in the study. One problem that became quickly
apparent was that the time periods for Groups 1 , 3, 6, and 8 conﬂicted
with the class schedules for many of the students. This problem
decreased the number of subjects participating in the study. The initial
experiment goup was comprised of 56 subjects, the control goup of 53
subjects.

Subjects reported at their assigned times to the microcomputer lab
at Room 12, Olds Hall, an MSU facility containing 22 DHK IBM-
compatible microcomputers. Subjects were seated one to a computer,
were each given a manila envelope containing study materials, and were
instructed to wait until the entire goup began. Each manila envelope
contained the following materials:

- Computer Anxiety Scale and two machine-scored answer forms

(Appendices C and I)

- Background Information form (Appendix D)

- Directions for Students (Appendix J)

- Time Sheet (Appendix K)

43
- Word Processing Tasks 1 and 2 (Appendices L and M)

- Either control or experimental Instructional Materials booklet
(Appendices A and B)

Subjects were ﬁrst instructed to open their envelopes and remove
the Computer Anxiety Scale and one machine scored answer form. After
ﬁlling in the answer form for their names, student numbers, and a
section number indicating the order (pre- or posttest, Phase I or Phase II)
of the CAS administration, subjects were asked to complete all twenty
items of the CAS. Subjects then ﬁlled out the Backgound Information
form.

Subjects then removed the Directions for Students, Time Sheet,
Word Processing Task 1 sheet, the Instructional Materials booklet from
the envelope. The researcher read the Directions for Students aloud and
asked subjects if there were any questions. Subjects recorded a starting
time announced by the researcher on the Time Sheet. Subjects then
attempted to complete Word Processing Task 1. Upon completion of
Word Processing Task 1 , each subject recorded the time of completion on
the Time Sheet and raised their hand as a signal to the researcher. The
researcher saved the output to disk, using a ﬁle naming scheme intended
to insure that each task would have a unique identiﬁer for each subject
(Appendix N).

Subjects were then instructed to remove Word Processing Task 2
from their envelopes, were told to record a second starting time on their
Time Sheets, and were further told to complete Word Processing Task 2
and to record their ﬁnishing time, much as they had with Word

44
Processing Task 1 . Again, subjects raised their hands after completing

Word Processing Task 2, and the researcher save their work to disk.

As subjects completed Word Processing Task 2, they were
instructed to complete the Feedback form and to retake the CAS.
Subjects were then allowed to leave the microcomputer lab.

The protocol just described was followed for Phase II of the study.
the only difference was the inclusion of only one Word Processing Task,
rather than two (Appendix 0). This task was similar irn form but different
in content from the Word Processing Tasks l and 2. The Directions for
Students (Appendix P) and Time Sheet (Appendix 9) were altered to
reﬂect this difference.

W

The ﬁrst task after data collection was to check to make sure that
all materials had been returned and all instruments had been completely
ﬁlled in. Disk ﬁles were checked against the appropriate Word
Processing Task Checklist (Appendices R. S, and T) to determine the
number of criteria met on each task. Scores for the Computer Anxiety
Scales and the Word Processing Checklists, as well as any data from the
Backgound Information form and the Feedback form, were entered into

a Statview SE + Graphics (Macintosh version) spreadsheet for analysis.

45
W

The Computer Anxiety Scale was machine scored by staff at the
Computer Center at Michigan State University. Scores could range from
20 - 100 points (1 - 5 points per question for 20 questions). Each Word
Processing Checklist consisted of a total of 9 items. Time to perform the
word processing tasks was measured in minutes and decimal equivalents
of minutes. The backgound and feedback questions pertaining to
desired time to learn computer skills and word processing were scored
using 5—point scales ranging from 1="none" to 5="> ﬁve hours." The
backgound and feedback question about interest in learning computer
skills was scored on a 4-point scale ranging from 1="not at all" to 4="a

lot."

QataAaalxsis

Data analysis in support of hypothesis testing focused on
comparing the experimental and control goups for differences in
Computer Anxiety Scores, number of Word Processing Checklist criteria
met, and differences in reported measures of motivation to learn
computer skills.

Repeated measures AN OVA is used to test for statistical
Signiﬁcance of mean difference scores for multiple outcome measures in

a one factor experimental design (Glass and Hopkins, 1984). This
Statistical procedure maintains the alpha-level, or probability of rejecting
a true null hypothesis of no difference between goup means (type I), at

46
the desired level. F-tests of mean difference scores were performed to

test for signiﬁcant differences between the experimental and control
goups with respect to computer anrdety.

Performance on the word processing tasks was compared using
repeated measures ANOVAs of mean goup checklist scores to detect any
signiﬁcant difference in performance on the word processing task.
Alpha-level for all tests of statistical signiﬁcance will be .05, which is
conventional for behavioral research (Cohen, 197 7).

Further analysis of the effect of the two treatments on computer
anxiety scores and on performance scores was undertaken by dividing
the experimental and control goups into high and low computer anxious
subgoups, and then testing to see if there are any differences
attributable to treatment for the subgoups. The splitting of the goups
was based upon Phase I pretest CAS scores, the upper half of CAS scores
being the "high anxiety" goup, and the lower half of CAS scores being
the "low anxiety" goup.

The Backgound and Feedback Questionnaires contained 3
questions that were intended as estimates of subjects' desire to learn
how to use computers. Assuming that the underlying variable - intention
to invest time in an activity - was continuous, ANOVA was chosen as the
analytical technique (Borg and Gall, 1983).

A secondary data analysis using the ANCOVA procedure was
performed to deterrrnine the effect of confounding variables on the
outcome CAS measures.

Data analysis was performed on a Macintosh Not using Statview SE
+ Graphics and Super ANOVA software. Scoring of subjects' word
processing tasks was performed using an IBM AT computer with

47
WordPerfect version 5. l , the same software used by subjects during the

study.

Emmnalxsis

According to Glass and Hopkins(1984), the "power" of a statistical
test is the probability of rejecting a false null hypothesis. The geater the
power of a test. the less likely one is to commit a type 11 error. or
retaining a null hypothesis when it is actually false. Ideally, a test would
have sufﬁcient power to minirrnize the probability of type II error.

One can increase the power of a test of a hypothesis testing
procedure by increasing sample size, increasing the alpha-level, and by
maximizing the effect size of the experimental treatment. Glass and
Hopkins note that increasing sample size and increasing the alpha level
of a test are within the control of the researcher. They advise taking the
largest sample that is practical, and then determine if the sample size
has suﬂ‘icient power for detecting a difference large enough to be of
interest. Cohen (1977) suggests a minimum power value of .80, (a beta-
level of .20) on the assumptions that a) researchers most commornly
specify an alpha level of .05, and b) the relative seriousness of a type I
error (i.e., rejecting a true null hypothesis) is four times as gave as that
of a type II error. Cohen assumes that behavioral scientists are more
concerned with type 1 than type II error: it is better to risk "failure to ﬁnd"
than ""ﬁnding something that is not there" (pg. 56). This concern is
reﬂected in the .05/ .20, alpha/beta level speciﬁcations.

F’

de

to

pl’

at

for

wl

48
The most common method of determining the power of a statistical

test such as an F-test is to calculate a "centrality parameter," p, and use
this parameter to locate the associated power of the test for a given alpha
level in the appropriate Pearson-Hartley power chart (Glass and Hopkins,
1984). Subkoviak and Levirn (197 7) note that the power of a test is
affected by the reliability of the instruments used to measure the
dependent variables: measurement error will decrease the power of a test
to reject a false null hypothesis. Subkoviak and Levin suggest a
procedure for adjusting the calculation of the centrality parameter, gven
an estimate of the reliability of a dependent measure, which will result in
a more realistic power estimate. They suggest usirng the followirng
formula:

 

 

¢ _ P» 71 V3
K
(v + 1) 2 a2
' i=1
where:
4: = Pearson-Hartley centrality parameter
p,,' = estimate of within-treatment reliability
n = number of subjects per treatment goup
We = linear contrast expressed in Milan-treatment standard
deviation units, or "effect size" (Subkoviak and Levin,
1977: Cohen, 1977)
(v + 1) = number of treatment goups
and,
K
2 of = sum of squares of the contrast coefﬁcients

49
as a means for correcting the power estimate to account for the imperfect

reliability of measurement of the dependent variable.

Previous studies using the CAS have estimated the test-retest
reliability of the instrument at .77. This study had approximately 50
subjects per goup, with two treatment goups. For comparisons of two
goups, the linear contrast coeﬂicients are " 1" and "- 1" (Subkoviak and
Levin, 1977). The desired effect size is .80 of a standard deviation.

Summarizing the values for the above variables:

p,,' = .78
n = 50
We = 80
(v + l) = 2
and,
K
2 a2 = (1)2 + (-1)2 = 2
k=l

and substituting these values into the Subkoviak and Levin equation:

 

(.77)(50)(.30)2

(2X2) = 2'48

 

one obtains a centrality parameter of 2.48. Locating this parameter on

the Pearson-Hartley power charts for alpha = .05, the power of an F-test
or t-test to detect a .80 standard deviation difference iii the means of two
groups is approximately .93. This compares favorably with assertions by

50
Cohen (1977) and Brewer (1972) of a minimum level of .80 as sufﬁcient

power to reject a false null hypothesis.

It is interesting to note how much power is lost when a smaller
effect size is posited. Assuming an effect size of .50 standard deviation
rather than .80, and again substituting the values in the Subkoviak and
Levin equation:

 

(.77)(50)(.50)2
(2X2)

 

= 1.55

a centrality parameter of 1.55 is obtained. Referring to the Pearson and
Hartley power charts, and again assuming an alpha-level of .05, this
value for the centrality parameter yields a power of only .49.

The power analysis reveals that this experiment was able to
reliably detect differences of .80 standard deviations between goup
means if they indeed existed.

514m

The study was a pretest and posttest control goup design with
random assignment of subjects with respect to measures of computer
anxiety and desire to learn computer skills, and a posttest control goup
design with random assignment of subjects with respect to the measure
of computer performance. The experiment was repeated to see of there
are any effects which endure over time.

The experimental treatment was word processing instruction for
teaching the WordPerfect application designed according to Carroll's
Minimalist Principles. The control treatment was the "Getting Started"

51
section from the WordPerfect documentation. Computer anxiety was

measured Marcoulides' Computer Anxiety Scale, desire to learn
computers was measured using three questions on Backgound
Information and Feedback forms, and performance of a word processing
task was measured using a task checklist.

Subjects were recruited from the Human Subjects Pool of the
Department of Psychology at Michigan State University. Desired
attributes included on subject sigr-up sheets for the study were no
previous experience using WordPerfect, and fear of using computers.
After testing subjects using the CAS, subjects reported approximately
two weeks later to a rrnicrocomputer laboratory for the study proper.

Data analysis involved analysis of descriptive statistics,
sigriﬁcance testing of the hypotheses using one-factor and repeated
measures ANOVA, and secondary analysis of the data to determine the
presence and possible effect of confounding variables using ANCOVA.

A power analysis revealed that, given an n of 50 subjects, an eﬁ'ect
size of .80 standard deviations would yield a power of .93 to detect a false
null hypothesis. Thus, there was sufﬁcient power to detect signiﬁcant
differences between control and experimental goups if they indeed
existed.

CHAPTER FOUR
RESULTS

Inimiuciion

In this chapter the results of the data analysis are presented. First
the sample will be described. This will be followed by an analysis of the
data to test the hypotheses. The chapter will conclude with an analysis
to check for the possible presence and effects of confounding variables.

WW

Of the original n= 109 goup, four subjects were removed from the
study due to incomplete data sets. T‘wo subjects from the control goup
were missing complete sets of CAS scores, and two subjectsfrom the
experimental goup were missing the word processing task document
ﬁles. The resulting goup consisted of n=105 subjects, with a control
goup of n=51 subjects. or 48.6% of the total, and an experimental goup
of n=54 subjects, or 51.4 of the total.

Table 1 presents the means and standard deviations of the age and
grade point average (GPA) of the control and experimental goups. The
mean age for the experimental and control goups was 18.55 years, and
the mean GPA for both goups was 2.68. An F-test comparison of goup
mean differences showed that they were not signiﬁcantly different at the

52

53
.05 level with respect to age (mean difference=-.083, chge=.28l, but

were signiﬁcantly different at the .05 level with respect to GPA (mean
diﬁ'erence=-.3 19, Fschca'e'JG. 157). GPA was not found to correlate
sigiiﬁcantly with the Phase I pretest CAS score (r=-.1 14), thus it was not
considered as a potential confounding variable.

 

 

Means and Standard Denrsgclnfns} of Subjects' Age and GPA
Control Experimental Totals
mean sd mean sd mean sd
Age 18.51 .903 18.59 .687 18.55 .796
GPA 2.52 .685 2.84 .627 2.68 .673

Table 2 presents the distribution of gender for the two goups.
Subjects were female by a signiﬁcant majority (Chi-square=12.34,
p=.0004). A one factor AN OVA revealed that females had signiﬁcantly
higher Phase I pretest CAS scores than did males (fable 3). Other
comparisons were performed to see if there were any other signiﬁcant
diﬁ'erences between gender goups. Speciﬁcally, males and females in
the sample were compared with respect to their GPA. current word
processirng skills, and number of semesters or terms of previous
computer courses. A one-factor ANOVA revealed no sigiiﬁcant
differences between male and female subjects with respect to GPA
(Fs¢.m=.049, p=.826), nor with respect to current word processing skills
(Chi-square=6.739, p=. 1503). Males and females differed signiﬁcantly

with respect to the number of semesters or terms of previous computer

54
courses (Chi-square=14.789, p=.01 13). Table 4 presents the number of

previous terms or semesters of computer study by gender. While there
are differences between males and females according to the contingency
table analysis, it is difﬁcult to perceive any pattern to these differences,
thus, the number of previous semesters or terms of computer study will

not be considered as a potential confounding variable related to gender.

Table 2
Counts and Percentages of Gender of Subjects

 

Control Experimental Totals
count percent count percent count percent

 

 

female 37 72.5 34 62.9 7 I 67.6

male 14 27.5 20 37.1 34 32.4
Table 3

AN OVA Table: Comparison of Phase I CAS Pretest Scores Based Upon

Gender

Source: df: SS: MS: F-test: p:

Between 1 l 1 18.016 1 1 18.016 5.108 .0259

Groups

Within 103 22544.974 218.883

Groups

Total 104 23662.99

55

 

Table 4
Number of Semesters/Terms of Subjects' Previous Computer Study
by Gender
Female Male Totals

count percent count percent count percent

none 12 16.9 3 8.82 15 14.3
< one 3 4.23 4 I 1.76 7 6.7
one 35 49.3 14 41.18 49 46.7
two 17 23.94 4 1 1.76 21 20
three 4 5.63 6 17.65 10 9.5
four 0 0.0 0 0.0 0 0.0
> four 0 0.0 3 5.6 3 2.8

Table 5 describes subjects' academic majors. The most prevalent
major was some form of social science. This category included
psychology, communications, and business administration. This category
was followed by engineering and "hard" science, including premedical
matriculation. The "not decided" category accounted for almost one-third
of the sample. In Table 5, within-goup percentages reﬂect within-goup
subtotals, and the percentages in the 'T‘otals" column reﬂect subtotals
for all subjects. A Chi-square analysis found the experimental and.
control goups to be signiﬁcantly different with respect to academic major
(Chi-square=10.372, p=.0346).

56

Table 5
Counts and Percentages of Academic Majors of Subjects

 

Control Experimental Totals
count percent count percent count percent

social 19 37.3 16 29.6 35 33.3
science

engineer/ 8 1 5.7 1 6 29.6 24 22.9
science

education 9 17.6 1 1.9 10 9.5
liberal 2 3.9 2 3.7 4 3.8
arts

not 13 25.5 19 35.2 32 30.5
decided

A review of Table 5 reveals that two categories of academic major,
education and liberal arts, had very few subjects, thus, it would be
unreasonable to use the ﬁve categories of acaderrnic major to determine
whether there were any signiﬁcant differences between the experimental
and control goups on the Phase I CAS pretest that were attributable to
academic major. In order to determine whether there were any
signiﬁcant differences between the experimental and control goups on
the Phase I CAS pretest that were attributable to academic major, the ﬁve
categories of this variable were therefore recoded into two levels. This was
done so that there would be sufﬁcient numbers of subjects irn each
category to provide for a reasonable analysis. The categories "social

57
science" and "engineer/science" were combined into one category of

academic major, and the categories "liberal arts," "education," and
"undecided" were combined into the other category. It should be noted
that subjects who listed "education" as their academic major indicated
that they were in the liberal arts domain. Table 6 presents the results of
a one-way ANOVA comparing subjects' Phase I CAS pretest scores based
upon the recoded academic majors. There was no signiﬁcant difference in

the scores based upon recoded academic major.

 

 

 

Table 6
ANOVA Table: Comparison of Phase I CAS Pretest Scores Based Upon
Recoded Academic Major
Source: df: S—S: MS: FL-test: p:
Between 1 2098.356 209.356 .919 .3399
Groups
Within 103 23453.634 227.705
Groups
Total 104 23662.99

Table 7 summarizes the data for subjects' class year. Over 80% of
the total goup reported being in their ﬁrst year of undergaduate
studies. A contingency table analysis revealed no signiﬁcant differences
between the experimental and control goups with respect to class
standing (Chi-square=3.49, p=. I747).

.J’

 

ﬁrs

SEC

58

Table 7
Counts and Percentages of Class Year of Subjects

 

Control Experimental Totals
count percent count percent count percent

ﬁrst 44 86.3 41 75.9 85 81
second 4 7.8 1 1 20.4 15 14.3
third 3 5.9 2 3.7 5 4.7

Table 8 is a summary of the number of previous semesters or
terms subjects had undertaken computer-related study. The question
for this item asked subjects to record whether or not they had had any
kind of course related to computers, including computer literacy, usage.
and progamming, and including any high school work. TWO-thirds
(66.7%) reported having taken one or two semesters or terms of previous
computer work. Almost 15% of the total sample reported having had no
previous computer instruction. Contingency table analysis revealed no
signiﬁcant differences between experimental and control goup with
respect to the number of previous semesters or terms subjects had

undertaken computer-related study (Chi-square=3.892, p=.5651).

 

Nun

none

(011

One

YWC

for

>1

59

Table 8
Number of Semesters/Terms of Subjects' Previous Computer Study

 

Control Experimental Totals
count percent count percent count percent

none 9 17.6 6 1 1.1 15 14.3
< one 3 5.9 4 7.4 7 6.7
one 23 45. 1 26 48. 1 49 46.7
two 1 1 21.6 10 18.5 21 20
three 5 9.8 5 9.3 10 9.5
four 0 0.0 0 0.0 0 0.0

> four 0 0.0 3 5.6 3 2.8

60

Table 9
Counts and Percentages of Subj ects' Current Word Processirng Skills

 

Control Experimental Totals
count percent count percent count percent

none 0 0.0 3 5.6 3 2.9
poor 15 29.4 14 25.9 29 27.7
average 30 58.8 24 44.4 54 51.4
good 5 9.8 12 22.2 17 16.2
excellent 1 2.0 1 1 .9 2 1.9

Table 9 summarizes subjects' self-evaluations of their current word
processing skills, irrespective of platform or software. Just over 5 1% of
all subjects reported their current word processing skills as "average."
Approximately 30% reported their word processing skills as either non-
existent or "poor." Given the fact that the criteria for students to sign up
for the study included an aversion to computers, it is interesting to note
that 19 subjects, or approximately 1 8% of the total sample, rated their
word processirng skills as "good" or "excellent." Contingency table
analysis revealed no signiﬁcant differences between the experimental and
control goups with respect to subj ects' self-evaluations of their current
word processing skills (Chi-square=6.503, p=.1646).

61

Table 10
Counts and Percentages of Subjects' Typing Skills Self-Ratings

 

Control Experimental Totals
count percent count percent count percent

poor 13 25.5 6 1 1.1 19 18.1
average 3 1 60.8 34 63.0 65 61.9
good 7 13.7 14 25.9 21 20.0

Table 10 presents a summary of subjects' self-ratings of their
typing ability. Nearly two-thirds of the total sample reported their typing
ability as "good." Contingency table analysis revealed no signiﬁcant
differences between the experimental and control goups with respect to
the subjects' self-ratings of their typing ability (Chi-square=4.969,
p=.0834).

Some subject attrition occurred between Phase I and Phase II of
the experiment. Table 1 1 presents the frequency counts for those
subjects who attended only Phase I and for those subjects who attended
both Phase I and Phase II. Just under 30% of the total number of
subjects dropped out of the experiment. A Chi-square test revealed no
signiﬁcant statistical differences in the dropout rates between the
experimental and the control goups (Chi-square=.239, p=.625).

62

Table 1 1
Subject Attendance at Phase I Only and Phases I and II

 

Control Experimental Totals
count percent count percent count percent

Phase I 16 31.4 15 27.8 31 29.5
only

Phases 1 35 68.3 39 72.2 74 70.5
and II

The Computer Anxiety Scale was checked for reliability using the
test-retest method. Subkoviak and Levin (1977) suggest calculating the
Pearson product-moment correlation coefﬁcient, r, as the appropriate
statistical procedure for checking test-retest reliability. The subjects'
ﬁrst CAS scores obtained during their initial testing were correlated with
the CAS scores obtained during the Phase I pretest, yielding a Pearson r
of .82.

 

 

Table 12
Subjects' Mean Computer Anxiety Scale Scores: Phase I
Control Experimental Totals
mean sd mean sd mean sd
n=51 =54 n=105
CAS 1 46.05 14.87 40.13 14.84 43.01 15.08

CAS 2 44.86 14.84 37.92 14.08 41.29 14.08

63
Table 12 summarizes data for CAS total scores for the Phase I

pretest (CAS 1) and posttest (CAS 2) administrations. Signiﬁcance
testing of mean differences (CAS 1 - CAS 2) comparing experimental and
control goups will be presented in the next section on hypothesis
testing. Overall, there appears to be a small decrement in CAS 2 scores
when compared to CAS 1 scores. A one-factor ANOVA comparing CAS 1
scores for the control and experimental goups revealed a signiﬁcant
difference at the .05 level between the two goups on the Phase I pretest
(Fachcgc=4.176, p=.043).

Table 13 summarizes the data for CAS total scores for the Phase II
pretest (CAS 3) and posttest (CAS 4). Again, signiﬁcance testing of mean
differences comparing experimental and control goups will be presented
111 the next section. Overall, there appears to be no decrement in CAS 4

scores when compared to CAS 3 scores.

 

 

Table 13
Subjects' Mean Computer Anxiety Scale Scores: Phase 11
Control Experimental Totals
mean sd mean sd mean sd
(n=35) (n=39) (n=74)
CAS 3 41.97 12.81 38.00 14.59 39.87 13.83

CAS 4 41.42 14.32 38.23 14.25 39.74 14.28

64
Table 14

Means and Standard Deviations of Subjects'

Desired Time to Learn Word Processing

 

 

Control Experimental Totals

mean sd mean sd mean sd
Phase I 3.25 .97 3.07 1.14 3.16 1.06
Pretest
Phase I 3.17 .97 3.13 1.02 3.15 .99
Posttest
Phase II 3.20 1.02 3.00 1.10 3.09 1.06
Pretest
Phase II 3.20 .96 3.02 1.08 3.10 1.02
Posttest

Table 14 presents mean values and standard deviations of

subjects' responses to the pretest and posttest question: "If an

opportunity were to arise, how much time would you be willing to devote

to learning word processing skills?" This item was answered on a 5-point

scale from "none" to "> ﬁve hours." In general, it appears that subjects

desired spending approximately 2 - 3 hours on learning word processing.

There appears to be a weak increase in mean scores for the Phase I

experimental goup and a weak decrease irn mean scores for the Phase I

control goup. Within goup Phase 11 mean scores appear to be

unchanged for both the control and the experimental goups.

65

Table I 5
Means and Standard Deviations of Subjects'
Interest irn Learning Computer Skills

 

 

Control Experimental Totals

mean sd mean sd mean sd
Phase I 3.09 .80 2.83 .90 2.96 .86
Pretest
Phase I 3.04 .87 2.83 .88 2.93 .88
Posttest
Phase II 3.08 .74 2.87 .76 2.97 .76
Pretest
Phase II 3.17 .75 2.76 .71 2.96 .75
Posttest

Table 1 5 presents means and standard deviations of subjects'
responses to the pretest and posttest question: "How irnterested are you
in learning computer skills, whatever they may be?" This item was
answered on a 4-point scale from "not at all" to " a lot," with lower values
indicating lower interest. Mean scores seemed to fall about the

"somewhat" value.

66

Table 16
Means and Standard Deviations of Subjects'
Desired Time to Learn Computer Skills

 

 

Control Experimental Totals
mean sd mean sd mean sd

Phase I 3.39 1.06 3.09 1.14 3.23 1.10
Pretest

Phase I 3.29 1.08 3.15 1.02 3.21 1.04
Posttest

Phase II 3.43 1.06 3.03 1.04 3.22 1.06
Pretest

Phase II 3.46 1.07 3.05 1.02 3.24 1.05
Posttest

Table 16 presents mean values and standard deviations of
subjects' responses to the pretest and posttest question: "How much time
would you be willing to devote to learrning computer skills, whatever they
may be?" Subjects responded to this question by choosing from a 5-
point scale with items ranging from "none" to "> ﬁve hours." Subjects
appeared willing to devote from 2-3 to 4-5 hours to learning computer
skills. The mean values for the control goup appear to be slightly higher
that those for the experimental goup.

67

Table 1 7
Means and Standard Deviations of Subjects' Word Processing
Task Performance Measured by Checklist

 

 

Control Experimental Totals

mean sd mean sd mean sd
WP 1 7.67 1.32 7.68 1.51 7.68 1.42
Phase I
WP 2 8. 16 1.30 8.46 l. 14 8.31 1.23
Phase 1
WP 3 8.37 1.09 8.26 1.09 8.31 1.08
Phase II

A summary of subjects' performance on the word processing tasks
is presented irn Table 1 7. Subjects' word processing performance was
measured according to a 9-point checklist. There appears to be an
increase in subjects' performance during Phase I when comparing the
practice task (WP 1) with the transfer task (WP 2). There also appears to
be a slight performance decrement with the second transfer task (WP 3)
which subjects performed one month after the ﬁrst two tasks.

68

Table 18
Means and Standard Deviations of Subjects'
Word Processing Task Time in Minutes

 

 

Control Experimental Totals
mean sd mean sd mean sd
WP 1 14.68 8.23 13.37 4.52 14.01 6.59
(Phase 1)
WP 2 7.02 2.78 7.49 2.88 7.26 2.83
(Phase 1)
WP 3 9.75 3.53 10.60 5.03 10.19 4.38
(Phase II)

Table 18 presents the means and standard deviations of time
subjects spent on their word processing tasks. The time is reported in
minutes and decimal equivalents of a nrinute. It is obvious that subjects
took signiﬁcantly less time to complete the second word processing task;
time for the second task was roughly half of the ﬁrst for both the
experimental and the control goups. Subjects also took less to perform a
similar task after one month when compared with the ﬁrst word
processing task; however, there appeared to be a "rebound eﬁ'ect" inn that
Phase II task time was generally longer than that for the second task of
Phase I.

69
Summanmtilnepcscapnmmiistics

The total goup consisted of n: 105 subjects, with a control goup
of n=51 and experimental goup of n=54 subjects. The subjects were
mostly female. Female subjects were found to have signiﬁcantly higher
Phase I pretest CAS scores than the male subjects. Experimental and
control goups differed signiﬁcantly with respect to their GPA: GPA did
not correlate signiﬁcantly with the Phase I pretest CAS, so it was not
treated as a confounding variable. The most prevalent acaderrnic major for
all subjects was some form of social science. Over 80% of all subjects
reported being in the ﬁrst year of their undergaduate studies. Sixty-six
percent of all subjects reported having taken 1 or 2 semesters or terms of
previous computer instruction.

Just under 30% of subjects dropped out of the study after Phase I.
There was no signiﬁcant difference between the experimental and the
control goups with respect to subject dropout rates.

A test-retest check of instrument reliability of the CAS yielded a
correlation coefﬁcient of :82. The control and experimental goups
differed signiﬁcantly on their pretest CAS score means.

Initially, subjects appeared to desire approximately 2-3 hours for
learning word processing skills. Subjects irnitially reported, on average.
being "somewhat" interested in learning computer skills, and willing to
spend approximately 2-3 to 4-5 hours learning computer skills.

Experimental and control goup subjects averaged 7.6 checklist
points on the ﬁrst word processing task. On the second word processing
task, experimental goup subjects averaged 8.4 points, and control goup
subjects averaged 8.16 points. On the third word processing task,

70
experimental goup subjects averaged 8.26 points, and control goup

subjects averaged 8.37 poirnts.

Subjects' average time to complete the word processirng tasks
decreased on WP 2 relative to WP 1 , and increased slightly on WP 3
relative to WP 2.

We

In this section, the experimental hypotheses presented in Chapter
One will be tested statistically. The following are the hypotheses of the
study:

1 . The experimental goup will score lower on a measure of
computer anxiety, the Computer Anxiety Scale, than will an
equivalent control goup after initial training in the use of word

processing.

2 The experimental goup will score lower on a measure of
computer anxiety, the Computer Anxiety Scale, than will an
equivalent control goup when measured during a word
processing follow-up session attended one month after initial

trairning.

3. The experimental goup will score higher on performance of a
word processing task than will an equivalent control goup.

71
Differences in performance scores will be stable after a follow-

up treatment.

. The experimental goup will indicate a geater desire to learn
more computer skills as measured by 1) subjects' estimates of
the number of hours they would be willing to devote to furtlner
word processirng skills study, 2) subj ects' estimates of their
interest in learning other computer progams, and 3) subjects'
estimates of the number of hours they would be willing to
devote to developing their computer skills, than will an
equivalent control goup, when measured immediately after
instruction.

. The experimental goup will indicate a geater desire to learn
more computer skills as measured by 1) subjects' estimates of
the number of hours they would be willing to devote to further
word processirng skills study, 2) subjects' estimates of their
interest in learning other progams, 3) subjects' estimates of the
number of hours they would be willing to devote to developing
their computer skills, and 4) subj ects' attendance at a second
computer session, than will an equivalent control goup, when

measured during a one-month follow-up computer session.

Computer Anxiety

The ﬁrst experimental hypothesis holds that subjects in the
experimental goup will experience more reduction irn their CAS scores
than will subjects in the control goup after an irnitial (Phase I) learrning
experience. In other words, mean difference scores (pretest: CAS 1 -
posttest: CAS 2) will be higher for the experimental goup than for the
control goup. The second experimental hypothesis was that the
experimental goup subjects would score lower on the CAS than would
the control goup subjects when measured during a word processing
follow-up session attended one month after initial training. In order to
test this second hypothesis, several comparisons will be made. The ﬁrst
will compare Phase II pretest CAS scores with Phase I pretest scores (CAS
1 - CAS 3) for signiﬁcant differences in means. This comparison will
determine whether subjects' computer anxiety as measured by the CAS
was different one month after their pretest measurement, and thus
determine the presence of any long-term effect on CAS scores after the
irnitial training. The second comparison will compare students' Phase II
pretest CAS scores with their Phase I posttest scores (CAS 2 - CAS 3), to
see if there was any change in subjects' post-treatment computer anxiety
as measured by the CAS which occurred over time and prior to the
second (Phase 11) word processing session. A third comparison would
compare subjects' pre- and posttest Phase II scores (CAS 3 - CAS 4) to
determine if there was any additional effect on CAS scores attributable to
the second word processing training experience. A fourth comparison
involving Phase I CAS pretest scores with Phase 11 CAS posttest scores
will deterrrnine the presence of effects on Phase I pretreatment scores

73
based upon two training sessions separated by one month, or as a result

of the entire experimental progam.

A repeated measures ANOVA is used to test the statistical
signiﬁcance of mean difference scores for multiple outcome measures in
a one factor experimental design (Glass and Hopkins, 1984). The Scheffe
method for multiple comparisons was employed, as it maintains the
alpha level at .05, for all comparisons and is the most ﬂexible of the
multiple comparison methods in its ability to determine signiﬁcant
contrasts (Hopkins and Glass, 1984). Tables 19 - 22 present repeated
measures ANOVAs for the CAS scores over time and within goup.

According to the results of the repeated measures analyses
performed on the CAS 1, 2, 3, and 4 scores for the control goup, their
were no signiﬁcant differences found for any one pairwise comparison
(Table 20). However, there was a signiﬁcant difference at the .05 level inn
CAS scores for all administrations of CAS to the control goup taken as a
whole (Table 19). This suggests that there was an effect on the CAS
scores of the control goup. but that the effect was spread out over the 4
admirnistrations. A review of Table 20 indicates that most of the
decrement occurred in the one-month interim period between Phase I
(CAS 1 and 2) and Phase 11 (CAS 3 and 4).

74

 

 

 

 

 

 

AN OVA Table for ComputerTfrtanCefyg Scores Repeated Measures,
Control Group
_Source: df: SS: M§?—_—F-test: p:
Between 34 21256186 625.182 19.463 .0001
Subjects
Within 105 3372.75 32.121
Subjects
treatment 3 308.65 102.883 3.425 .02
residuals 102 3064.1 30.04
Total 139 24628.936
Table 20
Computer Anxiety Score Mean Comparisons, Control Group
Comparison Mean Difference Schem—
CAS 1 - CAS 2 1.086 0.229
CAS 1 - CAS 3 3.114 1.883
CAS 1 - CAS 4 3.657 2.597
CAS 2 - CAS 3 2.029 0.799
CAS 2 - CAS 4 2.571 1.284
CAS 3 - CAS 4 0.543 0.057

Within the experimental goup, the pairwise comparisons involving

the Phase I pretest (CAS 1) and all subsequent admirnistrations of the
CAS were signiﬁcant at the .05 level (Tables 21 - 22). The ﬁrst

experimental hypothesis is upheld: subjects in the experimental goup

rated their posttest computer anxiety lower than their initial pretest

75
computer anxiety as measured by the CAS. Furthermore, this difference

remained signiﬁcant for all subsequent administrations of the CAS,
suggesting that the irnitial treatment effect was durable over time. All
other pairwise comparisons were not statistically signiﬁcant, therefore
the second experimental hypothesis, that the experimental goup
subjects would score lower on the CAS than would the control goup
subjects when measured during a word processing follow-up session

attended one month after initial training, is rejected.

 

 

Table 2 1
ANOVA Table for Computer Anxiety Scores Repeated Measures
Experimental Group
—SBurce: df: S‘S: MS: F-test: p:
Between 38 31692.474 834.012 52.96 .0001
Subjects
Within 117 1842.5 15.748
Subjects
treatment 3 341.385 1 13.795 8.642 .0001
residuals 114 1501.115 13.168

Total 1 55 33534.974

76

Table 22
Computer Anxiety Score Mean Comparisons, Experimental Group

 

 

Comparison Mean Difference Scheffe—FEE?—
CAS 1 - CAS 2 2.897 4.144‘I

CAS 1 - CAS 3 3.692 6.730.

CAS 1 - CAS 4 3.462 5.915.

CAS 2 - CAS 3 0.795 0.312

CAS 2 - CAS 4 0.564 .157

CAS 3 - CAS 4 -.231 .026

‘=signiﬁcant at .05

A second analysis was conducted to further determine which
subjects experienced a possible treatment effect. Subjects' scores were
divided at the 50th percentile and F—tests of CAS mean difference scores
were conducted for the experimental and control goups for both the
upper and lower percentile ranges. It should be noted that by dividing
the experimental and control goups into subgoups for analysis, the
power to detect signiﬁcant differences is signiﬁcantly diminished (Glass
and Hopkins, 1984). Nevertheless, subjects' mean difference scores
comparing their Phase I pretest CAS scores with all subsequent
administrations of the CAS in the upper 50% of the experimental goup
were signiﬁcantly different at the .05 level (Table 23). Mean differences
for any comparison of CAS scores in the upper 50% of the control goup
were not statistically signiﬁcant at the .05 level (Table 24). For the lower
50% range, there were no signiﬁcant differences at the .05 level for the
experimental goup (Table 25). Interestingly, there was one signiﬁcant

77
mean difference at the .05 level irn the lower 50% of the control goup

(Table 26). A review of the mean difference scores of the upper and lower
halves of the control goup again reveals the "distributed" effect on the
control goup CAS scores described earlier; CAS scores appear to be
lower for all control goup subjects irrespective of percentile gouping.
Overall, the desired treatment effect of lowered computer anxiety
appeared to be experienced by the more computer anxious students, as
measured by the CAS, in the experimental goup, and possibly by all
students, although to a lesser extent, in the control goup.

 

 

Table 23
CAS Mean Score Comparisons, Experimental Group, Upper 50%

Comparison Mean Difference Schemes-t—
CAS 1 - CAS 2 4.227 4.101‘

CAS 1 - CAS 3 6.091 8514*

CAS 1 - CAS 4 5.955 8137‘

CAS 2 - CAS 3 1.864 .797

CAS 2 - CAS 4 1.727 .685

CAS 3 - CAS 4 -.136 .004

"=signiﬁcant at .05

78

Table 24
CAS Mean Score Comparisons, Control Group, Upper 50%

Comparison Mean merence Eeffe F-test

 

 

 

 

 

CAS 1 - CAS 2 .812 .033

CAS 1 - CAS 3 4.25 .911

CAS 1 - CAS 4 3.375 .575

CAS 2 - CAS 3 3.438 .596

CAS 2 - CAS 4 2.562 .331

CAS 3 — CAS 4 -.875 .039

Table 25
CAS Mean Score Comparisons, Experimental Group, Lower 50%

Comparison Mean Wrence Schm
CAS 1 - CAS 2 1.176 .65

CAS 1 - CAS 3 .588 .162

CAS 1 - CAS 4 .235 .026

CAS 2 - CAS 3 -.588 .162

CAS 2 — CAS 4 -.941 .416

CAS 3 - CAS 4 -.353 .058

79

Table 26
CAS Mean Score Comparisons, Control Group, Lower 50%

 

—Comparison Mean Difference Scheﬁe F-test

CAS 1 - CAS 2 1.316 .458
CAS 1 - CAS 3 2.158 1.233
CAS 1 - CAS 4 3.895 4.016"
CAS 2 - CAS 3 .842 .188
CAS 2 - CAS 4 2.579 1.761
CAS 3 - CAS 4 1.737 .799

‘=signiﬁcant at .05

Recalling that approximately 30 subjects did not continue with
Phase II of the experiment, a comparison was made of Phase I pretest
CAS scores between subjects who attended only Phase I and subject who
attended botln Phase and Phase II. Table 27 presents Phase I pretest
CAS scores for subjects who attended the Phase I session only and the
Phases I and 11 sessions, control and experimental goups.

Table 27
Phase I Pretest CAS Scores: Phase I Only and Phases I and II

 

 

Control Experimental Totals
mean sd mean sd mean sd
Phase I
only 48.18 19.09 36.06 12.30 42.32 17.05
(n=31) (n=16) (n=15)
Phase I 8:

Phase II 45.08 12.71 41.69 15.56 43.29 14.29
(n=74) (n=35) (n=39)

80
A two-factor AN OVA comparing pretest CAS means of subjects

attending Phase I only with subjects attending botln Phases, for both the
control and experimental goups, revealed no statistically signiﬁcant
differences at the .05 level attributable to attendance (Table 28). Thus,
there appeared to be no differential dropout phenomenon among subjects

with respect to Phase I pretest CAS scores.

 

 

 

Table 28
ANOVA Table Comparing CAS 1 Score. Group by Attendance
Source: df: SS: MS: F-test: p:
Group (A) 1 1312.536 1312.536 5.947 .0165
guidance 1 34.736 34.736 . 1 57 .6924
A x B ‘ 1 415.358 415.358 1.882
Error 10 1 22292.42 1 220.7 1 7

It is interesting to note that there was a signiﬁcant mean diﬂcccncc
between Phase I pretest and posttest CAS scores among the subjects '
when considering attrition. In the control goup, i.e., those who dropped
out and those who did not drop out did not have statistically signiﬁcant
mean difference scores at the .05 level when the Phase I CAS pretest was
compared with the Phase I CAS posttest (non-dropout control goup
mean difference: 1.086, chgc=.864: dropout control goup mean
difference=1.438, Fsdwe=.623). In the experimental goup, the subjects
who dropped out, when taken as a goup, did not have statistically
signiﬁcant mean difference Phase I pretest/posttest CAS scores at the
.05 level (dropout experimental goup mean difference=.4, Fm=.04).

81
Experimental goup subjects who continued on to Phase 11, however, did

have statistically signiﬁcant mean difference scores at the .05 level (non-
dropout experimental goup mean difference=2.897, Fsmﬁ=21.319).
Thus, experimental goup subjects who experienced an immediate
reduction in CAS scores stayed with the experiment: all other subjects.
whether they stayed with the experiment or not, did not experience an
immediate reduction in CAS scores.

Word Processing Task Performance

The third hypothesis was that the experimental goup would score
higher on performance of a word processing task than would an
equivalent control goup. Phrtherrnore, differences inn performance
scores would be stable after a follow-up treatment. A comparison of two
variables, word processing performance as measured by a performance
checklist, and time spent on word processing tasks measured inn
minutes, was used to test the hypotheses. In order for the experimental
hypothesis to be upheld, one would expect a signiﬁcant difference in
performance means in favor of the experimental goup: experimental
subjects would be expected to score higher on the checklist measure, and
would also be expected to score lower on the time measure. The former
assumes that students who score higher on the checklist measure are
more effective word processing learners: the latter assumes that students
who take less time to complete the word processing task are more

efficient word processing learners.

82
It is useful to recall that there was a total of three word processing

tasks, a practice task and a transfer task during Phase I, and another
transfer task during Phase II. In order to test the hypothesis that the
experimental goup would score higher on performance of a word
processing task than would an equivalent control goup, a one-way
ANOVA was performed comparing mean performance checklist scores of
the experimental and control goups for each word processing task. In
addition, a repeated measures ANOVA of average amounts of time to
complete the word processing tasks measured in minutes were
performed.

One-way ANOVAs of the experimental and control goups
performance checklist mean scores on the three word processing tasks
revealed no signiﬁcant differences (fables 29-31). A review of the mean
scores (Table 17) shows that they are very similar for the control and
experimental goups. It appears that neither treatment was more
effective than the other in teaching word processinng.

 

 

 

Table 29
AN OVA Table: Word Processing Task One Performance as
Measured by Checklist
gurce: df: SS: MS: F-test: p:

Between 1 .009 .009 .004 .947
Groups

Within 103 208.981 2.029

Groups

Total 104 208.99

83

 

 

 

 

Table 30
ANOVA Table: Word Processing Task Two Performance as
Measured by Checklist
Source: df: SS: MS: F-test: p:
Between 1 2.458 2.458 1.642 .2029
Groups
Within 103 154.171 1.497
Groups
Total 104 156.629
Table 3 1
ANOVA Table: Word Processing Task Three Performance as
Measured by Checklist
_Source: df: SS: MS: F-test: p:
Between 1 .244 .244 .205 .6519
Groups
Within 72 85.607 1. 189
Groups
Total 73 85.851

It is interesting to note that the checklist means were relatively
elevated. [Wen the "practice" task scores averaged 7 .6 out of 9, the
highest possible score, for both goups, indicating that both treatments
were eﬁ'ective in teaching subjects how to use word processing to
complete the assigned task. The high mean scores for all three word
processing tasks also indicate the possible presence of a "ceiling effect" -
so many students were able to attain the highest possible score that
there was insufﬁcient variance available to determine the presence, if

84
any, of a differential treatment effect. Table 32 presents the percentages

of subjects irn the experimental and control goups scoring 9 on the
performance checklist for each of the three word processirng tasks, and
supports the notion of a ceiling effect at work in the study.

Table 32
Percentages of Subjects Scoring "9" on the Performance Checklist

 

control exp

WPTask 1 39.21 40.74
WP Task 2 60.78 66.67

WP Task 3 65.71 56.41

A repeated measures comparison of the word processing
performance checklist means of the practice task (WP 1) with the transfer
task (WP 2) provides a possible measure of training effectiveness, i.e., if
there is an increment in mean performance scores between the practice
and the transfer task, then the training can be said to be eﬁ'ective. For
both the control goup and the experimental goup subjects, performance
on the transfer task (WP 2) was statistically better than it was on the
practice task (WP 1) at the .051eve1, indicating that both treatments were
effective in training subjects to perform the word processinng tasks (Tables
33—36). There was no statistically signiﬁcant difference in the control or
experimental goup task perfornnance means when comparing word
processing task 2 with task 3.

85

 

 

 

 

Table 33
AN OVA Table for Word Processing Scores Repeated Measures,
Experimental Group
Source: df: SS: MS: F-test: p:
Between 38 77.231 2.032 2.202 .0017
Subjects
Within 78 72 .923
Subjects
treatment 2 14.308 7.154 9.424 .0002
residuals 76 57.692 .759
Total 1 16 149.23 1
Table 34
Word Processing Performance Mean Comparisons. Experimental Group
Comparison Mean DTff'erence Scheﬁe F-test
WP 1 - WP 2 -.846 9.196"
WP 1 - WP 3 -.538 3.729"
WP2-WP3 .308 1.216

‘=signiﬁcant at .05

86

 

 

 

 

Table 35
AN OVA Table for Word Processing Scores Repeated Measures.
Control Group
Source: df: SS: MS: F-test: p:
Between 34 1 16.99 3.441 4.014 .0001
Subjects
Within 70 60 .857
Subjects
treatment 2 12.248 6. 124 8.72 .0004
residuals 68 47.752 .702
Total 104 176.99
Table 36
Word Processing Performance Mean Comparisons. Control Group
Comparison Mean DiffErence Sober-f? F-test
WP 1 - WP 2 -.514 3.296"
WP 1 - WP 3 -.829 8.554“
WP2 -WP3 -.314 1.231

‘=signiﬁcant at .05

Tables 37-40 present comparisons of word processing performance
for both the experimental and control goups based upon subjects
percentile rank in eitlner the upper or lower 50% with respect to Phase I
pretest CAS scores. It is interesting to note that in the experimental
goup, subjects in both the upper and lower computer anxious
subgoups experienced their geatest performance increase on the Phase
I transfer task, while in the control goup subjects in both the upper and

87
lower computer anxious subgoups experienced their geatest

performance increase on the Phase II task. These phenomena partially

mirror that of subjects' CAS scores. in which the higher anxiety

experimental goup showed the treatment effect of lower anrdety

immediately after Phase I treatment, while the control goup as a whole

showed lower anxiety after the Phase 11 treatment.

Table 37
Word Processing Task Performance Comparisons,
Experimental Group, Upper 50% CAS

 

 

Comparison Mean Difference Scheﬁe F—test

WP 1 - WP 2 -.864 5.856“
WP 1 - WP 3 -.591 2.741
WP 2 - WP 3 .273 .584

‘=signiﬁcant at .05

Table 38
Word Processing Task Performance Comparisons,
Experimental Group, Lower 50% CAS

 

 

Comparison Mean Dﬁerence Scheﬁe F—test

WP 1 - WP 2 -.824 3271"
WP 1 - WP 3 -.471 1.068
WP 2 - WP 3 .273 .584

*=signiﬁcant at .05

88

Table 39
Word Processing Task Performance Comparisons.
Control Group, Upper 50% CAS

 

 

Comparison Mean Difference &f$ﬁ?F—test
WP l - WP 2 -.562 2.066
WP 1 - WP 3 -.875 5.000.
WP 2 - WP 3 -.312 .638

‘=signiﬁcant at .05

Table 40
Word Processing Task Performance Comparisons,
Control Group, Lower 50% CAS

 

 

Comparison Mean Difference Schema—
WP l - WP 2 -.474 1.308

WP l - WP 3 -.789 3.633.
WP2 -WP3 -.316 .581

I"=signiﬁcant at .05

One-way ANOVA comparisons of the control and experimental
goups of the average amounts of time to complete the word processing
tasks measured in minutes were performed, and the results were similar
to those for task performance as measured by the checklists. Tables 41-
43 present the results of AN OVAs comparing average time to complete
the word processing tasks expressed in rrninutes and decimal equivalents
of minutes. None of the comparisons was signiﬁcant at the .05 level.

89

Table 4 1
AN OVA Table: Word Processing Task One Time in- Minutes

 

 

 

 

 

—Source: df: SS: M§ F-test: p:
Between 1 45.512 45.512 1.049 .3082
Groups
Within 103 4470.294 43.401
Groups
Total 104 4515.806

Table 42
AN OVA Table: Word Processinng Task Two Time in Minutes
Source: df: SS: MS: F-test: p:
Between 1 6.078 6.078 .756 .3865
Groups
Withirn 103 827.811 8.037
Groups

Total 104 833.889

90

Table 43
AN OVA Table: Word Processing Task Three Time in Minutes

 

 

 

 

_Source: df: SS: MS: F-test: p:
Between 1 13.38 13.38 .694 .4076
Groups
Within 72 1388.58 1 19.286
Groups
Total 73 1 4O 1 .96 1

There were signiﬁcant practice effects with respect to time for both
control and experimental goups. Subjects in both goups took
signiﬁcantly less time to perform the transfer task (Time 2) than they did
the practice task (Time 1) (Tables 44-47). On the basis of checklist
scores and time measures, both the control and experimental treatments
could be said to be successful in training subjects to perform word

processing tasks.

91

 

 

 

 

 

Table 44
ANOVA Table for Word Processing Time Repeated Measures. Control
Group
_Source: df: SS: MS: F-test: p:
Between 34 131 1.241 38.566 1.124 .3332
Subjects
Withinn 70 2401.185 34.303
Subjects
treatment 2 955. 126 477.563 22.457 .0004
residuals 68 1446.059 21.266
Total 104 3712.426
Table 45
Word Processing Time Mean Comparisons, Control Group
Comparison Mean Difference Scheffe F—test
Time 1 - Time 2 7.234 21.532"
Time 1 - Time 3 4.915 9.942"
Time 2 - Time 3 -2.319 2.212

"=signiﬁcant at .05

The experimental goup experienced a signiﬁcant increase in task
time for the Phase 11 word processing task (Time 3) when compared with
the Phase I transfer task (Time 2), indicating some performance loss
during the one-month interim period. The control experienced some
apparent loss, but it was not statistically signiﬁcant. This loss was only
partial: comparisons of the Phase 11 word processing task time (Time 3)
with the Phase I practice task time (Time 1) revealed signiﬁcant

92
differences for botln the control and experimental goups, indicating that

the effects of task practice were still signiﬁcant when subjects were given
the opporturnity to practice one month after the initial training session.

 

 

 

 

 

Table 46
ANOVA Table for Word Processing Time Repeated Measures,
Experimental Group
_Source: df: SS: MS: ﬁest: p:
Between 38 1029.685 27.097 1.352 .1309
Subjects
Within 78 1563.415 ‘ 20.044
Subjects
treatment 2 495.378 247.689 17.625 .0001
residuals 76 1068.037 14.053
Total 1 16 2593.1
Table 47
Word Processing Time Mean Comparisons, Experimental Group
Comparison Mean D—iﬁerence Scheﬁe F-test
Time 1/Time 2 5.038 17.613"
Time 1/TTme 3 2.635 4.818"
Time 2/Time 3 -2.403 4.007"

"=signiﬁcant at .05

Comparison of time to complete task for the ﬁrst, second. and third
tasks for the upper and lower halves of the experimental and control
goups revealed that all subgoups showed signiﬁcant practice effects

93

during Phase I (Tables 48 - 5 1). Both upper and lower CAS percentile

subgoups for both the experimental and control goups took

signiﬁcantly less time to perform the transfer task (Time 2) than to
perform the practice task fl‘ime 1) during Phase 1. Within the control
goup, there were two other signiﬁcant comparisons. First, the higher
anrdety half of the control goup took signiﬁcantly longer to perform the
Phase II task (Time 3) when compared with the Phase I transfer task

(Plane 2). The lower anxiety half of the control goup took signiﬁcantly
less time to perform the Phase II task than they did the Phase I practice

task.

Table 48
Word Processing Time Comparisons,
Experimental Group, Upper 50% CAS

 

fomparison Mean Difference Scheffe F-test

Time l/TTme 2 5.101 10.734"
Time 1/Time 3 2.707 3.023
Time 2/Time 3 -2.394 2.364

"=signiﬁcant at .05

Table 49
Word Processing Time Comparisons,
Experimental Group, Lower 50% CAS

 

 

iomparison Mean DiFerence Schem
Time l/Time 2 4.957 6.59"
Time 1/Time 3 2.542 1.73
Time 2/Time 3 -2.415 1.562

"=signiﬁcant at .05

94

Table 50
Word Processing Time Comparisons,
Control Group, Upper 50% CAS

 

 

Comparison Mean Dﬁerence Sm
Time 1/Time 2 6.241 12.298"
Time 1/T‘ime 3 2.702 2.306

Time 2/Time 3 -3.538 3.953‘

"=signiﬁcant at .05

Table 51
Word Processing Time Comparisons,
Control Group, Lower 50% CAS

 

 

 

Comparison Mean Difference Schef-fe F-test
Time l/Time 2 8.071 1 1.209.
Time 1 [Time 3 6.779 7.909"
Time 2/Time 3 -1.292 .287

"=signiﬁcant at .05

Desire to Learn More Computer Skills

The hypotheses that the experimental goup subjects would
indicate geater desire to learn more computer skills as measured by 1)
subjects' estimates of the number of hours they would be willing to
devote to further word processing skills study, 2) subjects' estimates of
their interest in leamirng other computer progams, and 3) subjects'
estimates of the number of hours they would be willing to devote to
developing their computer skills, than would an equivalent control goup,

when measured immediately after instruction and when measured

95
during a one-month follow-up computer session was tested by comparing

subjects' mean responses to questions on the backgound and feedback
study instruments (Appendices D and F). The ﬁrst question asked: "If an
opportunity were to arise, how much time would you be willing to devote
to learrning word processing skills?" Subjects responded by choosing one
of ﬁve responses ranging from "none" to "> ﬁve hours." The second
question asked: "How interested are you in learning computer skills,
whatever they may be?" Subjects responded by choosing one of four
responses ranging from "not at all" to "a lot." The third question asked:
"How much time would you be willing to devote to learrning computer
skills, whatever they may be?" Subjects responded by choosing one of
ﬁve responses ranging from "none" to "> ﬁve hours." One factor ANOVA
goup comparisons were used to determine if there were any differences
between experimental and control goups with respect to any of the
measures: repeated measures ANOVA comparisons were used to see if
there were any differences within experimental and control goups over
time with respect to any of the measures.

The one factor ANOVA comparisons (Tables 52-63) revealed only
one signiﬁcant control goup / experimental goup comparison, that of the
Phase 11 Feedback question regarding "innterest in learning computer
skills," the second question listed above (Fable 59). The control goup
rated their interest inn learning computer skills, whatever they may be,
higher than did the experimental goup, when asked immediately after
the second word processing training session. All other
experimental/ control goup comparisons were not signiﬁcant at the .05

level.

96

Table 52
ANOVA Table: Desired Time to Learn Word Processing.
Phase I Pretest

 

 

Source: df: SS: MS: F-test: p:
Between 1 .858 .858 .753 .3877
Groups
Within 103 117.39 1.14
Groups
Total 1 04 1 1 8.248

Table 53

ANOVA Table: Desired Time to Learn Word Processing,
Phase I Posttest

 

 

-Source: df: SS: MS: F-test: p:
Between 1 .058 .058 .057 .8 1 13
Groups
Within 103 103.504 1.005
Groups

Total 104 103.562

97

Table 54
ANOVA Table: Desired Time to Learn Word Processinng,
Phase II Pretest

 

 

 

 

Source: df: SS: MS: F-test: p:
Between 1 .738 .738 .651 .4224
Groups
Witlnin 72 81.6 1.133
Groups
Total 73 82.338
Table 55
ANOVA Table: Desired Time to Learn Word Processing,
Phase II Posttest

Source: df: SS: MS: F-test: p:
Between 1 .561 .561 .527 .4701
Groups
Within 72 76.574 1.064
Groups

Total 73 77. 135

98

 

 

 

 

 

Table 56
ANOVA Table: Interest in Learning Computer Skills,
Phase I Pretest
-Source: df: SS: MS: F—test: p:
Between 1 1 .838 1 .838 2.49 . 1 176
Groups
Within 103 76.01 .738
Groups
Total 1 04 77.848
Table 57
AN OVA Table: Interest in Learning Computer Skills,
Phase I Posttest
Source: df: SS: MS: F-test: p:
Between 1 1. 1 12 1 . 1 12 1.442 .2326
Groups
Within 103 79.422 .77 1
Groups

Total 104 80.533

99

Table 58
ANOVA Table: Interest in Learning Computer Skills,
Phase II Pretest

 

 

—Source: df: SS: MS: F-test: p:
Between 1 .844 .844 1.479 .228
Groups
Witlnin 72 41.102 .571
Groups
Total 73 41.946

Table 59

ANOVA Table: Interest in Learrning Computer Skills,
Phase II Posttest

 

 

Source: df: SS: MS: F-test: p:
Between 1 2.984 2.984 5.669 .0199
Groups

Within 72 37.895 .526

Groups

Total 73 40.878

100

Table 60
ANOVA Table: Desired Time to Learn Computer Skills,
Phase I Pretest

 

 

_Source: df: SS: MS: F-test: p:
Between 1 2.354 2.354 1 .944 . 1662
Groups
Within 103 124.694 1 .2 1 1
Groups
Total 104 127.048

Table 61

ANOVA Table: Desired Time to Learn Computer Skills,
Phase I Posttest

 

 

Source: df: SS: MS: F-test: p:
Between 1 .559 .559 .508 .4778
Groups

Within 103 1 13.403 1.101

Groups

Total 104 1 13.962

101

Table 62
ANOVA Table: Desired Time to Learn Computer Skills,
Phase II Pretest

 

‘

 

—Source: df: SS: MS: F-test: p:
Between 1 2.995 2.995 2.7 1 1 . 104
Groups
Within 72 79.546 1.105
Groups
Total 73 82.541

Table 63

ANOVA Table: Desired Time to Learn Computer Skills,
Phase II Posttest

 

 

Source: df: SS: MS: F-test: p:
Between 1 3.038 3.038 2.784 .0996
Groups

Within 72 78.583 1.091

Groups

Total 73 81 .622

102

The repeated measures comparisons (Tables 64-75) revealed no

significant mean diﬂ'erences at the .05 level for any pairs of measures for

any single question within either the control or the experimental group.

Table 64

AN OVA Table: Desired Time to Learn Word Processing
Repeated Measures, Control Group

 

 

_Source: (11': SS: MS: F—test: p:
Between 34 1 15.386 3.394 17. 173 .0001
Subjects
Within 1 05 20. 75 . l 98
Subjects

treatment 3 .193 .064 .319 .8116
residuals 102 20.557 .202
Total 1 39 1 36. l 36
Table 65

Desired Time to Learn Word Processing Mean Comparisons,

Control Group

 

 

Comparison Mean DiﬂTerence Scheﬁe F-test
Ph I Pretest - Ph I Posttest .086 .213

Ph I Pretest - Ph II Pretest .086 .213

Ph I Pretest - Ph II Posttest .086 _2 13

Ph I Posttest - Ph II Pretest 0 0

Ph I Posttest - Ph II Posttest 0 0

Ph 11 Pretest - Ph II Posttest O 0

103
Table 66

ANOVA Table: Desired Time to Learn Word Processing
Repeated Measures, Experimental Group

 

 

 

 

 

Source: df: SS: MS: F-test: p:
Between 38 156.09 4. 108 22.353 .0001
Subjects
Within 1 17 21 .5 . 184
Subjects
treatment 3 .256 .085 .459 .7 1 17
residuals 1 14 21.244 . 186
Total 155 177.59
Table 67
Desired Time to Learn Word Processing Mean Comparisons.
Experimental Group
_Comparison Mean Werence Scheﬁe F-test
Ph I Pretest - Ph I Posttest .026 .023
Ph I Pretest - Ph II Pretest . 103 .367
Ph I Pretest - Ph 11 Posttest .077 .206
Ph I Posttest - P11 11 Pretest .077 .206
Ph I Posttest - Ph 11 Posttest .05 1 .092
Ph II Pretest - Ph II Posttest -.026 .023

104

Table 68
AN OVA Table: Interest in Learning Computer Skills
Repeated Measures, Control Group

 

 

 

 

 

Source: df: SS: MS: F-test: p:
Between 34 71.171 2.093 14.653 .0001
Subjects
Within 105 15 . 143
Subjects
treatment 3 . 171 .057 .393 .7583
residuals 102 14.829 . 145
Total 1 39 86. 1 7 1
Table 69
Interest in Learning Computer Skills Mean Comparisons.
Control Group
-Comparison MeanTiﬁErence Sm
Ph I Pretest - Ph 1 Posttest -.029 .033
Ph I Pretest - Ph 11 Pretest .000 .000
Ph I Pretest - Ph 11 Posttest -.086 .295
Ph 1 Posttest - Ph 11 Pretest .029 .033
Ph 1 Posttest - Ph II Posttest .057 . 131

Ph II Pretest - Ph 11 Posttest -.O86 .295

Table 70
AN OVA Table: Interest in Learning Computer Skills
Repeated Measures. Experimental Group

 

 

 

 

 

Source: df: f SS: MS: F-test: p:
Between 38 84.577 2.226 16.534 .0001
Subjects
Within 117 15.75 .135
Subjects
treatment 3 .224 .075 .549 .6497
residuals 114 15.526 .136
Total 155 100.327
Table 71
Interest in Learning Computer Skills Mean Comparisons.
Experimental Group
Comparison Mean Werence W
Ph I Pretest - Ph I Posttest -.026 .031
Ph I Pretest — Ph 11 Pretest -.051 . 126
Ph I Pretest - Ph 11 Posttest .051 . 126
Ph 1 Posttest - Ph II Pretest -.026 .031
Ph I Posttest - Ph 11 Posttest .077 .282
Ph II Pretest - Ph II Posttest .103 .502

106

Table 72
ANOVA Table: Desired Time to Learn Computer Skills
Repeated Measures. Control Group

 

 

 

 

 

Source: df: SS: MS: F-test: p:
Between 34 136. 171 4.005 23.043 .0001
Subjects
Within 105 18.25 .174
Subjects
treatment 3 .079 .026 .147 .9314
residuals 102 18.171 . 178
Total 139 154.421
Table 73
Desired Time to Learn Computer Skills Mean Comparisons,
Control Group
Comparison Mean Werence Scheﬁe F-test
Ph I Pretest - Ph I Posttest .057 .107
Ph I Pretest - Ph 11 Pretest .029 .027
Ph I Pretest - Ph 11 Posttest .000 .000
Phi Posttest - Ph 11 Pretest -.029 .027
Ph I Posttest - Ph II Posttest -.057 . 107

Ph II Pretest - Ph II Posttest -0029 .027

107
Table 74

ANOVA Table: Desired Time to Learn Computer Skills
Repeated Measures, Experimental Group

 

 

 

 

 

_Source: df: SS: MS: F—test: p:
Between 38 142.09 3.739 28.225 .0001
Subjects
Within 1 l 7 15.5 . 132
Subjects

treatment 3 .051 .017 .126 .9445
residuals 1 14 15.449 .136
Total 155 157.59
Table 75
Desired Time to Learn Computer Skills Mean Comparisons.
Experimental Group
Comparison Mean DWerence Scﬁeﬁe F—test
Ph I Pretest - Ph I Posttest .026 .032
Phi Pretest - Ph II Pretest .051 . 126
Ph I Pretest - Ph II Posttest .026 .032
Ph I Posttest - Ph II Pretest .026 .032
Ph I Posttest - Ph II Posttest _000 .000
Ph II Pretest - Ph II Posttest -.026 .032

108
W

This section summarizes the results according to each hypothesis.

Hypothesis #1: The experimental group will score lower on a
measwe of computer anxiety. the Computer Anxiety Scale, than will an
equivalent control group after initial training in the use of word processing.

Subjects in the Phase I experimental group rated their posttest
computer anxiety signiﬁcantly lower than their pretest computer anxiety.
Moreover, this eﬁect was found in the upper 50% of subjects as
determined by their Phase I pretest CAS scores. In other words,
experimental group subjects who were initially more computer anxious
reported feeling signiﬁcantly less anxious when they took the Phase I
CAS posttest. This treatment effect endured with all subsequent
administrations of the CAS. indicating that the initial drop in CAS scores
was durable over time. Pretest/posttest mean differences for subjects in
the Phase I control group were not signiﬁcantly different. The ﬁrst
experimental hypothesis is therefore supported by the data.

Hypothesis #2: The experimental group will score lower on a
measure of computer anxiety, the Computer Anxiety Scale, than will an
equivalent control group when measured during a word processing follow-
up session attended one month after initial training.

Neither the experimental group nor the control group Phase II
pretest and posttest CAS scores were signiﬁcantly different. The second
experimental hypothesis is therefore rejected. One phenomenon of
interest is the overall reduction of the control group's CAS scores. A

109
comparison of the Phase I pretest and Phase II posttest for the control

group was signiﬁcant, while no other pairwise control group comparison
was signiﬁcant.

Hypothesis #3: The experimental group will score higher on
performance of a word processing task than will an equivalent control
group. Differences in performance scores will be stable qﬁer a follow-up
treatment.

There were no signiﬁcant differences between the experimental and
control groups with respect to their scores on the three word processing
tasks as measured by the performance checklists. Within group
comparisons of checklist scores revealed that both experimental and
control treatments were effective in training subjects to perform the word
processing tasks. as the checklist scores for word processing task 2 were
signiﬁcantly higher than for word processing task 1 for both groups.
Subjects in the experimental group, irrespective of their percentile
grouping into upper and lower halves with respect to computer anxiety.
experienced their greatest performance increase on the Phase I transfer
task. which was the task immediately following their ﬁrst instructional
treatment. Subjects in the control group in both the upper and lower
computer anxiety halves performance increase on the Phase II task
compared with the Phase I practice task. These results somewhat
parallel the results for subjects computer anxiety scores. in which the
higher anxiety showed the treatment effect of lower anxiety immediately
after Phase I treatment, while the control group as a whole showed lower
anxiety after the Phase 11 treatment. There appeared to be a "ceiling

1 10
eﬁ'ect' as indicated by the high numbers of subjects who were able to

score "perfect 9’s" for the word processing tasks.

There were no signiﬁcant differences between the experimental and
control groups with respect to the time taken to complete any of the word
processing tasks. There were signiﬁcant diﬁ'erences within groups when
comparing time taken to complete the ﬁrst word processing task and
time taken to complete the two subsequent word processing tasks.
Subjects in both groups took signiﬁcantly less time to complete the
transfer tasks (WP 2) when compared to the practice task (WP 1).
Comparison of the Phase 11 word processing task (WP 3) with the Phase I
practice task was signiﬁcant for both the experimental and control
groups. indicating that at least some of the beneﬁts of task practice
endured over time. Comparisons of upper and lower pretest CAS half
groupings for both the experimental and control groups revealed that all
subgroups based upon treatment and computer anxiety took signiﬁcantly
less time to complete the Phase I transfer task than to complete the
Phase I practice task. Within the control group, the Higher anxiety half
of the group took signiﬁcantly longer to complete the Phase II task when
compared with the Phase I transfer task, and the lower anxiety half of the
control group took signiﬁcantly less time to complete the Phase II task
than they did the Phase I practice task.

Given the observation that there were no signiﬁcant differences
between the experimental and the control groups with respect to task
performance and time to complete the task for any of the word
processing tasks. the third experimental hypothesis is not upheld.

111
Hypothesis #4: The experimental group will indicate a greater

desire to learn more computer skills as measured by 1) subjects' estimates
ofthe number ofhours they would be willing to devote toﬁlrther word
processing skills study, 2) subjects' estimates of their interest in learning
other computer programs, and 3) subjects' estimates of the number of
hours they would be willing to devote to developing their computer skills.
than will an equivalent control group, when measured immediately after
instruction.

There were no signiﬁcant diﬁ'erences between the experimental and
control groups with respect to any of the questions pertaining to subjects'
desire to learn computer skills or word processing when measured
immediately following the Phase I treatment. The fourth experimental
hypothesis is therefore rejected.

Hypothesis #5: The experimental group will indicate a greater
desire to learn more computer skills as measured by 1) subjects' estimates
of the number of hours they would be willing to devote to ﬁrrther word
processing skills study, 2) subjects' estimates of their interest in learning
other programs. 3) subjects' estimates of the number of hours they would
be willing to devote to developing their computer skills, and 4) subjects'
attendance at a second computer session, than will an equivalent control
group, when measured dwing a one-monthfollow-up computer session.

There was only one signiﬁcant difference between the control group
and the experimental group with respect to the questions pertaining to
desire to learn computer skills or word processing when measured before
and after the Phase 11 treatment. The control group rated their interest
in learning computer skills, whatever they may be, higher that than did

112
the experimental group, when asked immediately after the Phase 11 word

processing training session. There were no signiﬁcant diﬁ'erences
between the experimental and control groups in the numbers of subjects
who dropped out of the study between Phase I and Phase 11. Based upon
both questionnaire and attendance data, the ﬁfth experimental
hypothesis is rejected.

0!-.-1' 0 PT ’_ 1'... .R' n ’ ,i-e_b‘. 015.121 '

The hypothesis testing performed thus far has examined within
W on repeated measures of the CAS to determine the
effect of the two treatments on subsequent CAS scores. In order to make
reliable tests of W with respect to the computer
anxiety measure, it is necessary to control for pretest group diﬁ'erences
on the Computer Anxiety Scale measure. Recalling the data summarized
in Table 12, pg. 62, a one-factor ANOVA comparison of the experimental
and control group revealed that there was a difference signiﬁcant at the
.05 level with respect to their Phase I pretest CAS scores (FM=4. 176,
p=.043). Thus, any attempt at between groups comparisons to determine
if one group scored signiﬁcantly more or less on a posttest CAS measure
that the other group would have to account for this pretest difference
between the groups. The analytical technique that allows for this
statistical control is the analysis of covariance, or ANCOVA (Glass and
Hopkins, 1984).

In order to apply the ANCOVA technique, the regression lines of
the covariate (CAS 1 , the Phase I pretest), and those of the outcome

113
measures (CAS 2, 3, 4) must have the same slope The ﬁrst step in the

test of this "colinearity assumption" is to regress each outcome measure
(CAS 2, 3, and 4) over the covariate (CAS 1) for the experimental and
control groups separately, yielding a total of six regression beta-
coefﬁcients. The next step is to calculate a Z-score for each pair of
experimental and control group beta-coefﬁcients according to the

 

following equation:
se3,+sec
where:
b. and bc = beta coefﬁcients for the experimental and control
groups
se. and sec = standard error of the experimental and control
group beta coefficients.

and check to see of the Z-score is signiﬁcant at the desired alpha level,
which in this case is set at .05.

The following example will clarify the test of the colinearity
assumption. Regressing CAS 2 over CAS 1 for the experimental group
yields the results in the ﬁrst beta-coefﬁcient table fi‘able 76). Regressing
CAS 2 over CAS 1 for the experimental group yields the results in the
second beta-coefﬁcient table (Table 77). Substituting the values for b., be,

se., and secin the above equation:

 

48863889
«/(.047)2+(.065)2

 

114

 

 

 

 

 

 

Table 76
Beta-Coefﬁcient for the Regression YCAS 2 over XCAS 1.
Experimental Group
variable: beta: S_E: t: p:
Intercept 2.353
Slope .886 .047 18.894 .0001
Table 77
Beta-Coefﬁcient for the Regression YCAS 2 over X05 1.
Control Group
variable: beta: ST): t: p:
Intercept 3.933
Slope .889 .065 13.696 .0001

we obtain a Z-score of —.0388 (p-.515), which is not signiﬁcant at the .05
level. Tests of the colinearity assumption for all relevant beta coefﬁcient
pairs were not signiﬁcant, thus the colinearity assumption is upheld and
the ANCOVA can be continued.

ANCOVA analysis was performed adjusting the experimental and
control group CAS outcome means to account for pretest CAS differences
between the two groups. Table 78 presents the sources of variance and
Table 79 the adjusted mean difference (control group mean minus the
experimental group mean) resulting from ANCOVA analysis adjusting the
Phase I posttest (CAS 2) group means.

115

Table 78
ANCOVA Table for Adjusting CAS 2 Means

 

 

 

 

‘Source: df: SS: MS: E-test: p:
Group (A) 1 6.789 6.789 .139 .6643
CAS 1 (B) 1 17901 .081 17901 .081 499.583 .0001
A x B l .027 .027 .001 .9781
Error 1 0 1 36 1 9.039 35.832

Table 79
Adjusted Mean Difference: CAS 2
Mean Diﬁerence (exp - ctrl) SE t: p:
1.674 1.192 1.404 .1634

The results of the ANCOVA analysis show that there were no
signiﬁcant differences between the experimental and control group CAS
Phase I posttest means when they were adjusted for the pretest CAS
differences between the two groups. Tables 80 and 81 present the
ANCOVA analysis adjusting the CAS 3 (Phase II pretest) means, and
Tables 82 and 83 present the ANCOVA analysis adjusting the CAS 4

(Phase II posttest) means.

116

Table 80
ANCOVA Table for Adjusting CAS 3 Means

 

 

 

 

 

 

gurce: (11': SS: M48: 7173mm: p:
Group (A) 1 1 1.928 1 1.928 .252 .6171
CAS 1 (B) 1 9593.694 9593.694 202.802 .0001
A X B 1 4.476 4.476 .095 .7593
Error 70 33 1 l .398 47.306

Table 81
Adjusted Mean Diﬁerence: CAS 3

Werence (exp - ctrl) §EL t: p:

1.142 1.614 .708 .4815

The results of the ANCOVA analyses revealed that, when
controlling for pretest differences in CAS scores, there were no signiﬁcant
differences between the experimental and control groups in the outcome
CAS scores measures either immediately after the Phase I treatment
(CAS 2), immediately before the Phase 11 treatment (CAS 3), or
immediately after the Phase 11 treatment (CAS 4) . Therefore, when
making Wu comparisons, it is clear that one cannot hold that
either the experimental or the control group treatments are better at
reducing computer anxiety. This clearly diﬁers from the results obtained
via the repeated measures analyses of the m comparisons, in
which diﬁ'erent patterns of treatment effects on computer anxiety scores

were noted for the two groups.

117

Table 82
AN COVA Table for Adjusting CAS 4 Means

 

 

 

 

—Source: (if? SS: MS: F-test: p:
Group (A) 1 4.284 4.284 .077 .7819
CAS 1 (B) 1 10252.045 10252.045 184.870 .0001
A x B 1 6.531 6.531 .1 18 .7325
Error 70 3881.886 55.456

Table 83
Adjusted Mean Difference: CAS 4

Werence (exp - ctrl) SE t: p:

.266 1.748 . 152 .8796

It is interesting to note the amount of adjustment of the
experimental and control group means which results from the ANCOVA.
Table 84 and 85 present a one-factor analysis of variance of the
experimental and control group CAS 2 means which would be the
preferred analytical technique for a between groups comparison had the
pretest CAS means for the experimental and control groups not been
signiﬁcantly different.

118

Table 84
ANOVA Table: Experimental vs. Control Group, CAS 2

 

 

Emce: (11': SS: M8: F-test: p:
Between 1 1262.105 1262.105 6.037 .0157
Groups
Within 103 21531.743 209.046
Groups
Error 104 22793.848 35.832

Table 85

Means, Standard Deviations, Standard Errors,
Experimental and Control Group, CAS 2

 

 

 

 

Group Means SD STE
experimental 37.926 ' 14.082 1.916
control 44.863 14.847 2.074

A comparison of the mean difference of 6.937 represented in Table
85 with the mean difference of 1.674 presented in Table 79 reveals the
extent of the adjustment of the CAS 2 group means resulting from the
ANCOVA.

WWW

There were no signiﬁcant differences between the experimental and
the control groups with respect to their Phase I posttest scores or their

Phase II pretest and posttest scores when one controlled for pretreatment

119
differences in the two groups with respect to computer anxiety. Thus.

the ﬁrst and second experimental hypotheses are not supported by the
13.111211 groups comparisons, while there is empirical support for the
ﬁrst hypothesis provided by the within groups, repeated measures

comparisons.

WW

It has been noted that there were signiﬁcantly more female
subjects in the study (Table 2). It was further noted that females scored
higher on the initial computer anxiety measure than did males. (Table 3).
By analyzing the CAS scores while attempting to account for the effects of
both goup and gender, it may be possible to determine if changes in
computer anxiety scores are attributable to gender, to treatment, or to an
interaction of gender and treatment. While such an analysis was not of
primary interest in this study, a brief exploration of possible gender
effects is warranted.

Tables 86 - 87 present the results of two-factor analyses of
variance comparing the experimental and control group CAS 1 and CAS
2 (Phase 1) scores while accounting for the effects of gender. In order to
perform this analysis, it was necessary to delete six male subjects from
the experimental group and three female subjects from the control group
so that cell counts could be equalized with respect to gender. As a
result, there were 34 female subjects each in the experimental and
control group cells, and 14 male subjects each in the experimental and

control group cells.

120

Table 86
Two Factor ANOVA Table for Gender and Group: CAS 1

 

 

 

 

 

-§ource: df: S? MS: F-test: p:
Gender (A) 1 943.46 943.46 4.219 .0428
Group (B) 1 914.941 914.941 4.091 .046
A x B 1 67.941 67 .941 .304 .5828
Error 92 20574.828 223.639

Table 87
Two Factor ANOVA Table for Gender and Group: CAS 2
Source: df: SSr: M8: F—test: p:
Gender (A) 1 863.61 863.61 4.038 .0474
Group (B) 1 1242.147 1242.147 5.809 .0179
AxB 1 41.314 41.314 .193 .6613
Error 92 19674. 139 213.849

It would appear that both gender and treatment have an effect on
CAS scores. There were, however, no interaction effects between gender
and treatment; the effects of treatment do not vary with gender. Thus,
any differences in CAS scores with respect to gender appear to be
attributable to the fact that female subjects started with higher CAS
scores than did males, and this difference appeared to persist after the
ﬁrst treatment.

Several limitations are readily apparent even from this cursory
analysis. First, given the prevalence of female subjects compared to male
subjects, the resulting imbalance in cell frequencies with respect to

gender renders this analysis suspect. Were we to attempt a two-way
analysis of variance that would use CAS 3 and CAS 4 as dependent
measures, then the male experimental and control group cells would
diminish in size to 8 per cell. It is obvious that this would result in
inadequate statistical power for any realistic analysis. In order to conduct
a more reasonable analysis that can account for possible gender effects,
then a balanced two-factor design is required (Glass and Hopkins, 1984),

and one would have to markedly increase the number of male subjects.

121

CHAPTER FIVE
DISCUSSION

landmine

In this ﬁnal chapter, the signiﬁcance of the study results is
discussed, followed by review of the limitations of the study, possible

directions for future study. and the ﬁnal conclusions.

W

The research idea presented at the beginning of this study was that
if brief instructional interventions designed according to Minimalist
Principles had positive effects on computer anxiety, leamers' abilities to
use computers. and leamers' desire to acquire computer skills, and that
those effects could be shown to endure over time, then this would
provide strong evidence for the efﬁcacy of using Minimalist Principles to
design computer skills training to teach computer anxious learners. An
implication of this line of reasoning was that if such instruction were
indeed efﬁcacious in the case of computer skills training, then it might
also hold promise in developing training not only for computer skills, but
also for technologies other than computers.

The evidence supplied by the results presented in the previous

chapter provides some evidence in support of the research idea.

122

123
The results offer some support for the hypothesis that the experimental

treatment would be more effective at lowering subjects' CAS scores than
would the control treatment. The between groups comparisons based
upon the ANCOVA did not support the experimental hypothesis. The
support came from one of the many repeated measures, within groups
comparisons. Learners in the experimental group who scored in the
upper 50% of their group on the pretest measure of computer anxiety did
indeed have statistically signiﬁcant, lower computer anxiety scores on
their subsequent CAS measures.

There is a very practical issue that must be addressed, however,
and that is the "so what?’ issue. There are no norms for the CAS,
therefore there is no basis for determining the meaning of a difference of
"x" points in computer anxiety. Depending upon the pairwise chosen,
subjects in the higher anxiety half of the experimental group experienced,
on average, between a 4 to a 6 point decrease in their computer anxiety
scores, or a 5%-7.5% "improvement." Until such time as CAS scores are
normed for this population of undergraduate students, it will be difficult
at best to ascribe meaning to such a difference in scores.

Nevertheless, the experimental treatment produced an immediate
decrease in anxious learners' computer anxiety when using easy, similar,
repeated tasks. For those instructional developers and educational
researchers who are interested in working to assist learners to feel more
comfortable with computers, in terms of reducing their computer anxiety,
this study appears to support the use of Minimalist Principles as a set of
guidelines for developing instruction responsive to the need of computer
anxious learners to reduce their computer anxiety as quickly as possible.

124
The results of the study with respect to the control group subjects'

computer anxiety scores must not be overlooked. There was evidence for
an overall decrease in computer anxiety for the control group subjects
based upon comparison of the Phase I pretest and the Phase II posttest
CAS administrations, especially for the half of the control group that
rated their computer anxiety lower on the pretest. This result suggests
that. with time and continued experience, leamers' computer anxiety will
abate somewhat, irrespective of the treatment.

A third factor should be considered in this discussion, the factor of
external reward. A signiﬁcant condition of the study was that subjects
received credit for their participation, credit which was required for
successful completion of their introductory psychology course. It was in
their interest as students to participate in the study sufﬁciently to receive
at least partial credit toward their course requirements. This condition
may have provided impetus for subjects to stick with the instructional
program sufficiently to learn some basic word processing skills, and to
feel more comfortable about using computers.

From this discussion of the study results thus far we can draw the
following conclusions:

1 . Computer instruction developed according to Minimalist
Principles of instructional design produced an immediate decrease in
anxious leamers' computer anxiety. While the meaning of that decrease
is not clear due to the lack of norms for the CAS measure, there is
nevertheless an immediate, measurable decrease.

2. Having computer anxious students use computers over time
appears to decrease their computer anxiety, irrespective of treatment. It

may be enough to get them to sit down in front of a machine and use it

125
to decrease their anxiety. While the effects of this kind of treatment may

not be immediately measurable, they do show up over the long run.

3. In order to get learners to sit down in front of the computer and
use it long enough to begin to experience a decrease in computer anxiety,
external rewards may play a signiﬁcant role. This conclusion is
necessarily weaker in that it was a byproduct of the study, and not the
result of a principle line of inquiry.

A fundamental issue needs to be addressed, and that issue lies
within the evidence for the eﬂicacy of mm the experimental and the
control treatments in training subjects to use word processing software
to complete a task. The results of the word processing task checklist
scoring and time to complete the tasks suggest that both treatments were
equally effective in training subjects to learn how to use word processing.
One may then ask, why is it necessary to consider the issue of computer
anxiety at all, when the end result is that learners learn how to use the
computer to perform word processing regardless of the type of instruction
used? On the basis of this study there is no simple answer. The ﬁrst
complication arises due to the presence of an external reward, that of the
credit subjects received for their participation in the study. Had this
reward not been present, then the intrinsic motivational effects of the
instructional treatments may have acquired more importance. This
study provided no means for addressing this issue. The point must be
made, however, that it is probably more frequently the case that external
motivators are at play in the case of learning computer skills than not at
play. People, including those who are computer anxious, learn how to
use computers for a variety of reasons: for example, to be able to qualify

for better jobs or positions, or to meet course or curricular requirements.

126
The presence of such external motivators may be sufﬁcient to "sit the

student in front of the machine" and get one to lay one's fears of the
technology aside in order to satisfy a strong motive.

A second consideration is the lack of difﬁculty of the tasks subjects
were required to perform. It has been noted that the tasks appeared to
be relatively easy for the subjects, given the prevalence of high
performance checklist scores. One may speculate as to the effect of task
difﬁculty on learners, and the need, if any, for addressing learner anxiety
when learners are faced with relatively easy tasks. Given that the
subjects in this study did not have to confront difﬁcult challenges with
respect to task, then perhaps the "user-friendliness" of the instruction
never arose as an issue. Upon reviewing the word processing task
materials. it is clear that subjects were strongly cued as the which
computer operations they needed to perform. Subjects were instructed
outright to use the tab function for indenting certain lines, for example,
or to use the centering function to center document titles. It may be
sufﬁcient to expose computer anxious learners to very easy, clearly
worded tasks early on in the learning process in order for learners to set
aside their initial computer anxiety.

A third consideration arises, that of the relative difference in the
designs of the instructional and control treatment. The fact that there
were no signiﬁcant differences between the experimental and control
groups with respect to the performance outcome measures leads one to
believe that the treatments, for all the differences described in the
methodology, may not have been that different in the eyes of the
subjects. Obviously, WordPerfect's "Getting Started" instruction was
sufﬁciently well designed to meet subjects' instructional needs effectively

127
for them to do well on the tasks. Whether or not the instruction was

designed to address the special needs of computer anxious learners is
ultimately irrelevant, given the fact that subjects acquired the skills.

Given that allstudy subjects were able to perform well on the word
processing tasks as measured by their checklist scores and time to
complete the tasks, irrespective of treatment, one may assert that any
instruction that is relatively well-prepared - clear in its procedural
explanations, setting easy tasks - may meet the instructional and
motivational needs of most individuals, computer anxious or not, well
enough such that computer anxiety need not be considered as a separate
issue.

However, there may always exist a number of individuals who will
try to avoid learning how to use computers or some other new technology
out of fear of that technology. It may not be enough merely to hold out
the reward of better job preparation or course performance in order to
"seat them in front of the machine." In those cases where anxiety
remains an issue, and where serious thought must be given to reducing
a person's anxiety, then the Minimalist Principles have been shown in
this study to provide a useful framework to glide the development of
instruction for such individuals.

With respect to the motivational questions to which students
responded, there was a lack of evidence supporting either of the
experimental hypotheses concerning the efﬁcacy of the experimental
treatment to promote subj ects' desire to learn word processing or other
computer skills. The only signiﬁcant difference found between the
experimental and control groups favored the control group: subjects in
the control group reported signiﬁcantly more interest in learning

128
computer skills irrespective of their kind, when asked immediately after

the Phase II experimental session. One explanation for this result could
be fatigue on the part of the experimental group, having been asked this
question for the fourth time, the time which produced the signiﬁcant
result. Another possible explanation could be boredom with the
experimental tasks, which were three variations on the same theme, and
this boredom could have been reﬂected in subjects responses to this
motivational question. Given the overall lack of variance in subjects'
scores on all the motivational question measures, one may set aside this

result as of little signiﬁcance.

iimiiaticnsnttheémdy

The study population was limited to undergraduate students
primarily in their ﬁrst year of matriculation. Any generalizations drawn
from this study must necessarily be limited to this population.

Female subjects were found to have signiﬁcantly higher pretest
computer anxiety than were male subjects as measured by the CAS. It
was never the intention of this study to explore gender differences with
respect to computer amdety and the effect of instructional treatments on
computer anxiety, thus, the design of the study never took this variable
into consideration.

While subjects were asked to work on a "realistic" task, or one that
they could possibly be called upon to perform in their -day-to-day lives,
the task was not a "real" task in the sense that it ﬁlled a real life need.
This injected a further measure of artiﬁciality into the study, as subjects

129
may or may not perform differently if they are called upon to perform a

real task. such as completing all their homework assignments for one
course, using word processing.

The tasks that subjects were instructed to complete were obviously
quite simple, given the prevalence of the "ceiling effect." This limited the
variance of checklist scores for the experimental and control goups and
made comparison of the efﬁcacy of the two treatments difficult at best.
Given the absence of large differences in treatment eﬁects, one must
question whether there were real differences in the experimental and
control treatments. The experimental and control treatments may have
been less different than initially assumed.

The strictly controlled conditions of the experiment, in which all
students were read the same instructions, and performed the same tasks
the same number of times, differ signiﬁcantly from work place or from
school environments. It is unclear how, if at all. the study results would
generalize to these other conditions.

The computer application used in the study was the WordPerfect
word processing software, and the platform was DOS-based.
Conclusions reached as a result of this study are not necessarily
generalizable to other software packages such as spreadsheets, nor to
other platforms such as the Macintosh.

CAS scores have a possible range from 20 - 100. Given the lack of
norms for the CAS, it is unclear what constitutes a signiﬁcant practical
difference in CAS scores. Does a 5-point decrement in CAS scores result
in a decrease in computer anxiety that has practical consequences for
the learner? Must a 10-point decrease be attained before there are

discernible effects on how a person feels about computers? Given the

130
lack of norms for the CAS, it is not possible to determine what

constitutes "high" computer anxiety from "moderate" computer anxiety.
and what the effects of different levels of computer anxiety are on
learners.

Analyses of 50th percentile half-goups based upon pretest Phase
CAS scores were subject to reduced statistical power due to halving the
numbers of subjects for purpose of within-percentile-goup comparisons.
Recalling the statistical power analysis presented at the end of chapter
three, true diiferences based upon percentile groupings may have been
missed due to insufﬁcient statistical power resulting in Type 11 error.

W

Future studies can take one of three directions: exploring further
the implications of the results of this study, addressing some of the
study‘s lirrnitations, or combining these two approaches.

Future exploration of the application of Minimalist Principles of
instructional design can focus on different computer applications, tasks.
and conditions. For example, does instruction designed according to
Minimalist Principles work as well for tasks and applications that are
primarily mathematical in nature, such as using spreadsheets or
statistical databases, as they do for word processing applications? Does
instruction designed according to Minimalist Principles work effectively
when tasks are more rather than less complicated? Does instruction

desigied according to Minimalist Principles work well in the workplace

131
with subjects working on real tasks, rather than in the computer lab with

subjects working on realistic tasks?

This question of whether or not the treatments were really different
can be addressed in at least two ways. First, the word processing task
difﬁculty could be increased, either by designing a more complicated task
such as composing a resume, imposing a time limit, or some
combination of features that would raise the level of difficulty. Second, a
more difﬁcult control treatment could be used in order to create the
widest possible potential difference between experimental and control
treatments. This could be accomplished, for example, by using an earlier
version of a software progam and its accompanying documentation,
documentation which most experts would agee has been poorly
desigred. By comparing Minimalist documentation with such a package.
one could determine the presence or absence of treatment effects. If
positive effects are attributable to Minimalist documentation under these
conditions, then parameters of the Minimalist documentation could be
varied in future experiments to determine which parameters best
account for the geater efficacy of the Minimalist documentation.

Future studies may want to use gender as a blocking variable in
order to study whether or not there are any differential treatment effects
that can be attributed to gender.

As mentioned in the discussion, there was insufficient power to
conduct the split-goup analyses based upon Phase I pretest computer
anxiety scores. Sufﬁcient samples of computer anxious learners are
needed to form experimental and control goups of adequate size in order
to maintain a desired level of statistical power when studying the eﬁ'ects
of treatments on subjects within this population.

132
Work needs to be done in standardizing the results of the CAS and

other computer anxiety measures. Until norms are discovered for the
CAS, it will be difﬁcult at best to determine the meaning, if any, of a
decrease in computer anxiety scores.

A question arises as to the applicability of the CAS to this
particular population of subjects. The copyright year of 1985 suggests
that CAS may have been developed at a time when microcomputers were
still relatively new to the market. Combined with the fact that a
signiﬁcant majority of the subjects reported previous experience using
computers. one must ask 1) was the CAS the appropriate instrument to
measure computer anxiety with this particular goup of subj ects. and 2)
how signiﬁcant an issue does computer anxiety remain in 1993?
Technologies that have attained widespread acceptance such as the
automobile or telephone were not initially embraced with open arms by
gateful consumers. Perhaps the microcomputer is becoming much less
of a novelty to the general public than it was ten years ago and,
especially among a generation raised with the microcomputer as a
ubiquitous tool, the phenomenon of computer anxiety is beginning to go
the way of anxiety previously associated with the "horseless carriage."

Lastly. research can be done to see of the Minimalist Principles can
be applied to technological domains other than computers. Mecharnics
who need to be trained on new automotive diagnostic technologies, or
factory workers who require new skills in manufacturing processes, are
only two examples of populations that could possibly beneﬁt from
instruction designed according to Minimalist Principles.

133
0992125199

This study provides support for the proposition that computer
skills instruction designed according to Minimalist Principles of
instructional design has a positive effect on computer anxious subjects
by reducing their computer anxiety while trairning them to acquire
computer skills. This results of this study also suggests that, given a
series of similar, very simple word processing tasks, subjects can learn to
perform those tasks irrespective of their level of computer anxiety or of
the types of instruction provided in the study. The possible lesson for
instructional designers and trainers who face the challenge of preparing
instruction for computer anxious students in the skillful use of current
technology and technologies as yet undeveloped, is to start your students
quickly on easy tasks, and do whatever it takes by way of reward to "sit
them down in front of the machine."

134
APPENDIX A

Control Treatment Instructional Materials

Instructional Materials

135
APPENDIX A

Control Treatment Instructional Materials

 

Table of Contents

 

will. SW The Basics ................................. 3
Registration ................................ 3
Install WordPerfect ........................... 3
The Template .............................. 3
Start WordPerfect ............................ 5
Clean Screen ............................... 6
Keys to Know .............................. 6
Function Keys .............................. 7
Pull-Down Menus ............................ 8
Menu Options .............................. 9
Codes ................................... 9
Exit WordPerfect ............................ 10

Getting Help ................................ 11
Help Feature ............................... 11
Reference Manual ............................ ll
WordPerfect Workbook ........................ ll
On-Line Tutorial ............................. 11
Customer Support ............................ 13
Ordering Templates ........................... l4

A Brief Lesson .............................. 15

136
APPENDIX A

Control Treatment Instructional Materials

 

 

The Basics

This section will introduce you to WordPerfect and some of its
basic functions and features. If you are new to WordPerfect, it
will be to your advantage to take a few minutes to read through
this section before you move on. If you are already familiar with
WordPerfect, you may want to glance through this section as an
introduction to some of the newer features that WordPerfect 5.1
has to offer.

 

MW" '

Before you move on, take a moment to fill out and send in the
Customer Registration card that came with your WordPerfect
package. Registering will make sure that you stay informed of the
latest releases, and entitles you to customer support as well as
WordPerfect’s quarterly newsletter, WPCorp Report. Keep the
upper portion of the form with your manual for quick reference to
your license number. If you are an update customer, you will
want to use the license number from your previous version of
WordPerfect. Your license number is needed for all disk updates.

 

Install
WIMP-moi

Included in your package is the WordPerfect Installation Program,
which guides you step-by-step through the WordPerfect installation
process. If you haven’t yet installed WordPerfect on your
computer. refer to the Installation Instructions card that came with
your package.

 

The template

If all journeys begin with a simple ﬁrst step, your experience with
WordPerfect should begin with the template, a WordPerfect menu
of feature options.

137
APPENDIX A

Control Treatment Instructional Materials

In your WordPerfect package, you received two color-coded
templates. one of which is designed to ﬁt over the function keys
on your keyboard. These templates correspond to both the
stande and enhanced IBM keyboards. Place the template on
your keyboard as shown in the photographs below.

 

> : '. : : e : i - - . , -
ﬂIIIIIIIIIII -
“IIIIIIIIII '-

 

 

 

 

 

The different colors are your guide to using the WordPerfect
function keys.

The color key is as follows:
Black Press the function key
Green Hold down Shift and press the function key
Blue Hold down Alt and press the function key
Red Hold down Ctrl and press the function key

Example: To Center. which is printed in green on the template,
you would hold down the Shift key and press F6.
If neither template ﬁts your keyboard. an address is provided for

requesting a new template in the Getting Help section immediately
following this section.

138
APPENDIX A

Control Treatment Instructional Materials

To remind you which key corresponds with each color. colored
stickers called keycals have been provided in your package. You
can, if you wish, peel the keycals from their backing and place
them on the corresponding keys (Shift=Green, Alt=Blue, Ctrl=Red)
on your keyboard.

 

WordPerfect

Once WordPerfect has been installed on your computer, you’re
ready to start the program. If you’re a new user. some things will
take some getting used to. but once you’re familiar with the
WordPerfect basics, the rest will fall into place.

Any time you are asked to Enter something throughout the documentation.
simply type the required information and press Enter.

III! lllsk
To start WordPerfect on a computer with a hard disk,

1 Turn on your computer and start DOS.

In most cases. the Disk Operating System (DOS) starts automatically when
you turn on your computer. DOS is software that helps your computer
communicate with WordPerfect. DOS must be started before any other
program can be used. including WordPerfect. If you need further
information. please see DOS and WordPerfect in Reference and refer to
your DOS manual.

2 At the DOS prompt, enter cd\directory name (directory name
meaning the name of the directory where WP.EXE is located).

WP.EXE is a file that helps you to start WordPerfect. You start
WordPerfect from the directory that contains WP.EXE. In most cases. the
Installation Program will have copied WP.EXE to a directory called

WP5 I, so you would type ch51 and then press Enter. WP.EXE will
be used every time you start WordPerfect and should not be deleted.

3 Enter wp
Remember. when you see Enter. type the required information (in this

case wp) and then press Enter.

Tm Disk llrlm

To run WordPerfect 5.1 on a two disk drive system, it is necessary that
each of your drives be at least 720K or larger. If you are not sure
whether your drives are at least 720K , please refer to your computer
manual or contact your dealer.

To start WordPerfect on a computer with a two disk drive system.

1 Start DOS. (See the paragraph on DOS in the Hard Disk
section above.)

2 Insert the WordPerfect l diskette into drive A.

139
APPENDIX A

Control Treatment Instructional Materials

3 Insert the diskette on which you want to store your ﬁles into
drive B.

Enter b: to change the default drive to B.
Enter map to start WordPerfect.

I When prompted. replace the WordPerfect l diskette with the
WordPerfect 2 diskette.

 

 

BIO. Screen Starting WordPerfect is like rolling a clean sheet of paper into a
typewriter. When you begin a new document. WordPerfect gives
you a clean screen to work from, which will be referred to
throughout the documentation as the editing screen or the normal
editing screen.

A CURSOR

A smus LINE ¢

A DOCUMENT

(1 OF 2)

A PAGE

A LINE

A POSITION

Doc 1 I l- 1' In. 1'

DA is A A
The dash near the top of your screen is the cursor. It points to
your current position on the screen. The status line at the bottom
of the screen displays the position of the cursor as well as
messages and warnings. When you save a document, the status
line can also display the ﬁlename of that document for as long as
you continue to work on it. (See Display Setup in Reference.)
Initial (default) formats, such as margins, tabs, and line spacing,
have already been set for you. You can, however, change them at
any time and as many times as you like throughout a document
(see Appendix G: Initial Settings for a complete list of initial
format settings).

Km to Km Before you begin experimenting, you’ll need to know where to

look for help in case you get lost or make a mistake. WordPerfect

140
APPENDIX A

Control Treatment Instructional Materials

is a remarkably forgiving program; here are a few keys to
remember in case you need them.

Backspace

The Backspace key is used to erase mistakes as you type. It
erases characters to the immediate left of the cursor.

Cancel

This key (Fl) backs you out of features that display a message on
the status line, such as Block, Exit, and Retrieve, as well as
backing you out of any menu. Cancel can also be used to restore
text that has been erased by any of the delete keys.

Delete
The Delete key (Del) deletes text at the cursor position.

Help

Pressing Help (F3) displays on-screen information about any
WordPerfect feature you are currently using. You can also use
this feature to display the WordPerfect template and/or a list of
features that begin with a particular letter. Pressing Enter or the
Space Bar will exit you out of help. (See Help in Reference for
more information.)

 

Faactlealeya

Function keys provide one gateway to WordPerfect features.
Throughout this manual and the WordPerfect Workbook, you will
often see key names. The guide to using these keys is simple:

- If a key name appears by itself (F8), press the key.

. When the key names are separated by a hyphen (Shift-F8),
hold down the ﬁrst key while you press the second key.

- If the key names are separated by a comma (Shift-F8,l),
complete the ﬁrst sequence (Shift-F8), release the keys, then
press the second key (1).

Function keys work in different ways, and present you with
different responses. Below are a few examples:

Feature Function Key How it Works

Bold F6 Turns the feature either on or
off

Print Shift-F7 Presents a menu of options

Center Shift-F6 Begins a feature that is

ended when Enter is pressed

Save F10 Requires e

141
APPENDIX A

Control Treatment Instructional Materials

The examples above refer to the original keyboard assignments.
You can. if you want to, change these assignments to suit your
needs. For example, the Help feature is currently assigned to the
F3 function key. You could change the assignment so that Help
would be assigned to the F 1 function key. For further information
on keyboard assignments. see Keyboard layout in Reference.

 

In addition to the function keys, there is another way you can
select WordPerfect features: Pull-Down Menus.

- wt leerch Layout lath Ibola I‘m: Graphic. lelp

noclnlul'mt‘

You can turn on the pull-down menus by holding down the Alt
key and pressing the equal sign key (=) (Alt-=). The pull-down
menu titles will appear in a bar at the top of the screen. You can
also set up the Alt key so it will display the pull-down menu bar
by pressing it alone (see Menu Options in Reference).

In WordPerfect, you can select features in the pull-down menus by
using the arrow keys (T,l,—+,+—) and pressing Enter, or by using a
mouse, if you have one connected to your computer. You can
also select these features by typing the mnemonic letter that is
highlighted in the feature name (e.g., S for Save). The pull-down
menus can also be turned on and off by clicking the right-most
button on the mouse.

If you don't have a mouse connected to your computer and set up to run
with WordPerfect, the mouse pointer (I) will not appear. See Mouse
Setup and Mouse Support in Reference for further information.

142
APPENDIX A

Control Treatment Instructional Materials

If you prefer to use the pull-down menus, look for this symbol:
CB. Throughout the documentation, this symbol will be followed
by the name of the pull-down menu where you can ﬁnd the
feature.

Example:
CB Select Append from the Edit menu.

 

Iaaanetlaaa

Earlier in this section, when you were introduced to function keys,
you learned that some features, such as Print (Shift-F7), present
you with a menu of options.

-Nllnoa-.e

‘ '09.

- m GI Diet

- Cmtml Printer

- multiple Peooa

- It“ 00cm:

- lntueltu Printer

COCOUNI-

ens-

- Detect Printer Bend-i Fri-e-
- Iindino 011.0: 0'

- m:- of Copiea I.

- hltipia Cooiea Caner-cad by m

- Graphica Quality w-

- hat Quality Ilﬁ

'IOQII.

lot-catc- : 9

Whenever you are presented with one of these menus, there are
three ways that you can select an option:

0 Type the option number, such as (l) for Full Document. (Use
the numbers at the top of the keyboard as opposed to those on
the number pad.)

- Type the mnemonic letter that is highlighted in the option
name (e.g., P for Page).

- Position your mouse pointer on the option and click the left
mouse button.

 

Codes are commands that tell WordPerfect and your printer what
to do. Whenever you use a feature such as Block or Underline,
WordPerfect inserts these hidden commands into your document.
You can examine these codes in the Reveal Codes screen. For
information, see Appendix C: Codes for a list of all codes, in
addition to Reveal Codes and Delete Codes in Reference.

143
APPENDIX A

Control Treatment Instructional Materials

 

lerﬂartect

When you ﬁnish using WordPerfect, you must exit before turning
off the computer.

1 Press Exit (F7) to begin exiting WordPerfect.

CB Select Exit from the File menu.

2 Press y to save the document, then enter a ﬁlename.
or

Press n to continue exiting without saving the document.

3 Press y to exit WordPerfect.

There are several saving options. For further information. see Save in
Reference.

144
APPENDIX A

Control Treatment Instructional Materials

 

Getting Help

If you have problems as you work with WordPerfect, help is
available from several sources.

 

Heb Featle The Help feature (F3) displays on-screen information about any
feature you are currently using. For more information, see Keys to
Know in The Basics section of Getting Started and Help in

 

Reference.
Bafﬂe-ca This manual contains an alphabetical list of WordPerfect features,
leelel describing how each feature works. Each section contains steps

and information pertaining to individual features. as well as more
detailed information under the Notes heading at the end of each
section. See Also will direct you to related information. A
comprehensive index is provided to direct you to the information
you need, as well as an Appendix devoted exclusively to
troubleshooting (see Appendix 0: Troubleshooting for further help).

 

W The WordPerfect Workbook contains several lessons which show

W you how WordPerfect features may be used and combined for
speciﬁc applications. Lessons include advanced applications as
well as basic fundamentals.

 

Ill-ﬂee Teterlal WordPerfect includes a tutorial that takes you step-by-step through
a number of basic skills. The tutorial is intended for use with
IBM personal computers and 100% compatibles only.

Important: The On Line Tutorial is installed with the WordPerfect
learning ﬁles. If you did not instal the Learning files when you installed
WordPerfect or have not subsequently done so. you must do so before
running the Tutorial (see the Installation Instructions card).

lll'd lllak
To start the tutorial on a computer with a hard disk.

1 Start DOS on your computer. (See the paragraph on DOS in
the Hard Disk section of Start WordPerfect in The Basics.)

2 Type path, then press Enter.

The path is a listing of the directories that are searched when you
want to execute a program from the DOS prompt. The directories
are searched in the order listed (see DOS and WordPerfect in
Reference for further information).

145
APPENDIX A

Control Treatment Instructional Materials

3 Check the path to see if it contains the directories where
WP.EXE, WP.FIL, and the Learning ﬁles are located.

The WordPerfect Auto-Install Program automatically asks you if you want
these directories entered into your path. If you inserted these directories
upon installation. they should now appear in the path. The directories
are C:\WP51 and C .'\WP5 I\LEARN. If your path contains these
directories, move on to step 6.

4 If your path doesn't contain the proper directories, type path:
then type the same directories that were listed in the original
path plus the directories where WP.EXE, WP.FIL, and the
Learning ﬁles are contained.

In the PATH command, the directories must be separated by a semicolon
( .') (e.g., PATH=C:\WP51 ,‘C:\WP51\LEARN).

This PATH command remains in effect until you turn off your
computer. If you want to use the PATH command permanently,
you should include it in an AUTOEXECBAT ﬁle (see DOS and
WordPerfect in Reference).

5 After you have ﬁnished typing the PATH command. press
Enter.

3 Enter tutor to start the tutorial.

Read the messages on the screen and follow the instructions as
directed. When you exit out of the tutorial, you are returned to
the same DOS prompt.

If you have problems starting the tutorial after following these
steps, try exiting out of any program that may still be running,
then try step 6 again. If you’re still having problems, go to the
DOS prompt before you enter tutor and enter the following
command:

set wp=lnklnc

These startup options disable the Cursor Speed feature and
enhanced keyboard calls and remain in effect until you turn off
your computer or use a new SET command. See your DOS
manual and Appendix N: Startup Options for more information.

Tue lllslr can
To start the tutorial on a two disk drive system,

1 Start DOS on your computer. (See the paragraph on DOS in
the Hard Disk section of Start WordPerfect in The Basics.)

Insert the WordPerfect I diskette into drive A.
Insert the Learning diskette into drive B.

146
APPENDIX A

Control Treatment Instructional Materials

4 Enter path=a:\;b:\ at the prompt.

The path lists the directories that are searched when you want to
execute a program ﬁle from the DOS prompt. The directories are
searched in the order listed.

This PATH command remains in effect until you turn off your
computer. If you want to use the PATH command permanently,
you should include it in an AUTOEXECBAT ﬁle (see DOS and
WordPerfect in Reference).

5 Enter tutor to start the tutorial.

8 When prompted, replace the WordPerfect l diskette with the
WordPerfect 2 diskette.

Read the messages on the screen and follow the instructions as
directed. When you exit out of the tutorial, you are returned to
the original DOS prompt.

 

WordPerfect is backed by a customer support system designed to
offer you fast, courteous service. If you’ve exhausted all other
Help avenues and need a friendly voice to help you with your
problem, follow these steps:

- Try to duplicate the problem, keystroke by keystroke, to see
exactly what was done.

' Be at your computer when you call Customer Support. Have
your manual and license number handy. Also know the model
of printer you are using.

0 Run the WPINFO ﬁle to gather information about your setup
that Customer Support may need (see Appendix 0:
Troubleshooting).

If you are within the United States, Puerto Rico, the US. Virgin
Islands. or Canada. toll-free? support is available by dialing:

(800) 533-9605 5.1—Installation
(800) 321-3383 5.1—Graphics
(800) 541-5097 5.1—Printers
(800) 541-5096 5.1—Features
(800) 321-3389 5.1—Networks

* This is for English versions of WordPerfect 5.]. If you have a French
version. consult your manual.

147
APPENDIX A

Control Treatment Instructional Materials

If you are in an area where the phone system does not handle toll-
free numbers, you can reach Customer Support by dialing:

(801) 226-4770 Graphics
(801) 226-7977 Printers
(801) 226-7900 Feature
(801) 226-4777 Networks
(801) 226-5444 Installation

You will be charged by the phone company for the call if you use the 801
number.

For IBM personal computers and compatibles, the customer support
department takes calls Monday through Friday, from 7 am. to 6
pm. Mountain time. For all other computers, support hours are
from 8 am. to 5 pm. Mountain time.

If you purchased this product within the US. and would like to update it
outside the U .S ., thereby receiving free customer support and update
notices locally, you will be charged a maximum of $150 (US) or 25% of
the local list price, whichever is greater.

 

tantataa

If neither template ﬁts your keyboard, return both of them along
with the name (and a photocopy or rough sketch) of the keyboard
you are using to:

WordPerfect Corporation
Attn: Information Services
1555 N. Technology Way
Orem, Utah 84057

WordPerfect Corporation may be able to create or provide a
template that will ﬁt your machine.

148
APPENDIX A

Control Treatment Instructional Materials

A Brief lesson

Now that you’ve gone through the basics, here is a brief lesson
designed to help you get to know WordPerfect. It will introduce
you to some WordPerfect basics and features that you might
typically use. To begin, start WordPerfect as previously described.

 

Greatlagthe

Equipped with the basics. you’re now ready to start typing. The
document you create will be a simple note. If you make any
mistakes, use the Backspace key to erase them.

You'll notice that when a line ﬁlls with text. the cursor returns to
the left margin in a new line. This automatic return is known as
word wrapping.

1 Type the following note without pressing Enter at the end of a
line.

Remember the discussion we had about starting a study
abroad program? I’ve just returned from a study abroad
conference in Illinois, and I have an outline I’d like you to
take a look at. Why don’t you stop by my ofﬁce the next
time you’re at this end of campus, and we can talk about
it.

 

You’ve just created your ﬁrst document. Looking it over, you see
a few things that you’d like to change. WordPerfect has several
features that make editing quick and efﬁcient.

Harlan the Career

You can move the cursor on your screen by using either the Up
and Down Arrows (T/l) or the Left and Right Arrows (-—/—i) on
your keyboard. To move the cursor with a mouse, simply position
the pointer where you want the cursor to be, and click the left
mouse button.

Some keyboards have a separate number pad that is also used for cursor
control in WordPerfect. The Nutn Lock key (or its equivalent) operates
as a toggle to turn Num Lock on and of. When Num Lock is turned off,
the number pad can be used for cursor control.

1 Using the arrow keys, move the cursor to the uppercase R in
Remember.

149
APPENDIX A

Control Treatment Instructional Materials

iaeert re. Typeeeer

WordPerfect uses Insert as the main form of editing. While in
Insert, any text you type is inserted at the cursor, moving the
existing text to the right. The alternative is to type over existing
text by using Typeover.

2 Type Do you and press the Space Bar.

3 Press Insert (Ins). Notice that the word Typeover appears at
the bottom left of the screen.

Type a lowercase r to replace the existing R.
Press Insert to end Typeover.
Move the cursor to the h in have in the second sentence.

Press Delete (Del) until the remainder of the sentence
(including the period) is erased.

NC“.

Many functions on the WordPerfect keyboard will repeat if you simply

hold down the key. Use caution in doing this until you get used to the
keyboard; holding down a key too long can insert unwanted codes into
your document and delete other necessary codes.

8 Type think I’ve come up with an outline that will make us
all very happy.

9 Move the cursor to the v in very.
lieck aal Underiiee

The Block feature is used to deﬁne portions of text for use with
other features.

10 Press Block (Alt-F4), then press Right Arrow (a) until the
word very has been highlighted.

CB Select Black from the Edit menu. then press Right Arrow (a) until
the word very has been highlighted. If you want. you can also hold
down the left button on the mouse and drag until the word is blocked.

11 Press Underline (F8) to underline the word.
Select Appearance from the Font menu. then select Underline.

You can block characters, words, sentences, and entire pages if
you wish. If you had blocked the rest of the note from the word
very forward, the entire block would have been underlined.

12 Move the cursor to the D in Do at the beginning of the note.

13 Press Enter two times to add blank lines at the beginning of
the note.

150
APPENDIX A

Control Treatment Instructional Materials

mun-mm

The Home key can be used for quick movement throughout your
document.

14 Press the Home key twice, followed by Up Arrow (T). This
returns you to the top of your document.

15 Type Mitch,
10 Return to the D in Do at the beginning of the note.
11 Press Tab.

lit.-

90 you Mr the dimeeien ue had about atarttnq a atudy
abroad proqrea? 1' ve juat returned (roe a atudy abroad conterence
in Illinoie, and I think 1' ve cone up with an outline that will
lake ua e11. am happy. why don' t you atop by ay office the next
tinem'h'eetthiaendotcaapue anduacantaihebouttt

Oocl'qllnl.)3"°l1.§'

Your edited document is complete.

 

Saving!“

You’ve typed the note and now you want to save it for future
reference. When you save a document in WordPerfect, all
document initial settings are saved as well, including printer
selection.

There is more than one way to save a document in WordPerfect,
but for now we’ll use the Save feature.

1 Press Save (F10).
a Select Save from the File menu.

A message appears on the status line at the bottom of the screen:
“Document to be Saved:”

2 Type study as the ﬁlename, and then press Enter.

The name of your document and the directory in which it is
located appear on the status line at the bottom of the screen. This
ongrnal document is now saved. Any changes you make

151
APPENDIX A

Control Treatment Instructional Materials

throughout the rest of the lesson are only temporary, and will not
affect the original document unless you save them under the same
ﬁlename at some future time.

 

Now that you’ve edited and saved your document, you can send it
to the printer. You can print from the screen or from disk. From
the screen, you can print a document, page, or any block of text.

For this section of the lesson, you will print directly from the
screen. To ﬁnd out how to print from a disk. refer to Print.
Document on Disk in the Reference section of this manual.

1 Press Print (Shift-F7) to display the Print menu.
CB Select Print from the File menu.

2 Select Full Document (1) from the Print menu to send the note
to the printer.

If your printer does not print, check to make sure it is turned on.
on-line, and that the printer cable is attached securely to your
computer and printer. If you did not select a printer when
installing WordPerfect, you may receive an error message.

For details on selecting a printer see Printer, Select in Reference.

 

Saleem
llaceaaetami

You’ve saved your original document once. To show you another
way to save your document, and also exit WordPerfect, you’ll save
the document under the same ﬁlename, using Exit.

For information on other saving and ﬁling options, see Save and
List Files in the Reference section of this manual.

To save the document and exit WordPerfect,
1 Press Exit (F7).
CB Select Exit from the File menu.

A message appears on the status line at the bottom of the screen
asking you if you want to “Save Document? Yes (No).”

2 Type y, and press Enter to save it under the STUDY ﬁlename.

Simply pressing Enter will automatically select the default option (the
option not in parentheses).

Since there is already a document with the ﬁlename STUDY, the
message on the status line asks if you want to replace the
previously-saved document. Though the text has not been

152
APPENDIX A

Control Treatment Instructional Materials

modiﬁed since you last saved the document, for the purpose of this
lesson. you'll want to replace it. If you had changed any text, the
new version of the text would replace the old version.

3 Type y.

The message on the status line asks if you want to “Exit WP? No
(Yes).”

4 Type y to exit WordPerfect.

Now that you've ﬁnished this lesson, you may want to look
through the Reference section for more information on individual
features. or turn to the WordPerfect Workbook for a more in-depth
introduction to WordPerfect.

153
APPENDIX B

Experimental Treatment Instructional Materials

Instructional Materials

um mar-1mm mil-«rm
hate-o tux-i Mr um

um
mu

mm
M

Memo
tut-i

.mrown.

ﬂuid

154
APPENDIX B

Experimental Treatment Instructional Materials

1. Getting Started - Opening the Application

a. At the C-prompt (the line where you type In commands), type "WP" g3
without quotations marks. It should look like this:

C:/>WP

or this:

C>WP

b. Wait a few seconds. Eventually a blank screen will appear, along with
some information in the lower right hand corner that looks like this:

 

Doclpgllin1”Pos"

 

 

 

You are now ready to begin work.

155
APPENDIX B

Experimental Treatment Instructional Materials

2. Entering Text, Using the "Cursor" to Move Around a Document

Entering text:

You enter text into the computer much like you would with a typewriter, but
with an exception. Hit the "return/enter" key only if you want a space between lines.
If you don't want a space between lines, then just continue typing when you get to
the end of a line - WordPerfect will automatically continue entering text on the next
line for you.

when m Bum
=1!th

For example, the text in the above paragraph was entered without using the
”return/ enter" key when coming to the end of a line. After the word "but" in the
ﬁrst line, the program automatically moved to the second line and continued
entering "with an exception" without the writer using the "return/enter" key.

To get the space between the ﬁrst and second paragraphs, however, the writer
hit the "return/enter" key two times after the phrase "the next line for you" and
began a new paragraph. That created the space between the two paragraphs.

Using the "cursor" controls:

A "cursor" is the blinking line which indicates your position in a document.

If you look at the right side of the keyboard, you'll ﬁnd some ”arrow keys,"
either by themselves, or as part of the #2, #4, #6, and #8 keys on the "numeric
keypad," which is the set of numbered keys on the far right of the keyboard. You can
use these keys to move around a document, either on line up and down, or one
character left or right, at a time. They control the position of the cursor.

156
APPENDIX B

Experimental Treatment Instructional Materials

3. Deleting Text
There are several ways of deleting text:

1. Use the ”back space" key. The cursor will move to the left, erasing letters
as it goes.

2. Use the "delete" key (sometimes marked ”del"). The cursor will stay in
place, and the letter to the right of the cursor will be erased.

Mai-o

157
APPENDIX B

Experimental Treatment Instructional Materials

4. Using "Function Keys” to Enter Commands.

You enter commands such as "BOLD” and "PRINT" by using "function keys."
These are special purpose keys that are located either along a row at the top of the
keyboard, or in a bank of two columns on the far left side of the keyboard. They are
labeled "F1" through "F10" or "F12," depending upon the keyboard. Here are some
examples commands and their associated function keys:

BOLD: F6
UNDERLI NE: F8

ATTENTION: Look in the appropriate section (cg., "printing") for
instructions on how to used specific commands.

ii

Function keys are pressed in order to enter commands. There are many more
possible commands to enter than there are function keys, however. WordPerfect
gets around this limitation by allowing the user to press mm of function
keys and, either the control key (abbreviated "ctrl"), or the alternate key (abbreviated
"alt"), or the shift key. Here are some examples:

CENTER TEXT: shift F6
PRINT: shift F7
SPELL CHECK: ctrl F2

In order to execute the previous commands, you would press and hold down
the "extra" key (ctrl, alt, or shift) ﬁrst, and then press the proper function key.

Throughout this tutorial, a command will be printed in capital letters, and the
keys you should press to enter that command will follow in parentheses, for
example:

PRINT (shift F7)

158
APPENDIX B

Experimental Treatment Instructional Materials

5. Remembering Commands: Using the Template

It is not easy, especially when learning, to recall the commands for
WordPerfect. Because of this, the program comes with a nifty little template that
lays over the function keys, and serves as a memory prompt for all the possible
function key commands. There is one for each type of keyboard:

Both templates are color coordinated to allow you to determine the wow
key combination to use for the desired command:

ctrl: red

shift: green

alt: blue

function key alone: black

For example, if you locate command ”PRINT,” it appars in green, meaning
that you press the shift key and the F7 key to print your document.

 

   
 

 

NOTE: The real templates are
color-coded to dbtingutsh betw
different key combinations.

 

Example:

 

Print 8 ft P7;

 

 

 

 

 

 

 

 

 

up
Who-mall-
s~.._: as: *-
us
& ... than“ .
an all lun- :- lums- }\ s... la a: m use.
nan- ~— mm an a su- um>mi an a s— I.
u - u I“ I.. it H I. u “U In “h h
n I! a n as l n n n n no m m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

159
APPENDIX B

Experimental Treatment Instructional Materials

6. Using Tabs and Centering Text

nging me tab key:

Whenever you have to indent to begin a new paragraph use the "tab key"
rather than the space bar. If you rely on the space bar for formatting your document,
the printer may make a mess of your careftu entered spacing.

Each press of the tab key will move the cursor over five spaces.

Centering text:
To center a word or a line of text that's already been entered:

1. Place the cursor at the beginning of the word or block.
2. Next, hit the CENTER (shift F6) command.

3. Finally, use the cursor keys to move to a new area of the page to continue
working.

To center a word or line of text before entering it: i?

1. Make sure the cursor is the beginning of the line where you will be
entering the text.

2. Hit the CENTER (shift F6) command.
3. Enter your text.

If you will be centering more than one line you will need to enter the
CENTER command at the beginning of each line.

160
APPENDIX B

Experimental 'h'eatment Instructional Materials

7. Undoing a Deletion, Escaping From a Command

Undoing text you've mistakenly deleted

If you delete some text by mistake, WordPerfect has a means of restoring it.
The program will store the last 3 deletions. You can ”call back" the deletions and
restore the desired one.

1. Enter the "CANCEL" (Fl) command. The most recently deleted text will
be displayed in its original position, highlighted.

2. Use the up arrow key to display the 2nd and 3rd most recently deleted text.
Each will be displayed in turn, and any other highlighted, deleted text will
disappear.

3. If you accidentally ”go too far,” eg., display the 3rd deletion when you want

to undo the 2nd deletion, use the down arrow key to re-display the desired
deletion.

4. Select "UNDELETE" (l) to restore highlighted text.

"Escaping" an unwanted command

If you enter a command (eg., ”PRINT” "CENTER? by mistake, one that you
don't want to execute, you can frequently cancel the command by pressing the "ESC"
or "escape" key immediately after entering the unwanted command.

Own.

161
APPENDIX B

Experimental Treatment Instructional Materials

8. Printing a Document

1. Enter the PRINT (Shift F7) command.
2. The print menu displays on the screen.

3. Select FULL DOCUMENT (1), PAGE (2), or MULTIPLE PAGES (3) to print
the whole document, the current page, or selected pages.

4. Wait for printer to print.

MIN

162
APPENDIX B

Experimental Treatment Instructional Materials

9. Exiting the WordPerfect Program
Once you've finished a work session and want to quit using WordPerfect:

1. Press theEXITUV) key.

2. This message will appear at the bottom of the screen:
Save document? Yes (No)

Press the "N" key, since you don't want to save your document at this
time.

3. A final message will appear at the bottom of the screen:

Exit WP? No (Yes)

Press "Y" and you will exit WordPerfect and return to the C> prompt..

mm
W3

163

APPENDIX B

Experimental 'h'eatment Instructional Materials

10. Summary

Desired Action

open Word Perfect

moving around document

deleting text

quitting WordPerfect

undoing

escaping an unwanted command

printing

Commandsl Resmnses to Prompts,
Min—Cm

WP
-arrowkeys

-del key
-backspacekey

- F7 (EXIT command)
- press "11" - document won't be saved

- press ”y” to exit program

- F1 (CANCEL command)

- use arrow keys to display deletion
want to undo

- select "Undelete" (1) to restore

-ESCkey

- shift P7 (PRINT command)
- select page, document, multiple pages

humans

164

APPENDIX C

Computer Anxiety Scale

COMPUTER ANXIETY SCALE (CAS)

The items in this questionnaire refer to things and experiences that may cause anxiety or
apprehension. For each item. darken the circle on the answer shoot corresponding to the

letterwhich incicates .
butbe suretoanswereach item individually.

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1. Thlnldngabouttakingaciassin A.Not B. A C.Falr D.Much E.Very
computer language (e.g. BASlC. at all little amount Much
Pascal. COBOL. etc).

2. Applying for a job that requires A. Not B. A 0. Fair D. Mich E. Very
some training in computers. at all little amount Much

3. Sitting in front of a home A. Not B. A 0. Fair D. Much E. Very
computer. at all little amount Much

4. Being around people who are A. Not B. A C. Fair D. Much E. Very
'into' computers. at all little amount Much

5. Watching a movie about an A. Not B. A C. Fair D. Much E. Very
intelligent computer. at all little amount Much

6. Looking at a computer print-out. A. Not B. A C. Fair D. Much E. Very

at all little amount Much

7. Getting ”error” messages from the A Not B. A C. Fair D. Much E. Very
computer. at all little amount

8. Watching or listening to news
programs about the increasing role A Not B. A C. Fair D. Much E. Very
of computers in society. at all little amount Much

9. Watching someone working at a A. Not B. A C. Fair D. Much E. Very
computer terminal. at all little amount Much

10. Being refused information because A. Not 8. A 0. Fair D. Mich E. Very
a terminal is “down”. . at all little amount Much

 

 

COMPUTER ANXIETY SCALE CONTINUED

165

APPENDIX C

Computer Anxiety Scale

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Used with penn‘sslon.

Page 2

11. Talking to a computer A. Not B. A C. Fair D. Much E. Very
programmer. at all little amount Much
12. Leaming to write computer A. Not B. A C. Fair D. Much E. Very
programs. at all little amount Much
13. Using a typewriter. A. Not B. A C. Fair D. Much E. Very
at all little amount Much
14. Visiting a computer store. A. Not B. A C. Fair D. Much E. Very
at all little amount Much
15. Attending a workshop on the uses A. Not B. A 0. Fair D. Much E. Vary
of computers. at all little amount Much
16. Erasing or deleting material from a A. Not B. A C. Fair D. Much E. Very
computer. at all little amount Much
17. Thinking about prepackaged A. Not B. A C. Fair D. Much E. Very
(soltware packages) programs for at all little amount Much

a computer.
18. Taking a class about the uses of A. Not B. A C. Fair D. Much E. Very
computers. at all little amount Much
19. Learning computer terminology. A. Not B. A C. Fair D. Much E. Very
at all little amount Much
20. Looking at a high speed computer A Not B. A C. Fair D. Much E. Very
printer. at all - little amount Much

Copyright 1985 by George A. Marcoulides. Calilomia State University Fullerton. California. USA

 

166
APPENDIX D

Background Information (Phase I)
Name

 

1. Please check the category that most closely describes your class standing:
first year
second year
third year
fourth year
____lifth year
sixth year

more than sixth year

2. Please write your age in years:

3. Please indicate your sex:

male female

4. Please record your grade point average:

5. Have you ever taken a course in computer literacy or usage. or a course in computer programming
(including high school)?

yes __no

6. If your response to question 5 was “yes.“ how many semesters of total course work in computer literacy.
usage or programming have you had (including high school)?

less than a lull semester or term

one semester or term

two semesters or terms

three semesters or terms

four semesters or terms

more than four semesters or terms
continued on other side

167
APPENDIX D

Background Information (Phase I)

7. How would you describe your current word processing skills (any type of computer)?

 

 

 

 

none poor average good __excellent
8. Please describe your current typing ability:
poor average 90°d

 

9. Please indicate your academic major. If not decided. please write 'not decided.“

 

10. it an opportunity were to arise. how much time would you be willing to devote to learning word
processing skills? .

none

 

1 hour or less

2-3 hours

 

__ 4-5 hours

more than five hours

 

11. How interested are you in learning computer skills, whatever they may be?

not at all

 

a little

 

somewhat

 

alot

 

12. How much time would you be willing to devote to learning computer skills, whatever they may be?

0000

 

1 hour or less

 

2-3 hours

 

4-5 hours

 

more than live hours

 

168
APPENDIX E

Background lntonnatlon (Phase II)

 

1 If an opportunity were to arise. how much time would you be willing to devote to learning word
processing skills?

none

1 hour or less

2-3 hours

 

4-5 hours

more than five hours

2. How interested are you in Ieaming computer skills. whatever they may be?
not at all

a little

 

somewhat

alot

3. How much time would you be willing to devote to Ieaming computer skills, whatever they may
be?

none

 

1 hour or less

 

2-3 hours

 

4-5 hours

 

more than five hours

 

169
APPENDIX F

Feedback Form

 

1. Were there any features of the instruction that you particularly enjoyed or appreciated?

2. Were there any features of the instruction that were particularly frustrating for you?

3. If an opportunity were to arise, how much time would you be willing to devote to learning word
processing skills?
none
1 hour or less

2-3 hours

 

4-5 hours

 

more than five hours

4. How interested are you in learning computer skills, whatever they may be?

not at all

 

a little

 

somewhat

 

a lot

 

5. How much time would you be willing to devote to learning computer skills. whatever they may be?

none

 

1 hour or less
__ 2-3 hours
4-5 hours

more than five hours

PM

PM

PM

PM

170

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.APTHENHDEKl3

Data Collection Schedule
Mon 1/25 Tues 1/26 Wed 1/27 Thu 1/28
5:30 - 11:30 9:30 - 11:30
Group 1 Group 4
5:00 - 7:00 2:00 - 4:00 6:00 - 8:00
Group 2 Group 3 Group 5
Mon 2/8 Tues 2/9 Wed 2/10 Thu 2/11
9:30 - 11:30 9:30 - 11:30
Group 6 Group 9
5:00 - 7:00 2:00 - 4:00 6:00 - 8:00
Group 7 Group 8 Group 10
Mon 2/22 Tues 2/23 Wed 2/24 Thu 2/25
9:30 - 11:30 9:30 - 11:30
Group 1 Group 4
5:00 - 7:00 2:00 - 4:00 5300 - 3:00
Group 2 Group 3 Group 5
Mon 3/3 Tues 3/9 Wed 3/10 Thu 3/11
9:30 - 11:30 9:30 - 11:30
Group 6 Group 9
5:00 - 7:00 2:00 - 4:00 6:00 - 8:00
Group 7 Group 3 Group 10

 

 

 

o w 0 ”WV

m w W U”U

171
APPENDIX H

Consent to Participate in a Research Study

I volunteer to participate in this study related to motivation to learn computers. I
understand that this study is being conducted by Christopher B. Reznich, a
doctoral candidate and instructor at Michigan State University. The purpose of
the study is to determine the opinions of adult learners regarding computers
before and after self-instruction. I understand that participation in this study is not
a part of a MSU course, is completely voluntary, and that participation will in no
way affect my grade. i understand that the information I give is completely
conﬁdential and complete confidentiality will be maintained throughout the
experiment. I understand that I am free to discontinue the experiment at anytime
during the process and that there are no guaranteed achievement results
because of my involvement in the study. I understand that I will not be penalized
for non-participation or for discontinuing my participation in the study. I
understand that the maximum estimated time required to complete the study is
approximately 4.5 hours.

 

 

 

 

 

Signature of Participant Signature of Researcher
Printed Name of Participant Printed Name of Researcher
Date

 

Telephone Number

172
APPENDIX I
Machine Scoring Form

.lPPPFPPh__—

_ -..l— -... _l._ .._l—.-P

 

 

 

 

 

 

 

 

 

 

 

 

 

$8.24 >80 {>88 <06, 85m 1 _ U
3:: 53 22.3 m... >_ h _ a HM ha

L_E_L_;_L Q®®©®®®®®
eeoeeoeoeee assesses
eecccccecee eeeoeeee
oooooooeooe assesses
eeeeceeeeee assesses
ooeeeoecooe eeeeeeee
oeooeeoeeoo ceeceeec
eecoeeeeeee assesses
eeeeeeeeeee eeeoeoee
eeoeeeeeeee eeeceooe
eeeeeeeeeee ass
oooooeooece e_
ooooooeoeae see a
eeeeeeseees see a
eeeeeeeeeee see ms
eeeeeeeocee see mo
oooooooooec see mm
oeeeoeseeee see mm
eeeeeeeeeee eee nr
eeeooeeoece see mm
oooeeeeeeee 663 am
eeeeeseeeee see sw
oeeeeeeeeee .3; m
eeeeeeeceee e e p
eooeoooooeo e a
fmoooooeeoee o @

ooeoeoooce F.e

 

 

 

 

 

 

.ssewa
.wmwma
.mwmmw
.mwmmw
.eeeee
.sassa
.eewww

.eeeee

)OOOm
.mwmmw
aeeeee
:eeeee

seeeee

.sssaa

)InOm

zeeeee

.essaa
.saaaa

)DOOm

aeeeee
seeeew

aeeeoe
sesame

)“nOm
geeeee

)wnOm

seeeoe

)unOm

.swwwa

.wwwww
.wwwaa

aeeece

seeeoo

)InOm

seeeee

.seewe
.wwwmw
access
a@@@@@
u@@@@@
geewwe
meewww
aeeeew
sewwww
seewwe
seewee
a@@@@@
:e@@@@

>mnOm

seeeee

lllllllllllllllllllllllllllllllllllllllllll s llllll

—-—-
a...
:eeeee
.eesea
.asaea
.eaesa
.aseaa
.ssaae
.saaaa
.saaaa
.aaasa
.aaaea
.aaeaa
a...
recess

173
APPENDIX I
Machine Scoring Form

a _ @335 p33... @333. @333 33c...
9 3%... £39.. £30.. £39. 889..
.m _ see- was... as? as? see.
.rK. m y. — @9999... @939... oeoeea @eeeoa essence...
a w009( w000( wOUO( wOUﬂ< wOUo<
a m m 92395 @923... @@@®@.. @269... @9069...
am uOUD< wOUO< woum< wOUO< wOU¢<
M ...... m... _ 8%... £3? $3? $89.. ease...
M mm ..m. cm — @eeees eoeees pesos... @eoees @ooees
fnec wOUu< uOUG< woun< wOUQ< wOUO<
..T mm ”..m.. @oeees posse... @993... @939... seamen.
aﬁar moon.‘ mac... woua.‘ wcoc.‘ wcoo.‘
8 Wm m .3“ @@@®0§ @OOQQB. @©@®Oa @OOQG: @OOQQs
n wmwm woo... moon... woon< moons. woou<
0 MM .m. m — @@@®@a. @®@®@§ @@@®@s @©@®@c @O@®@s
II "Ma wOUG< w000( w000( wOUO( woum(
e ..m rm“ M .m @939... causes... @oeee; @2692 959?.
Ci". .3 wOUD< wOUO< wOUQ< wouo< woum(
u e mamma — @eeeea @eeees @eeeea sesame... @eemés
r2 ."r wOUO< wOUm< woon( wooc( woua<
O m mmme — @@@®Oe. @@@®Q§ @@@®@s @@@®@z @@@®®s
arr—n.” OUm< wOU¢< wOUO< woum( wouu(
mm a @eeees @eeeea @eeees @@@e@2 @@@@@a
7 WWMM mouo< uOUO< woum( wGUﬂ< wOU¢<
m 123 4 — eeeees eoeees eeeeee @ooeeu eeeeee
_-_--__- -—-—-—_—_——-_—--—-__-—_———-_-__.-—-

174
APPENDIX J

DII‘BCIIOI'IS for Students (Phase I)

There are two word processing tasks in this experiment. Please follow these directions as
closely as you can, and please work alone. This is very important to assure the integrity of this
study. Your cooperation is greatly appreciated.

1. Please record the starting time announced by the proctor on your Time Sheet in the
space next to "Starting Time #1."

2. Please read the "Word Processing Task 1" sheet.

3. Use the "Instructional Materials" any way you see fit to try to complete the first word
processing task. You may stop working on it at anytime.

4. Please work alone and at a comfortable pace. You will not be graded on your
performance or completion of this task. You will have a maximum of one hour and
thirty minutes to complete the two word processing tasks. You may take less time if
you so desire.

5. When you feel you have completed "Word Processing Task 1," or wish to stop work

 

Then, please raise your hand. The proctor will give you instructions to begin "Word
Processing Task 2." Please record the second starting time in the space after
"Starling Time #2."

6. It you need to take a break, you may do so.

7. Use the "Instructional Materials" any way you see fit to try to complete the second
word processlng task. You may stop working on it at any time.

8. When you have completed "Word Processing Task 2," ”or wish to stop work on it,

 

please raise your hand. You will be given further instructions at that time.

9. If you do not finish the task, please stop working when instructed to by the proctor.
Place your keyboard on top of your computer. Record the time announced by the
proctor on your ‘lime Sheet. The proctor will give you further instructions.

175
APPENDIX K

TIME SHEET (Phase I)

STARTING TIME #1:

 

FINISHING TIME #1:

 

STARTING TIME #2:

 

FINISHING TIME #2:

 

176
APPENDIX L

Word Processing Task 1

Use WordPerfect to create a copy of the following text. Print out 1 copy of the completed
document. Do NOT exit the program when you finish or wish to stop work, but record the time on
the Time Sheet after "ﬁnishing Time #1:" and raise your hand. The proctor will give you further

instructions.

 

ANNOUNCEMENT

It's the end of the year. and all survivors are invited to celebrate!

Where: 4 South Asserrbly Hall
When: Friday, June 5
Music: "Stone Age"

All refreshments will be provided, courtesy of the Board of Student Govemrnent.
Designated drivers will be available for a token fee.

signed:

(your name)

 

Include the following:

enter all OI the text

use the centering function to center the title word "ANNOUNCEMENT". Only the title
word should be centered.

indent where lncﬁcated using the "tab" feature.

use upper and lower case characters as indicated.

use the preset computer settings (margins, tabs. type style) - don't try to change
them. Don't try to replicate the dotted lines at the top and the bottom of the
announcement.

enter your own name under the word "signed?

177
APPENDIX M

Word Processing Task 2

Use WordPertect to create a copy of the following text. Print out 1 copy of the corrpleted
document. Do NOT exit the program when you finish or wish to stop work, but record the time on
the Time Sheet after "ﬁnishing Time #2:" and raise your hand. The proctor will give you further

instructions.

 

MOVIE NIGHTS!

The classic: are here for your viewing pleasure:

"Casablanca" - May 21
"Beau Geste" - May 23
"Gone With The Wind" - May 25

All films will be shown in the Mufti-Purpose room, Cannon Hall, at 7:00 pm.
Admission is a buck.

Any questions? Call the Film Board Rep:

(your name)
3456789

 

Include the following:

enter all of the text

use the centering function to center the title words "MOVIE NIGHTS". Only the title
words should be centered.

indent where indicated using the "tab" feature
use upper and lower case characters as indicated.

use the preset corrputer settings (margins. tabs, type style) - don't try to change
them. Don't try to replicate the dotted lines above and below the text.

enter your own name under the words "Any questions?"

178
APPENDIX N

Scheme for Naming WordPerfect Document Files

This scheme provides unique identiﬁers for all WordPerfect word
processing document ﬁles:

ﬁle name: ﬁrst four letters of subject's last name

ﬁle extension: three numbers, the ﬁrst two representing the
subject's experimental group (1- 10), the third number
representing the order of the subject's word processing
task: 1, 2 (Phase I), or 3 (Phase 11).

Example;

If subject Robert Slickgard in group 9 has just completed his
second word processing task, then his document ﬁle would be labeled
SLIC.092.

179
APPENDIX 0

Word Processing Task (Phase II)

Use WordPerfect to create a copy of the following text. Print out 1 copy of the corrpleted
document. Do NOT exit the program when you finish or wish to stop work, but record the time on
the Time Sheet after "Finishing Time #1:" and raise your hand. The proctor will give you further

instructions.

 

CHECK THIS OUT!

I otter maximum discounts on all school supplies. My prices are second to none!

- 2500 sheets computer paper: $10 per box
- textbooks: 50% off list price (three weeks for delivery)
- 8.5 x11 ruled pads: 10 for $2

Stop by Room 17, Marshall Hall, Monday through Wednesday. I'm open for
business from 8 pm until 11 pm. I will not be undersold!

You can believe it, or my name isn't
(enter your name here).

 

Include the following:

enter all of the text

use the centering function to center the title words "CHECK THIS OUT!". Only the
title words should be centered.

indent where indicated using the "tab" feature.

use upper and lower case characters as indicated.

use the preset conputer settings (margins, tabs, type style) - don't try to change
them. Don‘t try to replicate the dotted lines at the top and the bottom of the
announcement.

enter your own name under the words "or my name isn‘t".

180
APPENDIX P

Directions for Students (Phase 2)

There is one word processing task in this experiment. Please follow these
directions as closely as you can, and please work alone. This is very important to
assure the integrity of this study. Your cooperation is greatly appreciated.

1. Please record the starting time announced by the proctor on your Time
Sheet in the space next to "Starting Time."

2. Please read the "Word Processing Task" sheet.

3. Use the "Instructional Materials" any way you see fit to try to complete
the word processing task. You may stop working on it at any time.

4. Please work alone and at a comfortable pace. You will not be graded
on your performance or completion of this task. You will have a
maximum of 45 minutes to complete the word processing task. You
may take less time if you so desire.

5. When you feel you have completed the Word Processing Task or wish
to stop work on it, ‘ '
" ’ ' ' " Then, please raise your hand. You will be given

further instructions at that time.

6. If you do not finish the task, please stop working when instructed to by
the proctor. Place your keyboard on top of your computer. Record the
time announced by the proctor on your Time Sheet. The proctor will
give you further instructions.

181
APPENDIX 9

TIME SHEET (Phase II)

STARTING TIME:

 

FINISHING TIME:

 

182
APPENDIX R

Word Processing Task 1 Checklist

Score 1 point for each checked item:

1. Entering at least 1 line of text of the task document ............. __
. Entering at least 5 lines of text of the task document ........... __
. Entering all the text of the task document ............................ _
. Centering the title of the document ...................................... __

. lndenting the line "It's the end of the year, and..." 1x ........... __

GUI-#0)“)

. lndenting the lines "Where:..", "When:...", and
"Music:..." 2x .................................................................

\I

.lndenting the line "All refreshments ..... " 1x .........................
8. Using upper and lower case letters where indicated ........... __

9. Printing out the document ................................................... _

total score:

 

183
APPENDIX 5

Word Processing Task 2 Checklist

Score 1 point for each checked item:

1. Entering at least 1 line of text of the task document ............

2. Entering at least 5 lines of text of the task document .......... __
3. Entering all the text of the task document ........................... __
4. Centering the title of the document ...................................... __
5. lndenting the line "The classics are here for ..... " 1x ........... __
6. lndenting the lines "Casablanca", etc. ..2x .......................... __
7. lndenting the line "All films ..... " 1x ...................................... __
8. Using upper and lower case letters where indicated ........... _
9. Printing out the document ................................................... __

total score:

 

184
APPENDIX T

Word Processing Task Checklist (Phase II)

Score 1 point for each checked item:

A

. Entering at least 1 line of text of the task document ............ _
. Entering at least 5 lines of text of the task document ......... _
. Entering all the text of the task document ........................... __
. Centering the title of the document ...................................... __
. lndenting the line "I offer maximum discounts ..." 1x ........... _
. lndenting the lines "- 2500 sheets", etc. ..2x ....................... _
. lndenting the line "Stop by room 17..." 1x .......................... _

. Using upper and lower case letters where indicated .......... _

(accurate-incurs

. Printing out the document ................................................... __

total score:

185

List of References

Bloom, A. J. (1985). An anxiety management approach to

computerphobia- WWW. 32(1). 9094.

Borg. W. R.. & Gall. M. D. (1983). W
19113313493329. New York: Longman.

Brewer, J. K. (1972). On the power of statistical tests in the American
Educational Research Journal. W
slournai. 9(3). 391-401.

Cambre, M. A. and Cook, D. L., (1985). Computer anxiety: deﬁnition,

measurement. and correlates W
Research. 1(1). 37-54.

Cambre, M. A. and Cook, D. L. (1987). Measurement and remediation of
computer anxiety. Winning. 21(12) 15-20.

Campbell. D. T.. and Stanley, J. C. (1963). W
W- Chlcagm Rand McNallY-

Carroll, J. M. (1984a). Minimalist design for active users. Human-
W 39-44

Carroll, J. M. (1984b). Minimalist training. W, 39(18), 125-136.

Carroll J M (1990). W
W Cambridge Mass: MIT

Press.

Carroll. J. M (1992). Minimalist documentation, in H. D. Stolovitch and
E. J. Keeps, (eds.) f
San Francisco: Jossey—Bass Publishers.

186

Cattell. R. B. and Scheier, I. H. (1961). r
W. New York: Ronald Press.

Chandler, P. and Sweller, J. (1991). Cognitive load theory and the format
of instruction. 99W. 8(4). 293-332.

COhen. J. (1977). .0. _ 1 ..W- M. 0._ b -_ 0 . ;.!.-. . v. ‘1‘! -‘0
New York: Academic Press.

Epstein, S. (1972). The nature of anxiety with emphasis upon its

relationship to expectancy. in CD. Spielberger (ed.), M
(Vol. 2). New York:

Academic Press.

Glass. G. V- and HOPkins. K D. (1984). W
W. Englewood Cliffs, NJ: Prentice-Hall.

Howard. G. S. (1986). ngputer Anxiety and the Us; 9f ngputggs 3;
Management, Ann Arbor, MI: UMI Research Press.

Howard, G. S. and Thomas. G. N. (1987). Computer anxiety
considerations for design of introductory computer courses.

Educational Research Quarterly, 11(4), 13-22.

Keller, J. (1983a).
v n i . Paper based upon a series of lectures given at
the Technische Hegeschool ’I‘wente. Enschede, Netherlands.

Keller, J. (1983b). Motivational design of instruction. In C. M. Riegeluth
(Cd.!). 40.11 P'I'-1!‘0 " 0"‘°; . “31’ 0

W. I-Iillsdale. NJ: Lawrence Erlbaum.

Keller, J. (1987). Strategies for stimulating the motivation to learn.
Won. 2.6(8). 1-7.

Mager. R. F. and Pipe. P. (1970). W.
Belmont. CA: Fearon Publishers.

187

Marcoulides. G. A. (1989). Measuring computer anxiety: The computer

anxiety seale- WW .49.
733 - 739.

 

Marcoulides, G. A. (1988). The relationship between computer
anxiety and computer achievement.

W412). 151-158.

Maurer. M. M. (1983). D e1 m n V ation f M
. unpublished master's thesis, Iowa State
University.

May, R. (1950). W. New York: Ronald Press.

Raub, A. C. (1981). rr 1
unpublished doctoral dissertation, University of Pennsylvania.

Simonson, M. R., Maurer, M., Montag—Torardi, M., and Whitaker, M.
(1987). Development of a standardized test of computer literacy

and a computer anxiety index. WWW:
Research. 3(2). 231-247.

Subkoviak. M. J. and Levin, J. R. (1977). Fallibility of measurement and
the power of a statistical test. W
W. 135(1). 47-52.

Resnick, L. (1989). Introduction, in L. Resnick, (ed.) M
and Insgyction. Hillsdale, NJ: Lawrence Erlbaum.

Rohner, D. J. and Simonson. M. R. (1981). Development of an index of
computer anxiety. Presented at the annual convention of the
Association of Educatidnal Communications and Technology.
April, Philadelphia. ERIC Doc. No. 207487.

Rosen, L. D. 8: Maguire, P. (1990). Myths and realities of
computerphobia: a meta-analysis. W, 3. 175-191.

188

Rossett, A (1992). Analysis of human performance problems, in H. D.

Stolovitch and E. J. Keeps. (eds.) WWW:
W. San PTancisco: Jossey-Bass Publishers.

Spielberger, C. D. (1966). Theory and research on anxiety, in CD.
Spielberger, (ed.), B h vi r. New York: Academic

Press.

Spielberger, C. D. (1972). The nature and measurement of amdety, in
..(Vol 1). New

York: Academic Press.

U.S. Congress, Ofﬁce of Technology Assessment. (1988). W

the American Economic Transitien: Qhetces for the Future.
Washington, DC: U.S. Government Printing Ofﬁce.

Warner. J. w. (1987). WWW

D'ITh‘O on w in A o.“e‘s qui '-u- _. r
git-stugy Deming Environmente. unpublished doctoral

dissertation, Michigan State University.

Weil, M. M., Rosen, L. D.. and Sears, D. C. (1987). The computerphobia
reduction program: year 1 Program development and preliminary

results. Behevier Research Methege, Instruments, egg Qempggeg,
19(2). 180-184.

White, S. W. (1991). The universal computer. W
W. 21(3).

WordPerfect Corporation. (1989). W
W. Orem. Utah: WordPerfect.