'vv-g. -W. ’..A....,v..-r . ,
‘: '_

.v *__— '— _—_ __ V
‘. u. y ,. 1%.. n vwgy‘p < ‘ n ‘ ‘ m .. ‘ ‘ Mm; .u “‘
‘ n‘. ‘ .. ‘ . . q. .

 

 

”work‘s-'4 - “R N‘.’ll-;CI~>|- “w“.- .- .. .» .m. ‘ H...‘ . ‘ ... ‘. _
. ‘ .
.4 n-" ""‘t-V:‘f€“\i"fil‘vIJy..,,g‘. ' , u ‘ ‘H y. . “

AUDIENCE COMMITMENT museums KNOWLEDGE ' _ _
or AUDIENCE AS DETERMINANTS or coenmva f ; ”
DISSGNANCE AND suasmum CHANGE, or _ > '

ATTITUDE AND SELF PERCEPTION. rouowms -
coummmnuomm ADVOCACY

’ mesis for the Dégree of PM).
mamam STATE UNIVERSITY
Roam NOEL WIDGERY.
1971

 

|r1c“) E

    
 
    

LIBRARY ’ ‘g.
Michigan State
Unircrsixy {A

A

:::3

W

W “III!

00
8
\l
(D
(D
01
w

This is to certify that the

thesis entitled

AUDIENCE COMMENT AND SOURCE KNOWLEDGE OF AUDIENCE
AS DETERMINANTS OF COGNITIVE DISSONANCE AND SUB-
SEQUENT CHANGE OF ATTITUDE AND SELF PERCEPTION
FOLLOWING COUNTERAITITUDINAL ADVOCACY

presented by

Robin Noel Widgery

has been accepted towards fulfillment
of the requirements for

Ph .D . degree in Speech

1327M d? MI/mv

Mo: professor

Date February 11, 1971

0-7639

AUDIENCE COMMITMENT AND SOURCE KNOWLEDGE OF AUDIENCE
AS DETERMINANTS OF COGNITIVE DISSONANCE AND SUB-
SEQUENT CHANGE OF ATTITUDE AND SELF PERCEPTION
FOLLOWING COUNTERATTITUDINAL ADVOCACY

By

Robin Noel Widgery

The purpose of this study was to assess the effects of audience
commitment and source knowledge of audience on attitude change and
self perception following the writing of oounterattitudinal essays .

Students enrolled in undergraduate communication courses at
General Motors Institute were administered a pretest attitude question-
naire containing several issues considered to be salient for them.

The issue producing the most skewed distribution on attitude intensity
called for two years mandatory military service for all college
students before being accepted by a college or university; thus it was
chosen as the experimental issue. About two weeks after the pretest,
experimental subjects wrote oounterattitudinal essays under varying
conditions of source knowledge of audience and audience commitment
toward the issue. Post—encoding measures were then taken of each
subject's attitude toward the issue and subject's self-image. The
effectiveness of each manipulation was assessed by asking subjects to
respond to questions designed to tap their perception of audience
attitude and the probability that they would know members of the

supposed audience. Subjects assigned to the Control group responded

Robin Noel Widgery
to the pretest and posttest, but did not write the essay.

The knowledge of audience variable was diChotomized between
the Known Audience and the Unknown Audience. Subjects in the Known
condition were informed that their supposed audience would consist of
students at General Motors Institute.. The audience for the Unknown
condition was to be students at MiChigan State University. Audience
commitment toward the topic in question was diChotomized into those
Who were strongly opposed and those who were undecided.

The major dependent variable, attitude Change, was assessed by
using pretest to posttest difference scores on four, seven—interval
semantic differential—type scales. In addition to attitude Change,
salience Changes were also measured.

Three hypotheses were tested in this experiment: First, that
attitude change would be significantly greater for the Known Audience
than fbr the Unknown; second, that attitude Change would be signifi-
cantly greater fOr the Uncommitted Audience than fOr the Opposed;
third, that the greatest amount of Change would be in the Known,
Undecided condition and the least amount of Change in the Unknown,
Opposed condition.

The results failed to support any of the stated hypotheses.
However, there was an inverse knowledge of audience effect, showing
significantly greater attitude Change for subjects in the Unknown
condition.

This unexpected result was explained by the supposition that

sUbjects encoding for the Known Audience perceived that message

Robin Noel Widgery
takeback would be possible if ever confronted by a member of the
supposed audience. The possibility of message takeback reduced the
arousal of cognitive dissonance thus militating against attitude
Change.

The results of this experiment were discussed within the
framework of cognitive dissonance theory and included suggestions for

fUture researCh using the oounterattitudinal advocacy paradigm.

AUDIENCE COMMITMENT AND SOURCE KNOWLEDGE OF AUDIENCE
.AS DETERMINANTS OF COGNITIVE DISSONANCE AND SUB-
SEQUENT CHANGE OF ATTITUDE AND SELF PERCEPTION
FOLLOWING COUNTERATTITUDINAL ADVOCACY

BY
Robin Noel Widgery

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of SpeeCh

1971

DEDICATION

To

Suzanne

ii

Accepted by the faculty of the Department of
Speech, College of Communication Arts, Michigan State
University, in partial fulfillment of the requirements
for the Doctor of Philosophy degree.

\
l J

rector of esrs

Guidance Committee: MW, Chairman

 

 

iii

ACMOMEIISMENTS

A doctoral dissertation is seldom the product of a single
individual's efforts. In the case of this research, the writer is
indebted to those empiricists upon whose work this one is predicated.
In this experiment no one has been more instrumental in introducing
me to the content of dissonance theory and the paradigm of counter-
attitudinal advocacy than has Dr. Gerald Miller. His enthusiasm and
sincere interest in this research have made the experience an enjoyable
one.

For their perceptive and cogent advice, I want to thank Dr.
Donald Cushman, Dr. Maryellen McSweeney and Dr. David Ralph.

To my colleagues at General Motors Institute, I owe much indeed.
'Iheir encormagement, and the climate of professionalism they have
created, have given me the desire and intellectual stimulation needed
to complete every step toward the doctoral degree. These men deserve
my warmest and sincerest gratitude: Dr. Michael Burgoon, Dr. Robert
Carter, Professor Harold Haskitt, Professor Tony Hain, Professor
David Hurt, Dr. Ronald Smith, Professor Cecil Stackpole, Dr. Stewart
Tubbs, Professor Roger Wilcox, and Dr. Joseph Zima.

Thanks, also go to Mrs. Delores "Scotty" Davidson, whose aid
has been appreciated so many times during the past four years.

Almost all ackriowledgients say something about a wife and how

much help she has been. All I can say is thank you Suzanne for your
iv

patience and tolerance during all these years. I am grateful that
you are still my wife, lover, confidant, and friend: even after

all this.

TABLE or CONTENTS

Page
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . iv

LIST OF'TABLES . . . . . . . . . . . . . . . . viii

LIST OF.APPENDICES . . . . . . . . ; . . . . . . x
Chapter

I INTRODUCTION . . . . . . . . . . . . . . 1

Statement of the Problem . . . . . . . . . l

Review of Relevant ResearCh . . . . . . . . 3

Audience Attitude Toward the Topic . 3

Source Knowledge of the Audience . . 6

Rationale and Hypotheses . . . . . . 7

II METHODS AND PROCEDURES . . . . . . . . . . . 9

Overview . . . . . . . . . . . . . . 9

Pretest . . . . . . . . . . . 9

Independent Variables . . . . . . . . . . 10
Audience Attitude . . . . . . . . . . 10

Source Knowledge of Audience . . . . . . . 11
Inductions and Posttest . . . . . . . . . ll
Dependent Variables . . . . . . . . . . 15

Attitude Change . . . . . . . . . . . 15

Message Content . . . . . . . . . . . 15

Self-image . . . . . . . . . . . . . 15

Harmful Consequences . . . . . . . . . 16

Encoding Anxiety . . . . . . . . . . . 16

Manipulation Checks . . . . . . . . . . 16

III RESUUTS . . . . . . . . . . . . . . . . l7
Pretest . . . . . . . . . . .
Experimental and Control Differences . . . . . 17
Manipulation Checks . . . . . . . . . . 19

vi

Chapter III (contd)

Chapter Page
Test of the Hypotheses . . . . . . . . . 20
Supplementary Analyses . . . . . . . . . 22

IV DISCUSSION . . . . . . . . . . . . . . . 28
Failure to ConfirmlHWpothesis l . . . . . . 28

Failure to Confirthypothesis 2 . . . . . . ' 29

The Locus of Dissonance Arousal . . . . . . 31

Aversive Consequences . . . . . . . . . 31

Threatened Self-image . . . . . . . . . 32

Implications fbr Future ResearCh . . . . . . 3”

Summary . . . . . . . . . . . . . . 36

REFERENCES . . . . . . . . . . . . . . . . . 37
APPENDICES . . . . . . . . . . . . . . . . . 39

vii

Table

10

11

LIST OF TABLES

Means, standard deviations, and analysis of variance
of pretest attitude scores for experimental and
control groups . . . . . . . . .

Attitude scores, amount of change, and analysis of
variance of attitude change scores of experimental
and control groups . . . . . . . . . .

Analysis of differences of attitude Change scores
between each experimental group and the control

group . . . . . . . . . . . . . .

Percentage and frequency of subjects correctly and
incorrectly perceiving audience attitude . . .

Means, standard deviation, and analysis of variance
of knowledge of source ratings . . . . . . . .

Attitude scores, amount of Change, and analysis of
variance of attitude change scores of experimental

mups O O I O O O O O O O O O O O I 0

Means, standard deviations and analysis of variance
of anxiety scores of experimental groups . . .

Means, standard deviations, and analysis of variance
of persuasiveness scores of experimental groups . .

Means, standard deviations and analysis of variance
of harmful effects scores of experimental groups .

Means, standard deviations and analysis of variance
of salience Change scores of experimental groups . .

Means, standard deviation and analysis of variance
of qualification scores of experimental groups

viii

Page

17

18

18

19

20

21

23

2H

2H

25

26

Table Page

12 Means, standard deviations, and analysis of variance

of trustworthiness scores of experimental groups . . 26
13 Means, standard deviations, and analysis of variance

of essay length of experimental groups . . . . . 27

LIST OF APPENDICES

Appendix Page
A PREIEST ATTITUDE QUESTIONNAIRE . . . . . . . . 39
B POSITEST ATTITUDE QUESTIONNAIRE . . . . . . . . H3

C CONTROL POS'ITEST ATTITUDE QUESTIONNAIRE . . . . . L+7

Chapter I

INTRODUCTION

Statement of the Problem

 

Experimental desigrs utilizing oounterattitudinal advocacy are
especially suited for testing consistency theories . This paradigm
requires that the source encode a belief—discrepant message, thus
becoming the receiver of his own communication. By using pre- and
posttest attitude measures the researcher is able to measure the
amount of persuasion the subjects have experienced. The basic assump-
tion of this model is that the individual will persuade himself rather
than being persuaded by another.

Festinger (1957) was one of the earliest proponents of the
persuasive efficacy of oounterattitudinal advocacy. According to him
an individual will experience cogiitive dissonance when he prepares a
message which is counter to his true beliefs. In short, dissonance is
aroused because the individual perceives the opposing cognitions (l)

I believe )5, and (2) I am advocating fl §. In their classic "dull
task" experiment Festinger and Carlsmith (1959) supported their con—
tention of a dissonance effect. They asked subjects to tell a
confederate that a dull task had been interesting to do. Subjects
receiving $1 as a reward for telling the confederate the task was
interesting demonstrated sigrificantly greater liking for the dull

task than subjects who were given $20 as incentive. Although this
1

2

study has since been attacked for several methodological weaknesses,
it, none-the—less, fired the interest of many consistency theorists
in the oounterattitudinal paradigm as a viable technique for testing
various aspects of dissonance theory.

More recent dissonance theorists such as Aronson (1968) and
Collins (1968) have taken issue with Festinger's original assumption——
that the locus of arousal is found in the conflict between the cog—
nitions "I believe )3," and "I advocated not 2g." Aronson (1968) has
refuted Festinger's earlier assumption:

In the experiments on oounterattitudinal advocacy for

example, I would suggest that it is incorrect to say

that dissonance existed between the coglitions "I

believe the task is dull," and "I told someone that

the task was interesting". . . . What is dissonant

is the cognition "I am a decent, truthful human being"

and the cognition "I have misled a person; I have A

conned him into believing something which just isn't

true: he thinks that I really believe it and I cannot

set him straight becarse I probably won't see him again."

(p. 2'4)

Carlsmith (1968) supports Aronson in his speculation as to the
locus of dissonance arousal. He underscores the importance of the
advocate's concern for his supposed audience, saying that "anytime a
person makes some statement counter to his attitudes, and a listener
(whose opinion is important to the speaker) is unaware of both the
speaker's attitude and his motivation for speaking against these
attitudes, dissonance will be aroused." (p. 806)

Summarizing the crux of the debate between the "early" and
"late" dissonance interpretations, Collins (1968) questioned whether

the discrepancies between what a person believes and what he says will

actually be sufficient to produce dissonance arousal. Instead he

3
believes that a dissonant state may result from the individual
feeling that his counter advocating was a moral or ethical trans-
gression, or he may believe that his belief—discrepant message may
cause some harmful consequences to the supposed audience . In short,
the central issue of concern here is whether or not dissonance will
be produced when an individual's self-image is threatened by the
counteratt itudinal act , or when he perceives that the act will cause
aversive consequences to others .

If the loci of the arousal state are found in the perception
of threatened self-image or possible aversive consequences , then it
is proper to assume that such hypotheses may be tested empirically.
Miller (1970) suggests:

To raise questions about the "late" dissonance inter-

pretation does not detract from the value of attempting

to specify precisely those situational factors which

serve to heighten or lessen the dissonance experienced

by a oounterattitudinal advocate . (p . 9)

In "attempting to specify precisely those situational factors"
the experimenter has manipulated two independent variables , audience

attitude toward the tgpic and source knowledge _o_f_ the audience , in

 

 

 

hopes of identifying more accurately the source of dissonance arousal.

Review of Relevant Research

 

Audience Attitude Toward the Topic. Underlying the use of this
variable is the assumption that if a supposed audience is perceived by
the oounterattitudinal encoder to be uncommitted on the issue in
question, the encoder may assume his message will have greater

influence than if the receiver is already committed. If such a

m
differential in audience commitment is perceived by an individual,

it is reasonable to assume it may affect the degree of cognitive
discomfort experienced after encoding a belief;discrepant message.

In an experiment by Bodaken (1970) sUbjects reported a significantly
greater amount of attitude Change When preparing belief—discrepant
essays for uncommitted receivers than. for those who were committed
favorably toward the message. This finding supports the theoretic
expectation that the subject's fear of aversive consequences to his
audience will create cognitive dissonance.

A study by Nel, HelmreiCh and Aronson (1969) supports the same
view; They asked subjects to make a video recording advocating the
legalization of marijuana to be shown to three supposed audience
groups-~those for the proposition, those against and those with no
opinion. They reported a substantially greater amount of attitude
change among subjects encoding fOr the uncommitted audience in the
low incentive condition (fifty cents vs. $5). SuCh a result confirmed
the dissonance effect predicted by Festinger and Carlsmith (1959):
when reward is low, dissonance will be maximal.

The preceding studies imply that When an audience is uncom-
mitted or has no opinion on a particular issue, it is believed to be
more vulnerable to persuasion. When this condition is perceived by
the oounterattitudinal communicator, dissonance should exist because
of anxiety over aversive consequences to the receiver.

Another study illustrating the relevance of aversive conse-
quences to dissonance arousal was done by HelmreiCh and Collins (1968).

These experimenters manipulated financial reward and audience

5
commitment. Subjects were asked to give speeChes based on arguments
given them on note cards. Their speeChes were then delivered in three
conditions: no takebaCk video, takeback video, anonymous audio. The
subjects in the no—takebaCk condition were not given an opportunity
to clarify their true opinion on the video tape recording, While those
in the takeback condition were allowed to explain after the SpeeCh
their true beliefs and Why they had recorded the oounterattitudinal
SpeeCh. Unlike the first two conditions, the anonymous audio SpeeCh
was delivered without subjects declaring their identities. The levels
of financial.reward.were either fifty cents or $2.50. A dissonance
effect (greater attitude Change in the low reward condition) was
reported for'subjects in the no—takeback group. It can be reasoned
that subjects in this group perceived that their beliefediscrepant
messages would be used to persuade others, perhaps having some aversive
effect upon them. SuCh a cognition may'have created greater dissonance
in the no—takebaCk condition accounting for the greater reported
attitude Change. Not being able to clarify their true positions, no—
takeback subjects were probably left with the feeling that the audience
would continue to misperceive their true attitude on the issue in
question. Compounding this cognition would be the feeling that the
audience had not been "set straight" on the issue.

Related to the issue of aversive consequences is the question
of self-concept. What happens to an individual's self—image When he
perceives that his belief-discrepant message may have harmful effects
upon.his audience? Bramel (1968) believes that dissonance may result

from fear of What others will think. He says:

6

The theoretical relevance of the self in this view of
dissonance theory now becomes clearer . . . dissonance
is a feeling of personal unworthiness (a type of anxiety)
traceable to rejection of oneself by other peOple either
in the present or in the past. Any information which
implies that one is incompetent or immoral arouses dis-
sonance. The reason dissonance is greatest when the
person feels personally responsible for his behavior is
that rejection by other people is usually greatest when
they believe the person voluntarily acted in an inap—
propriate way. (p. 365)

Bramel suggests that dissonance may be aroused by anxiety over the
disapproval of 935335. for committing a belief-discrepant act that may
be harmful. The present study assumes that such dissonance may also
be aroused by the oounterattitudinal advocate ' 8 fl cogrition that he
is "incompetent or immoral." The belief-discrepant act may violate
the individual's own self-concept to such a degree as to produce
dissonance.

Source Knowledge of the Audience. This experimenter believes

 

that if fear 9: aversive consequences to the audience or threatened

 

 

self-image help to arouse dissonance, then it is logical to assume

 

that the source's familiarity with the supposed audience may affect
the amount of (‘lissommce experienced. Tf, for instance, a counter-
attitudinal advocate believes that he lmcws or will be knam by
members of his audience, he may have greater concern lest they suffer
harmful effects because of his influence. Also, having such a rela—
tionship with those in the audience may affect the advocate's self—
concept . We may assume that the more familiar the source is with
another individual, the more sensitive he will be to any act that he
commits against that individual which is improper, immoral, or un-

ethical. The guilt (dissonance) aroused by committing such an act

7
may be even more acute if the advocate is unable to explain to those
in the audience his true opinions or the circumstances in which he
advocated his belief-discrepant message.

If, however, the advocate perceives that it is quite improbable
that he will know or be lcnown by anyone in the supposed audience, he
may tend to impersonalize the counterattitudinal act . He may tend to
think less of his audience E3 _s_e_, because he believes it higlnly im—
probable that he will have to confront anyone in the unknown audience .

In such a condition, the advocate should experience minimal dissonance .

Rationale and typothes es

 

Because this experimenter believes that the counterattitudinal
advocate will generally experience greater dissonance and subsequent
attitude change when his communication may threaten his self-image or
cause potential harmful consequences to his audience, the following
hypotheses are proposed:

H1: Persons encoding counterattitudinal messages for a

known audience will report greater attitude change

 

toward the direction of the position advocated than
persons encoding counterattitudinal messages for an

unknown audience .

 

Since previous studies indicate that dissonance will be maximal
when the supposed audience is perceived to be uncommitted on the issue,
the following prediction is made:

H2: Persons encoding counterattitudinal messages for an

uncommitted audience will demonstrate greater attitude

 

8
change toward the direction of the position
advocated than persons encoding counter-

attitudinal messages for a committed audience .

 

Considering the combined effects of both the commitment and
familiarity variables, it is reasonable to assume:
H3: Attitude change in the direction of the position
advocated in the counterattitudinal message will

be greatest when the communication is encoded for

the known , uncommitted audience and attitude change

 

will be least for the unknown, committed audience .

 

On the basis of Aronson's (1968) and Collin's (1968) specu—
lation about the locus of dissonance, Hypothesis 1 is predicated on
the belief that an encoder of a belief-discrepant message should
experience greater dissonance with the known audience than with the
unknown. His relative familiarity with the members of the audience
should regulate the degree to which he perceives his message as
having potential harmful consequences. Hypothesis 2 is based upon
the same speculation . When the counterattitudinal advocate perceives
that his supposed audience is uncommitted on the issue in question,
dissonance and subsequent attitude change should be maximal. Hypothesis
3 assumes that the combination of the known, uncommitted audience should
produce the greatest amount of dissonance and thus subjects should
report more attitude change than in the other three cells . Subjects
in the unknown , committed audience condition should experience the
least amount of cognitive dissonance, subsequently reporting less

attitude change than in any of the other three cells.

Chapter II

METHODS AND PROCEDURES

Overview

A pretest of student attitudes was administered to 83 from
classes in public speaking at General Motors Institute. Based upon
the results of this test an issue was chosen for 88 to Lee in counter-
attitudinal encoding. Ss were then randomly assigned to experimental
and control groups. The Ss in the experimental groups wrote belief—
discrepant essays for fifteen minutes under different conditions of
audience commitment and familiarity. A posttest was administered to
the experimental groups measuring attitude intensity and salience on
the issue in question. The posttest also contained items designed to
check for subject perception of audience attitude , and audience
familiarity. Items were also included in the posttest to assess 88'
sensitivity to any potential harmful consequences which may have
resulted from the belief-discrepant essay. Another posttest measure
was used to measure the Ss' self—concept. Ss in the control group
received the pretest and posttest instruments, but did not encode

belief—discrepant messages .

Pretest
The students were induced to complete a qLestionnaire on the

pretext that it was part of a statewide college survey of student

10
attitudes on various issues. (See Appendix A.) The pretest
instrument was designed for two measures: (1) attitude direction
and intensity for several relevant issues, and (2) attitude salience
for the same issues.

In the measurement of attitude direction and intensity for the
several issues, each issue was followed by four, seven-interval,
semantic differential scales bounded by the adjectives good—bad,
valuable-worthless, pleasant—unpleasant, and fair-unfair. These
scales were drawn from the evaluative dimension of meaning as identi-
fied by 05good, Tannenbaum, and Suci (1957). The issues were scored
by summing across all four scales using one as low and seven as high.
Salience was measured by using two seven-point semantic scales with
important-unimportant and crucial-trivial as anchors . The issue
which had the most skewed distribution and the highest salience was

chosen as the topic for the counterattitudinal essays.

Independent Variables

 

The two independent variables manipulated in this experiment
were audience attitude toward the topic and probable lcrowledge of the

audience by the source .

Audience attitude . In manipulating this variable , half of the

 

83 were told that the audience who was to read their essays was

 

already strongly opposed E the advocated position . The other half
were told that their respective audience was undecided on the issue

in question.

11

Source Knowledge of Audience . This variable refers to the

 

degree of probability that the S perceives that he is known or may
become known by those in the supposed audience. In the m con-—
dition, 83 were induced to believe that students in another class at
General Motors Institute would read their essays. Those in the 227
kpgwg_condition were told that their essays would be read by students
at MiChigan State University. The supposition in using these induce—
ments was that the probability of students at General Motors Institute
ever knowing other students at that college was relatively high, While
the probability of their ever knowing students at MiChigan State was

relatively low.

Induction and Posttest

 

TWO weeks after the pretest had been administered, and the
experimental issue had been determined, 83 were asked to write a
counterattitudinal essay. These instructions were given to the Ss in
the "opposed" conditions:

MiChigan College Information
Exchange Project

Psychology and communication faculty at MiChigan State
University, and seven other MiChigan colleges are
currently conducting a.researCh project aimed at

finding out what it takes to strengthen or Change
people's attitudes. What we do is to get from students
some indication of their attitudes on certain issues and
then, at some later date , we present them with certain
arguments in the form of essays which are designed to
get themito Change or strengthen their attitudes.

Sometime ago questionnaires were administered to students
at eight MiChigan colleges asking them their attitudes
toward two years of mandatory military service fer all
male college students before being accepted by a college
or university. Analysis of the questionnaires indicates

12

that nearly all of these students are either undecided
93 or strongly opposed t_o_ this particular policy.

 

Because this is a communication course, we are interested
in getting from students in this class the most r—
suasive arguments we can, supporting the policy %t "all
male college students should serve at least two years in
the armed service before being accepted by a college or
university."

 

THOSE WHO WILL READ YOUR ARGUMENTS

The individuals who will be reading and thinking about
your particular arguments will be forty students at
Michigan State University (GMI). These particular
students at MSU (GMI) are strongly opposed to the policy
in question. Because of their strong opposition to this
policy, this group of students will probably not be very
easily influenced by your arguments. _—

 

In short, you are being asked to write an essay that
presents the most persuasive arguments favoring this

propos ition :

 

All male college students should be required
to serve at least two years in the armed
service before being accepted by a college or
university.

As you write keep your reading audience in mind.
They are:

Forty students at Michigan State (GMI) who are
strongly Opposed :2 the above proposition.

 

You will now be given 15 minutes to complete your essay.
The following instructions were given to those 88 who were in

the "undecided" condition:

13

Michigan College Information
Exchange Project

Psychology and commmication faculty at. Michigan

State University, and seven other Michi an colleges

are currently conducting a research project aimed

at finding out what it takes to strengthen or change
peOple's attitudes. What we do is to get from students
some indication of their attitudes on certain issues
and then, at some later date, we present them with
certain arguments in fine form of essays which are
designed to get them to change or strengthen their
attitudes.

Sometime ago questionnaires were administered to students
at eight Michigan colleges asking them their attitudes
toward two years of mandatory military service for all
male college students before being accepted by a college
or university. Analysis of the questionnaires indicates
that nearly all of these students are either undecided
gr or strongly opposed t_o_ this particular policy.

Because this is a communication course, we are interested
in getting from students in this class the most persuasive
arguments we can, supporting the policy thatfirall male
coIIege students should serve at least two years in the
armed service before being accepted by a college or
university."

 

THOSE WHO WILL READ YOUR ARGUMENTS

The individuals who will be reading and thinking about
omn particular arguments will be forty students at
M15igan State University (GMI) . These particular
students at MSU (911) are undecided on the policy in
question. Because of their md’écrsi'o—n toward this
policy, this group of students will probably _bg very
easily influenced by your arguments.

 

 

In short, you are being asked to write an essay that
presents the most persuasive arguments favorig this

propos ition :

 

All male college students should be required
to serve at least two years in the armed .
service before being accepted by a college or
university.

As you write keep your reading audience in mind.
They are:

ll}

Forty students at Michigan State (Q11) who
are undecided on the above proposition.

You will now be given 15 minutes to complete your essay.

85 were asked to read their respective inducement messages.

The rationale for having the 88 read the inducement instead of having
it read to them by the E, was to heighten the individual encoder's
feeling that he would be responsible as an individual for any aversive
consequences to the receivers . Heightened realization by the 83 that
their collective messages would carry the weight of persuasion on the
audience might have facilitated avoidance of responsibility for any
possible aversive consequences .

After the 83 read the inducement , they were given paper on
which to write their essays. In order to insure that all 85 committed
themselves publicly to their counterattitudinal message , each 8 was
instructed to put his name and student number on the paper before he
began to write . C

After the essays had been completed, a post—encoding question—
naire was distributed (see Appendix B) . This instrument contained the
same four semantic differential scales used in the pretest to measure
attitudes toward the issue in question. Other measures in the posttest
instrument were designed to assess the subjects' perception of audience

familiarity and audience commitment on the issue. The posttest also

 

 

tapped subject perception of possible harmful consequences to the

 

audience resulting from the counterattitudinal message . In order to

note any changes in the subject's self-concept as a result of his

 

counterattitudinal act, five scales loading highly on the safety

15
dimension of credibility and five loading highly on the qualification
dimension were included (Berlo, Lemert, Mertz, 1970). Control 83
were given a posttest questionnaire containing several issues among
them the experimental issue with the same scales employed in the

experimental conditions . (See Appendix C.)

Dependent Variables

 

Attitude change . The major dependent variable was the amount

 

of attitude chapge measured by the difference between individual pre-

 

and posttest scores on the issue in question.

Message content. A word count was made of essays to assess any

 

 

significant differences between cells. It was hoped that this would
be an index of any differential in effort between the four cells.

Self-image was measured by using ten semantic differential

 

scales identified by Berlo, Iemert, and Mertz (1970). The instrument
included five scales drawn from each of the two major factors of

source credibility as reported in their study: Safety (trustworthiness)
and Qualification. Each S evalLated himself on each of the following
scales:

Safety (trustworthiness)-—safe-unsafe, just-unjust,
kind-cruel , friendly-unfriendly, honest-dishonest .

Qualification--trained—untrained, experienced-
inexperienced, skilled—unskilled, qualified—
unqualified, informed—uninformed.
The rationale for making this measurement was based on the
assumption that dissonance ought to affect the subject's own self-
perception. If, for instance, he believed his belief-discrepant

message might affect someone adversely, the dissonance thus aroused

16
might be reduced by lowering his own self-image as well as changing
his attitude on the issue in question. Thus, lower self-image was
expected in the group where dissonance was maximal.

Harmful Consequences. In order to assess whether 83 perceived

 

that their belief-discrepant messages had harmful effects upon the
supposed audience, the 83 were asked to respond to this statement:

"My essay may have a harmful effect upon tlnose who will be reading it."
They responded on a five—interval Likert scale with Strongly Agree,
Neither Agree Nor Disagree and Strongly Disagree as anchors .

Encoding Anxiety. In order to measure differences in anxiety
among the groups , 88 were asked to respond on a five-interval Likert
scale to this statement: "I felt anxious about preparing this message
for my audience." The notion here was to see if high anxiety existed
when dissonance was maximal.

Audience Commitment . The efficacy of this manipulation was

 

determined by having 83 respond to this qtestion: "Before you wrote
your essay, how did you think the students who will be reading it felt
toward the issue of mandatory armed service for all male college
students?" 88 responded on a scale consisting of four alternatives:
Favorable, Undecided, Opposed, and Don't Know.

Source Knowledge of Audience. The check of this manipulation

 

was determined by having 88 respond to the following statement: "Some
of the students in the classes who are to read my essay will know who
I am. " Ss responded on a seven-interval scale with Highly Probably,

Neutral, and Hignly Improbable as anchors .

Chapter 111

RESUUTS

Pretest

One hundred.seventy—five students at General Motors Institute
were randomly assigned to fOur experimental groups and one control
group. To ensure that there were no significant differences between
these five conditions, subjects' pretest attitude scores were used
for a simple analysis of variance. The results of this test indicate
that there were no significant differences (Table 1).

Table 1. Means, standard deviations, and analysis of variance of pre-
test attitude scores for experimental and control groups .

 

 

Condition Mean Standard Deviation
Known: Opposed 7.69 3,uu

Known: Undecided 7.73 3.u3
Unknown: Opposed 8.00 3.59
Unknown: Undecided 7.78 3.66

Control 7.57 3 17

Source SS df MS F

Between 3.55 R .89 < l n.s.
Within 20u0.88 170 12.01

Total 26nu.u3 15%

Experimental and Control Differences

 

In order to measure fOr differences between the experimental

and control groups, a oneaway analysis of variance on difference scores
17

18
was performed. Table 2 shows a significant F value indicating that
there were differences among the five groups .

Table 2 . Attitude scores , amount of change , and analysis of variance
of attitude change scores of experimental and control groups .

 

 

 

Condition Pretest Posttest Change
Known: Opposed 7.69 8.15 .us
Known: Undecided 7.73 8.H9 .76
unknown: Opposed 8.00 9.HH 1.uu
Unknown: Undecided 7.78 11.59 3.81
Control 7.57 8.20 .63
Source SS df MS F
Between 273.67 u 68.H2 5.29*
Within 2199.u5 170 12.9u

Total 2u73.12 17H

*p < .05

To identify precisely which cell(s) accounted for this effect, Scheffe's
comparisons were performed, comparing each experimental condition with
the control group . Table 3 indicates these comparisons and shows that
only the Unknown: Undecided condition differed significantly from the
Control group . A

Table 3 . Analysis of differences of attitude change scores between
each experimental group and the control group .

 

 

Condition Mean Scheffe ' s t
Control . 6 3

Known: Opposed . 1+6 . 20
Known: Undecided . 76 . 15
Unknown: Opposed 1 . nu . 96
Unlonown: Undecided 3 . 81 3 . 71*

 

*p < .05

19
Manipulation Checks

 

Audience Attitude. In order to assess the effectiveness of the

 

manipulation of audience attitudes , subjects were asked to identify the
attitude of their respective audience . Table 1+ shows the numbers and
percentages of correct and incorrect responses in the four conditions .
In all, 88 percent of the subjects perceived their audience's attitude
correctly, while 12 percent misperceived these attitudes. Although the
Undecided conditions were 19 and 22 percent incorrect respectively,
this induction was considered to be generally successful.

Table '4. Percentage and frequency of subjects correctly and
incorrectly perceiving audience attitude .

 

 

 

 

Condition Correct Incorrect
96 (No . ) 96 (No . )
Known: Opposed 91+ (33) 6 (2)
Knowrn: Undecided 78 (25) 22 (8)
Unkrnown: Opposed 100 (36) - (-)
Unknown: Undecided 81 (29) 19 (7)
Total “8'5 (12 35 “'17 "(T77

 

Audience Knowledge of Source. Subjects had been told that

 

their supposed audience would be students at either Michigan State or
General Motors Institute . It was supposed that the probability was
greater that the Q11 subjects would perceive themselves to be known by
other Q11 students than by the MSU students . To test for this per-
ceived difference, subjects were asked to reSpond on a seven-point
semantic scale. These scores were used in a two—way analysis of
variance (Table 5). Since this analysis yielded a knowledge of source

main effect (F = 3u.uu), it was assumed that this manipulation was

20
relatively successful. It should be noted, however, that While there
was a significant difference between the Known and Unknown conditions,
the mean fer both groups was below the neutral position.

Table 5. Means, standard deviations, and analysis of variance of
knowledge of source ratings.*

 

 

Condition Mean Standard Deviation
Known: Opposed 3.83 2.17
Known: Undecided 3.27 1.71
Unknown: Opposed 1.77 1.31
Unknown: Undecided 1.92 l.uu
Source SS df MS F
Audience .0” l .ou < 1 n.s.
Knowledge of

source 2.91 1 2.91 3u.uu**
Interaction .12 1 .12 l.u5 n.s.
Error 10.80 135 .08

Total 13.87 138

*Knowledge of source perception scores ranged from 1 (very improbable)
to 7 (very probable)

**p < .05

Test of the Hypotheses

 

Hypotheses l and 2, whiCh predicted main effects fOr audience
attitude and audience knowledge of source, were tested by a twoeway
analysis of variance. Hypothesis 3 was not tested because its con-
firmation was predicated upon the confirmation of Hypotheses 1 and 2:
neither'of'which was confirmed. Data used in testing these hypotheses
were mean pretest to posttest attitude Change scores. The significance

level used for all tests in this study was .05.

21

Hypotheses l and 2. The first hypothesis predicted a positive

 

relationship between audience knowledge of source and amount of encoder

attitude change. Specifically, this hypothesis predicted that:

Persons encoding counterattitudinal messages for a
known audience will report greater attitude change
toward the direction of the position advocated than
persons encoding counterattitudinal messages for an
unknown audience.

Hypothesis 2 predicted a positive relationship between audience

attitude and amount of encoder attitude Change. This hypothesis

stated

that:

Persons encoding counterattitudinal messages for
an uncommitted audience will demonstrate greater
attitude change toward the direction of the
position advocated than persons encoding counter-
attitudinal messages for a committed audience.

Table 6 indicates that there is only one main effect demonstrated.

Table 6. Attitude scores, amount of Change, and analysis of variance of

attitude change scores of experimental groups.

 

 

 

Condition Pretest Posttest Change
Known: Opposed 7.69 8.15 .H6
Known: Undecided 7.73 8.u9 .76
Unknown: Opposed 8.00 9.uu l.uu
Unknown: Undecided 7.78 11.59 3.81
Source SS df MS F
Audience 1.77 l 1.77 2.26 n.s.
Knowledge of

source ”.07 l H.07 5.l9*
Interaction 1.06 1 1.06 1.35 n.s.
Error 106.08 136 .78

Total 112.98 139

*p <.05

22

Audience knowledge of source does have a significant effect, but in a
direction opposite the prediction. Subjects in the Unknown conditions
experienced greater attitude change than their Known condition counter-
parts. Change scores for an audience attitude effect were in the
predicted direction, but the obtained F was not significant.

Hypothesis 3. This hypothesis predicted the cells having the

 

least and greatest amounts of attitude Change. It stated that:

Attitude change in the direction of the position

advocated in the counterattitudinal message will

be greatest When the communication is encoded for

the known, uncommitted audience, and attitude

change will be least fbr the unknown, committed

audience.
Support for this prediction was dependent upon the directional confir-
mation of both Hypotheses l and 2. Because these were both not
supported, Hypothesis 3 was, therefOre, not confirmed in this experi—

ment.

Sppplementary Analyses

 

Besides using attitude Change as a criterion measure, other
measures were taken in the posttest. Salience was the only other pre—
and posttest measure-—a11 others were posttest only scores. These
latter measures were: (1) subjects' anxiety, ( 2) subjects' evaluation
of essay persuasiveness, (3) subjects' perception of subsequent harmful
effects resulting from.the essay, and (u) subjects' self-perception
scores.

.Anxiety. Subjects were asked in the posttest how muCh anxiety
they experienced during encoding. These scores were tested in a two-

way analysis of variance. As reported in Table 7, there was a

23

Table 7. Means, standard.deviations and analysis of variance of
anxiety scores of experimental groups.*

 

 

Condition Mean Standard Deviation
Known: Opposed 2.00 1.2H
Known: Undecided 2.76 1.37
unknown: Opposed 2.1” 1.08
Unknown: Undecided 2.”? 1.19
Source SS df MS F
Audience .30 1 .30 6.79**
Knowledge of

source .01 1 .01 < l n.s.
Interaction .05 l .05 1.03 n.s.
Error 5.uu 136 .ou

Total 5.80 139

 

*Anxiety'was measured on a 1 (least anxiety) to 5 (greatest anxiety)
scale.

**p < .05

significant audience attitude effect. Those subjects writing essays

fer undecided readers experienced significantly greater anxiety than

those writing for the Opposed audience.

Persuasiveness. Subjects were asked to evaluate the persu-

 

siveness of their essays. A.two~way analysis of variance shown in
Table 8 indicates that there was a significant audience attitude effect.
Those subjects Who wrote for undecided readers, believed their messages

to be more persuasive than did those writing fer readers who were

opposed.

21+

Table 8. Means, standard deviations, and analysis of variance of
persuasiveness scores of experimental groups.*

 

 

Conditions Mean Standard Deviation
Known: Opposed 3.19 1.53
'Known: Undecided n.21 1.55
Unknown: Opposed 3.75 1.50
unknown: Undecided 9.22 1.27
Source SS df MS F
Audience .59 l .59 9.u3**
Knowledge of

source .10 1 10 1.51 n 8
Interaction .09 1 .09 1.01 n s
Ekttn~ 8.16 136 .06

Total 8.99 139

*Persuasiveness was measured on a 1 (very weak) to 7 (very strong)
senantic scale.
**p < .05

HBrnfiua Effects. Subjects were asked to evaluate the possible

 

harmful effects their'essays might have upon their'respective audiences.
Table 9 shows that there is a signigicant audience attitude effect.
Subjects writing for the undecided audience believed that their essays
would be more likely to have a harmful effect than did those Who
wrote for audiences opposed to the essay.

Table 9. Means, standard deviations and analysis of variance of
harmful effects scores of experimental groups.*

 

 

Conditions Mean _ Standard Deviation
Known: Opposed 2.29 1.06
Known: Undecided 2.73 .99
unknown: Opposed 2.08 1.0%
Unknown: undecided 2.69 1.27

 

*Harmful effects were measured on a 1 (least effect) to 5 (greatest
effect) scale.

25

Table 9 (contd)

 

Source SS df MS F
Audience .28 1 .28 7.87*
Knowledge of

source .01 1 .01 < 1 n.s
Interaction .01 l .01 < l n.s
Error 5.m+ 136 .014

Total 5.7M 139-

*p < .05

Salience. Besides assessing attitude direction and intensity,
attitude salience, or importance, was also treasured. Table 10 reports
the anount of salience change between the pre- and posttests .

Table 10. Means, standard deviations and analysis of variance of
salience change scores of experimental groups.

 

 

 

Condition Mean Change Standard Deviation
Known: Opposed -l. 31; u . 32
Known: Undecided — . 70 5 . 01
Urflcnown: Opposed .08 u . 314
Unknown: Undecided - . 19 5 . 1&2
(Control) (— .09) (9.80)
Source SS df MS P
Audience .03 1 .03 < 1 n.s.
Knowledge of

source .93 1 .93 1.38 n.s.
Interaction . 21 1 . 21 < 1 n . s .
Error 91 . 12 136 . 67

Total 9 2 . 29 139

 

Although there is a slight indication of negative change for the known

conditions, this change is not statistically significant.

26

Self-perception. During the posttest, subjects were asked to

 

evaluate themselves on ten seven-point semantic differential scales
designed to measure two dimensions of credibility-qualification,
(ability) and trustworthiness (character). Table 11 shows an analysis
of variance based on mean qualification scores. Though there is a
sligmt interaction effect, it is not significant.

Table 11. Means, standard deviation, and analysis of variance of
qualification scores of experimental groups.

 

 

Condition Mean Standard Deviation
Known: Opposed 29.97 5.27
Known: Undecided 25.85 3.80
Unknown: Opposed 27.08 9.95
Unknown: Undecided 25.92 9.20
Source SS df MS 1?
Audience .16 1 .16 1 n.s.
Knowledge of

source .71 1 .71 1.20 n.s.
Interaction 1.62 1 1.62 2.76 n.s.
Error 80.29 136 .59

Total 82.73 139'

 

In Table 12 an analysis of variance of mean trustworthiness scores

indicates no effect whatever.

Table 12. Means, standard deviation, and analysis of variance of
trustworthiness scores of experimental groups.

 

 

Condition Mean Standard Deviation
Known: Opposed 30 . 11 3 . 39
Known: Undecided 30 . 92 3 .19
Unknown: Opposed 29.89 3.90
Unknown: Undecided 29.56 3.75

27

Table 12 (contd)

Source SS df MS F
Audience .260 1 .260 < 1 n.s.
Knowledge of

source .002 1 .002 < 1 n.s.
Interaction .032 1 .032 < 1 n.s.
Ekmtm~ 51.599 ' 136 .379

Total 51.838 1337

 

werd count. To test fer differences in the amount of subject

 

effort among the experimental groups , am analysis of variance was per-
formed using total word count scores. Table 13 indicates that there

was no significant effect.

Table 13. Means, standard deviations, and analysis of variance of
essay length.

 

 

Condition Mean Standard Deviation
Known: Opposed 171.91 51.01
Known: undecided 163.61 99.66
unknown: Opposed 172.72 90.57
unknown: undecided 181.00 59.38
Source SS df MS P
Audience .00 1 .00 < 1 n.s.
Knowledge of

source 82.77 1 82.77 1.15 n.s.
Interaction 68.72 1 68.72 < 1 n.s.
Extra: 9529.27 139 72.12

Total 9675 . 76 ITT

 

Chapter IV

DISCUSSION

Although this experiment failed to demonstrate support for the
three hypotheses , plausible reasons for non—confirmation can be found
within the framework of dissonance theory. Despite this disappoint-
ment, the study has provided additional speculative evidence regarding

the loci of dissonance arousal.

Failure to Confirm Hypothesis 1

 

This hypothesis predicted that dissonance and subsequent
attitude change would be greater for the known audience condition. It
was no small surprise when the inverse of this prediction emerged.
After considering all aspects of the experimental procedure and the
obtained results, however, possible explanations for this effect can
be posited.

The most likely reason for the greater attitude change in the
Unknown Audience condition is based on the Known Audience subjects'
ability to "take back" their essays if they should ever be confronted
by an acquaintance. This possibility finds support in the study by
Helmmeich and Collins (1968). Subjects in that study who were given
an opportunity to take back their belief discrepant messages failed
to demonstrate the dissonance effect, while those who were unable to
to take back the message did experience dissonance .

28

29

Similar to this rationale is the one expressed by a student
dwing debriefing for the present study. He said that if his friends
were to read the essay, they would know that it did not express his
true opinion. Thus subjects in the Known Audience conditions may have
been unconcerned about the effects of their essays . When friends are
familiar with your attitudes in general, there is little danger of
your belief discrepant essay influencing them. For those in the Known
conditions, affixing of signatures to the essay may have been insur—
ance that the essays would not be taken seriously by the supposed
audience. .

Another possible reason for failure to obtain a knowledge of
audience effect may be found in the joint measuring of attitude change
and salience change at the same time . While subjects were given four
scales on which to note attitude intensity change, they were at the
same time given two scales on which to note change in salience. It
is likely that these two measures worked against each other. Changing
salience for some subjects may have been an alternate means of dis-
sonance resolution, thus enabling them to maintain their previous
attitude intensity and direction. Supporting this speculation is the
somewhat greater, although not significant, negative salience changes
recorded for the known audience than for the unknown. The opposite
occurred for attitude change: greater positive change for the mflcnown

audience than for the known .

Failure to Confirm Hypothesis 2

 

Even though the results of this effect were in the predicted

30
direction, the audience attitude hypothesis was not confirmed. Both
Nel gt El: (1969) and Bodaken (1970) reported significantly greater
attitude change fer those persons encoding fer'neutral audiences. In
light of suCh strong theoretical support for an audience attitude
effect, it was especially disappointing that it was not supported in
this study.

One factor Which may help to explain the failure to confirm
the audience attitude effect is the possibility that subjects in the
Known conditions believed.that they could "take back" their’essays,
thus reducing dissonance in both the Known: Opposed and Known:
Undecided groups. If subjects perceived the possibility for’message
"take back," the fear of aversive consequences to the supposed audience
might have been.rended, thus militating against dissonance arousal
and subsequent attitude change.

Despite the fact that the audience attitude effect did not
reach statistical significance, it is especially provocative to note
that subjects in the Undecided conditions reported that they had: (1)
experienced significantly greater‘anxiety'while encoding the belief—
discrepant essays (F1,136 = 6.79), (2) believed their~essays to be
significantly more persuasive (Pl,136 = 9.93) and (3) perceived that
their essays would have more harmful effects upon their supposed
audience (Fl,136 = 7.87). These indices suggest that the psychological
stresses were present fer greater~dfissonance arousal in both of the
Undecided Audience conditions, but the audience attitude effect fer

attitude change still was short of statistical significance.

31

Tie Locus of Dissonance Arousal

 

It was the hope of the experimenter to address in some mean-
ingful way the issue regarding the locus of dissonance arousal in
counterattitudinal advocacy. For it is by clarifying this issue that
more reliable predictions may be made about the types of cognitive
reorganizations that are likely to be made by individuals when con—
fronted by inconsistent situations .

Aversive consequences. The debate can be restated, in essence,

 

as existing between Festinger (1957), who believed dissonance was
aroused when the individual realized the inconsistency between what he
believed and what he said ("I believe 5, but I advocate pg: 29")

those recent dissonance theorists who see the inconsistency as
existing between what an individual says and the possible harmful
effects which may result from such statements. Aronson (1968) puts

it this way:

In the experiments on counterattitudinal advocacy for
example, I would suggest that it is incorrect to say
that dissonance existed between the cognitions "I
believe the task is dull [X]" and "I told someone the
task was interesting [not X]." . . . What is dissonant
is the cognition "I am a decent, truthful human being,"
and the cognition "I have misled a person; I have
conned him into believing something which just isn't
true: he thinks that I really believe it and I cannot
set him straight because I probably won't see him
again." (p. 29) (The brackets are mine.)

This view subscribes to the belief that dissonance arousal results from
concern for the well-being of receivers of belief—discrepant messages.
The present experiment lends support to this vieWpoint . As noted above,
subjects who encoded for the undecided audience believed that this

audience would experience greater harmful effects as the result of

32
reading the counterattitudinal essays. Subjects also perceived that
this audience would be more strongly influenced by the essays than
would those who were "strongly Opposed" to the issue in questio --
noted by the higher perceived persuasiveness of the essays for the
undecided audience . Although the audience attitude effect for atti-
tude change fell short of statistical significance, the above findings
support the conclus ion that source concern for aversive consequences
to the receiver does assist in the arousal of cognitive dissonance.
It could be reasonably argued, however, that the non-significance of
the attitude effect places the above proposition on shaky footing.
In light of this, one consideration should be offered. Whereas in the
Nel gt_a_1_. (1969) and Bodaken (1970) studies, the topics of "legal-
ization of marijuana" and "required living on campus by all students"
were used respectively, the topic here concerns a national policy of
mandatory military service for all men before college. It is possible
that the topic used in this experiment might have been perceived as
being beyond the scope of direct harm to the supposed audience.
Decisions to smoke marijuana or live for four years on a college campus
are questions that may be decided by the audience in question. However,
subjects may have perceived that the question of mandatory military
service was beyond the decis ion—making parameters of those in the
audience .

Threatened self-image . Another viewpoint regarding the locus

 

of dissonance arousal has been expressed by Bramel (1968). His view
is that dissonance is the result of threatened self-image. Expression

of belief-discrepant opinions is dissonant with one's own self-

33
perceptions. He says:
. . dissonance is a feeling of personal unworthiness

(a type of anxiety) traceable to rejection of oneself

by other people either in the present or in the past .

Any information which implies that one is incompetent

orimmoral arouses dissonance. (p. 365)

Part of this speculation is supported by this experiment. Consistant
with the belief that dissonance is traceable to "a type of anxiety,"
it was observed that subjects encoding for the undecided audience
experienced significantly greater anxiety than did those encoding for
the strongly opposed audience. The question, however, still remains
as to whether or not such anxiety is equivalent to "personal unworthi—
ness" and "traceable to rejection of oneself by other people . . . ."
The greater anxiety consistent with an audience attitude effect could
be attributable to causes other than feelings of personal unworthiness
by the subjects. Such anxious feelings could have resulted from
concern for aversive consequences to the receiver and not any sense of
personal guilt.

The possibility that dissonance is aroused by threatened self—
image is not supported by the findings of this experiment. There was
no reported effect based upon self-perception scores on either the
trustworthiness or qualification dimensions. There are, however, some
possible explanations for the lack of significant differences among
self-perception scores. One reason may be that self-image is too
firmly rooted to be easily changed by committing a single act of counter—
attitudinal advocacy . Another explanation could be that the Berlo gt 2.1.

(1970) scales are not sensitive enougn to detect any changes in self—

perception. To the experimenter's knowledge this study is the first to

39 i
use these scales for self-evaluation, all others have been for the
evaluation of other individuals or institutions . It is possible that
subjects asked to evaluate themselves , experienced a type of defensive
avoidance reaction . This may have accounted for generally higher
scores in the experimental groups than were observed in the Control
group.

Another possible explanation for not recording any differences
in self-perception may be the time of self-evaluation. Since dis—
sonance is a post—decisional phenomenon, it is reasonable to assume
that threatened self-image is greater immediately after the decision
has been made to encode a belief-discrepant message. Threatened self-
image should be reduced after attitude change scores have been taken.
In the posttest instrument, self-evaluation was the last item. By
the time subjects reached this part of the questionnaire, dissonance
may have been adequately resolved by means of attitude or salience

change .

Implications for Future Research

 

The failure to confirm the hypothesis concerning the knowledge
of audience variable probably resulted from no weakness in the theo-
retical suppositions, but rather from weaknesses of experimental
design. V Because of the takeback possibility inherent among subjects
in the known audience conditions , subjects may not have experienced
the dissonance necessary for attitude change. In a sense, this experi-
ment has amounted to a replication of the takeback study of Helmreich

and Collins (1968) . Future studies should continue to investigate the

35
efficacy of the knowledge variable . However, the essential method—
ological factor is subjects ' perceptions of probability of identifi-
cation with the belief—discrepant message. As long as the encoder
believes that the audience will identify him personally with the
message, the pressure toward dissonance will be reduced. The present
study has sham that some dissonance will be produced even when the
audience is Imknown to fine souroe. This leads the experimenter to
believe that he would have observed the predicted results had he not
asked subjects to place their signatures on the essays. If it is
assumed that dissonance is aroused through concern for aversive con-
sequences, then it is reasonable to presume that the desired effect
mignt emerge when the encoder's name is 3393 on the belief-discrepant
message. Omission of the name would lessen the possibility of a
subsequent takeback.

Although the threatened self-image theory of dissonance arousal
was not given any direct support in this experiment, the researcher
believes that there are certain refinements in methodology that should
be made when testing for threatened self-image. First, other instru-
ments for self—evaluation ought to be tested. It is possible that the
Berlo scales were not sensitive enough for self-examination. Second,
the researcher ought to test for changes in self-image immediately
after dissonance arousal and before other alternatives for resolution
have been presented.

The issue regarding the locus of dissonance arousal is critical
to the process of identifying and understanding those variables which

stimulate cognitive reorganization in inconsistent situations .

36

Under this assumption the present study may be justified.

Summary

The major purpose of this study was to explore certain aspects
of the issue regarding the locus of dissonance arousal in situations
involving counterattitudinal advocacy. The center of concern was on
the manipulation of audience attitude. toward the counterattitudinal
issue and encoder knowledge of the supposed audience. It was pre-
dicted that attitude change would be greater when subjects wrote
essays for undecided receivers. It was further hypothesized that atti—
tude change would be greater when subjects peroeived a greater prob-
ability that the audience was known. The final hypothesis predicted
that the greatest amount of attitude change would be for the Known,
Undecided Audience, while the least change would be for the Unknown,
Opposed Audience .

None of the stated hypotheses were confirmed in this experiment,
although an inverse relationship was found between lmowledge of audience
and amount of attitude change . Greater attitude change was reported
by subjects who encoded belief discrepant messages for the unknown
audience. This effect was explained on the basis that subjects believed
they would be able to take bad< their essays for the known audience ,
but not for the unknown. Subjects thus perceiving that message take
back was possible were given an out which militated against dissonance

arousal.

REFERENCES

Ableson, R. P. Psychological implications. In Ableson, R. R. ,
McGuire, W. J., Newcomb, T. R., Aronson, E., and Tannenbaum,
P. H. (Eds.), Theories of Cognitive Consistency: A Source—
book. Chicago: Ra—nd M13113; and (50., I968, 1I2-139.

 

Aronson, E. Dissonance theory: Progress and problems. In Ableson,
R. P., McGuire, W. J., Newcomb, T. R., Aronson, E., and
Tannenbaum, P. H. (Eds.) Theories of Cognitive Consistency:
_A_ Sourcebook. Chicago: WI? mm

 

Aronson, E. The psychology of insufficient justification: An analysis
' of some conflicting data. In Feldman, 8. (Ed.) Cognitive
Consistency: Motivational Antecedents and BehavioraI Con-

seguents. mew York and Tendon: AcademicTress, 1966,_II'5~136.

 

 

Berlo, D. K., Lemert, J. B., and Mertz, R. J. Dimensions for
evaltating the acceptability of message source. The Public
Qinion Quarterly, 1970, g, 563-576.

 

 

Bodaken, E. M. Choice and perceived audience attitude as determinants
of cognitive dissonance and subseqtent attitude change following
counterattitudinal advocacy . Unpublished dissertation,
Department of Communication, Michigan State University, 1970.

Bramel, D. Dissonance, expectation and the self. In Ableson, R. P.,
McGuire, W. J., Newcomb, T. R., Aronson, E., and Tannenbaum,
P. H. (Eds. ) , Theories 3i: Cognitive Consistency: A Sourcebook.
Chicago: Rand McNally and Co. , 1968, 355-365.

 

 

Brehm, J. W. and Cohen, A. R. Choice and chance relative deprivation
as determinants of cognitive dissonance . Journal 9: Abnormal
and Social Psyclnologl, 1959, _5_8_, 383-387.

 

Brehm, J. W. Explorations _°_11_n_ Cognitive Dissonance. New York: Wiley,
1962.

 

 

Carlsmith, J. M., Collins, B. E., and Helmtreich, R. L. Studies in
enforced compliance: I. The effect of pressure for compliance
on attitude change produced by face-to-face role playing and
anonymous essay writing. Journal of Personality and Social
Eychology, 1966, §_2_, 256-117—2 . —

 

 

37

38

Carlsmnith, J . M. Varieties of counterattitudinal behavior. In
Abelson, R. P., McGuire, W. J., Nechmb, T. R., Aronson, E.,
and Tannenbaum, P. H. (Eds.), Theories of Cognitive Consistency:
A Sourcebook. Chicago: Rand WTW W

Collins, B. E. The mediation of change due to counterattitudinal
behavior. In Abelson, R. P., McGuire, W. J., Newcomb, T. R.,
Aronson, E., and Tannenbaum, P. H. (Eds.), Theories of
Cognitive Consistengg: A Sourcebook. Chicago: Ra—fid Maially

3

an ., 6.

 

Collins, B. E. The effect of monetary inducements on the amount of
attitude change produced by forced compliance. In Elms, A. C.
(Ed.), Role Playing, Reward, and Attitude Change. Cincinnati:
Van Nostrand, I969, 2991—22-3.

 

Festinger, L. A Theo 9: Cognitive Dissonance. Stanford, Calif .:
Stanford University Press, 1957.

 

Festinger, L. and Carlsmnith, J. M. Cognitive consequences of foroed
compliance. Journal of Abnormal and Social Psychology, 19 59 ,
33, 203-210. *—

 

Helmreich, R. L. and Collins, B. E. Studies inn forced compliance:
Commitment and magnitude of inducement to comply as determi-
nants of opinion chance. Journal of Personality and Social
PsychologI, 1968, 10, 75-8I. '—

 

 

Miller, G. R. Counterattitudinal advocacy: A current appraisal. An
' unpublished paper, Department of Communication, Michigan State
University, 1970.

Nel, E., Helmreich, R. L. and Aronson, E. Opinion change in the
advocate as a function of the persuasibility of his audience:

a clarification of the meaning of dissonance. Journal _of
Personality and Social Psychology, 1969, l_2_, 11'7-1'99.

 

Osgood, E. E., Suci, G. J., and Tannenbaum, P. H. The Measurement of
Meaning. Urbana, Illinois: University of Illih'ois Press,19§"7.

 

Rosenberg, M. J. When dissonance fails: On eliminating evaluation
apprehension from attitude measurement . Journal 21: Personality
and Social Psychology, 1965, I, 28-92.

 

 

APPENDIX A
PRETEST ATTITUDE QUESTIONNAIRE

Dear GMI Student:

You are being asked to participate in a project that students
in all accredited colleges in Michigan are helping with. It
is an opinion survey concerning college student's attitudes
toward a variety of issues. We hope you will be willing to
help.

You might wonder why we are asking you to put your name and
college on this form. Very simply we will be correlating
these answers with other data available and will be doing
this project again at GMI. If you participate now you will
not be asked to again. All data is anonymous and will not be
seen by anyone at GMI .

I have been asked to coordinate this project at GMI and am
very grateful for your help .

Sincerely ,

Michael Burgoon

Assistant Professor of Communication
Department of Communication

and Organizational Behavior

39

90

COLLEGE OPINION PROFILE

Name Student Number-

 

College Class Time Instructor

 

The purpose of this survey is to solicit the opinions of students on
a variety of current issnes--campus and national. On eacln of the
following pages you will find a number of issues each followed by a
series of descriptive scales. For example, here is an item like those
you will see:

The United States should withdraw from the United Nations .
Bad : : : : : : Good
Your job is to place a check-mark (X) above the line that best indi-
cates your judgrent about the proposition. For example, if you feel
that U. S. withdrawal should be very bad, you would check as follows:
Bad X : : : : : : Good

If you feel that such a move (withdrawal would be quite bad, you
should check as follows:

Bad :X: : : : : Good

 

If you feel neutral or indifferent about the proposition, or if you
feel that the scale is irrelevant to the proposition, you would check
as follows:

Bad: : :X: : : Good

Remember: Never put more than one-check-mark on a single scale
and be sure that each check is in the middle of the
line, not on the boundaries . ’

ALL SCALES SHOULD BE CHECKED —- II) NOT OMIT ANY.

WORK RAPIDLY —— RECORD FIRST IMPRESSIONS -- DO NOT CHANGE MARKS.

The use and possession of marijuana should be legalized.

Trivial

Crucial

 

Good

Bad

 

Worthless

Valuable

 

Pleasant

Unpleasant

 

unfair

Fair

 

Imptmtant

Unimportant

Draft deferments should be eliminated fer college students.

Good

Bad

 

unfair

Fair

 

Crucial

Trivial

 

worthless

Valuable

 

Pleasant

Unpleasant

 

Unimportant

Important

 

Black students should be advised to majornin less rigorous curricula.

Unpleasant

Pleasant

 

Bad

Good

 

Important

Unimportant

 

Fair

Unfair

 

 

Valuable
Crucial

Vkmthless

Trivial

 

92

All students should be required to live on campus fer at least two
years of their college attendance. -_—-

 

 

 

 

 

Unimportant Important
Valuable worthless
Good : Bad
Unfair : Fair
Pleasant unpleasant
Trivial : Crucial

 

All college male students should be required to serve at least two
years in the armed service before being accepted by a college or

 

 

 

 

 

 

university.

Good Bad
Important : ‘Unimportant
worthless _____Valuab1e

Crucial Trivial
Unpleasant : Pleasant
Fair : Unfair

 

APPENDIX B

POS'I'IEST ATTITUDE QUESTIONNAIRE

Form 5-36C

93
Name ___________________________________________________ Date ....................................... .

College

 

_- Class Time ___________ Instructor ______________________

Now that you have written your essay, we would like to know how you feel toward
the issue. PLEASE RESPOND ON EACH OF THE FOLLOWING SCALES.

All male college students should be required to serve at least two years
in the armed service before being accepted by a college or university.

_______ Very Pleasant

________ Quite Pleasant

....... Slightly Pleasant

________ Neutral

........ Slightly Unpleasant

__--_-__ Quite Unpleasant

________ Very Unpleasant

All male college students should be required to serve at least two years
in the armed service before being accepted by a college or university.

........ Very Crucial

________ Quite Crucial

________ Slightly Crucial

........ Neutral

________ Slightly Trivial

-___-___ Quite Trivial

________ Very Trivial

All male college students should be required to serve at least two years
in the armed service before being accepted by a college or university.

________ Very Good

________ Quite Good

---_-___ Slightly Good

________ Neutral

________ Slightly Bad

--_-__-_ Quite Bad

________ Very Bad

MSU- 11- 70

Form 5-36C
an

All male college students should be required to serve at least two years

in the armed service before being accepted by a college or university.
........ Very Valuable
........ Quite Valuable

Slightly. Valuable

________ Neutral

________ Slightly Worthless

........ Quite Worthless

________ Very Worthless

All male college students should be required to serve at least two years
in the armed service before being accepted by a college or university.

________ Very Unfair

________ Quite Unfair

_______ Slightly Unfair

________ Neutral

___--___ Slightly Fair

________ Quite Fair

________ Very Fair

All male college students should be required to serve at least two years

in the armed service before being accepted by a college or university.
________ Very Important

-_-_ __ Quite Important

__- _ Slightly Important
________ Neutral

_______ Slightly Unimportant
........ Quite Unimportant

........ Very Unimportant

Before you wrote your essay, how did you think your reading audience felt
toward the issue in question?

........ They were favorable toward two years mandatorymilitary service for
male college students.

________ They were undecided toward two years mandatorymilitary service for
male college students.

MSU- 11- '70

Form 5-36C
1+5

........ They were opposed to two years mandatory military service for male
college students.

........ I don't know what their attitude was toward two years mandatory
military service for male college students.

RESPOND TO THE FOLLOWING STATEMENTS BY PLACING AN "X" ON
THE LINE WHICH MOST NEARLY REPRESENTS YOUR FEELINGS.

Statement One

 

My essay may have a harmful effect upon those who will be reading it.

........ Strongly Agree

________ Mildly Agree

--______ Neither Agree nor Disagree
________ Mildly Disagree

_______ Strongly Disagree

Statement Two

 

Some of the students who are to read my essay will know who I am.

 

________ Highly Probable
........ Quite Probable
________ Slightly Probable
........ Neutral

........ Slightly Improbable
________ Quite Improbable
________ Highly Improbable

Statement Three

 

How do you evaluate the persuasive effectiveness of the essay you have
just written?

________ Very Strong

________ Quite Strong

........ Slightly Strong

........ Neutral

________ Slightly Weak

________ Quite Weak

________ Very Weak

MSU- 1 1- '70

Form 5-36C
1+6

Statement Four

 

I felt anxious about preparing this message for my reading audience.
________ Strongly Agree
________ Mildly Agree
________ Neither Agree Nor Disagree
--__-__- Mildly Disagree
........ Strongly Disagree

Statement Five

 

I will be able to explain my true beliefs to those who will be reading my essay.

 

........ Strongly Agree

-------- Mildly Agree

________ Neither Agree Nor Disagree
_-____-_ Mildly Disagree

________ Strongly Disagree

Statement Six

 

Please evaluatejourself on the following scales.
WORK RAPIDLY -- MARK EVERY SCALE -- RECORD FIRST IMPRESSIONS:

 

 

 

 

 

 

 

 

 

 

Safe : : : : : : Unsafe
Untrained : : : : : : Trained
Just : : : : : : UHJUSt
lnexperienced : : : : :' : Experienced
Kind : : : : . : : Crud
Unskilled : : : : : : Skilled
Friendly : : : : : : Unfriendly
Unqualified : : : : : : Qualified
Honest : : : : : : Dishonest
Uninformed : : : : : : Informed

 

Note: Recheck to be sure you have responded to every item in the questionnaire.

WHEN FINISHED, TURN QUESTIONNAIRE OVER AND AWAIT FURTHER
INSTRUCTIONS.

MSU-ll- '70

APPENDIX C
CONTROL POS'ITEST ATTITUDE QUESTIONNAIRE

Form 3- 36D
1+7

MICHIGAN COLLEGE INFORMATION EXCHANGE PROJECT)“

Name Date

 

College Instructor Class Time

 

Recently, the activities of college students across the country have
provided a multitude of news stories for the mass media. Yet With all the
publicity surrounding student revolt and unrest, little is really known about
student attitudes toward a number of issues related to student life on our col»
lege campuses. On the following pages are. several issues that are of concern
to college students. Each issue is followed by a set of descriptive scales. We
would like to have you place a check (./) on the line that you think best indicates

your Opinion toward the issue. Each scale should be checked --~ do not omit

 

any.

*Sponsored by the Averell Foundation Attitude Study Fund.

MSI_I-11-'7O

Fonn 5-36D

The voting age should be lowered to 18 years.

 

1+8

 

 

 

 

Slightly
Very Bad Quite Bad Bad Neutral
Very Quite. Slightly
Important Important Important Neutral
Very Quite Slightly

Unpleasant Unpleasant

 

 

 

Very Fair Quite Fair
Very Quite
Trivial Trivial
Very Quite
Worthless Worthless

Unpleasant Neutral

 

' Slightly

 

 

 

Good Quite Good Very Good
Slightly Quite Very

Unimportant Unimportant Unimportant

 

 

 

 

 

Slightly
Fair Neutral
Slightly
Trivial Neutral
Slightly

Worthless Neutral

 

 

 

Slightly Quite Very
Pleasant Pleasant Pleasant
Slightly I Quite Very

Unfair Unfair Unfair
Slightly Quite Very
Crucial Crucial Crucial

Slightly Quite Very
Valuable Valuable Valuable

All male college students should be required to serve at least two years in the armed
service before being accepted by a college or university.

 

Very
Ihnfleasant

Qune
llnplt asant

 

 

 

 

 

Very Bad Quite Bad
Very Quite
Crucial CruCial
Very Quite
Valuable Valuable
Very Qu1te
Unfair Unfair
Very Quite
Important Important.

‘MSU- ll~ '70

 

Slightly
Unpleasant Neutral

 

' Slightly

 

 

 

 

Bad Neutral
Slightly
Crucial Neutral
Slightly
Valuable Neutral
Slightly
Unfair Neutral
Slightly
Important Neutral

 

 

“Si ightly

 

 

 

 

 

 

 

 

Qune Vtry
Pleasant Pleasant Pleasant
~ Slightly
Good Quite Good Very Good
. Slightly .. Quite Very
Trivial Trivial Trivial
Slightly Quite Very
Worthless Worthless Worthless
Slightly Quite
Fair Fair Very Fair
Slightly Quite Very

Unimportant Unimportant U nimportant

Form 5-36D

Draft deferments should be eliminated for college students.

1+9

 

 

Slightly

 

 

 

Very Fair Quite Fair Fair Neutral
Very Quite Slightly
Trivial Trivial Trivial Neutral
Very Quite Slightly

Worthless Worthless

Worthless Neutral

 

 

Slightly

 

 

Very Good Quite Good Good Neutral
Very Quite Slightly

Unimportant Unimportant Unimportant Neutral

 

Very
Pleasant

Quite
Pleasant

 

Slightly
Pleasant

Neutral

 

Slightly

Very

 

 

 

 

 

 

 

Quite
Unfair Unfair Unfair
Slightly Quite Very
Crucial Crucial Crucial
Slightly Quite Very
Valuable Valuable Valuable
Slightly Quite
Bad Bad Very Bad
Slightly Quite Very .
Important Important Important
Slightly Quite Very
Unpleasant Unpleasant Unpleasant

All students should be required to live on campus during their college attendance.

 

 

 

 

 

 

 

 

Very Quite Slightly
Important Important Important Neutral
Very Quite Slightly
Valuable Valuable Valuable Neutral
Quite Slightly
Very Good Good Good Neutral
Very Quite Slightly
Trivial Trivial Trivial Neutral
Very Quite Slightly
Unfair Unfair Unfair Neutral
Very Quite Slightly
Pleasant Pleasant Pleasant Neutral

MSU-11-7O

 

Slightly

Quite

Very

Unimportant Unimportant Unimportant

 

 

 

 

 

Slightly Quite Very
Worthless Worthless Worthless
Slightly

Bad Quite Bad Very Bad
Slightly Quite Very
Crucial Crucial Crucial
Slightly Quite

Fair Fair Very Fair

Slightly Quite Very

Unpleasant Unpleasant Unpleasant

Form 5-36D

50

Military activity in South Viet Nam should be significantly increased.

 

Quite

 

Very
Trivial Trivial
Very Quite

Worthless Worthless

 

.Quite

 

Very Fair Fair
Very Quite

Unpleasant Unpleasant

 

 

Slightly

Trivial Neutral

 

Slightly
Worthless Neutral

 

' Slightly

Fair Neutral

 

Slightly
Unpleasant Neutral

 

' Slightly

 

Very Bad Quite Bad Bad Neutral
Very ' Quite Slightly
Important Important Important Neutral

 

 

Slightly

Quite

 

 

 

 

 

 

 

 

 

Very
Crucial Crucial Crucial
Slightly Quite Very
Valuable Valuable Valuable
. Slightly - Quite Very
Unfair Unfair Unfair
Slightly Quite Very
Pleasant Pleasant Pleasant
Slightly Quite Very
Good Good Good
Slightly Quite Very

Unimportant Unimportant Unimportant

Draft card burners should automatically be reclassified l-A by their draft boards.

 

Quite

 

 

 

Slightly

 

 

 

 

 

 

Very
Unfair Unfair
Very Quite
Important Important
Very Quite
Valuable Valuable
Very Quite
Trivial Trivial
Very .Quite
Good Good
Very Quite
Pleasant Pleasant

MSU-11-‘70

 

 

Unfair Neutral
Slightly
Important Neutral
Slightly
Valuable Neutral
° Slightly
Trivial Neutral
' Slightly
Good Neutral
Slightly
Pleasant Neutral

 

 

Quite

 

 

Slightly
Fair Fair Very Fair
Slightly Quite Very

Unimportant Unimportant Unimportant

 

 

 

 

 

 

 

 

Slightly Quite Very
Worthless Worthless Worthless
. Slightly . Quite Very
Crucial Crucial Crucial
' Slightly .Quite
Bad Bad Very Bad
Slightly Quite Very
Unpleasant Unpleasant Unpleasant

Fo rm 5- 36 D

51

The college grading system should be abolished.

 

 

' Slightly

 

' Slightly

Quite

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU-ll- '70

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

' Very Bad Quite Bad Bad Neutral Good Good Very Good
Very Quite Slightly Slightly ° Quite Very
Pleasant Pleasant Pleasant Neutral Unpleasant Unpleasant Unpleasant
Very Quite Slightly Slightly Quite Very
Unimportant Unimportant Unimportant Neutral Important Important Important

Slightly Slightly Quite
Very Fair Quite Fair Fair Neutral Unfair Unfair Very Unfair
Very Quite Slightly Slightly Quite Very
Worthless Worthless Worthless Neutral Valuable Valuable Valuable
Very Quite Slightly Slightly Quite Very
Crucial Crucial Crucial Neutral Trivial Trivial Trivial
The United States should withdraw from the United Nations.
Very Quite Slightly Slightly Quite Very
Unpleasant Unpleasant Unpleasant Neutral Pleasant Pleasant Pleasant.
Very Quite Slightly 1 Slightly Quite Very
Trivial Trivial Trivial Neutral Crucial Crucial Crucial
Slightly Slightly Quite
Very Bad Quite Bad Bad Neutral Good Good Very Good
Very Quite Slightly Slightly Quite Very
Valuable Valuable Valuable Neutral Worthless Worthless Worthless
Very Quite Slightly Slightly Quite Very
Important Important Important Neutral Unimportant Unimportant Unimportant
Very 'Quite ' Slightly ' Slightly Quite
Unfair Unfair Unfair Neutral Fair Fair Very Fair

Form 5-36D
52

Presidential nominations should be based on a direct vote of the peOple.

 

' Quite ' Slightly

Slightly

 

Quite

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Very
Crucial Crucial Crucial Neutral Trivial Trivial
Quite Slightly Slightly Quite
Very Fair Fair Fair Neutral Unfair Unfair
Very Quite Slightly Slightly Quite
Worthless Worthless Worthless Neutral Valuable Valuable
Quite Slightly Slightly Quite
Very Good Good Good Neutral Bad Bad
Very Quite Slightly Slightly Quite
Unimportant Unimportant Unimportant Neutral Important Important
Very Quite Slightly Slightly Quite
Pleasant Pleasant Pleasant Neutral Unpleasant Unpleasant
All firearms should be licensed and registered.
Very Quite Slightly Slightly Quite
Worthless Worthless Worthless Neutral Valuable Valuable
Quite Slightly Slightly Quite
Very Fair Fair Fair Neutral Unfair Unfair
Very Quite Slightly Slightly Quite
Crucial Crucial Crucial Neutral Trivial Trivial
Very Quite Slightly Slightly Quite
Unpleasant Unpleasant Unpleasant Neutral Pleasant Pleasant
Quite Slightly Slightly Quite
Very Bad Bad Bad Neutral Good Good
Very Quite Slightly Slightly Quite
Unimportant Unimportant Unimportant Neutral Important Important

MSU-11-7O

 

Very
Trivial

 

Very Unfair

 

Very
Valuable

 

Very Bad

 

Very
Important

 

Very
Unpleasant

 

Very
Valuable

 

Very
Unfair

 

Very
Trivial

 

Very
Pleasant

 

Very Good

 

Very
Important

   

zes

"‘llfllllljllllllllllll'lllllljll