PLACE ll RETURN BOX to roman this chockom from your record.
To AVOID FINES Mum on or baton data duo.

DATE DUE DATE DUE DATE DUE

3...; t

   
   

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU IaAn Afﬂmnﬁvo WW Opponunlly Institution
Walla-9.1

 

ABSTRACT

THE EFFECTS OF PERCEIVED POTENTIAL REWARD
ON INDIVIDUATION
By

Michael Eugene Pacanowsky

For various reasons, actors attempt to control interactional
situations by managing their self—presentations. One aSpect of impres-
sion management that has been little studied is the extent to which
actors' self-presentations serve to direct attention toward or away from
themselves. This self-presentational tactic is labelled individuation.
Individuation is conceptualized as the extent to which a person, as a
social object, is distinguishable from other persons, as social objects.
It is hypothesized that individuation will be a positive linear function
of the perceived potential reward structure of interaction situations.

In an experiment, one hundred nine students wrote self-descriptive essays
in three conditions of varying potential rewards. Results indicate that
individuation is affected by reward structure, but that the relationship

is more complex than the simple linear one hypothesized. Verbal expres-
sions are responsive to the eXperimental manipulations, in that subjects
experiencing the possibility of a positive or negative outcome use

verbal eXpression similarly, and differently from subjects not exPeriencing
the possibility of a reward or punishment. Results also indicate that
individuation is relatively unaffected by the sex, Machiavellianism, or

self-esteem of the subject.

THE EFFECTS OF PERCEIVED POTENTIAL REWARD

0N INDIVIDUATION

By

Michael Eugene Pacanowsky

A THESIS

Submitted to

Michigan State University

in partial fulfillment of the requirements

for the degree of

MASTER OF ARTS

Department of Communication

1977

Accepted by the faculty of the Department of Communication,
College of Communication Arts, Michigan State University, in partial

fhlfillment of the requirements for the Master of Arts degree.

54.7%.” 4%

Director of Thesiér

 

(5:;22:”%”~477/ /(f:23¢)4(/ , Chairperson

Guidance Committee:

 

 

ACKNOWLEDGMENTS

Many peOple assisted me in this study. I wish to thank Professor
Lucy Rey, Western Michigan University, for her cooperation in conducting
the experiment. Stathi Afendoulis, Larry Gunsberg, Steve Prevost,
Lisa Beth Roth, Sue Schemmel, Liz Scott, Chip Steinfeld, and Jay Todd
served admirably as coders. I would also like to thank Joseph Woelfel
and Donald Cushman, members of my committee for their suggestions.
Special thanks go to Edward Fink, chairperson of my committee, for the
countless suggestions, questions, comments, and encouragements that

made doing this thesis 3 most worthwhile experience.

ii

TABLE OF CONTENTS

Table of Contents .

List of Tables

List of Figures .

CHAPTER ONE: THEORY .

l
1

l

1

1

.1
.2

.3

.4

.5

Introduction . . . . . . . . . . . . .
Theoretical Overview: Interaction as Control

1.2.1 The function of rules
1.2.1 Role of communication

Impression Management .

1.3.1 Conceptual framework . . . .
1.3.2 Ingratiation and self- ~presentation .
1.3.3 Individuation

Social contingency .

Procedural rules .

Operational rules and the nature of
control .

Role of Other Factors in the Individuation
Process
Summary .

CHAPTER TWO : METHOD

NNNN
AMNH

NNN
\IO‘U'I

Overview of Experimental Design .
Experimental Independent Variable .
Nonexperimental Independent Variables .
Dependent Variables

2.4.1 Verbal Measures: expression .

Syntax . .
Type of information eXpressed

2.4.2 Verbal measures: impression .
2.4.3 Nonverbal measures .

Coder Training

Ethical Considerations
Summary .

iii

Page

. iii

. vii

{ONO‘

. 12
. 13

. 15
. 16
. 17
. 19
. 19
. 22'
. 24
. 25
. 25

. 25
. 27

. 27
. 28

. 29
. 32
. 33

CHAPTER THREE: FINDINGS AND DISCUSSION .

muons
thu-t

3.6

Overview .

Results of Data Analysis Pertaining to Sample

Results of Data Analysis Pertaining to Measures

Results of Data Analysis Pertaining to
Experimental Manipulation . .

Results of Data Analysis Pertaining to the
Relationship Between Individuation and
Perceived Potential Reward

Summary

CHAPTER FOUR: IMPLICATIONS .

4.1 Summary of Results .
4.2 Failure to Reject the Null Experimental Hypothesis
4. 2.1 Failure to examine interaction effects
4. 2. 2 Use of untransformed data .
4. 2. 3 Multicollinearity . . .
4. 2. 4 Summary of problems of analytical methods .
4.3 Quadratic Effect of Experimental Condition .
4.4 Future Research . . .
4.5 Summary
APPENDIX A: FORMS E AND F
APPENDIX B: INTERACTION ANALYSIS PROFILE .
APPENDIX C: VALIDATION (PRETEST) RESULTS .
APPENDIX D: CODING FORM
APPENDIX E: DEFINITIONS OF SELF-PRESENTATIONAL TACTICS .
APPENDIX Pz' CODING SCHEDULE
FOOTNOTES .

BIBLIOGRAPHY

iv

34

34
36
41

55

57
77

79

79
8O

83
84
85
86

86
89
90

Table

Table

Table

Table

Table

Table

Table

Table

Table
Table
Table
Table
Table

Table

Table

Table

Table

9A:

QB:

9C:

9D:

9E:

10:

11A:

118:

12:

LIST OF TABLES

Description of Sample .

Intercorrelations Among Nonexperimental
Independent Variables .

Intercorrelations Among Experimental and
Nonexperimental Independent Variables .

Internal Consistency of Personality Measures

Descriptive Statistics and Intercoder Reliabilities
for Verbal Measures .

Means and Standard Deviations of Verbal
Dependent Indicators

Internal Consistency of QSC Battery .

Means, Standard Deviations, and Ranges of
Personal Geometry Profile Constructs

Intercorrelations Among Dependent Indicators

Intercorrelations Among Dependent Indicators

Intercorrelations Among Dependent Indicators

Intercorrelations Among Dependent Indicators

Intercorrelations Among Dependent Indicators
Means, Standard Deviations, and Analysis of
Variance for Manipulation Checks

Correlations of Sex, Self-Esteem, and
Machiavellianism with Dependent Indicators

Correlations of Sex, Self-Esteem, and
Machiavellianism with Dependent Indicators

Multiple Regression of Nonexperimental
Independent Variables with Indicators of
Individuation .

Page

37

39

40

41

43

44

45

45
47
48
50
52

53

56

6O

62

63

Table
Table

Table
Table
Table
Table
Table
Table

Table

APPENDIX C:

APPENDIX F:

13A:

13B:

14:

15:

16:

17:

18:

19:

20:

Correlations of Experimental Independent
Variables with Dependent Indicators .

Correlations of Experimental Independent
Variables with Dependent Indicators .

Multiple Regression of Experimental
Independent Variables with Indicators of
Individuation .

Canonical Correlations of Linear and Quadratic
Effects of EXperimental Conditions with
Expression Indicators of Individuation

Canonical Correlations of Linear and Quadratic
Effects of Experimental Conditions with
Impression Indicators of Individuation

Canonical Correlation of Nonexperimental
Independent Variables with EXpression
Indicators of Individuation .

Canonical Correlation of Nonexperimental
Independent Variables with Impression
Indicators of Individuation .

Canonical Correlations of EXperimental and

NoneXperimental Independent Variables with
Expression Indicators of Individuation

Canonical Correlations of Experimental and
Nonexperimental Independent Variables with
Impression Indicators of Individuation

VALIDATION (PRETEST) RESULTS

CODING SCHEDULE .

vi

. 65

. 66

. 68

. 7O

. 71

. 73

. 74

. 75

. 76

.C-1

.F-l

FIGURE 1:

LIST OF FIGURES

A COMPARISON OF QSC LEARNING PROGRAMS .

vii

A-4

CHAPTER ONE

THEORY

1.1 Introduction

 

We believe that much of human interaction can be fruitfully explained
by viewing social behavior as a process of control based on consensually
shared rules. In this thesis, we will begin by examining the function
of rules and the role of communication in the control process. We will
then look at those situations where control is concerned with managing
the distribution of rewards and punishments forthcoming from the
environment. We will focus on a specific control mechanism, individua-
tion, examining its importance, the rules underlying its use and Oper-
ation, and its communication aspects. We will then present the results
of a laboratory experiment that attempts to shed more light on the
nature of individuation. Finally, we will discuss the implications of

this research for further investigation of the process of control.

1.2 Theoretical Overview: Interaction as Control

 

1.2.1 The function of rules. We believe that much of social

 

behavior, intentional or nonintentional, makes sense when viewed as a
process of control. Pe0ple live in social situations that offer a
variety of outcomes. When we talk about control of an interaction, we
are talking about the constraints on the participants in the interaction
that lead to the likely occurrence of one particular outcome or set of
outcomes rather than others. Such control is possible because of, and

is exercised by the use of, consensually shared rules. Let us elaborate.

2

_We agree with Harré that ”there are no natural forces in men that
lead automatically to the formulation of a society when two or three
hang around together" (1974:143). Instead, social behavior comes about
because pe0p1e fashion and follow rules. A rule is a statement that
specifies behavior apprOpriate to situations. Each rule is composed of
two elements. The first element points out the context or circumstance
in which the rule applies. The second element indicates what action
may, should, or must be performed under the rule (Gottlieb, 1968:39).
An example of a social rule would be: "In order to be favored by
important pe0p1e, flatter them.” It specifies a context (when one wants
to be favored by important peOple) and an appropriate action (flattery).

The power of the rule is based on the social contingencies that
relate the action and the situation. For example, the rule about flat—
tering important pe0p1e is based on the social contingency: important
pe0ple respond favorably to flattery. This is not to suggest that there
is some inherent trait of important pe0ple that makes them favorably
predisposed toward flattery. This is not the important question. What
is important is the fact that the contingencies of the relationship
between flattery and important pe0p1e holds because pe0p1e "agree" to
them and manifest their agreement in their actions. If important
pe0ple did not "agree" to being favorably predisposed by flattery, then
the social contingencies relating flattery to outcomes would no longer
be functional and the rule eventually would cease to have potential

1
for control.

It is important to understand that when we ascribe agreement to a
rule, we base it on the manifestation of agreeing through action rather

than on the manifestation of agreeing through awareness. As long as

3
pe0ple "agree" to the rule in action, their actions maintain the func-
tioning of the social contingencies. As an extreme example, a person
may say "please" or "thank you" and fail to be aware of the rule that
specifies these as appr0priate behaviors in order to receive salt at
the dinner table. Yet these actions and those of the other interactants
will maintain these social contingencies. The continued functioning
of the social contingencies, not the individual's awareness of them,
gives rules their power.

The question of awareness has import, not for the power of the
rule, but for the exercise of control. A person can manifest agreement
with a rule through action at three different levels of awareness.
First, one can be unaware of the rule and the social contingency.
Secondly, one can be aware of the rule but not the social contingency.
Thirdly, one can be aware of both the rule and the social contingency.2
For example, suppose we consider the rule "in order to get people to
like you, be enthusiastic." This may be based on the social contin-
gency: (pe0p1e's affections depend on displays of enthusiasm. A person

may be enthusiastic, unaware of the rule or contingency, and still

receive the affections of others. In such cases, a non-intentional,

 

functional exercise of control takes place. In the second case, a
person may be fully aware of the rule but not the contingency. His
enthusiasm is displayed at apprOpriate times (when he desires the affec—

tions of others). In this case, intentional, reactive exercise of con-

 

trol takes place. In the final case, a person may be fully aware of
both the rule and the contingency. In such a position, the individual
can uncover new rules specifying new occasions when enthusiasm will be

appr0priate. For example, he may uncover the rule, "in order to get

4

pe0ple to do you favors, be enthusiastic." In such cases, an intentional,
3
proactive exercise of control takes place.

 

Thus, control refers to triggering functional social contingencies
that constrain outcomes in social situations. Sometimes this control
is exercised intentionally and we talk about goal-directed behaviors.
Sometimes the control is exercised non-intentionally and we talk about
functional behaviors. In either case, the exercise of social control
is accomplished through the use of consensually shared rules that,
based on social contingencies, specify behaviors appropriate to
situations. Now the question of interest is, what role does communica-

tion play in this process of control?

1.2.2 Role of communication. Goffman suggests that a person uses

 

communication to control others "largely by influencing the definition
of the situation which the others come to formulate" (1959z4).
Influencing the situational definition is effective because it con—
strains the number of behaviors that will be considered apprOpriate.
The behaviors one engages in, the second element of a rule, are contin—
gent on the situational definition, the first element of a rule. By
causing a situation to be defined one way rather than another, the
individual limits the number of rules, and therefore the number of
responses, that can be made.

For Goffman, the situation is a dramaturgical gestalt comprised of
setting and actors. The individual can influence the definition of this
gestalt, and thereby exercise control, in at least three different ways:
(1) He can influence the definition others have of the setting. "This

is a quiet place." (2) He can influence the definition others have of

5

his or their roles. "I am the teacher here. You are the students."
(3) He can influence the definition others have of his or their
characteristics. "1 am enthusiastic. You are materialistic."
Control arises from the rules that specify the apprOpriate behaviors
once one has accepted the definition of the situation. If this is a
quiet place, you will whisper. If he is the teacher, you will behave.
If she is enthusiastic, you will like her.4

Because the exercise of control depends on the definition of the

situation that others accept, the ways the individual attempts to

 

influence the situational definition are partially constrained by how
rigidly defined each of these domains already is. The more rigidly
defined any given domain, the more communicative effort it takes to

redefine it.5

If I wish to st0p you from sailing around the world,

I cannot argue that the world is flat and you will surely fall off.

The setting, in this case the shape of the world, is too rigidly
defined for you to accept my definition of the situation. Such an
attempt at control will be unsuccesstl. Similarly, if I try to get
you to accept a definition of our roles or our qualities when you
already have strongly contrary ones, my control attempts will still be
unsuccessful.6 The result of this state of affairs is that, in attempt-
ing to exercise control, the individual will likely seek to impose a

definition of the situation on that domain that is least well-defined.

Because in many of the interactions in which we enter the setting and

 

the roles are more or less unambiguously defined, potential control

 

often rests with the individual's ability to get others to accept a

 

 

definition of characteristics. Almost regardless of the setting and

the roles we may be in, I can attempt to control the outcome of the

6
interaction by structuring the communications I give to manage the
impression I give off. Let us look at impression management then to

understand how it works to facilitate control.

1.3 Impression Management

 

1.3.1 Conceptual framework. Our discussion of rules and social

 

control thus far provides us with a conceptual framework for analyzing
communication tactics of control. Let us state briefly some implica-
tions of such an analysis. Tactics of impression management are seen
as rule-governed and, as such, they function as mechanisms of control.
The rules governing these tactics are seen as being based on working
social contingencies which stipulate the apprOpriateness of the tac-
tic for a particular situation. In order to develop a comprehensive
theory of impression management and social control, we need to be able
to answer a number of related questions. First, what is the underlying
social contingency upon which the rule is based? Secondly, what tactic
is prescribed as appropriate in that situation? That is, what is the
procedural rule (Cushman G Whiting, 1972)? Thirdly, what behaviors
comprise the tactic? That is, what are the Operational rules? And
lastly, what is the nature of the control afforded by the tactic?

This gives us a framework that lets us compare tactics and understand
the interworkings between them.

A brief example at this point is in order. We would use this frame—
work to view embarrassment and humor as mechanisms for social control.
Underlying social contingencies: (a) embarrassment indicates that in-
apprOpriate behavior has occurred in the interaction and that the

smooth flow of interaction is to be at least temporarily interrupted;

7

(b) humor serves to reinitiate smooth interaction (Fink 8 Walker, 1975).
Procedural rules: (a) when an inapprOpriate act has occurred and the
interaction must be interrupted, be embarrassed; (b) when interaction
has been interrupted by an inappropriate act, in order to resume, be
humorous. Operational rules: (a) blush, stammer, avert eyes, turn
away; (b) point out something incongruous, tell a joke, chuckle. Nature
of control: through the use of these two tactics, embarrassment and
humor, the individual can control the flow or continuity of interaction.7

A We propose to use this conceptual framework to analyze tactics of
impression management. Of Special interest to us are those tactics of
impression management apprOpriate in situations of potential reward or
punishment forthcoming from the environment. Before we survey those
tactics, however, let us add a word of caution that may have more
relevance later. As we see it, tactics of impression management permit
individuals to control outcomes. One subset of these outcomes is
attaining rewards and avoiding punishments. But this is not the only
set. (Another set of outcomes is the flow of interaction, for example.)
When a person eschews rewards, it seems to make more sense to view this
as control of some other set of outcomes rather than to claim reductively
that everything a person does is rewarding. When we discuss tactics of
impression management used to control outcomes of potential reward or
punishment, we do not equate that set of outcomes with all the outcomes

that can be controlled.

1.3.2 Ingratiation and selfipresentation. Research on tactics of

 

impression management in situations of potential reward or punishment

has tended to focus on two aspects of impression control -- ingratiation

8
andstyles of self-presentation. In ingratiation tactics, the indivi-
dual's communications are oriented toward the other. 'Ingratiation may
be in the form of flattery, in which communication enhances the favor-
able qualities of the other. Or, ingratiation may be in the form of
opinion conformity, in which communication indicates agreement with
the Opinions of the other (Jones, 1964). The social contingency
underlying the use of this tactic is that others look favorably on
those who think highly of them. Thus the rule: in order to receive
favors from others, ingratiate.

Styles of self-presentation, on the other hand, focus on communica-
tion by the individual about himself.8 As Stires and Jones (1969) point
out, an individual can attempt to control the situation by accentuating
positive aspects of the self, or by being modest and underplaying those
aSpects. Different styles of self-presentation reflect different under-
lying social contingencies. Positive self-presentations work because
others look favorably on people with good qualities. Modest self-pre-
sentations work because others look favorably on peOple who are not
threatening. It should be obvious that there are as many self-presenta—
tional styles for attaining rewards as there are social contingencies
relating perceived personal qualities and favorable predispositions.

In the research reported in this thesis, we present a metatreatment
of this problem. By looking at individuation, the extent to which an
individual appears similar or different to others, we are concerned with
overall effect of different self-presentational tactics in situations of
potential reward or punishment. We intend to present a conceptual defi-
nition of individuation compatible with our notions of rule-governed con-
trol, discuss its scientific and social significance, and examine its

relevance for communication.

9

1.3.3 Individuation. We define individuation as the extent to

 

which an individual as a social object in a perceptual field is distin-
guishable from other social objects in that perceptual field. This
definition is markedly different from definitions used by other
researchers. For example, Ziller (1964) defines individuation as a
person's subjective feeling of the extent to which the self is differ-
entiated from other social objects. Zimbardo (1969) defines individua-
tion by contrasting it with deindividuation, the mindless anonymity of
crowds, bacchanalia, and mob violence. For him, individuation is a
state of feeling responsible for one's actions. Maslach, building on
Ziller and Zimbardo, defines individuation as the "state in which the
person feels differentiated, to a greater or lesser degree, from other
pe0p1e and objects” (1974:413). The difference between our definition
of individuation and these definitions is similar to the distinction
James (1892) makes between the "I" and the "me" conceptions of self.
Ziller, Zimbardo, and Maslach define individuation from the perspective
of the I, the experiencer. To them, individuation is a state of
feeling. We, however, are defining individuation from the perspective
of the me, the social object. This does not mean that individuation
can only be seen from outside the actor, that I can have no subjective
experience of my individuation. Rather, as we have defined it, the
subjective eXperience of individuation requires that we look at ourselves
the way others do and see to what extent we, as social objects, differ
from other social objects. (In fact, an interesting issue that should
be dealt with is the extent to which my estimate of my individuation

corresponds with your estimate of my individuation.)

10

The definition of individuation as the extent of an individual's
distinguishability from Others carries with it two implications. The
first is that distinguishability from other social objects depends on
the perceptual strength of the focal object relative to other objects
in the perceptual field. Perceptual strength is a function of how
different the focal object is from other objects along all perceptual
attributes being used. For example, in a box of 100 blue marbles, a
red marble has certain perceptual strength because of its color. But
in a box of 100 blue blocks, the red marble has perceptual strength
because of its color and shape. In dealing with pe0p1e, distinguish-
ability is a multi—attribute concern. The second implication of the
definition is that distinguishability anticipates a social consequence
for individuation as a control mechanism. That is, distinguishability
implies attracting greater or lesser amounts of attention from some-
body else. i

This definition of individuation has considerable significance
for the social scientist. Research concerned with the extent to which
an individual is the same or different from others can be fruitfully
reinterpreted by assuming that the individual's actions are exercises
of social control. Research on conformity, for example, assumes that
group pressures make peOpIe conform to group-held Opinions and adopt
them as their Own. The classic Asch study (1951) is exemplary of the
power of groups to get individuals to conform to even obviously mistaken
views. The logic of our analysis suggests another interpretation,
however. In the Asch study, the phenomenon manifested was not that
people succumb to the power of groups, but that peOpIe in strange situa-
tions behave in a functionally useful fashion. They camoflage themselves

and appear to be no different than others.

11

Similarly, we can use this perspective for a different interpreta-
tion of research on situations when people appear different from others.
In Schachter's (1951) research on deviance, for example, a confederate
whose opinions clashed with that of the group was singled out by group
'members for onerous tasks. Schachter concluded that by being different
in group situations one risks possible punishment. Although we do not
disagree entirely with this explanation, we offer a more general one.

We suggest that the confederate's differentness attracted attention.

The manner of differentness, however, earned the punishment. Similarly,
we would guess that the person in the group singled out for the presti-
gious tasks was also one who attracted attention by being different.

His manner of differentness earned a reward. We think that the implica-
tion of these examples is that a functional approach to individuation
has considerable fruitfulness for social science.

Even more important than these implications for social science
research is the social significance of individuation. In our culture,
we tend to exalt the individual and believe that a healthy personality
requires that we develop a sense of distinctiveness or uniqueness that
sets us apart from others (Maslow, 1962). Advertising often tries to
exploit this. Products are offered that promise to make us different,
distinguishable from others. Cachet perfume, for example, is concocted
to "react with each woman's body chemistry" and thus provide the wearer
with "a scent different from every other girl." The functional attrac-
tiveness of such a ploy should be obvious.

This striving to appear different may eventually serve to strengthen
our own sense of identities. Self-perception theory (Bem, 1969) suggests

that as others respond to the differentness of our social selves, we

12
tend to internalize these responses and develop a sense of uniqueness.
However, there are signs of counterforces in our society that seem
ominous to some.

Milgram (1970) comments on the anonymity of urban life and suggests
that the appearance of sameness gives rise to a social detachment among
city dwellers. In a study he reports, it was found that urban people
were less willing to help a stranger than non-urban people. We would
suggest that the act of offering help draws attention to oneself. (In
fact, many studies of helping behaviors require the individual to step
out of the crowd to offer help.) To the extent that cities are seen
as threatening places, we would expect that the functional response
would be to avoid attention, to not offer help, and thereby avoid the
hostility of the environment. Should the presentation of an undifferen-
tiated self continue in urban areas, perhaps the appearance of indistin-
guishability will become internalized, and Zimbardo's fears of the
violence of mindless crowds will come true.

Regardless of the extent of individuation in our society in the
future, this discussion should have provided anecdotal evidence to sug-
gest that individuation is in fact a mechanism for control.9 Earlier
we suggested a conceptual framework for analyzing such mechanisms that
required we identify (1) the social contingency underlying the use of
the tactic, (2) the procedural rule, (3) operational rules, and (4)
nature of control. We will devote the remainder of our discussion to

these tasks.

Social contingency: The social contingency underlying the use of

 

individuation seems to be this: distributors of environmental pay-offs

13
(whether these pay-offs are rewards or punishments) respond to those
social objects that attract their attention. Employers interview
those applicants whose resumes catch their eye. Editors publish stories
with catchy leads. Tyrants make martyrs out of dissidents. Sergeants
assign KP duty to those soldiers who step out of line. The recurrent

pattern is: distinguishability earns the pay-off.

Procedural rules: The procedural rules guiding the use of indivi-

 

duation are two. In order to attain rewards, attract attention. In
order to avoid punishment, avoid attention. These two rules suggest a
preposition for empirical investigation: I

The degree of individuation a person achieves is a

positive function of the actor's perception of potential

reward afforded in a situation.

We have argued that individuation is a process of seeing ourselves
as others see us. We have also argued that interaction is a process of
controlling outcomes and that behavior aims to limit outcomes to those
functionally useful to the individual. We believe that we deveIOp this
ability to see ourselves as Others see us through the reinforcement
(both positive and negative) of our attempts of controlling interaction.
We assume, perhaps wrongly, that the degree of individuation an actor
achieves functionally approximates the degree of individuation attempted.
There are, however, two important considerations that will confound the
problem: the amount of information about the other members of the field
available to the person, and the person's skill in manipulating his
communications to achieve desired effects.

Because the degree of individuation achieved depends on the degree

of distinguishability, or differentness, of the actor from others, the

14
amount of information the actor has about those others is crucial. If
the others are known to the actor and in his presence, then he can
observe their behaviors and determine the level of differentness (or
similarity) to be behaviorally achieved for a given level of distinguish-
ability. If, however, the others are not known to the actor and not in
his presence, then he must make assumptions about what their behaviors
may be. This clearly hampers the actor. He may be in a situation where
he knows what degree of individuation is apprOpriate for the pay-off,
but lacks the information about others' behaviors that would permit the
achievement of that degree of individuation. Not knowing how others
are acting, the actor does not know how to act. This clearly will have
effects on the actor's ability to individuate.

A second consideration must be given to the actor's skill in mani-
pulating his communications in order to achieve certain effects. The
actor may know the appropriate degree of individuation for a situation,
and may be in a position to know what the behaviors of the others are,
and still not be able to manage his self-presentation in such a fashion
as to achieve this degree of individuation. Thus, it seems reasonable
that those skilled in the nuances of communication art may be more
facile in achieving the appropriate degree of individuation than those
who lack this skill.

So far we have said that the degree of individuation will depend on_
'the actor's perception of the potential rewards or punishments in the
situation, and on his ability to individuate. There are, however, two
other perceptions of the situation that will first determine whether the
use of individuation is apprOpriate for the situation. First, the actor

makes a self-presentation to some target. The actor's perceptions of

15

the target's control over the distribution of the pay-off is important.
If the actor believes that the target, in fact, does not control the
distribution of the pay-off, the actor will not be motivated to make an
individuating self-presentation. Secondly, the actor's perceptions of
the extent to which the pay—off is to be distributed selectively is also
important. If the actor perceives that everyone will be rewarded (or
punished) equally, then there is no functional reason to individuate.
Therefore, the procedural rule guiding the use of individuation is rele-
vant in those contexts where not only the potential for reward or punish-
ment exists, but where these two antecedent conditions are also being
met. The target must be perceived as being in control of distributing
the pay-off and the pay-off must be perceived as being selectively
distributed.

This discussion of the procedural rule guiding the use of individua-
tion suggests the following hypothesis for empirical analysis:

In situations of potential rewards or punishment, when a

person perceives these pay—cffs to be selectively distributed

by a target of self-presentation, the degree of individuation

attempted will be a positive function of the person's percep-

tion of the amount of reward.

The relationship specified by this hypothesis is irreversible,
stochastic, sequential, contingent, and substitutable (Zetterberg, 1965).

We must now look at the Operational styles that will guide a person's

use of individuation.

Operational rules and the nature of control: Specifying the rules

 

that the actor will use to accomplish the individuation tactic is a
difficult task. Goffman (1959) suggests that in every interaction, the

individual "takes a line" or presents a version of the self that he

l6
hOpes will be accepted by others. This process of "negotiating an iden-

tity" involves both the verbal and nonverbal eXpressions used by the

 

individual to establish his identity, and the impressions of the indivi—
dual's identity that others come to hold. Both the expressions used by
the individual and the impressions that others come to hold are important
factors in establishing an individual's level of distinguishability.

For example, an individual will appear similar or different from others
depending on the way he uses language and on the things he says (eXpres-
sions). Similarly, an individual will appear similar or different from
others depending on the reactions he evokes (impressions).

Clearly, however, the impressions one creates depend on the eXpres-
sions one uses. "Impression management" is based on our ability to
manipulate our expressions for certain effects. We expect then that the
Operational rules constituting the tactic, individuation, are concerned
with both eXpressions, the use of language and the kinds of things said,
and impressions, the result that arises from the expressions used. It
is in this manner that we eXpect the individual to central the distribu-

tion of rewards and punishments from the environment.

1.4 Role of Other Factors in the Individuation Process

 

We also intend to consider three other factors that may be relevant
to the individuation process: sex, Machiavelliansim, and self-esteem.
Different sex roles cause us to approach control of situations in differ-
ent ways. We expect that the rules for appropriate behavior in a given
situation differ for each sex. Moreover, the operational rules that
constitute those behaviors also differ for each sex. We expect, there—

fore, that sex differences may be important in the individuation process.ll

17

Also important may be Machiavellianism, or the actor's skill in
manipulation. Earlier we mentioned that the actor's ability to achieve
apprOpriate degrees of individuation would depend on his ability to
manipulate the effects of his communications. Numerous studies by
Christie and Geis (1969) point to the skill of high Machiavellians to
effect control of situations through a number of tactics. We eXpect
that Machiavellianism may be important in the actor's ability to
individuate, and therefore, obviously, in the individuation process.12

A final factor that may be relevant to the individuation process
is the actor's self-esteem. If our personalities reflect our Own
sense of uniqueness, then a person with high self-esteem may generally
tend to behave in a more unique way than someone with low self-esteem.
The uniqueness of behavior will be important in attempts at individuation.

We expect, therefore, that each of these variables, perhaps singly
or perhaps in combination, will interact with perceptions of the reward
structure in our hypothesis. Hypotheses of interaction effects, however,

will not be tested in this thesis.

1.5 Summary

We have tried to approach social behavior as a process of control
wherein interaction is governed by rules. We have argued that indivi-
duals initiate control by trying to influence the definition of the
situation, which is the first element of a rule. We further pointed to
at least three ways the individual could try to influence this defini-
tion: (I) by defining the setting; (2) by defining the roles; and (3)
by defining characteristics. The third way was seen to be the process

of impression management and we have argued that, because settings and

18
roles are often rigidly defined, social control will often be initiated
through impression management.

We then discussed tactics of impression management used in situations
'of potential reward or punishment and identified individuation as a
"metatactic" of impression management. A conceptual definition of indi-
viduation compatible with a theory of social control was presented, and
the significance of individuation, both theoretically and socially, was
discussed.

We examined individuation in the framework of the theory of social
control. We argued that the fanctional reason that one individuates is
to maximize the potential for rewards and minimize the potential for
punishments forthcoming from the environment. We said that individuation
was contingent, however, on certain perceptions about the selective
distribution of the pay-off and the control of the distribution being
met. The ability to individuate depends on information about other mem-
bers of the field and skills in managing the effects of communication
that the actor has. Knowledge of the rules and contingencies underlying
the use of individuation are also seen as important. We discussed other
factors that might significantly affect the individuation process.

Finally, we suggested a hypothesis for empirical test.

CHAPTER TWO

METHOD

2.1 Overview of EXperimental Design

 

An exPeriment was designed to test the relationship between indi-
viduation and perceptions of potential reward or punishment. Students
.in two social psychology classes at Western Michigan University
participated. (For more complete description of the subjects, see
Section 3.2.) A cover story was developed to keep the students unaware
of the actual purpose of the experiment. In the cover story, the stu-
dents were told that the Department of Communication at Michigan State
University was conducting research into the possibility of pe0ple commu-
nicating without using normal audio-visual channels. Such communication
was called Quasi-Sensory Communication, or QSC.13 The students were
told that the MSU communication researchers had demonstrated that extra-
sensory communication, though crude, was possible and that means by
which pe0p1e could be taught to communication via QSC were being developed
and tested.

The students were then asked to take part in research designed to
test three different QSC learning programs: a Positive Reinforcement
Learning Program, in which the QSC subject could expect to be positively
reinforced while learning QSC; a Standard Environment Learning Program,
in which the QSC subject would be neither positively nor negatively
reinforced; and a Negative Environment Learning Program, in which the QSC
subject could expect to be negatively reinforced while learning QSC. The

students were randomly assigned to one of three conditions. Approximately

19

20

one third were told that research into the Positive Learning Process
was complete and that subjects would probably be assigned into the
Standard Learning Program, although some would be chosen for the Negative
Learning Program. Another third were told that research into the Nega-
tive Learning Program was complete and that subjects would probably be
assigned into the Standard Learning Program, although some would be
chosen for the Positive Learning Program. The final third, serving as
the control grOUp, were told that research into both the Positive and
Negative Learning Programs was complete and that they would be assigned
to the Standard Learning PrOgram. The possibility of being placed in
the Positive and Negative Learning Programs was thus supposed to repre-
sent to the students a situation of potential reward or punishment.

Two eXperimental instruments were used. The first instrument,
"Form B," consisted of the cover story, manipulation checks, and other
items to measure student perceptions of the QSC study. In the second
instrument, "Form F," students were asked to write introductory eSsays
describing themselves, and them to fill out a QSC Battery and a Personal
Geometry Selection Profile. (Forms E and F are provided in their
entirety in Appendix A.) The introductory essay gave students the Oppor-
tunity to attract attention or not attract attention to themselves by
writing individuating self-presentations. The QSC Battery and the Per-
sonal Geometry Selection Profile, in addition to maintaining the appear-
ance of being part of a QSC experiment, gave students another opportunity
to attract attention or not attract attention to themselves by answering
certain questions with conforming or nonconforming responses. Students
were told that QSC experimenters would read their responses to Form F,
and on the basis of those responses, select students for the various

learning programs.

21

In the administration of the eXperiment, students were directed
first to Form B, where they were told to read the cover story and
answer questions testing the effectiveness of the manipulation. When
the students completed half of Form E, they were directed to Form P
which they completed entirely and them returned to Form E to answer
remaining questions about their perceptions of the QSC study. Since
the students were told that the QSC experimenters doing the selection
for learning programs would only see Form F, we believe this procedure
may have allowed the questions in Form E to appear inappropriate for
attempts at control through self-presentational tactics. The average
time to complete both forms was about 55 minutes. When the last student
had completed the items in Form F (i.e., had completed all items to be
used in data analysis for this thesis), an immediate debriefing was
provided.

It should be emphasized that this experiment was designed to meet
explicitly the requirements of the antecedent conditions for individuat-
ing self-presentations. In order to ensure that the students would per-
ceive that the distribution of reward or punishment was to be selective,
those in the non-control conditions were told that mg§£_subjects would
be assigned to the Standard Learning Program, although s9me_would be
selected for the Negative or (as the case may have been) Positive Learning
Program. Secondly, in order to ensure that the students would perceive
that the target of their self-presentation would in fact control the
distribution of the pay—off, they were told that the eXperimenter who
read their introductory essay would be the one who would assign them to

the various learning programs.

22

In addition, a precaution was taken to reduce the amount of variance
in the introductory essays the students would write. Two "sample” essays,
purportedly written by other students participating in the QSC experiment,
were included in Form F. The essays were prepared by having students in
an introductory communication class write essays describing themselves.
On the basis of the content of those essays, the two essays were written
containing the kind Of information that students typically included in
introducing themselves. The two essays were them informally tested to
demonstrate that others would believe they could have been written by
an average student of either sex. These two essays were then included in
Form F to give the students a point of reference for writing their own
essays.

In summary, the experimental manipulation contained a cover story
in which students were led to believe that they wOuld participate in a
study on QSC where they might differentially receive rewards, punishments,
or neither rewards or punishments. The students were then given the
opportunity to attract attention or not attract attention to themselves
by writing individuating self-presentations and by giving conforming
or nonconforming answers to questions. The student was given the means
by which he or she could manipulate his or her responses in order to

attempt to control the distribution of potential rewards or punishments.

2.2 Experimental Independent Variable

 

In this study, the experimental independent variable is the perceived
potential reward structure in a situation. The Positive, Standard, and
Negative Learning Programs provide three different levels of potential

reward. The essential "learning process" in all three programs was

23
described as being the same. The differences in the programs were in
the settings and in the manner in which right and wrong answers during
learning were supposed to be treated. In the Positive Learning PrOgram,
the situation was described as being pleasant with blue carpeting, blue
curtains, and deeply padded furniture. Positive reinforcement techni-
ques were to be used to encourage correct responses. For each correct
response, the subject was to be rewarded with the sound of a pleasant
tone and was given a small monetary payment of $.25. In the Negative
Learning Program, the setting was described as being severe, with white
walls, bright lights, and a solitary bench for the subject to sit on.
Negative reinforcement techniques were to be used to discourage incorrect
responses. For each incorrect response, the subject was to be punished
with the sound of a harsh tone and was given a small electric shock
described as being like that when you cross a wool carpet and touch a
door handle. In the Standard Learning Program, learning was said to
take place in an ordinary room and no reinforcement techniques were to
be used.

Student perception of the reward structure was measured by three
sets of questions. Students were first asked to indicate how much they
- would like to be in each of the three learning programs. They then were
asked to rate how rewarding/threatening the settings and reinforcement
techniques of each learning program were. Finally, they were asked to
rate how rewarding/threatening each learning program itself was. These
three sets of questions allowed us to check on the success of the eXper-
imental manipulation.

Both linear and quadratic effects of OXperimental conditions are

used as independent variables in the data analysis. Because we are

24
concerned with a predicted linear relationship between perceived poten—
tial reward and individuation, we will use the linear effect of experi-
mental conditions as the experimental independent variable in testing
the experimental hypothesis. For eXploratory purposes, however, we
will consider the quadratic effect of experimental conditions. This
compares the effects of the control condition with the effects of the
negative and positive conditions taken together, and thus alerts us to
any curvilinear relationships that may exist. These two measures,
linear and quadratic effects of eXperimental condition, are the experimen-

tal independent variables in this study.

2.3 Nonexperimental Independent Variables

 

Several weeks prior to the QSC experiment, and in a different context,
the students were asked to provide background information on their sex,
their year of birth, year in school, birth order, and place of residence.
In addition, the students were asked.to fill out four personality
measures. These included the Christie et al. Mach IV test for Machiae
vellianism, the Rotter test for Internal-External Locus of Control, the
Rosenberg Self-Esteem test, and the Crowne and Marlowe Social Desirabi-
1ity Scale (all in Robinson and Shaver, 1969). The Mach IV test and the
self—esteem test would allow us to investigate the influence of these
factors on the individuation process. The other information was gathered
for purposes of exPloratory analysis. (Demographic questions and

personality tests are provided in Appendix B.)

25

2.4 Dependent Variables

 

The primary dependent variable in this study was the degree of
individuation, or the extent to which the individual's self-presentation
made him distinguishable from others. The eXperiment was designed so
that the students would have the opportunity to individuate while filling
out Form F. Ways had to be found that would allow us to assess the
degree of individuation in each essay (i.e., verbal measures) and on the

QSC Battery and the Geometry Profile (nonverbal measures).

2.4.1 Verbal measures: expressions. In her explanatory study on

 

individuation, Maslach (1974) found that the length of self-description
and the frequency of the use of the word "I" were related to individuation.
A reasonable explanation for this finding is that peOple who are trying
to distinguish themselves from others will do so by providing more infor-
mation about themselves. Those who do not want to distinguish them-
selves from others will avoid attention by not giving much information
about themselves. On the basis of this finding, we decided to see if
differential verbal behavior was related to individuation. Two main
categories of eXpressions were examined. The first category was con-
cerned with syntax. The second category was concerned with types of
information expressed.

Syntax: An exhaustive category scheme was designed to allow us to
examine the use of various parts of speech in self—presentational
essays. The categories were pretested in an introductory communication
class (p_= 40; 16 males, 24 females; it took approximately 20 minutes
to fill out the form). Approximately half the students were given

Form F and told to complete the form by trying to be different from

26

others and stand out. The other half were told to complete the form
by trying to be similar to others and to not stand out. We assumed
that the language use of people consciously trying to individuate in
the pretest would be the same as that for peeple in the experiment who
would not be told exPlicitly to individuate. The categories that mani-
fested different use for those peeple trying to attract attention from
those trying to avoid attention would be those categories sensitive to
different degrees of individuation.

In analyzing the pretest data, we discovered that virtually all of
the parts of speech were dependent on the length of the essay. That is,
the number of adverbs, adjectives, etc., in any given essay depended on
the total number of words in that essay. For this reason, we decided
to examine the rates of use of different parts of Speech. Significant
differences (p 5;.10, two—tailed14) between the two pretest groups were
found in the rates of use of first person singular pronouns, active
verbs, passive verbs, present tense verbs, and a category consisting
of nouns, gerunds, and infinitives. People trying to be similar to
others had a higher rate of use for all categories except passive verbs.
The reason may have been that people trying to be similar wrote more
terse sentences, free of descriptive words. Their sentences consisted
principally of a subject (often "1"), an active verb, and an object.

In addition, significant differences in the total number of words and
the number of words per sentence (p5 .0001 and p5 .007, respectively)
were found. (A complete breakdown of these and other pretest results

is provided in Appendix C.)

27

Type of information expressed: Subjective estimates of possible

 

correlates to individuation in the expressions used were also investi—
I
gated. Significant differences between the two pretest groups were
found in the rate of sociological (role dependent) information in the
essay (ppS .01. See Miller G Steinberg, 1975, for a discussion of
this type of information). A significant difference (pp£E.lO) was
also found in the rates of statements that referred to categories the
person was in. (No significant difference was found in the rates of
statements referring to attributes of the person. See Kuhn G McPart-
land, 1954.) In addition, because the two "sample" essays had been
constructed based on the content of typical student information, we
also compared the pretest essays to see how many included the same con—
tent dimensions as the "sample" essays and how many included additional
content dimensions. The rate of use of non-unique content dimensions
in the essay (like age, hometown, major in college) and the number of
unique content dimensions also manifested significant differences
(p:£;.0001 and p_£$.02, respectively). (For an explanation of the

coding procedure, see Section 2.5.)

 

2.4.2 Verbal measures: impressions. In addition to pretesting
measures of expressions used, we directed coders to rate the impressions
created. (For an explanation of the coding procedure, see Section 2.5.)
Estimates of a number of self-presentational tactics that would seem to
be related to distinguishing oneself from others were taken. Those
tactics that showed significant differences between pretest groups were:
differentness, positiveness, uniqueness, negativeness, enthusiasm, in-

tensity, and modesty. These seven self-presentational tactics served as

28
impression indicators of individuation. In addition, estimates of
several other self-presentational tactics that may function as other
forms of control mechanisms were made. These included flattery, stress,
embarrassedness, humor, calculatedness, defensiveness, and unsureness.
We heped that this would permit us to begin studying a problem we
discussed earlier, that of specifying which self-presentational tactic
would be chosen out of the set of potentially appropriate tactics for a

given situation.

2.4.3 Nonverbal measures. Finally, two other measures of indivi-

 

duation were obtained from the QSC Battery and the Personal Geometry
Selection Profile. In the QSC Battery, a geometric symbol headed a
list of 15 words selected to be of equal meaninglessness (Archer, 1960).
Students were told to associate the symbol with a word on the list and
then to circle the word. In addition, students were told that the words
were arranged according to their perceived similarity by others; the
more similar, the closer together on the list they appeared. On one
third of the lists, the students were given the supposed reSponses of
both the two "sample essay" students. On one third of the lists, the
students were given the supposed response of just one of the "sample essay”
students. On the remaining third of the lists, the students were given
no reSponses of the two "sample essay" students.

The students could appear to be similar to the responses of others
by circling a word close to the supposed reSponses of the "sample essay"
students. Discrepancy scores were calculated by finding the number of
words between the student's response and the supposed response of the
"sample essay" student (or the mean response of the two "sample essay"

students).

29

A pretest of the QSC Battery was done in several introductory
communication classesls. One hundred nine students participated.
Approximately half the students were given the QSC Battery and told to
try to be different from others; the other half were told to try to be
similar to others. Discrepancy scores for those items with two supposed
answers generally showed significant differences at the p S:.005 level.
Discrepancy scores for those items with one supposed response generally
showed significant differences at the p_££.01 level. Discrepancy scores
for those items with no supposed responses were generally not significant.
(A breakdown of the pftests for each item in the QSC Battery is provided
in Appendix C.)

The Personal Geometry Selection Profile was based on a measure
deveIOped by Ziller (1973) in his work on individuation. It consists of
a box with nine evenly spaced circles arranged in a line. The student
was instructed to put the initials of one of the "sample essay" students
in one circle and then his or her initials in another circle. The
student repeated the procedure for the other "sample essay" student and
again for the "average student." The difference between the student and
the three others is used as a measure of the degree of individuation.

These two nonverbal tests, along with the measures of expressions
and impressions, comprised the indicators of the dependent variable of

this study.

2.5 Coder Trainipg

 

Because of the difficulty of the task of making reliable subjective
estimates on such a large number of variables, rigorous coder training

was necessary. Eight undergraduate students, five males and three females,

30
served as coders. All were communication majors.v Four were freshmen,
and two each were sephomores and juniors. They were first given an
overview of the experiment; however, they were kept blind to the exper-
imental hypothesis. Then they were given a thorough explanation of the
process of measurement. The need for extreme care in assuring reliable
measurement was stressed.

Actual training began on the measures of exPressions. Coders re-
viewed English grammar and discussed different ways of identifying the
various parts of speech. All coders practiced coding an essay, and
then discussed their responses. This procedure was continued until it
seemed clear that the coders had a good grasp of the measures of
expressions.

All of the measures of impressions were coded using ratio scales,
and it was thus necessary to explain the process of ratio scaling to
the coders. (A sample of the coding form is provided in Appendix D.)
One of the two "sample” essays was used as the standard. As the standard,
it was set to 10 on all impression measures. Coders were then taught
to rate each essay in comparison to the standard essay. They were told
that, for whatever self-presentational tactic they were coding, if the
student's essay was twice as much (e.g., twice as flattering) as the
standard, to score it as 20. Similarly, they were told that, if the
student's essay was half as much as the standard along the dimension
being coded, to score it at 5. Any non-negative integer could be used
in rating impression indicators.

After the coders demonstrated facility using the ratio scale, they

 

were divided into four pairs and each pair was assigned four or five of

the impression variables as their responsibility. (One of the four pairs

31
was not assigned any impression variables. Instead they rated psycho-
logical and sociological information on scales from 0 to 9 and counted
the number of category and attribute statements in each essay.) The
coders first had to arrive at a mutual definition of the impression
variables they were to code. (Definitions are included in Appendix E.)
They then began to practice the ratio scaling techniques on a set of
33 essays (17 male, 16 female) written by students in an introductory
sociology class at Lansing Community College (LCC). Administration of
the pretest forms took approximately 20 minutes. The coders were told
to compare responses after every two essays, to discuss their discrepan—
cies, and to agree to a common response. Each coding pair spent approx-
imately 15 hours working on the LCC essays.

After the initial training period on the LCC essays, the coders
began work on 40 pretest essays (p. 25) written by students in an
introductory communication class. The coders were instructed to discuss
discrepancies if they were unusual (i.e., more than two points.off in
the range from 0 to 15, more than three points off in the range from
16 to 25, more than four points off in the range from 26 to 35, etc.)
The coders also divided the 40 essays and, with the coders working inde—
pendently, each coding pair scored the measures of expressions for ten
essays. Each coding pair spent approximately 20 hours working on the
pretest essays.

On the basis of the pretest, these measures of expression syntax
that showed significant differences between groups were selected for
the OXperimental data. In addition, the coders were given feedback on
the reliability of their subjective estimate measures. This feedback

proved useful. Because the coders were unsure of how successful they

32
had been up to this point, the feedback confirmed their areas of
strength and indicated those areas where more work was needed.

After final discussion of the task, coders began coding the eXper-
imental data. Coding pairs began each scoring session by doing their
impression questions on two LCC or pretest essays so that they could
reset the standard and work out any discrepancies. During the actual
coding, the coders of any coding pair worked in separate rooms. After
completing four essays, the coders would compare and discuss responses.
No responses were changed, however. If the coders' discrepancy was
unusual (as discussed before), they were instructed to reset their
standards by doing additional LCC or pretest essays.

The experimental essays were broken into approximate quarters.
Each coding pair scored the eXpression variables for their quarter as
well as their four or five impression variables across all subjects.
Each coding pair Spent approximately 30 hours coding the experimental
data. (A table showing the items that were coded by each coding pair

is provided in Appendix F.)

2.6 Ethical Considerations

 

In the negative condition, students were led to believe that if they
were chosen for and participated in the Negative Learning Program, they
might receive small electric shocks while learning QSC. This was part
of the experimental manipulation and at no time did we intend to carry
out any "QSC studies." Nonetheless, from the students' point of view,
the possibility of receiving electric shocks could definitely be anxiety-
producing. Therefore, immediately after the experimental instrument was

filled out, all students were fully debriefed. They were told that there

 

33
would be no electric shocks given to any of them. Furthermore, they
were given a complete eXplanation of the experimental hypothesis and

the eXperimental design.

2.7 Summary

In this chapter, we reviewed the methods used to test the eXperi-
mental hypothesis relating individuation to perceived reward structure.
We described an eXperimental manipulation that would permit students
to make individuating self-presentations to a target that controlled
the selective distribution of potential rewards and punishments. We
also described the ways the success of this manipulation would be
checked.

We discussed at some length the development of three sets of depen-
dent measures of individuation: expressions and impressions, which were
both measures of verbal behaviors, and nonverbal measures that allowed
students to give conforming and nonconforming responses. Finally,
because many of the dependent measures were scored by coders using
ratio scales, an extensive coder training program was followed to assure

reliable measures. This program was also discussed.

CHAPTER THREE

FINDINGS AND DISCUSSION

3.1 Overview
Any scientific investigation aims ultimately to provide a test of
some theoretic notion. But before such a test can be meaningful, the
investigator must have confidence in the internal and external validity
of his procedures. In this chapter, we will present an analysis of our
data that represents a systematic assessment of the evidence upon which
arguments for accepting or rejecting the eXperimental hypothesis are
based. In this analysis, we attempt to answer feur sets of questions.
1) What confidence do we have in our sample?
Are there problems of external validity? (i.e., is the
sample representative? are there unusual distributions
of any variables within the sample?) Are there problems
of internal validity? (i.e., has the randomization of
subjects across conditions successfully controlled for
other variables that might confound the results?)
2) What confidence do we have in our measures?
Are our measures of independent nonexperimental and
dependent variables reliable and internally consistent?
Are there anomalous relationships among dependent indi-
cators that would make some measures suspect? Are our
measures valid?

3) What confidence do we have in our experimental treatment?

Has our attempt to structure differentially subjects'
perceptions of reward been successful?

4) What evidence do we have for accepting or rejecting the
eXperimental hypothesis?

What is the functional form of relationship and magnitude
of effect of the experimental treatment on the dependent
indicators? Are the dependent indicators affected by
nonexperimental independent variables?

34

35
To answer these questions, we present the results of various
statistical tests and procedures.

1) In order to demonstrate our confidence that the sample does
not jeepardize the internal or external validity of the
study, we present frequency distributions of, and intercor-
relations among, a number of demographic and personality
variables. These statistical analyses are evidence of the
representativeness of the sample. In addition, because
these variables are not manipulated in this study (hence
their designation as noneXperimental independent variables),
we have controlled for them by randomly assigning subjects
to treatment conditions. Therefore, to indicate how suc-
cessful the randomization process has been, we also present
the correlations between the nonexperimental independent
variables and the experimental independent variables (i.e.,
linear and quadratic effects of eXperimental condition).

2) In order to demonstrate our confidence that the measures
used in the study are meaningful, we present internal con-
sistency scores for the multiple-item personality tests
and the QSC Battery. In addition, we present intercoder
correlations as measures of reliability for each of the
expression and impression indicators of individuation.
Intercorrelations among dependent indicators are also pre-
sented to reveal the nature of the dependent indicators
and to establish the construct validity of these measures.

3) In order to demonstrate our confidence that the subjects'
perceptions of reward in the situation has been differen-
tially structured across conditions, we present statistical
tests of the manipulation checks.

4) In order to assess the impact of the various nonexperimental
and experimental independent variables on the dependent in-
dicators, we present the results of correlational, regression,
and canonical analyses.

Because of the large number of measures involved in this study, we
have identified different sets of variables and indicators by different
prefixes. NX identifies noneXperimental independent variables; X iden-
tifies eXperimental independent variables; DE identifies expression
indicators of the dependent variable, individuation; DI identifies impres-
tion indicators of individuation; and DB identifies the QSC Battery and

Personal Geometry Profile measures. Measures without prefixes are those

used to create the various dependent indicators.

36

3.2 Results of Data Analysis Pertaining to Sample

 

Table 1 provides demographic and personality data for the sample.
Higher values for the personality scores indicate greater presence of.
the trait. Higher values on the Internal—External Locus of Control
scale indicate greater perception of external control. The means and
standard deviations of the personality scores reflect the lack of many
extreme values in the sample.

Most of the demographic variables also Show no anomalous distribu-
tions that would cause us to suspect the composition of our sample. Year
of birth (NX4) does reveal a somewhat greater diversity in ages than what
might be eXpected in most undergraduate classes, but we do not see this
as a problem. One aspect of the sample demographics that may pose a
problem for the study is the sex distribution (NXZ). In the sample, the
number of female subjects is twice the number of male subjects. In the
validation study (where the eXpression and impression indicators of indi-
viduation were selected), the female/male ratio was more even. If there
are sex differences in the use of expression or impression indicators
of individuation, then the indicators selected on the basis of the vali-
dation study (where sex differences were not controlled for) may not be
apprOpriate for the eXperiment. If inappropriate indicators were selected,
then it would be less likely that the experimental hypothesis would be
confirmed. More information relevant to this issue can be found in
Table 12, in which correlations of sex with expression and impression
indicators are presented.

Table 2 shows the linear intercorrelations among the demographic
and personality variables that constitute the noneXperimental indepen-

dent variables in the study. Table 3 shows the correlations of these

37

 

mcﬂmmﬂz

. ooo.ooa Ao>o some
ea oco.ooﬂ-ooo.oH mo Apao
ooo.oH-oom.N mo ezoe

89mm we: use xhuceoo some

mo.HH “moHV spasanmeammo

 

HNIH mo.e
HN-H mm.e om.HH
onuom Ho.m mo.em
vmﬁuom oo.eH NH.Nn
ewcmm :oﬁumﬁ>on can:
vampcmum

Hmsuom ﬂosxzv
ﬂNoHV
HOHucou mo mooog
Hecaouxmuﬁmcuech mmxzv
ﬂeesv amoemm-msom mmxzv

AeoHv
amacmssso>aaaua2 ﬂaxzv

 

mosoom xuwam:0mhem

e
e oom.m cage mama owmssa>
w
m

seam;

 

eoeoeﬁmom mo oomﬁm ﬂoxzv

0H
e

00
em

wcwmmﬁz
usage Asco
chop young
shoe umaﬁm

 

smeso cesﬁm ﬂmxzv

mcﬂmmﬂz
nemﬁ opomom
nemﬁ
mama
memﬂ
ommﬁ
Hmmﬁ
mmmﬁ
mmmﬁ
emmﬂ
mmmﬁ

 

Logan we see» mvxzv

.oH wcnmmﬂz

M H0390
mm Hoﬁcem
mm Hoﬁcsw
ea oaoeonmom
N smegmeAd

 

Hoonom :ﬂ Hme> ﬂmxzv

Na mcﬂmmﬁ:
mo OHNEOL
mm mam:
xmm mmxza
ea wcﬁmmﬁ:
oe mmmﬁu coocampmd
oo mmmﬁu maﬁ2902

amasu msxza

was sauce
om goon use masssum
0H cho uceesppmcﬂ Hmu

Icoeﬂaomxe poo mcﬁﬂﬁﬂd

«a xaco meASmmom
xuﬁﬁmcomheo use mcﬂﬁﬁﬂm

IJII

meumanzm

C

 

 

OHQEmm mo coﬁpawgomeo .ﬁ mﬂemh

38
nonexperimental independent variables with the eXperimental independent
variables, linear effect of experimental condition (X1) and quadratic
effect of experimental condition (X2; see Section 2.2). In Table 2, we
are especially interested in the correlations among sex (NX2), Machia-
vellianism (NX7), and self—esteem (NX8), because these are variables
that we suspect may affect the individuation process. The linear cor-
relations show them to be relatively independent of one another, with
the strongest relationship being a nonsignificant r_= —.16 between sex
and self-esteem.

In Table 3, correlations indicate that the random assignment of
subjects to conditions was fairly successful in controlling for the non-
experimental independent variables. Self—esteem (NX8) and social desir-
ability (NXlO) both Show small but significant associations with the
linear effect of experimental condition (X1; 3's = -.20 and -.25, respec—
tively, pfs=:-.05). The relationship between self—esteem and linear
effect of eXperimental condition will pose some slight problem in terms
of multicollinearity when both variables are entered into regression or
canonical analysis with the dependent indicators.

Let us summarize the implications arising from the investigation of
the first three tables. First, our sample appears fairly representative
of a college pepulation.16 Most of the nonexperimental independent varia-
bles show no abnormalities in their distributions. Moreover, they are
relatively independent of the experimental independent variables. Thus,
their effects on the dependent indicators have been effectively controlled
by the successful random assignment of subjects to condition.

Two potential problems are worth noting, however. First, if the

CXpressions used or impressions created in individuating differ by sex,

u...d-n-I-ﬂ nun.) usurJdJ-..a~.|

 

—:...— .u—~_-.-..-..-Aa-.r.p.-.~7N .i--.-:/\ n<-.v‘ -“.v.-..QHJ -.-',vs.~‘ . - .-.-..

39

.N

u onEom .H u oﬁmze

 

vaH n .m ESEHKmZ “mm H .m. EdEﬁCMZU

€833-33 mo. W m

Aveimuuocov Ho. W m .3.

.1

 

 

mo. tram. mo.- we. no. Ho. .ms.- mo. OH.- saussnmsﬂmoe Haauom on2
.ms.- .1wm. .ms.- co. ..m~. so.. me. mo.- Hopscou mo mzooﬁ
anchouxOuHmcpeucH mxz
mo. mo.- mH.- mo.- eo. les.- No.- amoemo-msmm wxz
eo. mo.- Ho.- ins. ms.- mH.- Emacmnssm>aaeoaz sz
mo. mo.- mo. HH.- eo.- mocoenmos mo momsa exz
eo. HH.- mo. ..mm.- Ameso Locum mxz
..em.- mo.- mo. Essen mo Amos exz
ms.. 00. Hoogum as sea» mxz
mo.- exam Nxz
mmmﬁo sz
aggaﬂmgﬂmgmﬁa
meaneﬂum> unbecomoecH Hmucoeﬂpomxocoz wcoe< mcoﬁpmHOHHOOAOch .N macaw

II .Il.|1ll.

Fuiuillh‘lhh

nU ‘.~I..

40

.N u 228. .H - gaze

as u m asses: .3 u m 22:22.:

5833-33 mo. W ..

m.Hm5Uo who: meNﬂm Haeo Hﬁm m“ 00. on cases coﬂpmHOAHoo week .No.- D Nx HxA ”ouozv
8. ......N.- 323238 338 2?.
Ho. vo. Homecoo mo msooﬁ

Hmcsouxo-Hmchech mxz

 

 

mH. Pom.u Eeeumo-mﬁem wxz
2.- mo. 52323369. 32
oo. 0H.- mocmwﬁmeh mo oomﬁm oxz
«0.- . NH.- geese :poﬂm mxz
vo.- mH.- cusﬁn mo sme> vxz
OH. No.- Hoocom Ca Mme» mxz
2. S.- m Sam 8?.
no. ﬁo.- mmmﬁu sz
coﬁpwwcoo :oﬁuﬁvcoo
Hmpcoeﬁhomxo mo poomwo awesopmoo Nx Hmecoeﬂpemxe mo poommo Hmecﬂq Hx

 

 

.moHemwam> pcepcomopcw Hmucoeﬁpemxecoz cam Hmucoeﬁsomxm mcoe< mcoﬂumﬁehhoowoch .m oﬁnmb

 

("no

ZR!
.-v

.U-

.‘i

l)!

'—d

C.‘

41
then the high female/male ratio in the sample may lessen the chance of
failing to reject the experimental null hypothesis. Secondly, there is
a slight problem of multicollinearity among one of the nonexperimental
independent variables, self-esteem, and the linear effect of experimental
condition. Nonetheless, Tables 1-3 lead us to be fairly confident that
the constitution of the sample does not jeopardize the validity of the

findings of this study. Let us now examine the strength of our measures.

3.3 Results of Data Analysis Pertaining to Measures

 

Tables 4 through 9 present evidence upon which we can establish a
level Of confidence in our various independent and dependent measures.
Table 4 Shows the internal consistency of each of the multiple-item
personality measures. We are concerned with the possible effects of
Machiavellianism (NX7) and self-esteem (NX8) on individuation, so we are
especially concerned with the reliability Of these two measures. Cron-
bach's «5: for Machiavellianism is .70 and for self-esteem is .80.

These values demonstrate moderate levels of internal consistency for

both measures.

Table 4. Internal Consistency of Personality Measures

 

 

 

 

p_ Cronbach's °< Number of items
NX7 Machiavellianism (83) .70 20
NX8 Self—esteem (86) .80 10
NX9 Internal-external
locus of control (84) .79 23

NXlO Social desirability (83) .77 3O

 

42

Tables 5 throngh 8 describe three sets of dependent indicators Of
individuation: eXpression indicators, impression indicators, and non-
verbal indicators from both the QSC Battery and Personal Geometry Pro—
file test. Table 5 presents descriptive statistics and intercoder
reliability for the expression and impression measures of individuation
coded by each coding team. The correlations between coders for expression
measures range from .35 to 1.00, with the majority Of correlations above
.87. Among the measures of impressions, intercoder correlations range
between .31 and .85. The .31 correlation for flattery (D13) is signifi-
cant at the p £§.005 level (one-tailed). All other correlations are
significant at the p_§5.001 level (one—tailed).18 The correlations
among the impression measures, demonstrate that complex subjective esti-
mates on ratio scales can be reliable, given sufficient coder training.
Except for the measures of number of category statements (DEIO) and the
amount of flattery (013), the expression and impression measures Of
individuation are fairly reliable, with many of them being highly
reliable.

Table 6 presents the means and standard deviations for the constructed
expression and impression indicators of individuation. (For mere infor-
mation on how these indicators were constructed, see Section 2.4.) Except
for number of words (DEl) and number of unique dimensions (DE9), the
expression indicators are all OXpressed in terms of rates per word.
Reliabilities were not recalculated for these measures because the corre:
lation between coders for number of words is .99+, and reliabilities of
these rate indicators would Show essentially no change from the reliabi-

lities Of the measures not expressed as rates.

43

.mo>ﬂuﬁcﬂmcﬂ paw meansmm .mcsoc mopsﬁocﬁ :mczoc: hepmoﬂecﬁ may .Hopomco mﬁzu a“ moﬁnmp ace :Hm

 

 

++mn. H.H m.oH
++mw. v.H v.0H
++Nm. 02—” m.OH
++mn. m.H o.oH
+Hm. m.o N.oH
++©o. w.H m.oH
++vn. m.m m.HH
++mm. o.H m.m
++mm. v.0 o.a
++om. e.H N.H
++nw. N.H H.v
++~m. m.v m.n
++mo. e.a n.o
++mw. m.m m.m
++mm. o.m H.wH
++wm. H.e o.m
++oo.H H.0m c.5n
++nm. v.m m.m
ﬂxpwﬁﬂnmﬁaom Hewoohouch .o.m coo:
Nam N same mcaeoo

v.H m.oH
n.H o.oH
m.H H.HH
m.H o.oH
v.H H.0H
m.H N.oH
o.m v.HH

1-1

I\

VONNHv—cmmvmv
Nxo'd'oomHVv—ir-ior-i
ODOLOOOOBBHNHN
mooooooooth-Hv—im

.a.m 53:
H same mcaeoo

8m .-. m asses: .wm u m 535223

2833-23 So. w m +-

$023.83 moo. Wm +

xuﬂmcepcﬂ
EmemSLOCO
mmecpoumﬁzoﬁmo
mmecescwas

xhouumﬁm
mmoce>ﬁuﬂmom

mmocpcepemmﬁp mo Opmeﬁumo O>HOOOmnsm

:oﬁpmsnomcﬁ HmOMmoHOHoom
wpcesepmpm kHOMOpmo
mcoﬁmceeﬁv pcooowmﬁp
mcoﬁmcoeﬁp amﬁﬂeﬂm

mnse> emcee ucmmesa
meso> O>ﬁmmmm
meso> O>ﬁpom
amazon
menocoam smﬂswcwm comsom pmnﬂm
memo:

meo20pcom

Osemmoz

we so €52

 

 

.mousmmez Hensm> How moﬁuﬂﬂﬁpmﬂﬁom HoeooomucH wcm moapmwemum o>wemﬂsommo

.m mﬁpmh.

44

Table 6. Means and Standard Deviations of Verbal Dependent Indicators.

 

 

 

 

Dependent Indicatora n Mean S.D.
DEl Number of words 99 77.81 36.00
DE2 Rate of sentences 99 0.08 0.02
DE3 Rate of first person singular pronouns 99 0.12 0.03
0E4 Rate of nouns ~99 0.24 0.05
DES Rate of active verbs 99 0.12 0.04
DE6 Rate of passive verbs 99 0.01 0.01
DE7 Rate of present tense verbs 99 0.10 0.03
0E8 Rate of non-unique dimensions 99 0.06 0.03
DE9 Number of unique dimensions 99 1.19 1.29
DElO Rate of category statements 99 0.02 0.01
DEll Rate of sociological information 99 0.08 0.03
011 Subjective estimate of differentness 98 11.46 2.09
012 positivenss 98 10.34 1.43
013 flattery 98 10.16 0.91
D14 uniqueness 98 10.62 1.50
015 calculatedness 99 10.96 1.53
016 enthusiasm 99 - 10.50 1.50
017 intensity 99 10.23 1.21

 

aDependent indicators are the average of the scores from the
two coding sets. Rate variables are standardized by dividing
by number of words in the self-descriptive essay.

In Tables 7 and 8, data about the nonverbal measures of individua-
tion are summarized. Table 7 shows the internal consistency of the QSC
Battery. The battery was developed to check on the extent to which
students would give conforming or nonconforming responses to a set of
test questions. The low s3 suggests that the QSC Battery is not inter—
nally consistent.

Table 8 presents the means, standard deviations, ranges, and inter?
correlations of the Personal Geometry Profile items. The evidence sug-
gests that we may have a problem with our measure. The Personal Geometry
Profile items were devised to allow subjects to appear similar to or

different from three others -- R.S., J.L., and ”the average student."

45
If the subjects used the items in such a fashion, we would eXpect
consistent scores across items and high correlations between items.
Although these items are significantly intercorrelated, it does not
appear that the subjects used the scale as a unidimensional test of
similarity. We will consider the possible problems of these two non—
verbal measures Of individuation in more detail when we discuss

Tables 90 and 9E.

Table 7. Internal Consistency of QSC Battery.

 

 

 

 

n_ Megn S.D. Cronbach'sgﬁ. Number of Items
Two Check Items 96 19.30 6.04 .16 4
One Check Items 96 16.32 5.10 .12 4
NO Check Items 96 14.89 4.70 .13 4
All Items
Bl QSC Battery Score 96 50.54 10.09 .28 12

 

Table 8. Means, Standard Deviations, and Ranges of Personal Geometry

Profile Constructs.

 

 

 

 

Item p. Mean S.D. Range
Discrepancy between student and R.S. 97 2.83 1.73 1.0 to 8.0
Discrepancy between student and J.L. 97 3.23 2.10 1.0 to 8.0

Discrepancy between student and
average student 96 2.42 1.72 1.0 to 8.0

 

7-4—. —-——-- _. ..___._‘_‘.__-~_‘ .—._~..._._... M<

 

-—.-_

46

Tables 9A through 9E present the intercorrelations among all de-
pendent indicators of individuation. In examining these intercorrela-
tions, we are looking for two things: 1) Are there anomalous relation—
ships that identify certain measures as problematic? 2) Are there
patterns of relationships that demonstrate the construct validity of
our measures?

Table 9A presents the intercorrelations among the expression indi—
cators of individuation. A problem in the rate indicators is immediately
apparent. Except for the rate of passive verbs (0E6), all the rate
indicators are significantly correlated with number of words (0E1).

These correlations range from -.18 for rate of active verbs (DES; p_fé.05)
to -.72 for the rate of similar dimensions (0E8; p_i£.001). The original
justification for computing the rate indicators was to minimize the
effects of essay length on the various CXpression measures. Whenever

a rate measure is computed, it correlates with both its numerator and
denominator. But in this case, the correlations of rates with number Of
words is substantial. Because of the nature of the data analysis used

to test the experimental hypothesis (the set of dependent indicators is
regressed onto a single eXperimental independent variable--see page 67),
these high intercorrelations mean that problems of multicollinearity

will make interpretation and Significance of some results problematic.
This problem will be discussed in more detail in the next chapter.

Table 98 presents the intercorrelations among the impression indica—
tors Of individuation. These seven self—presentational tactics were
selected as impression indicators on the basis of their differential use
in the validation study. Two of these indicators, differentness (011)

and uniqueness (014), have face validity as measures of individuation.

 

47

ﬂemHAaO-mcov Hoo. w a++ ﬁemsumo-mcov moo.iwnm+ hemHAaO-mcov Ho.mwr at ”assume-mace mo.mw.m.

ﬂea u EV

++vo. ++mm.u ++wn. ma. mH.n mo. ++ov. ++om. ++©v. ++Hw. coﬂumﬁhomcﬂ

HmOMmOHOAOOm mo Oumm Hﬁmo

amm.- ++no. ++mm. *mH.- +wm. ++om. ++mm. ++Ho. ++oo. mpceeeumum

xuomoumo mo oumm oﬁmo

++mm.u HH. eoN. OH. No. wo.u HH.- ++om. mcoﬂmceeﬁw
pceeommﬂw mo ponesz mma

++Hv. ravm.i ++wm. ++Hv. ++om. ++mm. ++Nn. mcowmceeﬂv

anodes-co: mo oumm mmo

0H.l ++wh. ram. ++Nv. ++Hm. «0H.I mﬂho> mmcmp
ucomehm mo Oumm nmo

rmm.- tom.- No. t.am.- ma. mneo>
O>Hmmmm mo Oumm owe

seem. ++mv. +nm. awH. mcwe> O>Huom mo Opmm mmo

NH. ++vv. ++om.n mason mo eumm va
++Ho. ++mm.- muse:OHQ amazmcﬁm
comaeg umhwm mo Opmm mma
++we.- mOOCOpcOm mo Opmm Nmo
memo; mo Aegean Hmo
mmm .mmm .mm .mm .mm .mm .mm .mm .mm .mm .mm

 

 

.mHOpmoﬁwcH “newcomoa mcoe< mcoﬁpmHOHHOOHOch .<m oﬁpmh

1e hi.

i

d

1

All

inc;
an -
:31:
to 0
l3,“

48
The high correlations of these two indicators with positivness (012),
enthusiasm (016), and intensity (017), suggest that these indicators
are tapping the same underlying construct. Surprisingly, flattery
(013) and caculatedness (015) are only slightly related to each other
(r_= .19, p_£§.05) and not too strongly related to the other five
indicators. This may reflect a second (and third) dimension underlying
our dependent variable, individuation. More likely, however, is that
this study reflects the low reliabilities of these two indicators.
The intercoder correlations reported in Table 5 reveal flattery and
calculatedness to be the two least reliable impression measures (Ejs =
.31 and .52, respectively). Although the averaging procedure used in
computing these indicators will improve reliability, the low correlations
of these indicators with the other impression indicators probably reflects
their still insufficient reliabilities. Thus, even though flattery and
calculatedness are problematic, the other impression indicators appear

to be valid measures of individuation.

Table 9B. Intercorrelations Among Dependent Indicators (cont.)

 

 

DIl 012 013 014 015 016 017

011 Differentness

012~ Positiveness .68++

013 Flattery .43++ .17*

014 Uniqueness .83++ .62++ .48++

015 Calculatedness .01 -.38++ .19* -.12

016 Enthusiasm .63++ .59++ .38++ .73++ -.14

017 Intensity .52++ .56++ .24** .67++ -.280 .81++

*RS. .05 (one—tailed) **p 5.- .01 (one-tailed)

+p_ f. .005 (one—tailed) ++p 5 .001 (one-tailed)
(Minimum p_= 98, Maximum p_= 99)

 

49

Table 9C presents the correlations between the expression and the
impression indicators Of individuation. The number of significant cor-
relations show the two sets to be richly interconnected and we might
assume that this is evidence for the earlier assertion that the impres-
sions we create arise from the expressions we use. Upon closer examina-
tion, however, we are forced to reserve our judgment. This is because
of the multicollinearity problem among number of words (0E1) and the
rate indicators that we mentioned earlier. Table 9E shows that number
of words correlates highly with each impression indicator. Correlations
range from .27 for calculatedness (015; p_EE.01) to .73 for enthusiasm
(016; p_fE.001). Except for the correlation with calculatedness, all
the impression indicators correlate with number of words at a signifi—
cance level of p_f£.001. Given the strength of these correlations, and
lremembering the multicollinearity problem, it is no wonder that the
rest of the eXpression indicators correlate so well with the impression
indicators. Thus the relationship between expression indicators and
impression indicators remains somewhat problematic. The extent to which
the impressions we create arise from the various expressions we use
remains unknown. We will discuss possible solutions to this problem in
the next chapter.

Tables 90 and 9E present the intercorrelations of the remaining
dependent indicators of individuation. In Table 90, the correlations
between the QSC Battery and the items from the Personal Geometry Profile
are shown. Earlier, in discussing Tables 7 and 8, we pointed to the low
internal consistency of both scales. Now we see that the correlations
between these scales are not significant. Given that both of these scales

U‘

attempts to measure the same variable, we can assume that at least one

50

.mo an m . my . . -
u m vommounxo one oum0ﬁnCw nozuo HH< oneness Museum mo cemmouoxo Ohm :mcoﬂmcoEwﬂ Oacwzc: 3cm :mncoxz whOpmoﬂuu“ exam

nmm u c EDwanz .mm n_m ennﬁcmzw
Ila
neoﬂweu-Ocov Hoo.sv e++
Andaman-ocov moo. wqro
. . naoﬂaau-ouov «3.”w as.

4 z . 1 .4.
AUG... mdUIOCO“ mo V no.“

P+mm.- ++mo.- ++om. ++mm.u «ma.u «Hm. 0H.n ++~v.u umH.a . ++Hv.n ++nm. Aummcoucm BHQ
i+mu.- ++mn.n P+mm. ++oo.u rHN.s mo. NH.- ++ov.i ++Nm.i . ++Nm.i . ++nn. ummmmnnucm who

imm.- +hm. NH. nob- NH. No.u mo. ++Hm. oo.n mo. +nm. mmmumouumnuﬁmu mHo
++me.- ++on.- ++wm. ++oo.n HN.| amm. som.n ++mo.u +mN.- ++am.i ++oo. mmOCOSUwcn VHO
r+wv.- ++mm.- eo.- ++mm.u aH.a mo.. 00.: ++om.n ++nm.i +nm.i ++mm. xhouummm .MHQ
++mm.- ++vm.- ++nm. ++ov.u aNN.u Ma. ana.i ++mm.i som.u ++ov.u ++mv. mmoco>ﬁpwmoa Nun
++mo.- ++om.- ++mv. ++oo.n ¢HN.- amm. SHN.- ++oq.- ++nm.n ++mm.i ++mo. mm0:u:0howwwo “HQ

 

 

 

 

 

 

nowuxanomcM mueoauuxum mcommuauae meoﬂmcoeﬂe mnao> omcou mnue> mnhO> maze: mcsocouo cmﬁawwwm moonouuom emnnox
Huuwuoﬁoﬂoom anomounu uswﬂca oscﬁ::-coc peemmhm o>ﬁmmmm o>muom cemuom umpwm
Han can an mo na om me we mm N: do

 

 

n.p:oou kapmoﬂncH econcomoo wae£< chAOmHehnoououcm .uo oﬂnmh

51

of them (and possibly both) fails as a measure of individuation. Table
9E provides more information relevant to this issue.

In Table 9E, we present the correlation of the nonverbal measures,
the QSC Battery and the Personal Geometry Profile items, with the verbal
expression and impression indicators of individuation. The QSC Battery
(081) Shows only one significant correlation (with calculatedness, DIS;
r_= -.20, p.£§.05) among the 18 verbal indicators. Given the evidence
that already has been presented to establish the reliability and validity
of the verbal measures, we can only conclude that, in this particular
instance, the QSC Battery does not measure individuation. The QSC Battery,
therefore, will not be used in further analysis.

For the first two constructs from the Personal Geometry Profile
(082, the discrepancy between self and R.S., and 0B3, the diécrepancy
between self and J.L.), we have a similar situation. The constructs are
uncorrelated with established measures of individuation and will also be
dr0pped from further analysis. The third construct from the Personal
Geometry Profile (DB4), the difference between self and the "average
student," presents a rather unusual problem. It is significantly corre-
lated with many of the rate expression measures, as we would like; however,
it is negatively correlated with number Of words and all of the impression '
indicators. This means that as students' essays reveal more positiveness,
differentness, uniqueness, etc., the students are reporting themselves to
be more, not less, like the average student. This is the opposite Of what
we expect. Except for the correlation with enthusiasm (016; £_= -.18,
p_£;.05), the relationship of this Personal Geometry Profile construct to
impression measures is not significant. Nonetheless, this indicator of

individuation will be used cautiously in further analysis.

52

 

vmo

«tvm.

mma

Aha n
++mm. as.
++oa. so.

as.
was Hmo

.m Edeﬂxmz .vm u.m Edeﬁcﬂzv
Guiana-28V 8o. w mt.

Awesome-ocoe so. .mii

V!

pcowsum
omene>m wee pompoum ceozpen zucwmoaomﬁa

.4.5 can pcewsum ceozuon zucmmonomwa
.m.m use anew3um cmozpee xocmmonomwo

mooom ssopemm umo

vmo

mmo

mmo

Hmo

 

 

.m.u:ooV meoumowwcH

ucepeemoa mcoe< mcoﬁumHOHuooneucH .am Oanmh

-.-...av u..~\....-4d—.-- u—..-—..-.'~A-.o.~ ud.....:(\ J...-..~ sns~.~.~.!.-.-p-.- ..-‘ .~\.~s,.\.

53

 

ham ".0 assaxaz .em ".0 asaacazv

 

 

00. .00.- 00.- m0. :OADMEHomcﬁ
neoﬂwmu-Ocov H00. W m++ HeOAmoHOAOOm mo comm ﬂame
”nausea-0:00 moo..wzm+ as. mo. Ho.- 00.- manageaaam
xnomoumo mo Oumm 0Hm0
ﬂeoHAaO-mcou as. w m-.
00. ma. vH. 00. mcoﬁmcesﬁe
8033-280 mo. Wm... puma-an. mo A352 as
sea. HH.- e0. 00. mcoﬁmcoeﬁv
Oscﬁcs-co: mo Opmm wmo
Pma. mo. m0. mH. mnao> emcee
pcomonm mo Opmm nmo
NH.- no. 00. 00. xuﬁmcoucH 5H0
No. +AN. No. mo. mnem> m>nmmaa mo mama emo
0H.- 00. 00. H0.- Emmﬁmsgucm 0H0
«mm. v0. *BH. 0H. meao> o>Huom mo comm mmo
mo. 00. mo.- aom.- mmeauepmasonu mHQ
.«NN. v0.- mo. 00.- mono: mo comm «mo
mH.- OH. No. as. mmmemseaca «Ha
a0N. H0.- no. ~0.- unaccepm Hmﬁsmcﬁm
v~.- 00.- e0.- no. anppmHm mH0 cemaom umewm mo ovum mmo
m0.- 0H.- 00.- m0. mmoee>ﬂpﬁmom «Ho Pmﬁ. H0.- NH. 00.- meoceucom mo Oumm «mo
00.- v0. N0. m0. mmocuconomwwo HHO 0H.- m0.- ﬁ0.- m0.- muse: mo senesz Hmo
v00 000 ~00 H00 v00 mmo ~00 H00
.ﬂ.ueoo0 mucumowch ocowcemo0 mcoe< mcoﬁpmﬁoahoonmucH .mm Oﬁnme

54

Let us summarize briefly the implications of this last set of
tables. Our purpose in presenting these tables was to demonstrate the
reliability and validity of our measures. The tables revealed mixed
success over the many measures we considered. Some measures proved
reliable and valid, and some measures less so.

The Cronbach's s: for Machiavellianism and self-esteem demonstrate
moderate internal consistency for these nonexperimental independent
variables. They will be used in further analysis. Intercoder correlaé
tions of the dependent expression and impression measures were generally
high enough for us to consider these reliable indicators of individuation.
Correlations among indicators were used to establish the construct vali-
dity of the impression indicators. Unfortunately, the high correlations
of number of words with the rate of expression indicators suggests that
the heped for independence of expression indicators was not achieved.
Problems of multicollinearity may confound subsequent analysis. Because
of its problematic relationship with the impression indicators, one of
the Personal Geometry items, the self-report of difference between the
individual and the "average student," will be used cautiously as an
indicator of individuation. The other two of the three items on the
Personal Geometry Profile and the QSC Battery revealed questionable
internal consistencies and/or relationships with other indicators that
made their validity as indicators of individuation suspect. These mea-
sures will not be used in further analysis. Having pointed out the
strengths and weaknesses of our measures, let us now consider the effec—
tiveness of our attempts to construct situations that the students would

perceive as differentially rewarding.

55

3.4 Results of Data Analysis Pertaining to EXperimental Manipulation

 

Table 10 presents the results of three questions used to check the
successfulness of our attempts to manipulate the students' perception
of the reward Situation. Students were asked to report how much they

liked each QSC training program, how rewarding they felt the treatments

 

were in each learning program, and how rewarding they felt each learning
program was. To check to see if the different learning programs repre-
sented situations of differing reward, we compared the ratings of condi-
tions by students in those conditions. That is, evaluations of the
Positive Learning Program made by students in the positive condition
were compared with evaluations of the Standard Learning Program made by
students in the control condition and with evaluations of the Negative
Learning Program made by students in the negative condition. For all
three manipulation checks, significant differences between conditions
(p S:.0001) were found. The Positive Learning Program is rated highest,
the Standard Learning Program next highest, and the Negative Learning
Program is rated lowest. Thus, student perception of the reward struc-
ture seems to have been successfully manipulated in the exPeriment.
Before we present the data analyses used in assessing the relation-
ships between our independent and dependent variables, let us review
briefly what we have already presented in this chapter. At the beginning
of the chapter, we said that before any test of a hypothesis could be
meaningfully interpreted, we needed to have confidence in the represen-
tativeness of our sample, in the reliability and validity of our measures,
and in the successfulness of our eXperimental manipulation. The results
presented so far are the bases for different levels of confidence in our

methods.

(a

W. 1.4-J—aﬂ4

Fqua Via - n:— . ...~—.)—

- n v-—

a

v

.Jn-~

A ..~ dd >

L »4

.-,-...a.~u>._ .4..<

ma: :

v...-AJ .2 His «>1-Au

uu-n uv—duuhn a.

(i

V--..rv.<

.u.

16 ~ n-..

-

S6

.mOHnmﬁnm> ucoecomov use we woesuﬁcmma Haemonm.oumoﬁecﬁ mucosa: :wﬂzm

 

 

 

 

0.0 00 0.000 soAAm
H000. 00.00 0.H~m N 0.~00 coapaecoo mcaomm 500
~.0 00 0.000 House
H000. ms.e0 0.H00 N H.00A aoaoaecoo 0e0300 A00
N.m 00 m.~00 Locum
H000. 00.00 0.5Hm N 0.400 00a0a0000 Manama ﬂa0
.mm .mw .mw ooasom :oﬂumoso
mm HA.N H0.N 0>A0awoz
00 0N.~ 00.0 essecaom 00A00A0 assesses 000
00 ~0.s 0H.0 asaosmoa co mcauam m00
mm 00.0 am.m o>aeawaz
em 00.H 0m.m wumecmpm Seaweed mcwcnmoq umo :30
OM mm.H mm.m O>Huwnmom Sun mHCmEHmmHH m0 ﬂthOm ﬁo>wmohmm mm:
mm 00.~ 05.N o>aommmz
em 0H.N 00.0 neaecaom aaewosa assesses 000
00 N0.H 00.0 0>Aoamoa 030 do assess e0 00A000 ﬂa0
.m .0.m coo: macaw coﬁpmoso

 

 

m.mxoocu :Owumﬁsmwcmz How oocmwam> mo momxﬂmc< 0:0 .mcoﬁumﬁ>o0 enmecmpm .mcmoz .0H mane»

57

First, the sample appears reasonably representative of the popu-
lation of "average college students" for which the experiment was
designed. However, we did notice a higher female/male ratio in the
experiment than in the validation study and cautioned that sex differ-
ence in expression and impression indicators could confound any of the
hypotheses. Second, our measures seem valid and fairly reliable, with
a few exceptions. There is some question about the strong relationship
of number of words with other eXpreSSion indicators. Multicollinearity
among expression indicators poses two problems: (1) testing the hypothe-
ses by relating the whole set of expression indicators with the linear
effect of experimental condition is likely to be biased conservatively
because degrees of freedom will be used up among the highly correlated
expression indicators, and (2) interpreting the causal links within the
set of eXpression indicators is made difficult by their rich intercorre-
lation. These issues make the testing of the experimental hypotheses
less than straightforward. Finally, the experimental manipulation
proved very successful. -Whatever complications may arise because of
the sample of the measure, we can be confident that students confronted.

situations of different reward.

3.5 Results of Data Analysis Pertaining to the Relationship Between

 

Individuation and Perceived Potential Reward

 

Having reviewed our findings so far, and having sensitized outselves'
to possible complications, let us now look at the results of data analyses
that reflect on the relationship of the independent and dependent varia-
bles of this study. In examining these relationships, there are several

things we want to do. Obviously we want to test the experimental

58

hypothesis that the degree of individuation will be a positive function
of perceived reward in a situation. To do this, we will examine the
relationships between the eXperimental independent variable, linear
effect of condition (X1), and the dependent indicators taken individually
and in sets. Moreover, to get a more complete picture of the overall
impact of experimental treatments on individuation, we will also conSider
the quadratic effect of experimental condition (X2) on our dependent
indicators. But in addition, we want to know the extent to which our
dependent indicators are the results of other nonexPerimental effects.
To do this, we will consider the effects of the nonexperimental indepen-
dent variable, sex (NX2), Machiavellianism (NX7) and self-esteem (NX8),
on our indicatOrs of individuation.

First, let us consider these noneXperimental effects. Tables 11A and B
Show the zero-order correlations of sex, Machiavellianism, and self-
esteem with the indicators of individuation. These nonexperimental inde-
pendent variables are differentially related to the dependent indicators.
Among expression indicators, sex (NX2) is significantly related to rate
of sentences (0E2; p_= -.25, p_£2.01), rate of similar.dimensions (0E8;
r_= -.28, p £3.05). Among impression indicators, sex is significantly
related to differentness (011; p_= .23, ppg;.05), flattery (013; r_= .20,
p_E;.05), enthusiasm (016; r_= .23, p_£2.05), and intensity (017; r_= .23,
11:5.05). This means that in the sample, women tend to write longer
sentences and have lower rates of similar dimensions and category state-
ments than men. They also are significantly more different, flattering,
enthusiastic and intense in their self—presentations than men. As we
noted above, the different sex compositions of validation and CXperimentalp

samples could be a confounding factor if these were significant sex

S9
differences in the expression and impression indicators. These find-
ings suggest such a sex difference. This issue will be examined further
in Table 12 below.

Like sex, self-esteem is also Significantly related to a number of
expression and impression indicators. Students with high self-esteem
tend to use a lower rate of active verbs (0E5; £_= -.21, pié .05) and
a higher rate of category statements (0E10; r_= .19, p.E¥.05). Inter-
estingly, students of higher self-esteem present themselves less positively
(012) and less uniquely (014) than those of lower self-esteem, but not
significantly 50. High self-esteem students appear significantly more
calculating (DIS; r_= .19, p5S}.05), less enthusiastic (016; r_= -.25,
p_:E.01), and less intense (017; £_= —.28, p:S;.005) than low self-esteem
students. This suggests that perhaps the high self-esteem students,
confident in themselves, appeared detached in their self-presentations
and that this detachment was seen as a calculated strategy.

Sex and self-esteem are both also significantly related to the
degree of individuation reflected in the Personal Geometry Profile item
(DB4; {)3 = -.20 and .20, respectively, pfs 55.05). This suggests that
women and students with higher self-esteem see themselves as being more-
different from the average college student, than do males and students
with lower self-esteem. Surprisingly, Machiavellianism (NX7) is not
significantly related to any of these indicators of individuation.

The reason for presenting the results of Tables 11A and B is to
allow us to assess some nonexperimental influences on individuation.
These results suggest that some of the expression and impression indica-
tors are significantly related to two of our nonexperimental independent

variables: sex and self-esteem. Our next concern is whether or not these

00

.N u OHmEow .H u OH

 

 

 

 

Ed
00 u.m essaxa: .50 u.m 005a0a20
ea. mo.- mH. codumEhomc«
3033-380 H00. Wm: 1.632303. mo 30m 34-00
moonmu-Ocov m00..Wﬂm+ cmﬁ. 00. «mm. muEOGOHmum
scomoueo mo Ouum 0Hm0
. Apogee-980 H0. \H a:
mo. N0. Ma. mcoﬂmcoeﬂw
ﬂeosaau-oco0 mo. mum. mean: 00 reason 000
no. ”0. am“. mcoﬂmcoewn
Ozewcs-co: mo Oumm mum
NH.- m0.- . 00. mean»
emcee ucomoem we owe“ nm0
-mm.- HH.- .mu. suamcmch 0H0
ea. m0.- m0. mnuo> o>wmmma mo oumm 0mm
xgmm.- 0H.- emu. Emmwmsgucm 0H0
.H~.- 00. 0s. m0A0> 0>s00a mo 00am mma
.0H. 00. 00.- mmocuoumHsOAeu mHn
0H. m0. ma. mason mo Ouem «mm
AH.- 0a.- H0. mmocosuaca 4H0
m0.- N0.- oﬁ. masoc0hm hausmcﬂm
m0.- HH.- xom. xuouumﬁw mHa cemuog umpwm mo comm mmo
o“.- co. . 5H. mmono>ﬂuﬁmom NHQ wo.- mo.- «emm. mOOCOHCOm mo oumm mun
m0.- NH.- ymm. mmOCpcouommﬂa HHQ v0.- mo.- 5H. mpaox mo conﬁdz amm
inoumu-mﬂom ﬁwwuwwmﬁo>mwﬁom2 xom EOOHmm-mmom Emﬁcmﬁﬂﬁo>mw£0m2 mxom
max 0x2 mxz 0x2 . nxz mxz

 

 

.muouOOAECH unaccomoo spa: EmwGOMHHO>mﬂnomz one .Eooumm-mﬂom .xom mo mCOwueaoaaou .<H~ oHnme

61

,.m n onEom .H u oﬂmzm

 

 

 

row. Va.

NH. H0.-

00. 00.

No. ram.
Eooumm-mﬁom Emﬁceﬁﬁao>mwcomz

0x2 nxz

:0 n m 033082 .00 u .0. 5&0an

«.ON...

NH.-

mo.

 

mxom
Nxz

meoﬂﬁmu-ocov m0.mn.min

ucopsum ommeo>m 0cm
ucowsum coozpon zocmmoaomﬁo

.4.0 new
ecowSHm coozpon xocmmoaomﬂo

.m.m 0cm
ucowzpm coozuon socmmoaomwo

oaoom xhoupmm umo

vmo

mmo

NmQ

H00

 

 

.mnoumowvcH ucoecomoa new: Emﬁcmﬁﬁao>mﬁcomz 0cm .Eooemm-mﬁom .xom mo meowpmHOHnou

nyucoov

.mHH oHan

62
nonexperimental independent variables are significantly related to the
entire set of expression or impression indicators.

Table 12 presents the results of multiple regression analysis
of the relationships between the nonexperimental independent variables,
sex (NX2), Machiavellianism (NX7), and self-esteem (NX8), taken singly
with the sets of expression and impression indicators. Some of these
relationships are moderately strong. The R? of sex with impression
indicators is .14. 52's of self—esteem with eXpression and impression
indicators are .19 and .15. None of the relationships are significant,
however. We are forced to conclude that individuation, as reflected
in subjects' self-presentation essays, is only Slightly influenced by
our nonexperimental independent variables.

So far we have found that individuation, as reflected in the
expression and impression indicators, is relatively unaffected by sex,
Machiavellianism, or self-esteem. A third indicator of individuation,
however, the Personal Geometry Profile construct is significantly
affected by sex and self-esteem. This contradiction may be due to
problems of multicollinearity that make suspect our assessment of the
expression and impression analyses. Or it may be due to the question-
able validity (see p.5il) that makes the Personal Geometry Profile
item suspect. In either case, no strong general evidence for direct
effects of noneXperimental independent variables on individuation has
been found. Let us now examine the effects of our experimental variables.
on individuation.

Table 13 presents the correlations of dependent indicators with
the two experimental independent variables, linear and quadratic effects

of eXperimental conditions. Linear effect of experimental conditions

63
Table 12. Multiple Regression of Nonexperimental Independent Variables

with Indicators of Individuation.

 

 

 

 

 

 

 

 

 

NXZ NX7 NX8
Sex Machiavellianism Self-Esteem
Beta Wt. Beta Wt. Beta Wt.
Expression Indicators .
0E1 Number of words -.02 -.31 .12
- 0E2 Rate of sentences -.29 -.26 , .08
0E3 Rate of first person
singular pronouns .11 .18 -.07
0E4 Rate of nouns, gerunds,
and infinitives .07 .05 .13
DES Rate of active verbs -.16 -.01 -.40
DE6 Rate of passive verbs -.10 —.13 .19
0E7 Rate of present tense
verbs -.09 -.18 .05
0E8 Rate of non-unique
dimensions -.08 .03 .27
DE9 Number of unique
dimensions .11 .17 .11
DElO Rate of category
statements -.15 .19 .20
0E11 Rate of sociological .
information .09 -.29 -.08
32 .10 .07 .19
Significance >u10 . >.lO ‘>.10
Impression Indicators
DIl Differentness .55 -.27 .47
012 Positiveness -.19 .25 -.17
013 Flattery .11 ’ -.08 .15
014 Uniqueness -.56 .10 -.22
015 Calculatedness -.08 .13 .00
016 Enthusiasm .14 -.12 -.19
017 Intensity .21 .03 -.23
32 .14 .04 .15
Significance <.10 .>.10 -<.10

 

(Minimum p.= 87, Maximum p_= 89)

 

64
is significantly related to only one expression indicator, rate of
category statements (0E10; £_= -.17, p_:2.05). It is significantly
related to three of the impression indicators, positiveness (012;
r_= .19, p_f§.05), enthusiasm (016; :_= .22, p 53.05), and intensity
(017; r_= .20, psS .05). Surprisingly, linear effect of experimental
conditions is negatively and almost significantly related to the
Personal Geometry Profile indicator of individuation (DB4; p_= -.16,
p_f;.10).

The experimental hypothesis predicted a positive relationship between
the linear effect of eXperimental conditions and individuation. These
results are not sufficient to confirm or reject the hypothesis. Certainly
some support for the hypothesis is found in some of the relationships
between linear effect and exPression and impression indicators. The
negative relationship between linear effect and the Personal Geometry
Profile indicator may be evidence for rejecting the hypothesis.
Alternately, it may further question the validity of that item as an
indicator of individuation. Further analysis of linear effect will be
presented in Table 14.

Tables 13A and B also show the correlation of the dependent indicators
with the quadratic effect of experimental conditions. Quadratic effect
(X2) is significantly related to number of words (0El;lr_= .22, p.$-.05),
rate of first person singular pronouns (0E3; p_= -.18, p:£§.05) and
calculatedness (DIS; £_= .20, p_££.05). This means that compared to
students in the standard condition, students in the negative and positive
conditions wrote longer introductory essays, used a lower rate of first

. . 19
person s1ngular pronouns, and were more calculat1ng.

65

 

Ho.-

v0.

Na.

ma.-

Ho.

NH.-

HO.-

OH.

swan..-

N0.

«mm.

80 .- m 03508: .00 .- .m 055520
08:00-23 .0. Wm...

00.- nowumshomcw
HmonoHOMOOm mo Oped

inﬁs- mucoeououm
xsomoumo mo ovum

00.- mcoﬁmcoswu
awe oscwca mo popes:

0H.- mcoﬁmcoewv
magmas-co: mo ouam

V0. mnnO> emcee Hammond mo ovum
NH.- meno> o>ﬁmmmg mo ovum
NH. mnuo> O>wuow mo ovum
m0. mono: mo Opmm
0H.- masocOHQ Hmaowcwm

:Omaom umuﬁm mo comm
00.- moo:Op:om mo mama

so. muse: mo honasz

 

 

mcoﬂuﬂvcoo Heucoeﬁhommo
mo uoommo oﬁpmhwmzo .Nx

mcowuﬂvcoo Hmucosﬁnummo
mo poommo Hmoawa .Hx

Hana

Damn

0mm

mmo
hma
0mm
mmo

vmo

mmo
mma

Hun

 

 

.mnouwowocu «covcomoa new: meanmﬁnm>ucoucomoecH Hmpcoeﬁnomxm mo mcoﬁpmﬁoanou .<m~ oHnoh

66

 

all. I

H50 0 essaxmz .00 u c essacsz0 HemHH00-0000 m0.m“ m.

 

m0.- 0H.- “savoum ommso>m 0cm
ucowSHm :oozuon xocmmOhomHo em

00. 0H.- .4.0 wan
«savoum coozuon seemQOHOmHa mm

mo.- 00. .m.m 1cm
peopspm coozuop HaemQOHOOHo N0
.0H.- 00. msoom sumpomm 000 H0

H00 -.m engage: .00 -.m eneacaz0

50. .00. suamcmoeH AH0
HH. sum. Emonscuem 0H0
«0N. oo.- mmocvoumHsonu mHQ
00. 00. mmoeoscacs 000
0H. NH. AnouuaHm 000
0H. .mH. . m0000>Honoa NH0
0H. mo. . mmOCpCOHommHo HHQ

 

mcoHancoo HmucoEHaomxo mcoHqucoo HmucoeHuomwo
mo poommo oHumuvmzo .Nx we woommo HmocHH .Hx

 

 

H.p:oo0

.maoumoHecH ucoecomoo :HH3.mOHanHm> unaccomoecH HmucoeHhomxm we mcoHumHonhou .mmH OHcmh

67

These results, for both experimental independent variables, con-
sider the relationships between OXperimental effects and individual
dependent indicators. A more complete analysis would have us look at
the relationship between each experimental independent variable and the
sets of dependent indicators.

Table 14 presents the results of multiple regression analysis of
the effects of experimental conditions taken singly on expression and
impression indicators of individuation. The regression relating linear
effect of experimental conditions (X1) with the sets of expression and
impression indicators is our strongest test of the experimental hypothesis.
The multiple 5? for linear effect with expression indicators is .41
(2_>>.10). The multiple R? for linear effect with impression indicators
is .31 (p ;>.20). Because we hypothesized a positive linear relationship
between the reward structure of the eXperimental condition and individua-
tion, we cannot reject the null hypothesis of no difference by eXperimental
condition (linear). Again, however, we are unsure of what to attribute
this failure. On the one hand, the failure may signify a failure of the
theory. On the other hand, the failure to find a significant relationship
between linear effect of eXperimental conditions and expression and im-
pression of individuation may be due to the conservative significance
level produced by having highly related dependent indicators. We will
discuss this issue in greater detail below.

Table 14 also presents multiple regressions of the sets of dependent
indicators with the quadratic effect of experimental conditions. These
reveal a significant relationship with expression indicators (R? = .52,
p_£§.01) and a non-significant relationship with impression indicators

(5? = .29, p_:>.20). This means that although there were no significant

 

“'7‘
I U

H00 - 0 0000002 .00 u 0 00000020

 

 

 

 

68

 

 

0H.A. 00..u 000000wmmwﬁm

00. 00. mm
0H. 0H. .N0H000ch 0H0
m0. mm. EmmHmznuem 0H0
NN. mo. mmocvoumHsonu mHo
00.- 00.- 0000000000 000
00. 00.- suouuaHm 000
mm. mm. mmoco>HuHmom «Ho
mH. 00.- mmoeecouommHo HHO
muoumOHveH :onmousH

H0. W 0H . A 3:03chme

00. 00. mm.
00. 00. 00000000000 000000000000 00 0000 0000
00. mm.- mueoaOpeum xaomoumo mo oumx 0Hm0
0H.- 0~.- meoneoaﬂe oscHes mo sonsoz 0mm
00.- H0.- meoneoeHv oschs-co: mo ovum 0mm
00. N0. mnuo> emcee ecomOHm mo ovum nma
00.- oo.- mnuo> O>Hmmmm mo ovum own
on. me. mnno> O>Huom mo oumm mmo
m0.- 00. mo>HuHeHmcH one .mucopow .mcsoc MO 0000 0mm
mm.- HN.- mcaonHQ umHsmch cowhom umhﬂm mo oumm mmo
Hm. mH. . - . moocou000 mo ovum mmo
m0. 0H. .. . meuo3 me 909052 Hum
mnoumoHveH :onmonmxm

.03 «pom .uz «pom
mcoHpHeeoo Hep:OEHAOmxo meoHuHecoo Hmucoanomxo
mo uoommo oHumawmso .Nx mo ueommo HmocHH .Hx

 

 

.coHumsz>HecH mo mHOpmochH 20H: moHneHHm> acoucomovcH HmueosHHomxm mo :onmonmom onHuHsz .0H oHan

69
differences in impressions across conditions, the expressions of students
in the negative and positive conditions were significantly different
from those of students in the control condition. One of the significant
among the expression indicators is number of words (p_= .73, std. err. =
.01). We will discuss this finding in greater detail below.

Although the multiple regression analysis with linear effect of
experimental condition would have us fail to reject the null hypothesis,
further investigation may give a more complete picture of the sources of
variation in individuation. Tables 15 through 20 present the results of
canonical correlation analysis of the overall effects of the noneXperi-
mental and experimental independent variables on the sets of eXpression
and impression indicators.

Table 15 presents the results of a canonical correlation analysis
Of the overall linear and quadratic effects of experimental conditions
on the set of eXpression indicators. The table shows a significant canon-
ical relationship (RC = .58, p_€é.01). This is certainly strong evidence
suggesting that the experimental manipulation had considerable impact on
the verbal behavior of the students. As with the relationship between
quadratic effect and expression indicators, rate of sociological informa-
tion, number of words, and rate of similar dimensions seem most instru-
mental in the relationship.

Table 16 presents the results of a canonical correlation analysis
of the overall linear and quadratic effects of experimental conditions
on the set of impression indicators. The table shows a nonsignificant
canonical relationship (50 = .37, p >.1o). This implies that although
the eXperimental manipulation affected verbal behavior, it did not affect

the specific impressions that arose from the verbal behavior.

70
Table 15. Canonical Correlations of Linear and Quadratic Effects of
Experimental Conditions with Expression Indicators of

Individuation.

 

 

Expression Indicators

 

 

 

 

Canonical Variate Number Canonical Correlation XE_ df_ Significance
1 ' .58 41.36 22 :£.01

First Set (Independent) Variate 1

X1 Linear effect of experimental conditions .53

X2 Quadratic effect of experimental conditions .83

Second Set (Dependent)

 

0E1 Number of words 1.18
DEZ Rate of sentences .55
DE3 Rate of first person singular pronouns -.66
DE4 Rate of nouns, gerunds, and infinitives .03
DES Rate of active verbs .82
DE6 Rate of passive verbs -.08'
0E7 Rate of present tense verbs .11
0E8 Rate of non-unique dimensions -.98
DE9 Number of unique dimensions , -.46
0E10 Rate of category statements -.17
DEll Rate of sociological information 1.50

 

(Minimum 2': 87, Maximum p_89)

.

71
Table 16. Canonical Correlations of Linear and Quadratic Effects of
Experimental Conditions with Impression Indicators of

Individuation.

 

 

Impgession Indicators

 

Canonical Variate Number Canonical Correlation ‘12 df Significance

 

 

 

1 .37 16.01 14 )>.20
First Set (Independent) Variate 1
X1 Linear effect of experimental conditions .75
X2 Quadratic effect of experimental conditions .64

Second Set (Dependent)

 

011 Differentness .20
012 Positiveness .97
013 Flattery -.05
014 Uniqueness -1.40
015 Calculatedness .41
016 Enthusiasm .60
017 Intensity .47

 

(Minimum p_= 87, Maximum p_- 89)

72

Tables 17 and 18 present the results of canonical correlation
analysis of the overall effects of sex, Machiavellianism, and self-
esteem on the sets of dependent indicators of individuation. Table 17
shows the canonical relationship of these nonexperimental independents
with the expression indicators to be nonsignificant (RC = .42, p J>.10).
Similarly, Table 18 shows the relationship of these noneXperimental
independents with the impression indicators to be also nonsignificant
(RC = .40, p_>>.10). Both analyses suggest that the nonexperimental
independent variables are not powerful predictors of individuation.

Tables 19 and 20 present the results of canonical correlation
analysis of the overall effects of both the eXperimental and nonexperi-
mental independent variables on the sets of dependent indicators.
Table 19 shows the relationship of the independent variables with the
expression indicators to be nonsignificant (Rc = .60, p ;>.10). 5c
”for the experimental effects alone is .58, suggesting that little addi-
tional explanatory power is gained by adding the noneXperimental inde-
pendent variables. Table 20 shows the relationship of the independent
variables with the impression indicators to be nonsignificant (RC = .47,
p_)>.10). In general, it appears that the nonexperimental variables are
better predictors of impression indicators, while the experimental

variables are better predictors of expression indicators.

73
Table 17. Canonical Correlation of Nonexperimental Independent

Variables with Expression Indicators of Individuation.

 

 

Expression Indicators

 

Canonical Variate Number Canonical Correlation _X2 df Significance

 

 

 

1 .42 28.06 33 J>.10
First Set (Independent) Variate 1
NXZ Sex -.07
NX7 Machiavellianism -.06
NX8 Self-esteem .99

Second Set (Dependent)

 

0E1 , Number of words .27
0E2 Rate of sentences .19
0E3 Rate of first person singular pronouns -.l4
DE4 Rate of nouns, gerunds, and infinitives .25
0E5 .Rate of active verbs -.86
0E6 Rate of passive verbs .48
DE7 Rate of present tense verbs .13
0E8 Rate of non-unique dimensions .57
059 Number of unique dimensions .25
0E10 Rate of category statements .55'
0E11 Rate of sociological information -.19

 

(Minimum n_= 87, Maximum n_= 89)

74
Table 18. Canonical Correlation of Nonexperimental Independent

Variables with Impression Indicators of Individuation.

 

 

Impression Indicators

 

Canonical Variate Number Canonical Correlation XE_ gf_ Significance

 

 

 

l .40 28.25 21 >u10
First Set (Independent) Variate 1
NXZ Sex -.59
NX7 Machiavellianism .12
NX8 Self-esteem -.89

Second Set (Dependent)

 

011 Differentness -1.86
012 Positiveness .67
013 Flattery -.51
014 Uniqueness 1.28
015 Calculatedness .09
016 Enthusiasm .16
017 Intensity .21

 

(Minimum n_= 87, Maximum n_= 89)

75
Table 19. Canonical Correlations of Experimental and Nonexperimental
Independent Variables with Expression Indicators of

'Individuation.

 

 

Expression Indicators

 

2
Canonical Variate Number Canonical Correlation X__ df_ Significance

 

 

 

1 .60 67.31 55 >.10
First Set (Independent) Variate 1
X1 Linear effect of experimental conditions .53
X2 Quadratic effect of experimental conditions .77
NX2 Sex -.14
NX7 Machiavellianism -.l8
NX8 Self-esteem -.28

Second Set (Dependent)

 

0E1 Number of words 1.11
0E2 Rate of sentences .61
DE3 Rate of first person singular pronouns -.65
DE4 Rate of nouns, gerunds, and infinitives -.05
0E5 Rate of active verbs .97
0E6 Rate of passive verbs —.11
0E7 Rate of present tense verbs .09
0E8 Rate of non-unique dimensions -1.0l
DEQ Number of unique dimensions -.56
0E10 Rate of category statements -.29
DEll Rate of sociolegieal information 1145

 

(Minimum n'= 87, Maximum n_= 89)

76
Table 20. Canonical Correlation of Experimental and Nonexperimental
Independent Variables with Impression Indicators of

Individuation.

 

 

Impression Indicators

 

2
Canonical Variate Number Canonical Correlation X__ g:_ Significance

 

 

 

1 .47 47.18 35 .>.10
First Set (Independent) Variate 1
X1 Linear effect of experimental conditions .56
X2 Quadratic effect of experimental conditions .31
NXZ Sex A .70
NX7 Machiavellianism . -.10
NX8 Self-esteem -.00

Second Set (Dependent)

 

011 Differentness .92
012 Positiveness .16
013 Flattery .15
014 Uniqueness -1.54
015 Calculatedness .02
016 Enthusiasm .57

017 Intensity , .55

 

(Minimum n_= 87, Maximum n.= 89)

77
3.6 Summary

At the beginning of the chapter, we said that in order for us to
have confidence in the meaningfulness of our test of the hypothesis, we
first had to have confidence in the internal and external validity of
our methods. This meant we had to have confidence in the representative-
ness of our sample, in the reliability and validity of our measures, and
in the successfulness of our experimental treatments.

We then presented the results of a number of data analyses relevant
to these issues. We concluded that our sample was representative of the
pOpulation of average college students. We also concluded that our
measures, for the most part, were reliable and valid. In addition, we
concluded that our experimental treatment was successful. Nonetheless,
we noted several problematic areas in our methods. (This will be dis-
cussed in more detail later.)

We then began examining the relationships between the independent
variables and the dependent indicators. We were especially interested
in testing the eXperimental hypotheses that individuation is a positive
linear function of experimental conditions. However, we were also
looking for other sources to explain the variance in individuation.

The results of these analyses was to fail to reject the experimental
null hypothesis. Although some individual indicators of individuation
were significantly related to the linear effect of experimental condition,
the sets of expression and impression indicators were not. We also
failed to find significant explanatory power in our nonexperimental
independent variables: sex, Machiavellianism, and self-esteem. We did
find, however, significant relationships between quadratic effect of
experimental conditions and overall effects of experimental conditions

on expression indicators.

78
In the next chapter, we will try to offer OXplanations for these
findings, both methodological and theoretical. We will also consider

different implications arising from our analysis.

CHAPTER FOUR

IMPLICATIONS

4.1 Summary of Results

 

Before we begin a discussion of the implications of our research,
let us briefly review the findings reported in Chapter Three. In that
chapter, we presented the results of a number of data analyses performed
to provide evidence for assessing the internal and external validity of
the research design and for testing the experimental hypothesis. We
found that for the demographic variables we measured, the sample was‘
relatively free of extreme values or unusual distributions. Moreover,
random assignment of subjects to conditions controlled for the possible
effects of sex, Machiavellianism, and self-esteem, three variables we
believed might influence individuation. We also found that most of our
measures, including the ratio-scaled measures of self-presentational
impressions, were reliably coded. Intercorrelations among indicators
established the validity of these measures, and pointed to possible prob-
lems in interpreting some significance tests for the experimental
hypothesis. Manipulation checks showed that the experimental treatments
were very successful in getting students to perceive differential reward
in eXperimental conditions.

Of the 18 measures in the two sets of dependent indicators, few
were sensitive to the linear effect of experimental conditions. Positive-
ness, enthusiasm, and intensity, all impression indicators of individua-
tion, were the only ones showing significant relationships with the linear

effect of eXperimental conditions. When taken as sets, neither expression

79

80
nor impression indicators of individuation were significantly related
to the linear effect of eXperimental condition. Expression indicators
of individuation, however, were Significantly related to both the qua-
dratic and the overall effects of experimental conditions. No signifi-
cant effects On the set of impression or expression indicators were found
for sex, Machiavellianism, or self—esteem.

The net result of these findings is twofold:

1) The findings cause us to fail to reject the exPerimental
null hypothesis that degree of individuation is posi-
tively and linearly related to perceptions of potential
reward.

2) The findings raise the question of why quadratic and
overall effects of experimental conditions are related
to the expression indicators of individuation.

In the rest of this chapter, we will consider methodological and
theoretical explanations that may account for our failure to reject the
experimental null hypothesis. We will also offer some interpretations
of the surprising relationships of quadratic and overall effects of

experimental conditions on individuation. Finally, we will suggest areas

for further research.

4.2 Failure to Reject the Null Experimental Hypothesis

The results of the experiment leave'us with at least two interpre-
tations: either 1) the hypothesis as stated is incorrect, or 2) problems
of method failed to subject the hypothesis to a true test. In the theore-
tical discussion of individuation, sufficient anecdotal evidence was pre-
sented to suggest that the hypothesis is grounded in social behavior.
In the results section, evidence was provided to suggest that there were
no problems with the sample, that the indicators used were valid and

reliable, and that the manipulation was successfu1. Let us look at both

81

experimental and analytical aspects of the methods to see whether
weaknesses there could confound the results.

Two arguments can be raised to suggest that the experimental methods
did not subject the hypothesis to a true test. The nature of the valida-
tion study and the nature of interaction in the experiment both may have
been inappropriate for testing the hypothesis.

The first argument suggests that the hypothesis was not subjected
to a true test because of weakness in the set of dependent indicators
used to test it. The dependent indicators were chosen in a validation
(pretest) study in which students writing self-descriptive essays were
differentially directed either to try to attract attention or to try to
avoid attention. No such explicit directions were given in the experiment.
Thus, the conscious pretest verbal behaviors and the perhaps not so con-
scious verbal behaviOrs in the experimental situation may not be compatible.
This is certainly a reasonable argument, and one that we anticipated in
our methods section. But our response is that the indicators chosen in
the validation study are sensitive to differences in the positive and
negative experimental conditions compared to the control condition, and
that they are therefore sufficiently sensitive to reflect a linear rela-
tionship in the experimental situation, if it is there.

I'A second descrepancy between exPeriment and validation (pretest)
study can be noted. In the experiment, there were three conditions:
negative, control, and positive conditions. In the validation study,
we told students to avoid attention or to attract attention, strategies
we felt would correSpond with the negative and positive conditions. Thus
the dependent eXpression and impression indicators were chosen on the

basis of differences in two conditions, not three. The possible effect

82
of this is that by looking for differences only between two extreme con-
ditions, we may have included indicators that would not have Showed sig—
nificant differences across three conditions. Thus, weak indicators may
have been selected for the experiment, and these indicators caused us
to fail to reject the null hypothesis. But if this were the case, if
our dependent indicators only reflected differences at the extremes of
our conditions,then we would be even less likely to find a quadratic
effect due to eXperimental condition. Among exPression indicators,
however, a quadratic effect is precisely what is found. Thus, we are
reluctant to ascribe our failure to confirm the experimental hypothesis
to problems in the nature of our validation study.

A second argument suggests that the weak nature of the interaction
between student and supposed QSC experimenter may have contributed signi—
ficantly to the results. The interaction in the eXperiment, this argument
would point out, was not face—to-face. Instead, the students wrote our
self-descriptive essays that they thought would later be read by QSC
experimenters. Writing self-descriptive essays is not an ordinary event,
and the rules that govern such an activity are probably more formal,
making responses conform to social eXpectations. This may have limited
the variance in behavior across conditions. Nonetheless, regardless of
how extraordinary written essays are, and regardless of how formally rule-
governed they are, significant differences in behaviors were observed in
the validation study and in the negative and positive experimental condi-
tions as compared to the standard (i.e., control) condition. Although
face-to-face interaction may provide a different test of the hypothesis,
we believe that the written self-descriptions are sensitive enough to

reflect a linear relationship between experimental condition and

83
individuation, if it is there. Thus, we are also reluctant to ascribe
the failure to confirm the experimental hypothesis to the nature of
interaction in the experimental situation. Having rejected these cri-
tiques of our experimental methods, let us consider some possible problems
in our analytic techniques.

We can identify at least three possible problem areas in our analy—
sis that may have contributed to our failure to reject the experimental
null hypothesis. These are: l) the failure to examine interaction
effects among experimental and nonexperimental independent variables,

2) the use of untransformed data in the expression and impression indica-
tors of individuation, and 3) the use of multiple regression techniques
that may have given biased significance estimated because of the multi-
collinearity among dependent indicators. Let us consider each of these
problem areas in turn.

4.2.1 Failure to examine interaction effects. Our analysis

 

revealed no main effects for sex, Machiavellianism, or self-esteem on
dependent indicators of individuation. This could be due to problems

of multicollinearity (to be discussed below), or it may reflect an actual
absence of main effects. Nonetheless, it is reasonable to assume that
these nonexperimental independent variables may interact with experimental
conditions. (Perhaps because of’their'Hmachismo", males feel the need to
stand out in spite of the threateningness of the negative condition.

Such an interaction effect for sex is certainly consistent with Maslach's
[1974] findings.) Similar arguments could be made for interaction effects
of Machiavellianism or self-esteem with experimental condition. We
recommend that future research be conducted to investigate these poSsibi-

lities.

84

4.2.2 Use of untransformed data. In the data analysis, OXpression

 

and impression indicator data were untransformed in any way. (The expres-
sion indicators themselves, however, were constructed as rates.) That is,
.the values of the expression indicators represented actual counts of
parts of speech or rates of parts of Speech in each essay. The values of
impression indicators were the actual ratio-scaled judgments made by the
coders. For each set of indicators, but for different reasons, we can
argue that instead of untransformed data, logarithmically transformed
data should have been used.

Research on writing samples from a single individual shows that the
use of different parts of speech is not normally distributed.20 This
non-normal distribution violated the normality assumption required in
our correlational analyses. A logarithmic transformation of our eXpres-
sion data may more closely approximate a normal distribution.

Similarly, research shows that ratio-scaled judgments of psychophy-
sical phenomena are often scaled logarithmically. Perhaps the ratio-
scaled judgments of impression indicators are also based on an underlying
logarithmic scale. If so, then the assumption of intervality required
by our correlational analyses is violated. A logarithmic transformation
of the impression data may more closely approximate an interval $0318-

Nonetheless, we believe that the problems solved by such transforma-
tions may not be all that crucial to the test of the hypothesis. The
problem of non-normal distribution of eXpression indicators is based on
writing samples from a single individual. In the eXperiment, we are
dealing with a sample of 99 individuals. Similarly, the problem of non-
intervality among impression indicators is most important when one's

scale has a considerable range. In the experiment, this problem is

85
attenuated by the relatively short range used in scaling impression
indicators. For each impression indicator, the mean is near 10 (the
standard) and the standard deviation is less than 2.5 (see Table 5).
Thus, although additional analyses can be easily conducted using the
logarithmic transformations of expression and impression indicator data,
such a move does not in our opinion warrant high priority.

4.2.3 Multicollinearity. We believe that the single greatest prob-

 

lem in our analytical methods is the multicollinearity among dependent
indicators. Our hypothesis is rejected not solely on the basis of the
strength of the relationship uncovered, but also on the basis of the
significance of that relationship. In the regression analyses, the
expression and impression indicators of individuation are highly related.
to one another. We do not have eleven (or seven) independent indicators.
The result is a very conservative estimate of the significance of the
strength of relationship uncovered.

Ironically, the problem of multicollinearity was one we considered
and attempted to circumvent by standarizing the expression indicators
accoring to number of words. Unfortunately, these rate indicators
proved to be highly related to number of words and to one another. One
way of ameliorating this problem would be to standardize on number of
sentences rather than number of words. The resulting rate of indicators
will probably be more independent of one another.

Such a procedure, however, does not change the multicollinearity
problem among impression indicators. Thus, we are forced to look for
another solution. One likely possibility is to create impression and
expression indexes of individuation based on a new analysis of validation.

data. We used the validation study only to select the dependent indicators.

86
The validation study could be used more powerfully by entering the
selected validation indicators into a regression analysis to predict the
validation condition (i.e., attract attention/do not attract attention),
and then using the resulting beta weights to create individuation indices
for the experimental data. We heartily recommend that further data
analyses employing this procedure be conducted.

4.2.4 Summary of problems of analytical methods. We have considered

 

three different problems of analytical methods that may have contributed
to our failure to confirm the experimental hypothesis. Of the three,
problems of untransformed data seem least crucial. Failure to examine
interaction effects, and problems of multicollinearity, however, do seem
to be relevant.

Recognizing these possible shortcomings (and the resulting tentative-
ness of our results), let us now consider the findings of significant
relationships between expression indicators of individuation and quadratic

and overall effects of exPerimental conditions.

4.3 Quadratic Effect of Experimental Condition

 

Having considered some of the possible methodological reasons for
our failure to reject the exPerimental null hypothesis, let us now turn
our attention to an issue that may be considered a failure of our theory.
Results of the data analysis of expression indicators suggest that people
in both the positive and negative conditions were individuating to the
same extent, while we had predicted that people in the negative condition
would attempt to avoid attention. What plausible interpretations can.

we offer to account for this finding?

87

One possible explanation for this is that in the manipulation of
the standard condition, students were told that both the positive and
negative conditions are filled and that they would be assigned to the
standard condition as a matter of course. In essence, they are being
told that no matter what they do, they cannot control the outcome of
selection to a learning program. Possibility of control is left Open,
however, to students in both the positive and negative conditions. The
quadratic effect found in the data may reflect differences due to this
control/no control possibility. The question then is why the students
in the negative condition seem to engage in the same behaviors as those
in the positive condition.

The answer may be that our definition of the situation in the nega-
tive condition was not the same as the students'. We assumed that the
student would see the situation as being one where behavior ought to be
exercised to avoid a negative outcome, the negative learning program.

We reasoned that because most peOple are supposedly going to be selected
for the standard program, the appropriate move for the student would be

to avoid attracting attention. This may have been an erroneous assumption.
Perhaps the student thought that, as long as there was the possibility

of being selected for either the standard or the negative learning program,
he had better engage in behaviors that would earn the preferred outcome,
the standard learning program.

What we are suggesting is that we must not be misled by the labels

given to the learning programs. From the point of view of the student

I

‘3
in the negative condition, the standard learning program may have been

rewarding compared to the negative learning program. In fact, by examining

the results of the manipulation checks (Table 8), we can see that for

. 88
question (a), the difference in perceived reward between the positive
and standard condition is 3.06, while the difference in perceived reward
between standard and negative is 3.20. For question (b), the differences
are 3.83 and 3.17. For question (c), the differences are 3.16 and 3.09.
If we eXpect someone in the positive condition to individuate for a 3-
point increase in perceived reward, we should also eXpect the same from
students in the negative condition.

Thus, the failure to confirm the hypothesis may be related to this

problem. In the theoretical discussion of individuation, we assumed that
we were considering situations where a single outcome was to be distributed
among members of a perceptual field. In these single outcome situations
(depending on the perceived reward value of that outcome), we argued that
the individual's self-presentation would be directed by a single rule.
A positive outcome would be met by behavior that attracted attention; a
negative outcome would be met by behavior that avoided attention. Thus,
we posited the linearrelationship between perceived reward and level of
individuation. A

In the experiment, however, it is possible that we were not dealing
with a single outcome situation, but with a two outcome situation. That
is, students could have locked at the situation from either one of two
different perSpectives. (l) I want to be selected for the preferred out-
come, therefore I ought to attract attention. Or (2), I do not want to
be selected for the aversive outcome, therefore I ought to avoid attention.
Students in both the positive and negative conditions acted similarly in
regard to the verbal expressions used. Thus, the findings show that there
is a significant relationship between perceived reward structure and the
indicators of level of individuation. However, this relationship is more

complex than the simple linear one we had posited above.

89

4.4 Future Research

 

In this discussion of the implications of our research findings, we
have uncovered two areas that seem especially fruitful for future research.
First, we recommend that expression indicators be standardized according
to number of sentences, that indices of the expression and impression in-
dicators be created by using beta weights for each indicator obtained
from multiple regression analyses of validatiOn data, and that these
indices of expression and impression indicators of individuation be used
in retesting the experimental hypothesis and in testing for interaction
effects with the nonexperimental independent variables.

The second area emerges from our discussion of the quadratic effect
found in the data. We Suggested that the reason the students in both
the positive and negative conditions were similar in regard to their
expression indicators was that students in both conditions may have been
trying to attract attention to achieve the preferred outcome Open to them.
However, this eXplanation leaves Open the question of why there is not a.
similar quadratic effect among impression indicators. The relationship
between expression and impression indicators is a yet unexplicated. It
seems to us that intensive efforts at elaborating the ways in which cer-
tain impressions arise from the use of certain expressions would be very
theoretically rewarding. By conducting further research in these two
areas, perhaps we shall gain even greater insight into the individuation

process.

90
4.5 Summary

We began this thesis by considering the rule-governed nature of
communication behavior. We argued that such a rules approach was the
foundation for conceiving of self-presentations as ways of controlling
interaction. We then discussed the use of self-presentations in control-
ling the distributions of potential reward or punishment in situations.

We argued that individuation was a crucial dimension of such interpersonal
behavior. Individuation was formally conceptualized as the extent to
which a person, as a social object, is distinguishable from other social
objects in a perceptual field. This conceptualization indicated that a
crucial function of individuation was to manage the amount of attention
drawn to the individuation in interaction. We then considered the commu-
nication behaviors related to individuation, and the conditions under
which we expected individuation to vary systematically.

Based on the insights of Goffman (1959), we pretested and selected
two sets of communication indicators of individuation: expression indica-
tors (differences in language use and information provided), and impression
indicators (differences in self-presentational styles). We then argued
that these indicators of individuation should vary systematically with
the individual's perceptions of the reward structure in a situation. In
fact, because peOple would tend to avoid attention in situations of poten-
tial punishment and attract attention in situations of potential reward,
we argued that individuation would be a positive linear function of the
perceived amount of reward.

An experiment was designed to test this hypothesis. One hundred nine
students from introductory social psychOlogy classes were randomly assigned

to three conditions, a positive condition, a standard condition, and a

91
negative condition, where the potential reward in each condition was
experimentally manipulated. Students were asked to write self-descrip-
tive essays and fill out additional questions. The essays were coded
based on a category scheme devised through validation (pretest) studies.

Data analysis revealed that the sample appeared to be representative
of the college population, and that the effects of certain nonexperimental
independent variables (sex, Machiavellianism, and self-esteem) were
effectively controlled for by random assignment of subjects to condition.
Indicators of individuation were checked for their reliability and
validity. Some indicators were dropped from further analysis. Manipu-
lation checks showed that subjects' perception of the reward structure
in the experimental conditions had been successfully manipulated.
Although significant relationships between linear effect of experimental
condition and some of the individual impression indicators were found,
no evidence of significant linear relationships between experimental con—
ditions and the eet§_of expression or impression indicators were found.
We were thus forced to fail to reject the experimental null hypothesis.
Significant relationships were found, however, between the expression
indicators of individuation and the quadratic and overall effects of
experimental condition.

In discussing the results, we considered a number of possible explar
nations for our failure to reject the experimental null hypothesis. We
concluded that some failures in the analytical methods, especially with
problems of multicollinearity in both sets of dependent indicators of
individuation, were possible obstacles in testing the hypothesis. We
suggested improvements in the analytical methods that should provide for

a stronger test of the hypothesis.

92

We also considered the significance of the quadratic findings in
light of our original theoretical discussion. We concluded that perhaps
student perception of the experimental situation differed from our own
perception of it, and that perhaps students in both the negative and
positive conditions were similarly acting in an attempt to attract atten-~
tion so that they might earn the preferred outcome Open to them.

This study represents our initial attempt to understand more fully
the process of negotiating identities that is part of every interactional
situation. We believe that impression management requires that the
individual be concerned with the extent to which he is distinguishable
from others. Therefore, we also believe that understanding interpersonal
communication requires that we, as social scientists, concern ourselves

with actor's afforts at individuation.

APPENDICES

APPEND IX A

FORMS E AND F

NOT to be used

(For description of subject perceptions only.

by QSC experimentcn)

 

A-l

*** BEGIN READING THIS FORM FIRST ***

osc STUDY
SUBJECT EXPERIMENTAL ATTITUDES
(FORM E)

Funded by All-University Grant
Number 11-6893

For UseyAt:

Quasi-Sensory

Communication Laboratory
Department of Communication
Michigan State university
1974-1975

‘
‘5- c
rap}:-
c:i’L
.57.
I
"“”i"
‘T‘
“.7

*** BEGIN READING THIS FORM FIRST ***

E-l

Extrasensory perception or ESP is a phenomenon that has long been
associated with mysticism and the occult. Most social scientists, therefore,
have been inclined to dismiss ESP as outside the realm of appropriate study.
Some legitimate researchers, however, are still intrigued with the way some
people can apparently communicate using a transmission channel other than
speech, writing, or sign language. To distinguish this area of study from
the unscientific realm of ESP, researchers have coined the term "quasi-
sensory communication" or QSC.1

Recent advances in physio—psychology are Opening the QSC area to new
efforts of intensive study. Biofeedback training mechanisms (which are
often used in the study of transcendental meditation) have been used in sen-
sitizing people to parapsychological experience. Moreover, in certain situ—
ations, galvanic skin response (GSR) and electroencephalographic activity
(EEG) have been linked to people's ability to learn fbreign languages.

Researchers in the Department of Communication at Michigan State
university have begun adapting these techniques to teach people how to
communicate via QSC. Of course, the "language" of QSC is a rudimentary one
and is limited to communication about fundamental feelings and objects in
our experience. However, the successes to date are encouraging.

What we are now trying to do is figure out the best way to teach people
how to communicate via QSC. We have been conducting a study of different
Quasi-Sensory Communication Learning Programs at colleges and universities
throughout Michigan. Several hundred students have participated. Your in-
structor has agreed to our using the students in this class as part of our
subject pool.

In this study, we are investigating three different QSC learning pro-
grams. These are the Positive Environment, Standard Environment, and Nega-
tive Environment Learning Programs. Each requires the subject to attend
three twenty-minute sessions. In all three programs, during the first
session, the subject meets with an experimenter and learns to monitor his
or her mental activity through the biofeedback techniques described earlier.
It is during the second and third sessions that the three programs differ.

Positive Environment LearningProgram

 

It has been hypothesized that when a subject's galvanic skin response
and electroencephalographic activity show minimal disturbance, the subject
is more receptive to cognitive learning, such as QSC. Therefbre, in this
learning program, the subject learns in a pleasant environment. A specially
designed "warm room" is used. Blue carpeting, Blue curtains, and deeply-
padded fhrniture have been found to be very pleasing.

During the second session, the subject sits in the "warm room" while
the experimenter is in a separate room. The experimenter then tries to

 

1 H.N. McBain et al. "Quasi-Sensory Communication: An Investigation
Using Semantic Matching and Accentuated Affect." Journal of Personality and
Social Psychology, 1n (April, 1970), 281-291

 

 

B-2

communicate to the subject via QSC a series of feelings (like hunger, love,
sleepiness, etc.). Each time-the—subject correctly interprets the QSC
message, a pleasant tone will be sounded. The subject will also be given a
small monetary reward ($.25) for each correct interpretation. After the
subject can correctly identify the feelings, the experimenter tries to com-
municate a second series of common objects (food, chair, pencil, etc.). In
this fashion, the subject begins to learn the "vocabulary" of QSC. In the
final session, the subject is once again placed in the "warm room" while

the experimenter is in a separate room. The experimenter and the subject
review the "vocabulary" of QSC. Then the experimenter tries to relate the
feelings and objects together into "sentences" and tries to communicate

these sentences to the subject. In this session, the subject is again posi-
tively reinforced with pleasant tones, and small monetary rewards for correct
interpretations. The subject can expect to be rewarded about no times during
the second and third sessions. Subjects in this program have reported QSC
learning to be quite enjoyable. -

Standard Environment Learning Program

In this program during the second and third sessions, the subject is
placed in an ordinary room while the experimenter is in a separate room.
The experimenter then tries to communicate to the subject via QSC a series
of feelings, objects, and "sentences." At no time is the subject positively
or negatively reinforced for correct or incorrect interpretations of QSC
messages. This program allows the researchers to test the effectiveness of
teaching QSC in everyday environments.

Negative Environment Learning Program

 

It has also been hypothesized that a subject's ability to learn is
greater when his or her galvanic skin response and electroencephalographic
activity show mild disturbances. Therefore, in this QSC program, rather
than being placed in an ordinary room or a "warm room" for the second and
third sessions, the subject is placed in a more severe environment. White
walls, bright lights, and a solitary bench have been found to produce a
mild degree of anxiety.

As in the other two programs, the experimenter tries to communicate
feelings, objects, and "sentences." However, each time the subject
incorrectly interprets the QSC message, a harsh tone will be sounded. The
subject will also be given a small electric shock (like that when you cross
a wool carpet and touch a door handle). The subject can expect to be
shocked about no times during the second and third sessions. Subjects in
this program have reported QSC learning to be slightly discomforting.

 

A-4

FIGURE 1

A COMPARISON OF QSC LEARNING PROGRAMS

E-3

 

[earning Program

Positive Environment
Learning Program

Standard Environment
Learning Program

Negative Environment
Learning Program

 

 

Setting "Warm Room" - blue Ordinary Room White walls, bright
* carpeting, blue cur- ' lights, solitary

tains, deeply padded bench
furniture

Reinforcement For each correct No reinforcement For each incorrect res-

techniques response: pleasant techniques ponse: harsh tone,
tone, small monetary small electric shock
reward ($.25) (like that when you

 

 

 

cross a wool carpet and
touch a door handle)

 

 

Research to Date

 

Research on the effects of the Negative Environment on QSC learning
has been completed. Subjects are now needed for the Standard Environment
and the Positive Environment Learning Programs. iThe subject pool of:whiCh
you are a member has been designated forifhe most part to be in the Standard
Environment Learning Program. However, there is a need for some subjects
in the Positive Environment Learning Program. Some of you wiII-be selected
for this group.

 

 

Whether you will be selected for the Standard Environment Learning
Program or for the Positive Environment Program depends entirely on your
responses on the yellow form, Form P} :Yau 5111 be saked to write a brief

'introduction‘describing‘yourself and also given two standardized QSC exercises.

QSC experimenters will then look through the responses on all the ellow forms
filled out by all the subjects and will decide who will be in whic learning

program. To avoid schedule conflicts, we will schedule the sessions with
each of you individually. Male and female subjects will be matched with
experimenters of the same sex.

 

 

 

Please follow the directions carefully and answer the remaining
questions on this form as instructed. This will permit us to assess the
perceptions you have of the experiment before you participate. To avoid
biasing the study, the QSC experimenters will not be allowed to see your
answers to questions on this form. The QSC experimenters will only see
Form F (yellow).

If you have any questions, please re-read this section of the form.
Then turn the page.

(This version of page E-4 was given to students in the positive condition.)

A-6

Research to Date

 

Research on the effects of the Positive and Negative Environments on
QSC learning has been completed. Subjects are now needed for the Standard
Environment Learning_Prgg£am. The subject pool of which you are a member
has been designated to be in the Standard Environment Learning Program.
All_of you have been selected for this group.

 

 

You will be asked to write a brief "introduction" describing yourself and
also given two standardized QSC exercises. QSC experimenters will then look
through the responses on all the yellow forms filled out by all the subjects.
To avoid schedule conflicts, we will schedule the sessions with each of you
individually. Male and female subjects willlbe matched with experimenters
of the same sex. _

Please follow the directions carefully and answer the remaining questions
on this form as instructed. This will permit us to assess the perceptions
you have of the experiment before you participate. To avoid biasing the study,
the QSC experimenters will not be allowed to see your answers to questions
on this form. The QSC experimenters will only see Form F (yellow).

If you have any questions, please re-read this section of the form.
Then turn the page.

(This version of page E-4 was given to students in the control condition.)

Research to Date

 

Research on the effects of the Positive Environment on QSC learning
has been completed. Subjects are now needed for the Standard Environment
and the Negative Environment Learning Programs. The subject pool dfiwhich
you are a member has’been designated for the most part to be in the Stan-
dard Environment Learning Program. However, there is a need for some sub-
jects in the Negative Environment Learning Program. Some of you wiII be
selected for this group.

 

Whether you will be selected for the Standard Environment Learning
Program or for the NegativeEnvironment:Rrogram depends entirely on your
responses on the yell ow form, Form F. You will be asked to write a brief
'introduction” describing yourself and also given two standardized QSC exercises.
QSC experimenters will then look through the responses on all the yellow forms.
filled out by all the subjects and will decide who will be in which learning
program. To avoid schedule conflicts, we will schedule the sessions with
each of you individually. Hale and female subjects will be matched with
experimenters of the same sex.

 

 

Please fellow the directions carefully and answer the remaining ques-
tions on this form as instructed. This will permit us to assess the per-
ceptions you have of the experiment before you participate. To avoid biasing
the study, the QSC experimenters will not be allowed to see your answers
to questions on this form. The QSC expgrimenters will only see Form P (yellow).

If you have any questions, please re-read this section of the form.
Then turn the page.

(This version of page E-4 was given to students in the negative condition.)

PLEASE ANSWER THE FOLLOWING QUESTIONS:

1. What is your name?

 

 

2. What is your student number?| I I i I i i i i i

 

3. What is your sex? (circle one) male female

u. What is today's date?

 

5. List two twenty-minute periods during the next two weeks when you would
be available for the first session of the QSC study. If you aren't
sure of your schedule at the moment, please write in your phone number
so that we may contact you.

 

 

 

 

 

 

Date Time (between 8:00 AM and 10:00 PM)
. ""' Circle one
a- ' AM PM
I) . , AM PM
Phone Number:

 

6. What is a Positive Environment Learning Program? (answer briefly)

 

 

 

 

7. What is a Standard Environment Learning Program? (answer briefly)

 

 

 

 

 

8. What is a Negative Environment Learning Program? (answer briefly)

 

 

 

PLEASE CONTINUE .

A-
9 17.-7

ekanaaaesaeaeaaaee

20. The subject pool of which you are a member has been designated
for the most part to be in the Standard Environment Learning
Program. Some of you, however, will be selected for the Positive
Environment Learning Program. What percentage of people in the
subject pool do you feel are likely to be selected fOr the
Positive Environment Learning Program? . . . . . . . . . . . .

 

0‘9

 

PLEASE CONTINUE.

(This version of page E-7 was given to students in the positive condition.)

A-lO
T‘.-7

'Ju'u’: *kkkékkkkéf:‘k***

20. The subject pool of which you are a member has been designated to
be in the Standard Environment Learning Program. Some subjects,
however, have been selected for the Positive and Negative Environment
Learning Programs. What percentage of all people in all subject,
pools do you feel are likely to be selected fer the Standard Environ-
ment Learning Program? . . . . . . . . . . . . . . . . . . . . . . . %

 

PLEASE CONTINUE .

(This version of page E-7 was given to students in the control condition.)

A-ll
E-7

eaeeaaenaaaeaeeee

20. The subject pool of which you are a member has been designated for
the most part to be in the Standard Environment Learning Program.
Some of you, however, will be selected fer the Negative Environment
Learning Program. What percentage of people in the subject pool do
you feel are likely to be selected for the Negative Environment
Learning Program? . . . . . . . . . . . . . . . . . . . . . . .

 

PLEASE CONTINUE .

Ghis version of page E-7 was given to students in the negative condition.)

********** **
*************

nae

*ﬁﬁ

*************

*************
***
***

*************
*************

STOP!!!

eases
eases

********
********
**A
*k*
ﬁnk
***
are
***
***
are

DO NOT TURN THE PAGE.

STOP!

A—12

*************
************ﬁ

*k* are
*k* are
*** *kﬂ
nee ***
*** see
can ***
*************
*************

0‘
CI.

*************

********** as
net nae
*k* *k*

*************
*************
*kk
***
***
***

PLEASE OPEN FORM E AND BEGIN READING.

A-13

Now that you have completed Form P, we are interested in your reactions
to the form and the selection procedures used in the study.

Please read the following statements. Then indicate how much you agree
or disagree with each statement by rating the statement on a scale from
0 to 10.

If you absolutely disagree with the statement, mark down a 0.
If you absolutely agree with the statement, mark down a 10.

 

 

Use any number from 0 to 10 that best describes your feelings.

Think of the "average

21.

22.

23..

24.

25.

26.

27.

28.

29.

30.

I feel that R.S. is similar to the average student . . . . .

 

I feel that J.L. is similar to the average student . . . .

 

I feel that I_am similar to the average student . . . . .
I feel that R.S. is similar to J.L. . . . . . . . . . . .
I feel that I_am similar to R.S. . . . . . . . . . . . . .

I feel that I_am similar to J.L. . . . . . . . . . . . . .

I predict that the average student will be chosen for the
Standard Environment Learning Program. . . . . . . . . . .

 

I predict that R.S. will be chosen for the Standard
Environment Learn1ng Program. . . . . . . . . . . . . . .

I predict that J.L. will be chosen for the Standard
EnViI‘Oﬂment lea-min g Program 0 e e e e ‘ e e e e e e e ' e e e

I predict that I_will be chosen for the Standard
Environment Learning Program. . . . . . . . . . . . . . .

PLEASE CONTINUE.

student." Now rate the following statements.

A-l4
E-lO

How do you feel that people will be selected fer one learning program or
the other? Below is a list of possible reasons for being selected for one
program or another. Rate how important you think each reason is on a scale
from 0 to 10.

If you think the reason is extremely unimportant, mark down a 0.
If you think the reason is extremely important, mark down a 10.

 

 

Use any number from 0 to 10 that best describes your feelings.

In selecting you for a given learning program, how important would you
estimate the following things to be:

31. The school one is from . . . . . . . . . . . . . . . . . . . . ______
32. One's "introduction" . . . . . . . . . . ______
33. One's sex . . . . . . . . . . . . . . . . . . . . . . . ._____
3n. One's handwriting . . . . . . . .'. . . . Q . . . . . . . . . . ______
3S. One's responses on the QSC Battery . . . . . . . ... . . . . .

36. One's reSponses on the Personal Geometry Selection Profile .
37. The impression the experimenters get from reading Form P . .

38. Chance (i.e., the choice will be made randomly) . . . . . . . .

************ﬂh*k**k***

Please read the following statements. Then indicate how much you agree or
disagree with each statement by rating the statement on a scale from 0 to 10.

If you absolutely disagree with the statement, mark down a 0.
If you EBsolutely agree with the statement, mark down a 10.

 

 

Use any number from 0 to 10 that best describes your feelings.
39. I think this QSC research is valuable and useful . . . . . .
no. I would describe myself as being highly manipulative . . . . . .

al. I have a pretty good opinion of myself . . . . . . . . . . . .

w************#********

PLEASE CONTINUE.

A-lS
E-ll

W2. 0f the two learning programs for which we need subjects, which

learning program would you like to be in?

 

Please answer the following question on a scale from 0 to 10 where 0 means

"none of it is due to my own efforts" and 10 means "all of it is due to my
own efforts ." ’

#3. Sometimes whether or not we are chosen for things depends on our own

efforts. Sometimes our efforts don't make a difference. If you are

chosen for the learning program you listed in Question “1, how much
of it do you feel will be due to your own efforts?

PLEASE CONTINUE .

A-16

E-12
naeeensnnnnneaeaee

Please answer the following question on a scale from O to 10 where 0 means
"I did not try at all to influence the experimenters" and 10 means "I tried
very hard to influence the experimenters."

us. How much did you try to write your "introduction" and answer
the QSC Battery and Personal Geometry Selection Profile as as to
influence the experimenters?

 

#5. How did you try to influence the experimenters?

 

 

*Aﬂﬁkﬂ*k*******s‘d¢*

Please answer the fellowing question on a scale from 0 to 10 where 0 means
"I was completely unsuccessful" and 10 means "I was completely successful."

us. Earlier we asked you to imagine that you were with R.S. and J.L. while
completing the QSC exercises. How successful do you feel you were in
imagining that situation? . . . . . . . . . . . . . . . . . . . . .

 

47. Do you have any other reactions to this study?

 

 

 

 

 

 

STOP!!! THANK YOU FOR YOUR COOPERATION.

For use of QSC experimenters only.)

(Learning Program Selection.

A-17
*** DO NOT BEGIN READING THIS FORM UNTIL INSTRUCTED TO DO SO ***

QSC STUDY

SUBJECT COGNITIVE CONFIGURATION

 

(FORM F)

Funded by All-University Grant
Number ll-6893

For Use At:

 

Quasi-Sensory
Communication Laboratory

Department of Communication
Michigan State university
197u-1975

 

 

 

*** DO NOT BEGIN READING THIS FORM UNTIL INSTRUCTED TO DO SO ***

A-18

E-l

You‘will be selected for a given learning program for this QSC study

based solely upon the responses youfgive to thegquestions in this form.

 

You.will be asked to write a brief "introduction" describing yourself

and to take two standardized QSC exercises. The introduction and the QSC
exercises have been given to several hundred college students in Michigan.

I.
2.
3.

u.

Please fill out the questions on this page, and then continue.

‘What is your’name?

 

What is your student number?

What is your sex? (circle one) male female

What is today's date?

PLEASE CONTINUE.

A-19
P-2

You are about to respond to three exercises that have been found useful
to QSC researchers. When you answer these exercises, please try to imagine
that you are actually with other people of the ease sex who you have just
ht.

. First, they will "introduce" themselves to you. Then you will be
asked to write down what you would tell them about yourself. Imagine that
you are speaking to than when you "introduce" yourself.

Now, we would like you to "meet" these people, R.S. and J.L. (These
"introductions," from actual introductions by subjects in earlier 060 re-
search, are not necessarily good nor bad. They are presented to give you
a: idea of what others have written. R.S. and J.L. are awaiting selection
to a QSC learning program.)

meanness
«museum

"I am a sophomore at Northern Michigan. My home tom: is Nanistique, Wichi-
gm. Right now I'm an English major although I have a strong interest in
journalism and news reporting. Last sumer I worked part time for a small
weekly and really enjoyed my experience. After I graduate, I would like
to write for a newspaper, possibly doing movie and theater reviews."

(by R.S.)

"Ny .name is J.L. I'm an eighteen-year-old student at Central Michigan
university. I've lived all my life here in Mt. Pleasant. I really enjoy
travelling and sightseeing. My main academic interest is in American
history -- particularly the colonial and revolutionary period. I hope
that I will eventually be able to find a job teaching somewhere after I'm
through school."

“AARMWMM
MAM““N

We ask you now to turn the page and "introduce" yourself. Please
write legibly.

A-20

"INTRODUCTIOW'
(Please Write legibly)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A-21
r-u

osc COGNITIVE BATTERY

This is a standard form in quasi-sensory communication, or QSC. We
are trying to find out how people put symbols and "words" together. There

are no right or wrong answers. The form compares the way you put sym53I§-'
an wo s toget er wit ~t e way other people have put them together.

an the following pages are lists of "nonsense words." The words have
been arranged in order by a large number of subjects on the basis of how
similar one word is to another. The closer any twO'words are on the list,
the more alike those words have been"?3und to be, Likewise, the IEFEEEF"
away any two words are, the less alike they have been fbund to be. Thus,
a.word that is sixth on the list is more similar to the seventh word than
to the thirteenth. '

 

At the top of each list is a symbol. What we'd like you to do is look
at the symbol, then look down the list of‘words. Try to associate a word
on the list with a symbol. Circle that word. In completing this form,
imagine that you, R.S., and 3.L. will answer each list aloud in turn. If
either R.S. or J.L. has answered before you, their responses are marked
on the form. .R.S.'s responses are indicated by a "I," and J.L.‘s responses
are indicated by an "x."

 

On some lists, you.will see both of their choices. Sometimes, however,
you will see the choice made by only one of them. And sometimes, you‘won't
see either of their choices.

Remember, there are no right or wrong answers to this form. You may
choose any word on the list, even one that has already been chosen. After
you have chosen the word, circle it, and then go on to the next list.

PLEASE READ THE DIRECTIONS ON THIS PAGE ONCE MORE, THEN CONTINUE.

PIN
ZAR
SAZ
SYG

' x HIP
FID

NYK

KUL
VOS

 

NAP
NAM

DIV
DAG
GAR
JHAR
LOR .
BOL
NOG
PON
PUC
PUK
KUM
NEY

PLEASE CONTINUE.

 

 

A-ZZ

GAW
QAK
CAG
KOS

PUZ
PID
RYN
FYL
Jrvx
ZYP
YBZ
wsv

x sov

 

 

1.1“
m

Jxon
JOF
m
sax
cux
our
oox
QUK

 

DEB
DEL
JEL
HEY
HUG
HUM
YUM
TUX
TAX
YAK
WAS
BAH
GAD
NOX
POD

PEN
use
YUH
MUY
nor
M06
x 602
POV
JOB
Jox
ons
mm

WYG

HYG
QIC

 

 

[7

BEN
QIK
QIN
KYX
JIS

JVID
HYL
CYR
PYR
ROH

won
Pow
SAI-I
LUP

PLEASE CONTINUE.

 

I \
I n
x /'

QID
MIP
PIP
FIH
FEX
GEB
GON
GUD
GUV

PUQ
NEQ
BEN
VAZ
VOK

 

A-23

X

JYC
KIJ
ZIY
ZUV
ZUF

NUJ
XWUQ
XUB
XUT

QEF
QIY

 

NUJ
RAJ
(arr
RUK
m
JOR
JAH
BBH
asz
SEQ
PEQ
PIV
x13

YOC '

 

TEQ
QER
MEP
BUP
ZOP
HOQ
DYQ
NIH

A-24
Po?

PERSONAL GEOMETRY SELECTION PROFILE

 

Now think about R.S. In the box below, there are nine circles. Put
an "R.S." in one circIe to represent "R.S." and an "I" in a different circle
to represent yourself.

0 O o 0 O O 000]

Now think about J.L. In the box below, there are nine circles. Put
a "J.L." in one circle to represent "J.L." and put an "I" in a different
circle to represent yourself.

[00 O O OO o 0 0|

Finally, think about the "average student." In the box below, there
are nine circles. Put an "A.S." in the circle to represent the "average
student" and put an "I" in a different circle to represent yourself.

LO O O 00 GO 0 0|

 

 

 

 

 

 

 

* PLEASE CLOSE THIS FORM AND OPEN FORM E To PAGE E- 9, '

APPENDIX B

INTERACTION ANALYSIS PROFILE

B—l

INTERACTION ANALYSIS PROFILE

INTER-UNIVERSITY RESEARCH

Part I

Please read the fellowing questions carerlly. Then answer each

to the best of your ability.

1.

2.

7.

Name:

 

Student Identification Number:

 

Sex (please check): Male

 

Female
Year of Birth / /
Year in School (please check): Freshman
Sophomore
Junior
Senior
Other

 

In your family, you are (please check): First-born
Later-born
Only child

I live (please check): On a farm

In the open country but not on a farm
In a village under 2,500

In a town of 2,500-10,000

In a city of 10,000-100,000

In a city over 100,000

B-Z

ID#
PART II

Listed below are a number of statements. Each represents a commonly
held opinion and there are no right or wrong answers. You will prob-
ably disagree with some items and agree with others. We are interes—
ted in the extent to which you agree or disagree with such matters of.
opinion.

Read each statement carefully. Then indicate the extent to which you
agree or disagree by circling the number below each statement.
The numbers and their meaning are indicated below:

If you agree strongly, circle +3
If you agree somewhat, circle +2
If you agree slightly, circle +1

 

 

 

If you disagree slightly, circle -1
If you disagree somewhat, circle -2
If you disagree strongly, circle -3

 

 

 

First impressions are usually best in such matters. Read each state-
ment, decide if you agree or disagree and the strength of your opinion,
and then circle the appropriate number below the statement. Give your
opinion gn_every statement.

 

If you find that the numbers to be used in answering do not adequately
indicate your own opinion, use the one which is closest to the way
you feel.

8. Never tell anyone the real reason you did something
unless it is useful to do so.

+3 +2 +1 -1 -2 -3
9. The best way to handle people is to tell them what
they want to hear.
+3 +2 +1 -1 -2 -3

10. One should take action only when sure it is morally right.
+3 +2 +1 -1 —2 -3

11. Most people are basically good and kind.
+3 +2 +1 -1 -2 -3
12. It is safest to assume that all people have a vicious streak
and it will come out when they are given a chance.
+3 +2 +1 -1 -2 -3

13.

1M.

15.

16.

17.

18.

19.

20.

21.

22.

23.

2M.

25.

B-3

II-2

Honesty is the best policy in all cases.
+3 +2 +1 —1 -2 -3

There is no excuse for lying to someone else.

+3 +2 +1 -1 -2 «3
Generally speaking, men won't work hard unless they're fbrced
to do so.

+3 +2 +1 -1 -2 -3
All in all, it is better to be humble and honest than to be
important and dishonest.

+3 +2 +1 -1 -2 —3
When you ask someone to do something for you, it is best to give

the real reasons for wanting it rather than giving reasons which
carry more weight.

+3 +2' +1 -1 -2 -3

Most people who get ahead in the world lead clean, moral lives.
+3 +2 +1 -1 -2 -3

Anyone who completely trusts anyone else is asking fer trouble.
+3 +2 +1 -1 -2 -3
The biggestdifference between most criminals and other people
is that the criminals are stupid enough to get caught.
+3 +2 +1 -1 -2 -3

Most men are brave.

+3 +2 +1 -1 -2 -3

It is wise to flatter important people.
+3 +2 +1 -1 -2 -3

It is possible to be good in all respects.

+3 +2 +1 -1 -2 -3

Barnum was wrong when he said that there's a sucker born every
minute.

+3 +2 +1 -1 -2 -3

It is hard to get ahead without cutting corners here and there.
+3 +2 +1 -1 -2 -3

26.

27.

B-4

II-3

People suffering from incurable diseases should have the choice
of being put painlessly to death.

+3 +2 +1 -1 ' -2 —3
Most men forget more easily the death of their father than the
loss of their property.

+3 +2 +1 -1 -2 -3

B-S

ID#
PART III

Please read the fbllowing pairs of statements. Then circle the letter
beside the statement that you believe is most true. Give your opinion
for every pair of statements.

28. a. Children get into trouble because their parents punish them
too much.
b. The trouble with most children nowadays is that their parents
are too easy with them.

29. a. Many of the unhappy things in people's lives are partly due
to bad luck.
b. People's mistrtunes result from the mistakes they make.

30. a. One of the major reasons why we have wars is because people
don't take enough interest in politics.
b. There will always be wars, no matter how hard people try to
prevent them.

31. a. In the long run people get the respect they deserve in this
world.
b. Unfbrtunately, an individual's worth often passes unrecognized
no matter how hard he tries.

32. a. .The idea that teachers are unfair to students is nonsense.
b. Most students don't realize the extent to which their grades
are influenced by accidental happenings.

33. a. Without the right breaks one cannot be an effective leader.
b. Capable people who fail to become leaders have not taken
advantage of their opportunities.

an. a. No matter how hard you try some people just don't like you.
b. People who can't get others to like them don't understand
how to get along with others.

35. a. Heredity plays the major role in determining one's personality.
b. It is one's experiences in life which determine what they're
like.
36. a. I have often fbund that what is going to happen will happen.
b. Trusting to fate has never turned out as well fbr me as
making a decision to take a definite course of action.

37. a. In the case of the well prepared student there is rarely if
ever such a thing as an unfair test.

b. Many times exam questions tend to be so unrelated to course
work that studying is really useless.

38.

39.

‘ no.

#1.

W2.

#3.

an.

“5.
#6.
H7.

#8.

#9.

B-6

III-2

Becoming a success is a matter of hard work; luck has little

. or nothing to do with it.

Getting a good job depends mainly on being in the right place
at the right time.

The average citizen can have an influence in government
decisions.

This world is run by the few people in power, and there is
not much the little guy can do about it.

When I make plans, I am almost certain that I can make them
work.

It is not always wise to plan too far ahead because many
things turn out to be a matter of good or bad fortune anyhow.

There are certain people who are just no good.
There is some good in everybody.

In my case getting what I want has little or nothing to do
with luck.

Many times we might just as well decide what to do by flipping-
a coin.

Who gets to be the boss often depends on who was lucky enough
to be in the right place first.

Getting people to do the right thing depends upon ability;
luck has little or nothing to do with it.

As far as world affairs are concerned, most of us are the
victims of forces we can neither understand, nor control.

By taking an active part in political and social affairs the
people can control world events.

Most people don't realize the extent to which their lives are
controlled by accidental happenings.

’There really is no such thing as "luck."

One should always be willing to admit mistakes.
It is usually best to cover up one's mistakes.

It is hard to know whether or not a person really likes you.
How many friends you have depends on how nice a person you are.

In the long run the bad things that happen to us are balanced
by the good ones.

host misfbrtunes are the result of lack of ability, ignorance,
laziness, or all three.

With enough effort we can wipe out political corruption.
It is difficult fer people to have much control over the things
politicians do in office.

50.

$1.

52.

53.

5Q.

55.

56.

Inn-'37

Sometimes I can't understand how teachers arrive at the grades
they give.

There is a direct connection between how hard I study and the
grades I get.

A good leader expects people to decide for themselves what
they should do.
A good leader makes it clear to everybody what their jobs are.

Many times I feel that I have little influence over the things
that happen to me.

It is impossible fbr me to believe that chance or luck plays
an important role in my life.

People are lonely because they don't try to be friendly.
There's not much use in trying too hard to please people; if
they like you, they like you.

There is too much emphasis on athletics in high school.
Team sports are an excellent way to build character.

What happens to me is my own doing.
Sometimes I feel that I don't have enough control over the
direction my life is taking. ‘

Most of the time I can't understand why politicians behave
the way they do.

In the long run the people are responsible fer bad government
on a national as well as on a local level.

B-8

‘ ID#

 

PART IV

Read each of the following statements. Then circle the number in the
column at the right according to how much you agree or disagree with
it. Please answer every statement. The responses are as fellows:

l. I disagree very much 5. I agree a little
2. I disagree on the whole 6. I agree on the whole
3. I disagree a little 7. I agree very much

u. I neither agree nor disagree
DISAGRBE AGREE
VERY MUCH VERY MUCH

57. I feel that I'm a person of worth, 1 2 3 N S 6 7
at least on an equal plane with .

others.

58. I feel that I have a number of l 2 3 u 5 6 '7
geod qualities.

59. All in all, I am inclined to feel ‘1 2 3 u 5 6 7
that I am a failure. ‘

60. I am able to do things as well as l 2 3 u 5 6 7
most other people.

61. I feel I do not have much to be 1 2 3 u 5 6 7
proud of.

62. I take a positive attitude 1 2 3 u 5 6 7
toward myself.

63. On the whole I am satisfied with 1 2 3 u s 6 7
myself.

OH. I wish I could have more respect far 1 2 3 u 5 6 7
myself.

65. I certainly feel useless at times. 1 2 3 u S 6 7

66. At times I think I am no good at l 2 3 u 5 6 7
all.

PART V

Read each of the fellowing statements.

67.

68.

69.

70.

71.

72.

73.

7Q.

75.
76.

77.

78.

79.

80.

81.

82.

B-9

ID#

 

Then indicate in the column at
the right whether you feel the statement is true or false.
answer every statement.

I can remember "playing sick" to get out of

something. . . . . . .

On a few occasions, I have given up doing something
because I thought too little of my ability . . . . ,

I sometimes feel resentful when I don't get my way .

I have never deliberately said something that hurt
someone's feelings . . . .

There have been times when I felt like rebelling
against people in authority even though I knew they
were right 0 O O O O O O O O O O Q 0 O O O O O . O . 0

If I could get into a movie without paying and be
sure I was not seen, I probably would do it. . , . .

My table manners at home are as good as when I
eat out in a restaurant. . . . . . . . . . . .

I never hesitate to go out of my way to help
someone in trouble . . . . . . . . . . . . . .

I am always careful about my manner of dress ,

I have never intensely disliked anyone . . . .

It is sometimes hard fer me to go on with my work

if I am not encouraged. . . . . .

No matter who I'm talking to, I'm always a good
listener . . . . . . . .

I am sometimes irritated by people who ask favors!t,r”

of me . . . . . .

I have never felt that I was punished withhht cause..

.
. .

..'O O O

,’

I have never been irked when peeple expressed ideas

Avery different from my own . . .

There have'been occasions when I felt like smashing

things.......

 

Please
CIRCLE ONE
.TRUE FALSE
.TRUE FALSE
. TRUE FALSE
.TRUE FALSE
TRUE -FALSE
.TRUE FALSE
. TRUE FALSE
.TRUE FALSE
.TRUE FALSE
.TRUE FALSE
TRUE FALSE
.TRUE . "FALSE
TRUE FALSE
.TRUE FALSE
.TRUE FALSE
.TRUE FALSE

83.

85.

86.

87.

88.

89.

90.

91.

92.
93.
9a.

95.

96.

8—10

V—2

I am always courteous, even to people who are
disagreeable . . . . . . .

I have almost never felt the urge to tell someone off..TRUE

When I don't know something I don't at all mind
aMitt ing it 0 O O I I O O O O O ' O O O O TRUE

Defers voting I thoroughly investigate the qualifi-
cations of all the candidates . . . . . . . . . . . . .TRUE

I sometimes try to get even, rather than fbrgive
”d forge t . . . . C C . O C C . . C . C . . . . . . . TRW

There have been times when I was quite jealous of the
good fertune of others . . . . . . . . . . . . . . . . TRUE

I sometimes think when people have a misfbrtune
they only got what they deserved . . . . . . . . . . . TRUE

At times I have really insisted on having things my
m way 0 O O O O I O O O O O O O O O O O O O O O O O O TRW

I would never think of letting someone else be
punished for my wrongdoings . . . . . . . . . . . . . .TRUE

I never resent being asked to return a favor . . . . . TRUE
I always try to practice what I preach . . . . . . . . TRUE
I'm always willing to admit it when I make a mistake...TRUE

I don't find it particularly difficult to get along
with loud-mouthed, obnoxious people . . . . . . . . . .TRUE

There have been occasions when I took advantage of
smme O O O O O O O I O O O O O O O O O O O .0 O O O 0 TM

0 O O O O . TRUE.

FALSE

FALSE
FALSE
FALSE
FALSE
FALSE
FALSE
FALSE

FALSE
FALSE
FALSE

FALSE

FALSE

’FALSE

APPENDIX C

VALIDATION (PRETEST) RESULTS

C-l

 

 

oe. ea.m on.” oH.n «AH.HH no.~ mma.o asamaosoo moo
No. oo.m on.” om.o aom.~a Ho.~ moa.o amaamsapcm oHo
mo. om.o on.o an.“ ooo.HH ao.~ omm.oa mmoooouusouoau moo
Hoo. am.os om.a ~o.~ «No.ms oo. ona.oo manganese: oHo.
no. oo.m om.o oo.H mao.oﬂ om. ooa.o Shoshana moo
Hoo. Ho.oe om.“ ma.~ ooH.NH oo.H Hoo.o mmooo>ashmoo Noo
Hoo. oo.o~ om.H mm.~ mom.ma oo. omo.oH amoeooowomon HHo
mwoueowpcu sowmmowman

moo. oo.o on.H oo. moo. mo. one. coasaaeomea Hecauoooaoom mo one: Homo
oo. ao.~ on.” Ho. omo. No. omo. successes» anemones mo opus osmo
No. -.o om.a oo.H oom.o on. omm. mooamooaao ozone: mo Looeoz omo
Loo. oo.oH on.” no. omo. oo. moo. Accumooaao osoaoo-aoo mo some omo
so. Ho.o om.o oo. moo. oo. one. mvo> once» anemone no case amo
oo. on.m on.a Ho. ooo. Ho. Noo. moue> o>ammao no one“ omo
oH. oo.~ om.a oo. has. so. an“. mowo> u>mpoa no case mmo
so. oo.o on.s mo. Hem. mo. Hoo. mo>aoa=amaa one .mooasou .mosoo no opus «mo
mo. oo.o on.“ no. use. No. mm”. moaooono Amoomoam cameos pause no case mmo
moo. oo.o on.” no. moo. no. was. . moooouoom mo opus «mo
soo. on.oo on.H oo.am oom.oo oe.nH Hao.mo mono: mo Aooaoz Hmo
meopoOMvem coﬁmmowmxm

oocaoemaomhm _ a mo .o.m coo: .o.m coo: ooowass>
N grown a macho

 

 

meoommm Aemmemmao one<oHo<>

C—2

 

 

mo.Aq no. eon.” mo.o oo~.e oo.» mao.o NH sous
moo.“ aa.em ooH.H no.~ ”Ho.a oo.~ one.» so sous
moo.“ oo.mo ooH.H oo.~ mmo.o no.~ aum.~ ow sous
mo.A oo.s ooo.e mo.m meo.a om.n ooo.o o amps
moo.w oH.o~ ooH.H oo.~ oo~.m mo.~ new.» o sou“
moo.m mo.so ooH.H so.“ nmo.o No.m ama.~ 5 amps
mo.4 on.» ooH.H No.» hoo.o om.m ooo.o 3 sons
moo.m oo.o~ ooH.H o~.~ moo.m oo.~ ooo.» m sous
moo.w mm.om ooo.ﬁ on.» asa.o No.o oom.o o acne
mo.« no. eon.” oo.m ~mo.a -.m mos.a n seam
moo.m. mo.Hm coo." No.o oao.m oH.~ ooo.~ N acne
moo.m oo.ao ooo.H oo.a moo.m Hm.~ mmo.~ H aosH
33H inseam 80
oooaoaeaomnm .m .mm .o.m one: .o.m as»: oHouAsa>
N msowu a moose

 

n.saooo meoommm ﬂemmemaao one<oHo<>

APPENDIX D

CODING FORM

1. Record Respondent ID NUmber.
2. Condition (Leave Blank)
3. Record Coder Set (1 - Liz, Steve, Lisa, Chip
2 - Larry, Jay, Stathi, Sue)
4. Count the number of sentences.
5. Count the number of words.
6. Count the number of first person singular pronouns.
7. Count the number of nouns, gerunds, and infinitives.
8. Count the number of active verbs.
9. Count the number of passive verbs.
10. Count the number of verbs in the present tense.
11. How many of the following dimensions are used in the essay?
Name Age __ School _ Home town—
Academic interest _____ Hobbies
Goal after graduation
12. How many additional dimensions are used in the essay?
***+******++
13. Count the number of statements that put the person into a
category.
14. Count the number of statements that refer to attributes of
the person.
15. Estimate the "psychological" content of the essay. (0 = no
psychological content, 9 = high in psydhological content.)
16. Estimate the "sociological" content of the essay. (0 = no
sociological content, 9 = high in psychological content.)
17. Estimate the size of the handwriting on a scale from 0 to 9.
(0 = very small, 9 = very large)
18. Is the essay printed or written? (0=printed, 1=written,

D—l

CODING FORM - FORM F ESSAYS

 

 

2 = both.)

D-2
You will be asked to rate the essay along several dimensions. Read the
essay by R.S. (page F-2). Assume that someone rating R.S.‘s essay would
rate each dimension as 10. If you feel that the essay you are reading
is greater in respect to the dimension than R.S.‘s, give it some number
greater than 10. If you feel it is less, then give it some number less
than 10.
Remember, for each question, R.S.‘s rating would be 10.
19. How different is this person from the typical student?
10. How positive is this self-presentation?

21. How much does this selfapresentation attempt to flatter
the reader?

22. How unique is this self-presentation?
**********
23. How negative is this self-presentation?

24. How much stress does there appear to be in this
-self- -presentation?

25. How embarrassed is this self-presentation?

26. How humorous is this self-presentation?

27.. How calculated is this self—presentation?

**********

28. How defensive is this self-presentation?

29. How enthusiastic is this selfopresentation?

30. How intense was the language in this self-presentation?
31. lHow modest was this selfupresentation?

32. How unsure was this self-presentation?

**********

33. Card Number (Leave Blank)

APPENDIX E

DEFINITIONS OF SELF-PRESENTATIONAL TACTICS

Differentness

Positiveness

Flattery

Uniqueness

Negativeness

Stress

Embarrassment

Humor

Calculatedness

Defensiveness

Enthusiasm

Intensity

Modesty

Unsurely

E-l

DEFINITIONS OF SELFvPRESENTATIONAL TACTICS

extent to which student's life history is unusual
relative to R.S.

extent to which student discloses attributes that
would be positively valued by society.

extent to which student's communications are
expressly directed toward a positive response on
the part of the reader.

extent to which the student's style of presentation
is unusual.

'extent to which student discloses attributes that

would be negatively valued by society.

extent to which the overall tone of the essay
revealed an uncomfortableness or feeling of being
in a bind in writing the essay.

extent to which the student made explicit or near
explicit references to situations he/she regarded
as "bad" or made apologies for past, present, or
future situations.

extent to which student intentionally tries to be
funny.

extent to which essay appears to be written in a
manner as to effect a specific response.

extent to which student withholds personal informa-
tion or attempts to justify statements.

extent to which student eXpresses positive affect.

extent to which student uses language intensifiers
or de-intensifiers.

extent to which student presents information without
further comment.

extent to which student uses indecisive language or
expresses doubts about feelings or situations.

APPENDIX F

CODING SCHEDULE

F—l

CODING SCHEDULE

One hundred fourteen students filled out the Interaction Analysis

Profile and/or Form F.

imately equal groups.

Their responses were divided into four approx-

Four coding pairs read their essays and used

 

 

 

 

 

 

 

a coding form (see Appendix D) to code different questions. The actual
coding assignments are indicated below:

Coding Form Subject

Questions 1.0. 1-28 29-57 58—86 87-114

1—12 Pair 1 Pair Pair 3 Pair 4

13-18 Pair 1 Pair Pair 1 Pair 1
19-22 Pair 2 Pair Pair 2 Pair 2
23-27 Pair 3 Pair Pair 3 Pair 3
28—32 Pair 4 ,Pair Pair 4 Pair 4

 

‘v

FOOTNOTES

FOOTNOTES

The question arises, functional for what? The individual? The
interaction? Society? We understand that the interplay of fUnc-
tional dynamics between individuals, interaction, and society is
complex. However, in this thesis, we will discuss social behavior
as it is functional for the individual.

A fourth situation, where one is aware of the contingency but not
the rule, is somewhat meaningless inasmuch as the contingency is
that which relates the rule to the outcome.

This issue of awareness of rules and social contingencies has a prob-
lematic implication for research (especially experimental research)
on control. In presenting subjects with a situation, even if we

can guide all subjects to preferring a given outcome, we cannot be
sure that they all have the same level of awareness of rules of
soCial contingencies that impinge on the situation. We can thus
expect different behaviors due not only to experimental manipulations,
but also due to differential exercise of control.

These three ways of influencing the definition of the situation can
be used singly or jointly. Depending on the outcome or set of out-
comes the individual seeks, he may also try to define himsélf as

"an enthusiastic teacher," or as "impressed with this magnificent
place."

This seems to be a relatively straightforward proposition which many

theories would predict. See, for example, Woelfel and Saltiel (1974).

 

10.

11.

12.

For a discussion of one aspect of getting others to accept your
definition of their roles or qualities, see the literature on
altercating (e.g., Weinstein, 1967).

Implicit in this model are the actor's social skills, the social
skills of other actors, and relative status factors that may be
exogenous determinants of control.

Jones considers the presentation of positive aspects of the self as
a third form of ingratiation. This obscures the differences in the
underlying social contingencies between the first two forms and the
third. Moreover, because of the assumption that one presents positive
aspects of the self, it is limiting in the kinds of self-presenta-
tions that it covers. For these reasons, we prefer to see positive
self-presentations as being different from ingratiation tactics and
only a subset of all possible self-presentational tactics.

The problems of control in urban societies are of great importance
to classical social macro-theorists.

One may object to mixing the "observed" and "observer" modes of
analysis in the same prOposition. We assume that, given the tremen-
dous amount of feedback individuals have about their performance

in interactional situations, they are fairly capable of seeing them-
selves as others see them.

In her study on individuation, Maslach (1974) indeed found this to
be the case.

Snyder (1975) has develOped a test for monitoring awareness.of self-
expressive behavior. A person who has a high awareness of these beha-

viors may be able to manipulate them to effect an apprOpriate degree

13.

14.

15.

16.

17.

18.

19.

of individuation. The test for awareness of self-expressive behavior
correlates significantly with the Mach IV test for Machiavellianism
(3; = .19, 2.3 .05).

To strengthen the credibility of the story, the students were referred
to a journal article on Quasi-Sensory Communication (McBain, 1976).
The significance levels of all tests of differences in pretest or
validation studies reported in this chapter are two-tailed.

The pretest for the QSC Battery consisted of a sample of students

and a set of measures different than those of the validation study
concerning verbal measures discussed above.

By including the "essays" from R.S. and J.L. in Form F, and by making
the members of the "field" those college students who might also be
selected for one of the learning programs, we felt we would sharpen
the students' sense of whom they needed to be similar to or different
from. These techniques somewhat limit our eXperimental concern to
individuation in college p0pu1ations. However, we don't believe that
the individuation process is theoretically limited to college
p0pu1ations.

It is important to note that the intercoder correlation for each
measure is not as high as the reliability for the corresponding in-
dicator. The indicator is the average of the two coders and, as
such, it is more reliable than the intercoder correlation indicates.
The significance levels of all tests reported in this chapter are
one-tailed.

Because of the multicollinearity among these indicators, these tests
(and all other multiple-indicator tests in this chapter) are not inde-

pendent, and the experiment—wise error is not maintained at p;§ .05.

20. Personal communication with Katrina Simmons, April, 1976. She
also notes that rates differ depending on essay length and essay

length did vary in this study.

BIBLIOGRAPHY

 

BIBLIOGRAPHY

Archer, E.J. Re-evaluation of meaningfulness of all possible
CVC trigrams. Psychological Monographs, 1960, 24, whole
no. 497.

 

Asch, S.E. Effects of group pressure on the modification and
distortion of judgments. In H. Guetzkow (ed.), Grou 5,
Leadership, and Men. Pittsburgh: Carnegie Press, 1951.

 

Bem, D. Beliefs, Attitudes, and Human Affairs. Belmont, California:
Brooks/Cole, 1970.

 

Christie, Richard G Florence Geis. Studies in Machiavellianism.'
New York: Academic Press, 1969.

 

Cushman, Donald G Gordon Whiting. An approach to communication
theory: Toward consensus on rules.‘ JOurnal of Communication,
1972, 22, 217-238.

 

Fink, Edward 6 Barbara Walker. Relative status, anticipated
interaction, and social facilitation as determinants of
humorous responses to embarrassment. Unpublished, 1975.

Goffman, Erving. The Presentation of Self in Everyday Life.
Garden City, New York: Doubleday, 1959.

 

Gottlieb, G. The Logic of Choice: An Investigation of the Concept
of Rule andiRationality. New York: Macmillan, 1968.

 

 

Harre, R. Some remarks on "rule" as a scientific concept. In
Theodore Mischel (ed.), Understanding Other Persons. Oxford:
Blackwell, 1974.

 

James, William. Psycholggy: Briefer Course. New York: Henry
Holt, 1892.

 

Jones, E.E. Integration: A Social Psychological Analysis. New
York: Appleton-Century-Crofts, 1964.

 

Kuhn, M.H. G T.S. McPartland. An empirical investigation of self
attitudes. American Sociological Review, 1954, 19, 68-76.

 

Maslach, Christina. Social and personal bases of individuation.
Journal of Personality and Social Psychology, 1974, 29, 411-425.

 

 

Maslow, Abraham. Toward a Psychology of Being, New York: Van
Nostrand, 1962.

McBain, W.N. Quasi-sensory communication: An investigation using
semantic matching and accentuated effect. Journal of Person-
ality and Social Psychology, 1970, 14, 281-291. ‘

 

 

Milgram, Stanley. The experience of living in cities. In Edwin
Hollander 6 Raymond Hunt (eds.), Current Perspectives in
Social Psychology. New York: Oxford University Press, 1971.

 

 

Miller, Gerald 6 Mark Steinberg. Between People: A New Analysis
of Interpersongl Communication. Chicago: Science Research
Associates, 1975.

 

 

Robinson, John 6 Phillip Shaver. Measures of Social Psychological
Attitudes. Ann Arbor, Michigan: Survey Research Center, 1969.

 

Schachter, Stanley. Deviation, rejection, and communication.
Journal of Abnormal and Social PsycholOgy, 1951, 16, 190-208.

 

Snyder, Mark. Self-monitoring of expressive behavior. Journal
of Personality_and Social Psychology, 1974, 39) 526-531.

 

Stires, Lloyd 8 Edward E. Jones. Modesty versus self-enhancement
as alternative forms of ingratiation. Journal of Experimental
Social Psycholpgyg 1969, 53 172-188.

 

 

Weinstein, E.A. Altercasting and interpersonal relations. In
P. Secord E C. Blackman (eds.), Readings in Social Psychology,
1967.

 

Woelfel, Joseph G J. Saltiel. Cognitive processes as motion in
a multidimensional space: A general linear model. Unpublished
manuscript, 1974.

Zetterberg, Hans. 0n Theory and Verificagion in Sociolggy,
Totowa, New Jersey: Bedminster, 1965.

 

Ziller, Robert. The Social Self. New York: Pergamon Press, 1973.

 

Ziller, R.C. Individuation and socialization. Human Relations,
1964, 12, 3414360.

 

Zimbardo, Philip. The human choice: Individuation, reason, and
order versus deindividuation, impulse, and chaos. In W.J.
Arnold G D. Levine (eds.), Nebraska Symposium on Motivation:
1969. Lincoln University of Nebraska Press, 1969.

 

 

HICHIGQN STRTE UNIV. LIBRRRIES

I” ”II III! H7 WI" I“ III II llill II”
III III llllllllll lllll l'l

 

31293008302931