\‘z.

“"21”. '
. u . J: :::- ‘ '

. . - IV. 'ogai 'd. ’5 .

' ’3‘ W37 3. 1

“(I I 5‘2“, 5'?

W’M

W; I,
'l p .-
EEC, .,.

.-(:
I "o
55'.»
f ’0‘." I

v.42 .
I "N."
.‘

M.
-3; a .:
YJ‘

gt}; "‘1‘. 2332‘?
J fl. . I.” :1926"%
SIT... 15'

V

f'“ “5-“: r ..

If . .«A "raw

“‘2 . m“)
"1’

.gljif’

"3"?“
“V:

M'.:
£32}: on
"1"‘4-m

J7,
a; $‘L‘f”!f’.~f ‘01:?

a; if? g”
‘53:! g";\’£:f’ “>341;

in”;

"7mlf:
3715?“

’5. ‘

" J-
‘7 (Nu, { .‘
' .r-V f:/‘::-jwu, a ; ‘ ‘

I‘:“_‘\.: 6.11:5, I'

ofid

r “ f
.f’f‘w ‘3"4’ “W";J 1/!"
’4' "WI ~31; “:1“:sz I:
'jlpr’c' (I. I. A:

., ':v .
a? "Ma
{5'1"- in
4,: ,
“I?” “Q ‘2“

““9132

AM?
If,

$43

‘ ’g ’5’}; ”112'.

l :1) :lgl‘v‘t)

1'.
/.
(2.1..-
”1'

.a
fg’; :z‘fl‘?‘

 

 

3’.”
2259.633
(41%?

'IE‘Y‘
1 "9.113: .-

3:3};

;’l
ffc—‘z-agl 3‘5
fig} $623. Yffv’v.

 

.. 9*

Date

0-7639

73377308

MICHIGAN STATE UNIVERSITY LIBRA

mu :1 @ymmmmmm mu’iz'

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ in:

1 31 3 006119709

 

LIBRARY

Michigan State

University

 

This is to certify that the

thesis entitled

IMAGES OF SELF, PEERS' PERCEPTIONS,
AND REFLEXIVE SELF-IMAGES:

NEW SUPPORT FOR THE SYMBOLIC INTERACTIONISTS

presented by
BRETT A. MA}!

has been accepted towards fulfillment
of the requirements for

MASTERJZILABIL degree in ____PSYCHOL0_G_ Y

Mflm

T
/ Major professor

8/5/88

MS U is an Affirmative Action/Equal Opportunity Institution

 

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

DATE DUE DATE DUE DATE DUE

 

 

 

 

‘
4 O
I \ ‘ " I,’
a.

 

Fungi 30033 1
E '.'425 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative ActIorVEquel Opportunity Institution

IMAGES OF SELF, PEERS’ PERCEPTIONS. AND
REFLEXIVE SELF-IMAGES:
NEH SUPPORT FOR THE SYMBOLIC INTERACTIONISTS

By

Brett A. May

A THESIS

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

MASTER OF ARTS

Department of Psychology

1988

xx;
(3*

‘49

\f3

ABSTRACT
IMAGES OF SELF, PEERS' PERCEPTIONS, AND

REFLEXIVE SELF-IMAGES:
NEW SUPPORT FOR THE SYMBOLIC INTERACTIONISTS

By

Brett A. May

Following the symbolic interactionist viewpoint, it was
hypothesized that others’ actual perceptions relate to
self-perceptions through individuals’ beliefs about how
others perceive them (reflected perceptions).
Self-perceptions were also expected to be influenced by the
individual’s power and locus of control. Perceptions were
assessed by ratings on the two central. and orthogonal
interpersonal dimensions of self- and other-acceptance in
small groups on two occasions. 0f four hierarchical
regression analyses, three supported the main hypothesis
strongly. Reflected perceptions, as compared to others'
perceptions, accountd for about twice the variance in
self-perceptions. Others’ perceptions also related
significantly to reflected perceptions. Power and locus of
control effects were also often significant. Unexpectedly
large differences were found when using measures of self-
versus other-acceptance. Supplemental evidence suggested
that individuals thought it was more important to be viewed
as other-accepting than self-accepting, .although their

schemata for self-acceptance tended to be better defined.

Table of Contents

List of Tables . . . . . . . . . . . . . . . . . . . .
Introduction and Concept Review . . . . . . . . . . . .
The Symbolic lnteractionist View of the Self-Concept

Personal Power . . . . . . . . . . . . . . . . . . .

Locus of Control . . . . . . . . . . . . .

Use of

Self-Schemas,

Method .

Self-Acceptance,

Concept Synthesis and Hypotheses

Participants and Setting . . .

Instruments
The Group Behavior
The Personal
Rotter’s
Spheres of Control

Procedure

Results . . . . .

Reliability

Rating

l-E Locus

Scale .

Battery:

Evaluation of Hypotheses . . .

Discussion
Conclusion
Appendices

Appendix A:

The SDC’S

lnterpersonally Oriented Groups .

and Other-Acceptance

of Control

Scale

Interpersonal

Interpersonal

iii

Scales for ARS and ARD

O O O O

Scale

11

13

14

18

20

20

21

21

24

26

27

28

30

30

31

40

52

SS

55

Appendix B:

Appendix C:

The Personal Power Scale
Group Behavior Rating Scales

for ARS and ARD . . . . . .

List of References . . . . . . . . . . . .

iv

56

59

62

LIST OF TABLES

Table 1: Internal consistency and interset
correlations for measures of acceptance versus
rejection of self (ARS) and other (ARO) as rated by
self, others, and reflected self at Times I and II.

Table 2: Hierarchical regression equations with self-,
other-, and reflected-ratings regressed on measures of
acceptance versus rejection of self (ARS) and other
(ARO) at Times I and II.

Table 3: Hierarchical regression equations for measures
of acceptance versus rejection of self (ARS) and others
(ARO) with self-ratings as the dependent variable and
locus of control (LDC), personal power (PPS), reflected
ratings, and ratings by others as the independent
variables.

Table 4: Shifts in the variance of self-ratings
accounted for from Time I to II on measures of
acceptance versus rejection of self (ARS) and others
(ARO) with reflected ratings, locus of control,

personal power, and ratings by others as independent
variables.

32

35

38

46

INTRODUCTION AND CONCEPT REVIEW

The present study investigates the use of locus of
control and personal power in elucidating the nature of the
relationship between perceptions by others, reflected
perceptions, and self-perceptions. Following the symbolic
interactionist perspective of Cooley (1902) and Mead (1934),
it is argued that the impact of other’s perceptions on self
is mediated by an individual’s reflected perceptions. In
this work, an attempt is made to clarify the
operationalization of this relationship, which seems to have
been commonly misinterpreted in the past. The manner in
which others differentially affect one’s self-concept is
also addressed. The influence of reflexive perceptions of
others on the self-concept are thought to vary as the other
is seen as more or less powerful and as one tends toward an
internal versus an external conception of control over
interpersonal events. In addition, where past research
approaches have typically examined perceptions between two
relatively unacquainted individuals at only one point in

time, the present research investigates the relationship

between other, reflected, and self-perceptions using several
others within a small group context where time of

acquaintance was varied.

2
I__ Symbolic Interactionist View of the Self-Concept

One of the first individuals to synthesize and
systematically develop ideas involving the reflected self
was George Herbert Mead. As the orientation and constructs
of the present study are founded largely upon Mead’s seminal
ideas, his work will be reviewed in some detail. Later
attempts to empirically verify Mead’s ideas and the
shortcomings of these efforts are next reviewed before
turning to recent studies that purport to have corrected
some of these difficulties. Limitations in the present
state of understanding of the symbolic interactionist
perspective will also be explored.

According to Head (1934) communication is the
fundamental process that makes society possible and that
makes human beings really human. How one comes to
communicate and Head’s view of the individual within this
process is crucial to the understanding of his theory. Mead
argued that persons initiate activity relating them to their
environments. In this, it is the act which is the unit of
existence. To Mead, individuals do not respond to stimuli
as objects outside of a response they may make, but rather
the stimuli and response are thought. of as a unit, so
interpenetrating that neither can be said to exist without
the other. Objects become stimuli as they function in the
context of acts and as they come to be defined in the acts’
completion. It is through experience and activity that

objects come to be viewed as separate from the self. To

3
Mead, it is the capacity to utilize and think about activity

and to conceptualize the object as object which sets humans

apart from other life forms. It is one’s conceptualization
of the object, rather than the object itself, which has
meaning for an individual.

Through activity, then, one realizes the object, which
may eventually include other individuals. Mead stated that

communication emerges when an activity has shared meaning

between persons. Thus, when two or more "selves”
conceptualize an object similarly, the potential for
communication exists. Mead referred to the sharing of

meaning of activity as comprised of significant symbols.
"The significant symbol emerges when the one who makes it is
aware of its meaning to the other, i.e., when one can
anticipate the response it will evoke in another" (Miller,
in Mead, 1982, p. 10). When we communicate by significant
symbols, we break out of the present. The communication is
not one of activity, but of the meaning of that activity}
When one breaks out of the present and into
conceptualization and significant symbols, one may also
realize a sense of self. That is, one has the capacity to
separate time and space and to see his or her self as an
object with a past and a future. For Mead, all things are
socially constructed, the self included.. One comes to
define one’s self through communication with others. In
this communication, attention is shifted from the present to

the meaning of behaviors of both oneself and other(s). The

4

self-concept is said to arise as a result of one’s idea of
another’s idea of oneself as an object or thing. we come to
think of ourselves as we think others see us. The self
develops and is emergent from symbolic interaction with
others. It is a socially defined structure existing in the
activity of viewing.one’s self reflexively. This idea has
been termed the reflexive or reflected self.

In later work conducted within Mead’s framework, the
self-concept has been thought to have three primary
components (Kinch, 1963; Rosenberg, 1981): [a] the
self-perception--how the individual views one’s self, [b]
the reflected self--how the individual believes another
views him or her, and [c] the accorded self--how others
actually see the individual.

In an extensive review of the research on symbolic
interactionist theory, Shrauger and Schoeneman (1979) found
considerable evidence that people’s self-perceptions agree
with the way they think others perceive them. In another
review, Rosenberg (1981) found similar support for this
relationship. 'Evidence for the relationship between the
accorded self and these constructs, however, has been mixed.
Rosenberg (1981) stated that "there is a consistent, though
imperfect, relationship between the social self (what others
actually think of us) and the reflected self” (p. 597). He
went on to state that "more often than not, people’s views
of what others think of them are accurate" but that ”in many

cases they also misread the attitude of the other toward

5
themselves" (p. 597). Shrauger and Schoeneman (1979) were
pessimistic in their evaluation of the relationship between
self- and accorded perception; ”There is no clear indication
that self-evaluations are influenced by the feedback
received from others in naturally occurring situations" (p.
549). Thus the relationship between self-perception and the
reflexive self appears to be solidly supported, whereas the
evidence for a relationship between these constructs and the
accorded self remains uncertain.

Schafer and Keith (1985), largely in rebuttal of
Schrauger & Schoenemans’ 1979 review, stated that
researchers who expect to find a significant and parallel
relationship between self-perceptions and reflected self,
and between self-perceptions and others’ actual perceptions
may be making ”a rather uncritical use of the symbolic
interactionist model of the self-concept" (p. 964). They
stated that the interactionist perspective does not
necessarily imply that actual and perceived appraisals of
others toward the self need to be treated similarly in
explaining the construction and organization of the
self-concept. Other authors were noted that made a similar
distinction, stressing the importance of the individual’s
perception of other’s appraisals over these others' actual
appraisals in affecting self-concept. In Mead’s original
work, it was the reflexive aspect of an interaction which
was the material for the idea of the self-concept. It is

the meaning the individual makes of the other’s behavior

8

that creates the significance of the symbol. Kinch (1983)
addressed this issue by defining a causal chain where the
actual responses of others have a direct effect on a
person's reflected perceptions which then effects the
self-concept. Schafer and Keith (1985) supported Kinch’s
view and cited earlier research showing that what we believe
others think of us is more closely related to our
self-concept than what the others actually think. In their
own research, Schafer and Keith (1985) used a path analysis
design which showed that others’ actual appraisals did
influence the self-concept, but that this influence took
place via the reflexive self.

Rosenberg’s (1981) review of this literature
acknowledged that though Mead's theory is generally
supported, it is imprecise and in need of refinement. He
noted that though we tend to see ourselves as we think
others see us, we have not identified which others influence
us and which do not. Schafer and others (Schafer, Keith, &
Lorenz 1984; Schafer & Keith, 1985) explored this issue by
studying married couples. This had the advantage of
insuring significance in the relationship between self and
one other. Schafer and Keith (1985) stated that another
limitation of their own (and much past) research was that
measurement of the relationship between actual, reflected,
and accorded selves were taken at only one point in time.
"As with many naturalistic studies, there are no repeated

assessments of self and other perceptions whereby the

7
effects of feedback over time from significant others on the
self could be determined. Although in this analysis we have
statistically examined effects, the influence of the
independent variable _on the self-concept can only be
assumed" (p. 989).

The present study redresses both of the aforementioned
limitations. The question of differential impact of others
on the self-concept is addressed through the constructs of
power and locus of control, while the limitation of number
of measurements and the issue of importance of the other to
the self is addressed through using repeated measures in

small lnterpersonally oriented groups.

Personal Power

Personal power is broadly defined in terms of influence
between related individuals. Kaplowitz (1978) sees power as
operative through a "set of propositions" about the
consequences of attributions an individual makes of another
in an interpersonal situation. From the symbolic
interactionist perspective, another’s personal power might
influence an individual through his or her reflected
perceptions. Others seen as more powerful would influence
an individual’s self-concept to a greater extent because the
reflected perceptions of more powerful individuals would
likely be weighted more than the reflected perceptions of
those others viewed as less powerful. Most treatments of

power, however, have largely neglected empirical

8

verification and have concentrated on broad theoretical
conceptualizations (Dahl, 1957; Nagel, 1988; Kaplowitz,
1978). Indeed, it is interesting that the power construct
has received extensive and complex theoretical treatment
while numerous investigators have stated that the field
suffers from overly simplistic operational definitions and
limitations in empirical measurement (Allen, 1984;
Gray-Little & Burks, 1983; Olsen & Rabunsky, 1972;
Safilios-Rothschild, 1970). Though many studies utilizing a
power construct have been conducted, the instruments used
often seem to measure different aspects of power and
typically correlate quite poorly with one another
(Gray-Little, 1982; Olsen & Rabunsky, 1972). In spite of
these limitations, the literature on personal power may be
roughly separated into two groups, those studies that focus
on narrowly-defined personal attributes and those that focus
on the power strategies utilized between individuals in an
interaction.

The idea that specific attributes of individuals have
variable impact in an interpersonal context has been widely
researched. Tedeschi (1972) called this impact personal
power and defined it as "...a relatively consistent
attribute of the person across situations." (p. 104).
Jacobson (1972) referred to "resources of the agent” which
include status, authority, education, communication skills,
and interpersonal abilities. Heider (1958) defined personal

power as the ability to influence the social and physical

9

environment of another person, and noted that it is
something deriving from what an individual possesses.
Minton (in Maher, 1988), in summarizing the work of Heider
(1958), identified ability, knowledge, intelligence,
strength, status, and competence as personal power
variables. Minton himself referred to a category of power
characteristics that are ”organismic”. Elements of this
category include skills, intelligence, knowledge, and
education. Tedeschi, Schlenker, and Bonoma ~(1973) found
that primary sources of personal power included expertise,
attraction, status, and prestige. Other attributes found to
influence interpersonal interaction include physical
attractiveness (Barnes & Rosenthal, 1985; Benson,
Karabenick, & Lerner, 1978; Cash, Begley, McCown, & Wise,
1975; Miller, 1970; Warner & Sugarman, 1988), speech ability
and fluency (Bord, 1975; Erikson, Lind, Johnson, & O’Barr,
1978; O’Sullivan, Ekman, Friesen, & Scherer, 1985; Warner &
Sugarman, 1988), assertive expression (Reicken, 1958), eye
contact (Cherulnik, Neely, Flanagan, & Zachau, 1978; Dovido
& Ellyson, 1982; Hurley & Marsh, 1988), attire (Schnieder,
1974), and body carriage (Fast, 1977; Schwartz, Tesser, &
Powell, 1982).

The literature on marriage and the family is also
concerned with personal power. It is within this research
sector that 3 strategies approach has been most often
utilized in attempting to understand power processes between

family members. In a recent review of this literature,

10
Gray-Little and Burks (1983) reported that power strategies

are usually investigated through decision-making. In past

research, decision-making was often decided simply by asking
each individual in the couple who made the decisions over a
variety of areas such as vacationing, grocery shopping, and
housing (see Safilios-Rothschild, 1970, for review). More
recently, the approach has been to develop a number of
strategies that are thought to represent the process of
power in an interaction. Strategies commonly thought to be
indicative of power in relationships include: number of
interruptions (Farina, 1980; Gray-Little, 1982; Mischler &
Waxler, 1988), frequency of concessions (Cromwell &
Cromwell, 1978: Gray-Little, 1982; Scanzoni, 1971), and
talking time (Gray-Little, 1982). Falbo (1977), Falbo and
Peplau (1980), and Cowan, Drinkard, and Macgavin (1984) have
developed lists of power strategies, each based largely upon
the farmer’s work, that purportedly account for a large
portion of power strategies .actually utilized in
interaction. Falbo and Peplau (1980) developed a list of 13
strategies that accounted for 98% of the strategies that
occurred among a total sample of essays on personal
decision-making. The power strategies utilized in the most

recent revision of this list (Cowan et al., 1984) include;

asking, begging/pleading, telling/assertion, reasoning,
demanding/arguing, stated importance, bargaining,
persistence, negative affect, positive affect, verbal

manipulation, eliciting reciprocity, using an advocate,

11
evasion, and lassez-faire.

As past research on personal power has been criticized
for only partially operationalizing the power construct, the
present study will include a measure of power called the
Personal Power Scale (PPS). Developed by this author, the
PPS involves rating others on a statistically supported
subset of the attributes and strategies noted above (see
Method section). It is hoped that the inclusion of both
attributes and strategies in a single instrument will more

fully represent the personal power construct.

Locus of Control

Rotter (1954, 1968) first introduced the
internal-external locus of control construct. He considered
it a generalized expectancy, operating over a large number
of situations, which related to whether or not the
individual perceives the self to possess or lack control
over what happens to him or her. Lefcourt (1966) noted
that, as a general principle, ”internal control refers to
the perception of positive and/or negative events as being a
consequence of one’s own actions and thereby under personal
control; external control refers to the perception of
positive and/or negative events as being unrelated to one’s
own behaviors in certain situations and therefore beyond
personal control" (p. 207).

Of particular relevance to the present research is the

idea that internal-external variance in attributing control

12
may be salient specifically in interpersonal contexts.
Paulhus (1983) found that locus of control could be
attributed by an individual in three distinct areas;
personal efficacy, sociopolitical control, and interpersonal
control. The interpersonal domain of control was said by
Paulhus to pertain to individuals as they interact with
others in dyads and group situations. ”An internal locus of

control indicates the expectation that one can, through

their gig behavior, control outcomes of personal
significance in interpersonal situations, whereas an
external locus indicates the expectation that in an
interpersonal situation one will feel a lack of personal

control over significant processes or outcomes.’ Since one
comes to define their self in part through their perceptions
of what others think of them, one might assume that the
locus of control construct acts as a mediating variable in
determining the importance of one’s reflected appraisal of
others on the self-concept.

The present research investigates the relationship
between locus of control and the influence of accorded
perceptions on self-concept via the reflexive self by
utilizing both Rotter’s (1966) internal-external locus of
control measure and the interpersonal control scale from
Paulhus’s (1983) work (see Method section). It is thought
that as one has an increasingly external sense of
interpersonal control, one’s reflected appraisals of others

toward them will have a greater influence on one’s

13

self-concept.

Use of lnterpersonally Oriented Group;

The use of the small group setting has many advantages
over the frequently utilized single measurement schema.
Within such small groups, individuals typically move from
being strangers to becoming relatively well acquainted. At
the close of these groups, individuals will have around 50
hours of direct interpersonally oriented contact (see Method
section for detail). Thus, where much past research has
looked for the presence of the reflexive self at only one
point in time, the use of the small group enables control
over the time of acquaintance, making it possible to
investigate the salience of the reflexive-self construct at
successive stages in the development of relationships.
Another advantage to the small group format is that it
allows measurement of how more than one "other" perceives
each individual. The present groups have from six to seven
members.

Individuals have also shown significant change over the
course of the groups’ development on the measure used to
represent the self-concept (Hurley & Rosenberg, 1986). This
is in part because these groups focus on developing
interpersonal skills and the self-concept measures used in
the present study pertain to interpersonal aspects of
personality. Thus, the small group setting is one where the

possible impact of others on the self-concept may be

14
measured because these aspects of the self-concept have been

shown susceptible to change.

Self-Schemes. Self-Acceptance, and Other-Acceptance

The self-concept gains its meaning through
interpersonal behavior. This has recently been
conceptualized is in terms of self-schemas (Markus, 1977;

Markus, Moreland, & Smith, 1985; Rogers, Kuiper, & Kirker,
1977). Markus, Crane, Bernstein, & Siladi (1982) state that
"Self-schemas are assumed to be summaries and constructions
of past behavior that enable individuals to understand their
own social experience and to organize a wide range of
information about themselves" (p. 38). Self-schemes are
cognitive generalizations about the self, derived from
experience, that organize and guide the processing of
self-related information. The union of summaries and
constructions across differing focal domains comprises the
self-concept. One self-schema relevant to the present
research concerns the domain of interpersonal experience.
Surprisingly enough, it has been found that self and others’
perceptions of interpersonal behavior can reasonably be
classified under only two principle and orthogonal
dimensions: acceptance versus rejection of self and
acceptance versus rejection of others.

”The evidence that two principle dimensions encompass a
broad variety of interpersonal behavior seems impressive but

unfamiliar to many contemporary researchers in the

15

psychotherapeutic and T-group areas” (Hurley, 1976b, p. 88).
Work presented by Foa (1981), Schaefer (1981), Adams (1964),
Peterson (1985), Bierman (1989), Carson (1969), and Kiesler
(1983), shows that one dimension plainly concerns ones
responses to other persons. Both Symonds (1939) and Bierman
(1969) labelled this dimension acceptance--rejection while
others have quite similarly labelled it:
affiliation-~hostility (Freedmen, Leary, Ossorio, & Coffey,
1951); positive-~negative (Chance, 1954); loving-rejection
(Roe, 1957); love--hostility (Schaefer, 1959); Iove--hate
(Carson, 1969); and affiliation (Wiggins, 1982).

As Hurley (1976b) noted, finding agreement in a label
for the second dimension has been more difficult. Symonds
(1939) called it dominance--submission as did Freedman et
al. (1951) and Carson (1969), while Bierman (1969) and
Chance (1954). both labelled it activity-passivity. Foa
(1981) Identified acceptance-rejection as the primary
content component in bath dimensions and held that in the
second dimension, this related to both social and emotional
behaviors of the subject toward the self. Wiggin’s (1982)
overview of the literature in this area noted that theorists
most commonly speak of the second dimension as dominance.

Allport (1981) observed that a two-dimensional

structure , has been utilized as a conceptual base of
personality since the time of Hippocrates, as well in the
works of Galen, Kant, Wundt, Herbert, and Pavlov, among

others. Gibb (1964) stated that "a person learns to grow

18

through his increasing acceptance of self and others" (p.
279). 'Argyris (1962) depicted interpersonal competence as
deriving from ”a basic need of man to increase his sense of
self-acceptance and acceptance of others" (p. 20). Hurley
(1986b) noted that the concept of acceptance versus
rejection of others and of the self were major features of
H.S. Sullivan’s (1953) interpersonal theory, and are
independently seen in the works of Carl Rogers (1951) and
Abraham Maslow (1970). This seems congruent with both Foa’s
perspective and Adams (1964) view that ”an interpersonal act
may be regarded as the Cartesian product of these two sets”
[acceptance-rejection of self/others] (p. 195).
Additionally, these concepts plainly formed the conceptual
base for Eric Berne’s (1966) four ’life positions’ (p. 270):
I’m OK,You’re OK; I’m OK, You’re not OK; I’m not OK, You’re
OK; and I’m not OK, You’re not OK.

Operationally, many instruments have been constructed
which directly utilize these orthogonal dimensions
(Benjamin, 1979; Hurley, 1976a; Kiesler, 1983; LaForge &
Suczek, 1955; Lorr & McNair, 1965). Though the instruments
range from those relatively simple and easy to administer
(Hurley, 1978) to those extremely complex and involved
(Benjamin, 1979), they repeatedly find acceptance versus
rejection of self and other(s) as primary and salient.
Furthermore, this salience has been demonstrated across a
wide variety of settings (Hurley, 1986b).

Interpersonal behavior, then, seems adequately

17
conceptualized as essentially undergirded by the dimensions
of acceptance versus rejection of self (ARS) and acceptance
versus rejection of others (ARO). Though a third dimension,
task orientedness-expressivity, (Bales, 1970) has been
identified as salient in some interpersonal contexts, Hurley
(1986b) has noted that most agree that it plays a minor
role. Conte and Plutchik (1981) stated that, "For
interpersonal personality data, however, any factors beyond

the first two account for very little of the total variance”

(p. 707). In the present study, ARS and ARO, as
self-schemata, represent important facets of the
self-concept. Though it may be argued that one’s

self-concept will vary depending on subject and setting, or
that many other self-schematas also exist which could be
measured, it is thought that the dimensions of ARS and ARO
not only represent significant aspects of most individuals’
self-concept, but they also represent the interpersonal
medium through which one gains information regarding other
areas of oneself. In that these dimensions are the primary
means by which an individual’s interpersonal world is
represented, and in that it is through interpersonal means
that an individual comes to develop a sense of self, it is
thought that the dimensions of acceptance versus rejection
of self and acceptance versus rejection of other(s)
appropriately represent pertinent features of the

self-concept.

18

Concept Synthesis and Hypothgggg

The present research is aimed at clarifying and
extending recent work on the relationship between the
reflexive-, accorded-, and self-concepts. The basic
relationship as defined by the symbolic interactionist
perspective is tested. Two other constructs, personal power
and locus of control, are also investigated. It is thought
that the influence of others on self-perceptions will vary
as the individual perceives these others to be more or less
powerful. In addition, as one has increasingly external
expectations of control, the others’ importance to oneself
increases, and reflected appraisals in general are
hypothesized to have a greater impact on one’s self-concept.
One might further expect that these constructs act
complementarily, as personal power pertains to perceptions
of another and locus of control pertains to the importance
of those perceptions to one’s self. Finally, the present
research utilizes small interpersonally oriented groups to
assess the status of the symbolic interactionist
relationship both early (Time I) and late (Time II) in the
groups’ development.
1. (a) It is hypothesized that individual’s impressions of
how others perceive her or him (the reflexive self) will
account for more of the variance in self-ratings than will
these other’s actual or accorded ratings. Self, reflexive,
and accorded perceptions will be measured on ARS and ARO

ratings from the Group Behavior Rating Scales. Individuals

19
will rate themselves (self-ratings), and also predict how
they will be rated by others (reflexive ratings) on these
measures. All will also be rated by each other group member
(accorded ratings) as well.

(b) It is also hypothesized that accorded ratings on
these same measures will account for a significant amount of
the variance in reflected ratings.

2. It is hypothesized that the above-stated relationships
will be significantly stronger after about 40 hours of small
group interaction (Time II) than after only about 20 hours
(Time I).

3. It is hypothesized that individuals perceived as less
powerful by others (as rated by the Personal Power Scale)
will have reflexive perceptions which account for more of
the variance in self-ratings when compared to those
individuals viewed as more powerful by others (note comments
in Discussion, p. 41).

4. It is hypothesized that the relationship in hypothesis 3
will be supported to a significantly stronger degree at Time
II than at Time I.

5. It is hypothesized that the reflexive perceptions of
others will account for more of the variance in self-ratings
with an increasingly external versus internal locus of
control (as measured by Rotter’s I-E Locus of Control Scale
and the Interpersonal Scale from the Spheres of Control
Battery (Paulhus, 1983)).

8. It is hypothesized that the relationship in hypothesis 5

20
will be supported to a significantly stronger degree at Time
II than at Time I.
7. It is hypothesized that locus of control and personal
power will jointly increase the self-ratings variance
accounted for by reflected ratings and that this hypothesis

will be significantly stronger at Time II than at Time 1.

Method

Participants and Setting

Participants were 27 male and female students (three
groups of 7 and one group of 6) enrolled in an upper level
undergraduate psychology course (PSY 400) at Michigan State
University. All class members consented to participate,
though one person from the group of six dropped the course
before any ratings were collected. The purpose of the
course, as stated in a description was, ”to build an
atmosphere of concern and respect for each member’s
personhood while also attempting to respond both
constructively and honestly to each participant’s behavior
within a here-and-now context”. Small groups were formed
during the initial meeting of the full class. While there
were no formal placement criterion, members were strongly
encouraged to enter a group in which they were unacquainted
with the other members. Attempts were also made at gender
balancing within each group, which consisted of three or
four members and two facilitators. Groups were co-led by

facilitators who had previously completed both this course

21
and at least one subsequent term of leadership training.
Groups typically met a total of about 50 hours, including
two 90-minute sessions weekly. Each group also held 12-hour
uninterrupted ”marathon" sessions near the third and seventh
weekends.

Students were informed that their course grades would
be based mainly on a detailed journal of their thoughts,
feelings, and experiences about self and each other group
member and secondarily on their course attendance and their
successful completion of two quizzes on the textbook,
Intgrpersonpl Living (Egan, 1976). Journals were regularly
reviewed and graded by the instructor. This separation of
course grades from actual within-group behavior was intended
to minimize artificial conformity or "good student” behavior
within the small group sessions. Attendance at all class
and groUp sessions was mandatory and absences were
relatively rare, averaging less than one per student per

term.

Instruments

IDS Group nggvior Rating Sop; g for ARSfignd ARO

The Group Behavior Rating Scales (GBRS), featuring
measures of acceptance versus rejection of self (ARS) and
acceptance versus rejection of others (ARO), consisted of
nine subscales, four of which constitute ARS, four which
constitute ARO, and one independent scale. All subscales

were bipolar and in semantic differential (Snider & Osgood,

22

 

 

 

1969) format. ARS subscales included:
Shows feeling§::Hidg§ feelipgg; Expressive-~Gugrded;
Activg--Passive; and Dominant--Submissive. ARO subscales
included: Warm-~Cold; Helps others--Harms others;

Gentle--Harsh; and Accepts others--Rejects others. The

independent Likgd--Diglikgg scale was presented first as a
buffer for the remaining measures (Smith, 1979).

The subscales were presented on separate pages of a
minibooklet with ten spaces separating each pair of anchors.
After the initial Likgd--Di§likgp_scale, each ARS subscale
was followed by an ARO subscale with favorable and
unfavorable anchors staggered to minimize response sets.
Ratings on each scale were translated into scores ranging
from zero to nine, yielding potential ARS and ARO scores
from zero to thirty-six. The scales were self-administered
and were preceded by standardized instructions. These
instructions were specific and requested "your personal
impression of each member's actual behavior within the group
sessions up to now” (Hurley, 1986a). They also encouraged
raters to use the full range of possible ratings.

Recent evidence these measures’ stability was
demonstrated by Hurley (1986b) using members of the American
Group Psychotherapy Association (AGPA) who participated in
two-day, 12-hour interpersonal groups. Members completed
GBRS ratings both early (2 1/2-hours) and late (12-hours) in
these groups. Though the difference between mean group

correlations early and late revealed an increment of

23

approximately 10% in interpersonal competence, as assessed
by multiplying mean ratings on ARS and ARO for each
occasion, ARS and ARO composite scores still showed
reasonable stability (ARS = .66, p < .01; ABC = .72, p <
.01). Thus, even when confounded with individuals’
behavioral changes, these measures showed appreciable
interoccasion stability. Evidence that these ARS and ARO
measures were distinct and homogeneous is also reported in
this work. Among the four ARS subscales, it was found that
members’ mean ratings of their group’s leader had median
correlations of .75 early, .74 late, and .74 early-to-late.
For ARO’s four subscales, the parallel median correlations
were .54 early, .62 late, and .42 early-to-late. In
contrast, the interquartet median correlations were much
weaker (early = .21, late = .21, and early vs. late = .20),
establishing that intraquartet linkages markedly exceeded
their interquartet counterparts, supporting these measures’
independence.

Theoretical and conceptual evidence presented earlier
showed that the ARS and ARO measures were well-grounded in
the literature. Additionally, several other measures have
shown the expected patterns of correlation with Hurley’s ARS
and ARO indicators. These measures are all longer and more
comprehensive than Hurley’s measures and require much more
time to administer and score. In light of this, it is
appropriate to use these as a criteria for Hurley’s much

shorter and quicker measures (Anastasi, 1982). Some of this

24
validational evidence (Hurley, 1986b) is briefly summarized

below:

"Very similar ARS and ARO measures had shown
construct validity in a study of the components’
autocorrelations (Hurley, 1978a). Strong evidence
of both convergent and divergent validity of ARS
and ARO indicators was also demonstrated by their
correlational patterns with different prototypical
measures (Wiggins, 1982) of the affiliation and
dominance dimensions. Thus, Gerstenhaber (1975)
reported that the Interpersonal Checklist’s
(LaForge & Suczek, 1955) LOV factor, commonly
taken as an affiliation index, correlated .55 (p <
.001) with ARO but .00 with ARS, while its DOM
factor, widely used to assess dominance,
correlated with ARS .70 but only .18 (pg) with
ARO. In another study, 47 members of six small
personal development groups described self and
each other group member on Lorr and McNair’s
(1965) Interpersonal Behavior Inventory (IBI)
after SO-hours of interaction. All had earlier
completed ARS and ARO ratings after both 22- and
45-hours of group participation. lndividual’s mean
ARS ratings were found (Hurley, 1983) positively
correlated (.41 & .83) with their parallel ratings
on the 181’s five-scale Dominance factor but
inversely (-.39 a -.44) with IBl’s four-scale
lntropunitive factor. Similar ARO rating did not
correlate significantly with these two IBI
factors, although ARO did correlate strongly (.73
a .74) with IBI’s six-scale Affiliation factor."

The Personal Pqur chig

The Personal Power Scale (PPS) is a measure intended to
assess a wide range of personal power attributes and
strategies. Developed by this author, the PPS consisted of
15 subscales subsumed under the categories of: Globglngpgp
(amount of power, amount of influence, control of others);

Power Strategies (telling/assertion, talking time, frequency

of concessions, interruptions, demanding/arguing,

25
persistence); Socigl Attributg§_ (physical attractiveness,

attire,» apparent socioeconomic status); and Interpersonal

Attributes (speech fluency, expression of ideas,
interpersonal expertise). The subscales were presented on

separate pages of a minibooklet with ten spaces separating
each pair of anchors. Positive and negative anchors were
staggered.to minimize response sets. Ratings on each scale
were translated into scores ranging from zero to nine. The
scales were self-administered and were preceded by a
standard instruction, ”Rate each person in your group on the
following dimensions. Use your own perceptions of each
other person and try to use the full range of possible
ratings. Remember, the research will hold these ratings in
complete confidence so rate as honestly and as accurately as

you can.”

The PPS was developed from a previously existing
measure called the Personal Power Functions Profile (Reyher,
in Gavrilides, 1980) and a number of power strategies found
throughout the literature. Originally, the PPS consisted of
18 subscales. A pilot study was conducted with 14 students
enrolled in Psychology 400 at Michigan State University in
the spring of 1987 to assess the principles of homogeneity,
internal consistency, and parallelism (Hunter, 1977). In
this work, the subscales height, build, posture, and
frequency of eye contact were deleted from the instrument
and the subscale control of others was added. Cronbach

alphas for individual subscales ranged from .67 to .97 with

28

a mean alpha of .90 for the revised group.

Rotter’s l-E Locus of Control Scale

Rotter’s (1966) Internal-External Locus of Control
Scale consisted of 23 forced-choice items plus 6 filler
items. Possible scores range from 0-23, on the basis of one
point for each external choice. Rotter purported that this
scale assesses a generalized expectancy of locus of control
over a wide variety of situations.

Rotter (1966) reported split-half reliabilities in the
.70s and test-retest reliabilities ranging from .56 to .83,
averaging .63. Recent studies have reported Cronbach alphas
of .77 and .72 for spouses (Sabatelli, Buck, & Dreyer, 1983)
and .70 for college undergraduates (Paulhus, 1983).
Additionally, the l-E scale has been generally found to
correlate only weakly with social desirability. Most
studies report this correlation as accounting for less than
1% of the total variance (Askanasy, 1985; Parkes, 1984),
although Paulhus (1983) reported a correlation of .32
between the measures.

A recent confirmatory factor analysis study by Askanasy
(1985) found that among four differing approaches to the
factor analysis, only 20% of the variance could be accounted
for. He stated (p. 1337) that ”the high levels of
obliqueness between factors seem to support [a] Rotter’s
(1966, 1975) stance that the measure is tapping a

generalized construct and [b] Collin’s (1974) observation

27

that ”there is a common theme of internal-external control
of reinforcement running through the 48 alternatives" (p.
387). Other empirical support for the measure’s validity
come from Joe’s (1971) extensive review of earlier
literature. He stated that "the data tend to support
Rotter’s contention that the internal-external control
concept is a generalized expectancy operating across many
situations” (p. 755). This was also supported by Strickland
(1977).

Rotter’s I-E scale also shows evidence of construct
validity through it’s significance as a mediating variable
over a wide variety of contexts (See Askanasy, 1985; Joe,
1971, for review), and in light of its continued use (see
Concept Review). Overall, Rotter’s I-E Locus of Control
Scale has been found to be a valid measure of a generalized

expectancy of control over a wide variety of settings.

Spheres of Cpntrol Battggy: Interpersonal Scale

The Interpersonal Scale ‘(IS) from the Spheres of
Control Battery developed by Paulhus (1983) is a 10-item
scale purported to assess an individual’s locus of control
specifically with regard to interactions with others. Items
are rated on a seven-point Likert scale, with half of the
items reverse scored to minimize response sets.

Paulhus (1983) reported Cronbach alpha reliabilities of
.75-.80 on cross-validation samples for the IS. Test-retest

correlations over four weeks were above .90 and at six

28

months were above .70. Paulhus found that the Interpersonal
Scale’s correlation with social desirability as measured by
the Marlowe-Crowne accounted for only 1.2% of the total
variance of the scale. The Interpersonal Scale has also
been found to have predictive validity across three separate
studies utilizing samples of college varsity football
players, tennis players, and nonathletes (Paulhus, Molin, &
Schuchts, 1979).

Finally, the Spheres of Control Battery (SCB) as a
whole was found to correlate .75 with Rotter’s I-E scale,
supporting the SCB’s construct validity. As a separate
scale within the battery, the Interpersonal Scale’s
correlation with the I-E scale accounted for only 8% of the
variance of Rotter’s scale, showing possible support for the
notion that it measures a specific facet of the generalized

expectancy assessed by Rotter’s scale.

Procedure

At PSY 400’s first session of the term, potential
participants were advised of the nature of this study and
invited to participate. They were told that this work would
explore individual’s perceptual accuracy in an interpersonal
context which entailed making ratings of self and others in
their group on a variety of dimensions. Also, they were
advised that their participation would hopefully assist
future group members by making perception within the group

better understood. They were further informed that all

29

responses would be confidential and that no identifying
information would be available to others. They were told
that for their participation they would receive confidential
feedback on aspects of their interpersonal style and, toward
the end of the term, on their perceptual accuracy in the
group itself. They were further advised that this author
would be available toward the end of the term to discuss
more specifically what the research involved and to assist
them in understanding their individualized feedback.
Finally, they were informed that participation was voluntary
and that there would be no penalty should they choose not to
participate.

Those choosing to participate were then asked to fill
out the I-E Locus of Control Scale and the Interpersonal
Scale from the Spheres of Control Battery. The GBRS and the
PPS were administered several weeks later, after
approximately 22 (Time I) and 45 (Time II) hours of group
interaction. Two GBRS versions were administered on both
occasions. One version asked each member to simply rate
both their self and each other member on the behavior they
have observed in the group up to that point in time. These
data yielded both the self-concept ratings and others’
actual ratings (accorded selves) of the individual. The
second GBRS version differed from the first in that each
person was asked to rate how you think epch member rates

your behavior within the group up to now. This provided a

measure of the reflexive self for each individual. Order of

30
presentation of the four measures at each administration was
randomly determined to minimize order effects.

After returning the final set of such ratings,
participants were given the option of receiving
individualized feedback on their personal perceptual
accuracy. This feedback was in the form of a graph showing
the proximity of reflected appraisals on the GBRS from those
ratings which were made by others. Participants were also

given the opportunity for debriefing at this time.

Results
Reliability

For the present sample, the Personal Power Scale (PPS)
items had a full-scale Cronbach alpha reliability of .97 at
both Times I and II. Individual item alphas ranged from .52
to .87 at Time I, and from .41 to .89 at Time II. Mean
internal consistency for individual items was .77 for Time I
and .76 for Time II. The PPS was also highly stable over
time as shown by an interoccasion Pearson p = .95.

Rotter’s Locus of Control (LOC) Scale had a Cronbach
alpha of .85 as contrasted with the parallel value of only
.55 for Paulhus’s Spheres of Control (SOC) scale. These
measures correlated significantly (p = .53), and alpha for
the combined measures was .77. Rotter’s LOC measure was
also correlated highly with the combined measure (; = .96)
and did not correlate significantly with the Personal Power

Scale at either Times I or II. Paulhus’s SOC, on the other

31

hand, was less well correlated with the combined measure (p
= .73) but correlated significantly with PPS (p =.50) at
Time II. Since Rotter’s LOC measure had greater internal
stability than both the SOC and the combined measure and
accounted for 92% of the latter’s variance, but was found
unrelated to the PPS, it was used in subsequent analysis,
rather than SOC or their combination.

The Group Behavior Rating Scales (GBRS) reliabilities
reflected the orthogonal nature of the two acceptance
dimensions. Within both ARS and ARO, internal consistency
was high regardless of source or occasion of ratings. Thus
Cronbach alphas ranged from .84 to .93 for the ARS items and
from .77 to .93 for the ARO subscales. In contrast, the
correlations between measures of ARS and ARO were not

significant on either occasion as shown in Table 1.

Evaluation pf Hyppthgggg

Multiple hierarchical regression analyses were utilized
to assess the tenability of the hypotheses. Since four
separate groups provided data, mean ratings were computed
and standardized by group. Measures were then pooled across
groups utilizing the, group’s standardized scores. Group

means and variances were equated to minimize treating

Table 1

Internal consistency and interset correlations for
measures of acceptance versus rejection of self (ARS) and
other (ARO) as rated by self, others. and reflected self at

Times I and II.

 

 

 

 

 

 

ARS‘ ARO‘ ARS g AROb
Time I .93 .92 .28 (pg)
_§lf R‘tingg
Time II .93 .93 .13 (pg)
Time I .92 .92 .03 (pg)
figflgcted Rating;
Time II .91 .89 -.28 (pg)
Time I .84 .85 .02 (pg)
Rating; by Others
Time II .89 .77 -.38 (pg)

‘lnternal consistency as measured by Cronbach’s alpha.

I'Pearson p between ARS & ARO.

33
between-group variance as within-group variance due to the
tendency of different groups to use different parts of the

scale for their ratings (i.e., some groups tended to rate

all members high, others tended to rate all members low,
etc.). Because of the orthogonal character of the two
acceptance dimensions, all analysis were independently

performed for each dimension.

Hypothesis 1a stated that the reflexive ratings would
relate more strongly to self-ratings than would the“
individual’s ratings by others. This was well confirmed by
the ARS data. Reflected ratings accounted for at least 62%
(L = .78) of the variance in self-ratings at Time I and 72%
(p = .85) at Time II. In contrast, others ratings of
individuals only accounted for about 35% (5’3 of .58 and
.60) of the self-ratings variance on each occasion (for p =
27, p < .01 when £1: 1.57). For the ARS dimension, then,
reflected ratings accounted for about twice as much of the
variance in self-ratings as did the parallel ratings by
others. The comparable ARO findings were mixed. Reflected
ratings accounted for only about 22% (L’s of .42 and .49) of
the variance in self-ratings. Ratings from others accounted
for 25% (p = .50) and 11% (p = .33) of the same variance at
Times I and II. Hypothesis 1a was supported by the ARC data
at Time II, but not at Time I when ratings of individuals by
others accounted for 7% more of the variance in self-ratings
than did reflected ratings.

Hypothesis 1b stated that reflected ratings would

34
mediate the relationship between ratings by others and
self-ratings. This relationship was confirmed on each
occasion by ARS data. Ratings by others accounted for
significant amounts of the outcome variance of both self and
reflected ratings. More specifically, ratings by others
accounted for an average of 35% of the variance in
self-ratings and an average of 33% of the variance in
reflected ratings. When reflected ratings were the first
independent variable in the regression, however, ratings by
others did not significantly contribute to the
predictability of self-ratings. If averaged for Times I and
II, reflected ratings accounted for 31.5% more of the
variance of self-ratings than did ratings by others. It is
important to note that this finding satisfies Kinch’s (1983)
criterion of the relationship between self, other, and
reflected ratings from the symbolic interactionist
perspective. Ratings by others did correlate significantly
with reflected ratings, but reflected ratings accounted for
nearly twice as much of the variance in self-ratings than
did ratings by others. In addition, the variance accounted
for by others’ ratings was reduced to zero when reflected
ratings were the first independent variable entered in the

regression equation.

Table 2

Hierarchical

and reflected-ratings regressed
versus rejection of self (ARS) and
and II.
Self Reflected

Ratings Ratings'
ARSi dep .62"
ARS2 dep .72H
AROi dep .18
AR02 dep .24'
ARSi dep ---
ARSZ dep ---
AROi dep ---
AR02 dep ---
ARSi --- dep
ARSZ --- dep
AROi --- dep
AR02 --- dep

‘ Entries show cumulative adjusted

“ p < .08.

' p < .05.

"p < .01.

35

regression equations with

self-,

on measures of acceptance

other

Ratings
by Others‘

.83

.34"

.36°'

.25”

.11'

.35"

.13'

.12'

R2.

(ARO)

at Times I
Full
Eguation
F = 20.86H
F = 32.19H
F = 5.75'
F = 4.69‘
F = 12.78H
F = 14.93"
F = 8.45'
F = 4.26“
F = 11.34H
F = 16.25"
F = 4.45'
F = 6.74'

36

Similar to the findings for hypothesis 1a, ARO again
accounted for less variance than ARS and provided only mixed
support for hypothesis 1b. Ratings of individuals by others
did account for significant portions of the variance in both
self and reflected ratings, though the Time II value for the
full equation for ratings by others regressed on
self-ratings closely (p < .06) approached significance. At
Time II, the same relationship between self, reflected, and
other ratings was found as was described for the ARS
dimension, supporting the stated hypothesis. Again,
however, the variance accounted for was much below that
found for ARS. Also, the Time I ARO data did not support
hypothesis 1b, as reflected ratings did not account for a
significant portion of the self-ratings variance, while
ratings by others did.

Hypothesis 2 stated that the association between self,
other, and reflected ratings described in hypothesis 1b
would be significantly stronger at Time II than at Time I.
For ARS, reflected ratings accounted for an additional 10%
of the variance over Time I. When regressed on both self
and reflected ratings, ratings by others showed a slight
tendency to account for more variance at Time II (an
additional 2% and 4% on self and reflected ratings,
respectively). For ARO, support was again mixed. Although
reflected ratings accounted for an additional 6% of the
variance at Time II, ratings by others accounted for less

variance at Time II when regressed on both self and

(n
\I

reflected ratings.

Hypothesis 3 stated that individuals will attend more
to the reflected ratings of those persons they perceive as
more powerful as compared to those of individuals they view
as less powerful. Operationally, individuals rated as less
powerful by others should attend to the reflected ratings of
those rated as more powerful when compared to those powerful
others (see Discussion). Thus we would expect that the
amount of variance in self-ratings accounted for would
increase as individuals decreased in power. For ARO items.

power accounted for an additional 13% of the outcome

variance at Time II, raising the total variance in
self-ratings accounted for to 37% (p = .61). Power was not
significant, however, at Time I. Interestingly, for ARS

items this relationship was highly significant, but contrary
to the direction predicted. Whereas decreasing power led to
additional ARO variance accounted for, increasing power led
to additional ARS variance accounted for. For ARS, power
significantly (p < .01) accounted for an additional 13.5%
and 7.5% of the variance in self-ratings at Times I and II,
respectively. This raised the total variance accounted for

to 74% (p = .86) for Time I and 78% (L = .88) for Time II.

Table 3

Hierarchical regression equations for measures of
acceptance versus rejection of self (ARS) and others (ARO)
with self-ratings as the dependent variable and locus of
control (LOC), personal power (PPS), reflected ratings, and

ratings by others as the independent variables.

Self Reflected Other Full
Ratings Ratings‘ LOC‘ PPS: Ratings‘ Eguation

 

 

ARSi dep .62““ .73““ .81““ .80 F = 24.49““
ARS2 dep .72““ .73 .79““ .79 F = 25.11““
AROi dep .18 .28 .27 .32 F = 3.83“
AR02 dep .24“ .25 .41“ .40 F = 5.03“
ARSi dep .62““ --- .74““ .73 F = 21.69““
ARS2 dep .72““ --- .78““ .78 F = 28.86““
AROi dep .18 --- .15 .30“ F = 4.35“
AR02 dep .24 --- .37“ .39 F = 6.15'
ARSi dep .82““ .73“ --- .72 F = 20.51““
ARS2 dep .72““ .73 --- .72 F = 22.54““
AROi dep .18 .28 --- .30 F = 4.29“
AR02 dep .24“ .25 --- .22 F = 3.30

‘ Entries show cumulative adjusted R‘-
“ P < .05.

““p < .01.

38

39

Hypothesis 4 stated that self-ratings would change
toward reflected perceptions of those viewed as more
powerful as compared to those who are less powerful. If
this were the case, the variance in self-ratings accounted
for by the power variable at Time II should significantly
exceed that at Time I. This hypothesis was not supported
for ARS. For ARO at Time II, however, a trend toward
support was found, with power accounting for an additional
22% of the variance of self-ratings.

Hypothesis 5 stated that individuals having a more
external locus of control will have reflected perceptions
more closely linked to their self-perceptions than will
persons having a more internal locus of control. No support
was found for this hypothesis. Though a significant (p <
.01) relationship was found at Time I for ARS, this was
opposite to the predicted direction. It was found that
those with a greater internal locus of control had reflected
perceptions which more closely paralleled self-perceptions.
In this case, locus of control accounted for an additional
11.5% of the variance in self-ratings with reflected ratings
first in the regression equation.

Hypothesis 8 stated that the relationship in hypothesis
5 would be significantly stronger at Time II than at Time I.
This hypothesis was not supported.

Hypothesis 7 stated that locus of control and power
would jointly increase the predictive ability of the

reflexive relationship, and that this would be more apparent

40
at Time II than at Time I. The first part of this
hypothesis was supported for ARS at Time I, with partial
support at Time II. At Time I, the four independent
variables accounted for 80% (p = .89) of the variance in
self-ratings, with significant contributions by each
variable except ratings by others. One drawback to this
finding was that the contributions of locus of control and
personal power were significant, but contrary to the
direction predicted. At Time II on ARS, the full equation
accounted for 79% (p = .88) of the variance of self-ratings,
but the LOC measure failed to contribute significantly (p =
.10). The hypothesis was not supported for ARO.
Furthermore, neither the ARS or ARO data supported the
hypothesis that significantly more variance would be
accounted for at time II than at Time I when all variables

were present in the regression equation.

DISCUSSION

The symbolic interactionist perspective asserts that we
come to view ourselves as we think others see us. Stated
another way, others’ views of us influence self-perceptions
through reflected perceptions. An empirical test of this
formulation must show that reflected perceptions are related
to self-perceptions. It must further be shown that
perceptions by others are related to reflected perceptions
as well. This implies an important difference in the

strength of these relationships which has been largely

41

neglected and misunderstood (Kinch, 1963; Rosenberg, 1981;

Schafer & Keith, 1985). Reflected and self-perceptions
should be linked more strongly than accorded and
self-perceptions. Indeed, accorded and self-perceptions
need not be directly related at all. Rather, accorded

perceptions must correlate significantly with reflected
perceptions if the other’s actual views are to have any
influence upon the individual’s concept of self.

The present findings amply support these aspects of the
symbolic interactionist viewpoint. For ARS and ARO (only
Time II ARO), the correlations between reflected perceptions

and self-perceptions were about twice as strong as those

between accorded perceptions and self-perceptions. It was
also found that others’ actual ratings of individuals
correlated significantly with reflected perceptions.

Especially convincing was the finding that others’ actual
ratings failed to account for any of the variance in
self-ratings when reflected ratings were the first
independent variable in the regression equation and ratings
by others was second.

This study also investigated the use of personal power
ratings. Initially it had been hypothesized that
individuals would be more influenced by the reflected
perceptions of those whom they viewed as more powerful as
compared to those seen as less powerful. Due to an

unforeseen statistical complication this hypothesis was not

directly testable because it required that the data be

42

organized by raters, since it is the perceived power of each
other group member that was hypothesized to influence which
reflected perceptions the individual would give higher
priority. The problem arose from the organization of data
by raters rather than by ratees. Unless the data are
evaluated by individual case, the power ratings will lack
,internal consistency. If individuals ratings of peers’
personal power agree, however, the data may be organized by
rates and the hypothesis can be rewritten. High interrater
agreement about PPS ratings permitted this reformulation and
it was hypothesized that individuals rated as less powerful
by others will link their perceptions of self more closely
to their reflected perceptions than will individuals rated
as more powerful. This is because less powerful individuals
will tend to look more to others whom they regard as
powerful for their self-definition than to those others whom
they view as less powerful.

This hypothesis was supported for the ARO items at Time
II, but interestingly, was directionally contrary to that
predicted for ARS items. On the self-acceptance dimension,
persons viewed as more powerful by others tended to have
self-definitions that corelated more strongly with the
reflected perceptions of others than did those viewed as
less powerful. Apparently, highly self-accepting persons
tended to more readily accept their own views of how others
see them, while less self-accepting persons were more

hesitant about accepting their own views of how others

43
regard them. This finding seems congruent with Sullivan’s
(1953) interpersonal theory and also fits related findings
from recent studies by Hurley (1986a) and Hurley & Myers
(1987).

Sullivan (1953) stated that people become anxious when
they perceive external threats to their self-system. These
individuals then utilize a variety of interpersonal security
operations to maintain or reestablish a feeling of safety

and to ward-off anxieties aroused by these adverse messages.

The present findings suggest that persons low in
self-acceptance tend to experience more interpersonal
anxiety than those higher in self-acceptance and

consequently shift more quickly into security operations
than do more self-accepting persons. These quick shifts are
thought to lessen the impact of adverse or threatening
reflected perceptions as these persons likely either
discount or distort reflected information. Hurley & Myers
(1987) provided indirect support for this notion. They
found that individuals’ ranges of ratings of peers and self
correlated inversely with defensiveness when it was known
that all ratings would be fully disclosed. Restricted range

seemed an index of personal security in this context, with

narrower ranges, indicating a reduced set of
discriminations, accompanying a greater need for
self-protection. Hurley (1988a) earlier reported that users
of restricted ranges also tended to inflate their

self-ratings as compared to others’ ratings of them. He

44

stated, "Considering that such persons [low range users]
commonly show personal adjustment problems, this observation
suggests that very narrow range users tend to manifest an
exaggeratedly self-protective stance" (p. 226). In terms of
security operations, Hurley suggested that very narrow range
raters seek to avoid interpersonal tensions through a
nonassertive and overly-deferential stance. This contrasted
with their rating themselves as much more assertive and
self-accepting than they were rated by their group peers'.
Taken together, these findings provide generalized support
for the notion that individuals regarded as relatively low
in power by others tend to more readily resort to security
operations to distort or discount their reflected
perceptions than do persons regarded as more powerful.

No support was found for the hypothesis that the
reflexive self would correlate more highly with the
self-concept as individuals adopted an increasingly external
locus of control. Instead, an increasingly internal locus
of control was found significantly associated with the
variance in self-ratings. Thus, increased confidence that
one has more control over the outcome of interpersonal
events apparently enhanced one’s ability to accept their
reflected perceptions of others for the ARS items. This
finding seems congruent with the argument cited for personal
power on the self-acceptance measure. That is, feeling more
in control of one’s experiences was associated with a lesser

need to distort or discount the reflected perceptions of

45
others, whereas the perception of external control
apparently enhanced interpersonal anxiety because others
appear to have greater control in determining one’s
self-definition.

One aspect of this study that reaches beyond prior
works was the attempt to measure changes in the self-concept
as a function of reflected perceptions of others.
Specifically it was thought that all variables would account
for more of the variance in self-ratings at Time II than at
Time I. Of eight relevant trends in the data, five
supported and two opposed this hypothesis, although none
reached generally accepted standards of statistical

significance.

There seem at least three plausible reasons for this
outcome. First, the small number of study participants (27)
may have made it difficult for the between-group differences
to exceed the within-group variance. Real differences
between the measures at Times I and II may have been
overshadowed by instabilities inherent in this research
design. Second, the first assessment was taken after 22
hours of rather intense interpersonal interaction within

these groups. This measurement may have occurred after the

Table 4

Shifts in the variance of self-ratings accounted for
from Time I to II on measures of acceptance versus rejection
of self (ARS) and others (ARO) with reflected ratings, locus
of control, personal power, and‘ ratings by others as

independent variables.

 

 

 

Time 1‘ Time II“ Shift
ARS .82 .72 +.10
Raflapgad Rating;
ARO .18 .24 +.06
ARS .73 .73 .00
Locus of Control”
ARO .28 .25 -.03
ARS .74 .78 +.04
Personal Power“
ARO .15 .37 +.22
ARS .34 .38 +.02
Ratings by Others
ARO .25 .11 -.14

“ Entries show cumulative adjusted R2.
“ These variables were second in the regression equation

after reflected ratings.

46

47
reflexive relationship had already developed. Third, there
may not have been any real changes.

In the present context, the first two alternatives seem
more plausible. That five of eight possible shifts were in
the predicted direction (reflected ratings on ARS & ARO,
personal power on ARS & ARO, accorded ratings on ARO) versus
only two contrary shifts (LOC on ARO and accorded ratings on
ARO), suggests that the differences were real but suppressed
for the above-mentioned reasons (see Table 4). Obviously, a
larger study is needed to clarify this issue.

Possibly the most interesting and unexpected finding
was the differential patterns of correlation yielded by the
self-acceptance and other-acceptance scales. While the
basic interactionist relationship held as predicted for ARS
items at Times I and II, and for ARO items at Time II, an
appreciable difference (p < .01) existed between the two
kinds of acceptance. Reflected self-acceptance ratings
accounted for nearly twice the variance in self-ratings as
did the other-acceptance relected ratings. While this
difference was unexpected, a perusal of pertinant literature
concerning these measures suggested that individuals
commonly respond quite differently to these self~ and
other-acceptance measures. The Hurley & Marsh (1986) study
of the mutual eye contact (MEC) correlates of self- and
other-acceptance found that early and late peer-based
ratings of self-acceptance and total MEC both correlated .55

although MEC correlated only .34 early and .38 late with

48

other-acceptance ratings. MEC also correlated somewhat more
firmly with self-rated self-acceptance and peer-rated
self-acceptance (.76 early and .83 late), than with
other-acceptance ratings by self and peers (.74 early and
.48 late). Agreement between peer and self-based ratings on
other-acceptance items also showed a puzzling decline as
members became more familiar and had more experience with
one another.

There is also indirect evidence that individuals’
regard being perceived as quite accepting of others as more
important than being seen as quite self-accepting. A study
by Hurley and Rosenberg (1986) found that member-based
ratings of leaders on the ARO items warm--cold and
gentle--harsh were the best predictors of their group’s

members’ net gains on both self-acceptance (respective L’s =

.61 & .62) and other-acceptance (respective p’s .71 &
.63). An earlier study of two-day psychodynamic groups of
mental health professionals (Hurley, 1986b), yielded

strikingly similar results as member-based ratings of

leaders on the warm--cold ARO subscale also generated the

strongest significant correlations with members’ net gains
on both self-acceptance (a = .46) and other-acceptance (p =
.67).

Furthermore, it appears that members of small groups
are significantly less accurate in predicting how their
group peers view them on items assessing other-acceptance

(L’s = .36 early & .35 late) than on self-acceptance items

49
(p’s = .56 early & .59 late). Yet, the behavior of their
group leaders on items assessing other-acceptance have been
found to be consistently linked more strongly to group
members’ interpersonal gains than are leader behaviors
linked to self-acceptance. Two possible reasons for this
discrepancy come to mind. One is that because it is more
important to individuals to be perceived by others as
other-accepting than as self-accepting, group members may
have a greater need to distort their ideas of how others see
them for other-acceptance than for self-acceptance. Various
ego-involvement effects, reviewed by Greenwald and Pratkanis
(1984), such as beneffectance are illustrative of
individuals’ general bias to protect their self-esteem. In
the interpersonal context, it may be more important for
individuals’ to be seen as warm, gentle, helping, and
accepting of others than as showing of their own feelings,
active, expressive, or dominant.

Another possible explanation for the lesser
correlations between self, other, and predicted ratings for
other-acceptance (as compared to the relationships for
self-acceptance) is that people may have vaguer notions of
what comprises other-accepting behaviors. One way of
stating this is to say that people may not possess very
clear schemata for the ARO scale items. Interestingly,
Markus’s primary evidence for self-schemata across various
publications (Markus, 1977; Markus, Crane, Bernstien. &

Siladi, 1982; Markus, Smith, & Moreland, 1985) include

50

schematization of independence--dependence and masculinity,
traits (especially independence) which appear highly
congruent with ARS, but irrelevant to ARO. Markus’s work
has also focused on differences among individuals on a
single self-schema. No research addressing the possibility
that various classes of evaluative traits may vary in their
ability to be schematically integrated is known to this
author. One possible reason for the presently noted
differential relationships for self- and other-acceptance,
then, is that individuals may be more cognitively vague or
less well schematized for other-acceptance variables than
for self-acceptance variables.

Markus (1977) stated that if a person has a developed
self-schema, she or he should be able to: (1) process
information about the self in a given domain with relative
ease, (2) retrieve behavioral evidence from that domain, (3)
predict future behavior in the domain, and (4) resist
counter-schematic evidence about himself. The present
findings indicate that individuals process information about
self-acceptance more accurately than they process
information about other-acceptance. This indirectly
satisfies the first of Markus’s criteria. Obviously, if you
cannot predict how others see you, you cannot do it easily.
To see if persons satisfy the remainder of these criteria
differentially for ARS and ARO, and to see if the
other-acceptance dimension is more important to people in

their interactions with others than is the self-acceptance

51
dimension, the following exploratory study was conducted.

17 students enrolled in Psychology 400 at Michigan State
University (Spring term, 1988) were simply asked to rate how
important it would be to them that their close friends
perceive them positively on each item of the ARS and ARO
scales. These eight bipolar items were intermixed and
presented on the blackboard in class. It was found that
students rated the other-acceptance items as significantly
more important (p < .05) than the self-acceptance items.
Two weeks later, without intervening discussions of this
topic, the same group of students were asked to record three
of their own behaviors in the last two weeks that
represented each of the eight ARS and ARO items. They were
then asked to rank-order these eight items for how easy it
was to supply a behavioral example for each item (these data
were gathered the following week due to time constraints).
Two members’ responses were unscorable. For the remaining
15 students a trend was found (p < .06) favoring the ease of
identifying the self-acceptance items.

This simple study provides some support for both the
hypothesis that it is more important for individuals to be
seen positively on other-acceptance than on self-acceptance
and that it is more difficult for them to define what
behaviors actually constitute the other-acceptance
dimension. Though these findings are preliminary, they do
suggest an explanation for the large discrepancy between the

variance-in self-ratings accounted for by ARS and ARO in the

52
initial study. Certainly a more comprehensive test of these

hypotheses is warranted.

Concluaion

Between self, other, and reflected perceptions, the
present findings supported the symbolic interactionist
perspective. It was found that individuals’
self-representations can apparently be explained largely by
how they think others see them. How individuals’ think they
are seen can be partially (and significantly) explained by
how others actually say they see them. It was also found
that the concepts of personal power and, to a lesser extent,
locus of control clarified these relationships. Less
powerful individuals apparently look to more powerful
persons to confirm their other-accepting view of themselves.
In contrast, for the self-acceptance dimension as an
individual is more powerful he or she is more able to
accurately utilize reflected perceptions of others. In this
case the powerful individual may not feel as anxious in
accepting reflected feedback from others and subsequently

has less need to distort or discount this information.

Locus of control seemed to function similarly for
self-acceptance. A more internal locus of control seems to
allow the individual to feel less anxious about accepting

reflected perceptions as valid and contributory.
A striking and unexpected difference was found between

the measures of self-acceptance (ARS) and other-acceptance

53

(ARO). The personal power variable yielded opposite
patterns of correlation with these measures. There was also
a large difference between the variance in self-ratings
accounted for by items of the self-acceptance and
other-acceptance measures. On both occasions the full
regression on self-acceptance accounted for about 80% of the
variance in self-ratings, while the parallel

other-acceptances regressions accounted for only 32% and 40%

of the variance. In. an exploratory follow-up study, two
possible explanations for this discrepancy were
investigated. It was found that individuals thought it was
more important to be seen positively by others on

other-accepting qualities, while at the same time there was
a tendency for them to have more difficulty in defining
other-accepting behaviors than self-accepting qualities.
Thus, individuals may have vaguer and less schematic
perceptions of the behaviors undergirding other-acceptance
items. Perhaps because this dimension is more important to
individuals, it is less well defined. Anxiety and security
operations may disrupt accurate feedback. Certainly this is
a question future research should decide.

The present study had several advantages over previous
research in this area as well as some limitations.
Utilizing small, interpersonally-oriented groups allowed
data to be gathered representing the perceptions of
individuals by and toward several others. This created a

great deal of stability when perceptions were combined and

54
this agreement among individuals in their perceptions of

others Isupported the reliability and validity of the

constructs used. This research also benefitted from prior
studies which identified correctable errors in
operationalizing the symbolic interactionist construct.

Using the additional variables of personal power and locus
of control also clarified of our understanding of factors
which might facilitate a more accurate use of the symbolic
interactionist paradigm by individuals. Probably the
biggest drawback in the present study was the small number
of participants from which data were gathered. It would
also have been more useful to gather the first set of
ratings earlier in the development of these groups, when
reflected perceptions were still in a very early stage of

development.

APPENDICES

Appendix A

The SOC’a Interpersonal Scale

 

 

Instructioqa: In the space next to each statement, indicate
the extent to which the sentence applies to you using the
following key:

NEITHER AGREE
STRONGLY DISAGREE NOR DISAGREE STRONGLY AGREE
1 2 3 4 5 6 7

1. Even when I’m feeling self-confident about most
things, I still seem to lack the ability to control social
situations.

2. I have no trouble making and keeping friends.

3. I’m not good at guiding the course of a conversation
with several others.

4. I can usually establish a close personal relationship
with someone I find attractive.

5. When being interviewed I can usually steer the
interviewer toward the topics I wish to talk about and away
from those I wish to avoid.

6. If I need help carrying off a plan of mine, it’s

usually difficult to get others to help.

7. If there’s someone I want to meet I can usually
arrange it.

8. I often find it hard to get my point of view across
to others.

9. In attempting to smooth over a disagreement I usually
make it worse.

10. I find it easy to play an important part in most
group situations.

 

SS

Appendix B

 

Scale

Powar

 

The Personal

Rate each person in your group on each of the

dimensions.
other person and try to

Instructions:
ratings.

following

own
full

perceptions of each

 

your

Use
use

of possible

hold these ratings in

range

the
the research will

Remember,

complete confidence so rate as honestly and as accurately as

you can.

Influance

 

Amount of

 

Amount of Power

aaaaaaaaaa

3333883888.

b b b b b b b b b b

I

p
0
W

b b b b b b b b b b

p
0
U

c c c c c c c c c c
d d d d d d d d d d
e e e e e e e e e e

N
N F
O L
T

c c c c c c c c c c
d d d d d d d d d d
e e e e e e e e e e

E L

E

E

UEN

f f f f f f f f f f

U
E
N

R
L
f f f f f f f f f f E

R
F
U

TIA
as h
8 h
8 h
8 h
8 h
as h
as h
8 h
8 h
8 h
T A
S

8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
L

L

i i i i i i i 1 i i

L

i i i i i i i i i i

J J J J J J J J J J

J J J J J J J J J J

Tellin /Assertion

of Others

Control

 

a a a a a a a a a a
b b b b b b b b b b

a a a a a a a a a a
b b b b b b b b b b

U
N
c c c c c c c c c c A

A
S
S

C

O O
N T
T H

c c c c c c c c c c
d d d d d d d d d d
e e e e e e e e e e

CON

SSERTIVE

d d d d d d d d d d
e e e e e e e e e e

E

f f f f f f f f f f

R
T
I

R E
O R
L S

f f f f f f f f f f

R O
O T
L H

as h
8 h
8 h
8 h
as h
8 h
8 h
as .n
8 h
8 h
VE
S

8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
as .n
ERS
LED

1 i i i i i

i
J J J J J J J J

i i i

J

i i 1 i i i i i i i

J

J J J J J J J J J J

57

 

Freguency of Concessions

Time

Talkin

(gives in to others)
a a a a a a a a a a
b b b b b b b b b b

I

F
c c c c c c c c c c R

N
F

CCCCCCCCCC

L
I

EQUE
d9 .1
de .1
.09 f
d e f
.08 .1
.09 f
de at
de 01
de f
de es
REQU
LOT
.09 f
.09 f
d O f
.09 at
.09 f
d e f
.09 f
de 4..
do. f
do. at
TTLE

8888888888 N
T

E
N

h h h h h h h h h h

1 1 1 1 1 1 1 1 1 1

T

h h h h h h h h h h

1 1 1 1 1 1 1 1 1 1

JJJJJJJJJJ

JJJJJJJJJJ

Interru tions

Demand

ing/Arguing

 

a a a a a a a a a a
b b b b b b b b b b

D
E

a a a a a a a a a a
b b b b b b b b b b

I

D

0:.
Cd
Cd
Cd
Cd
Cd
Cd
Cd
0.0
Cd
Cd

TE

0:.

RE

AN
Cd
Cd
Cd
Cd
Cd
Cd
Cd
Cd
Cd
Cd

"A

DO

RRUPTS
TEN

e .18 h
e f8 h
e foe h
9 fan .h
e foe h
8 f8 h
e foe h
9 f3 h
e 413 h
e 9.8 h
RRUPTE
TEN

QUENTL
DS/ARG
e foe h
8 .18 h
a fee h
8 f8 h
9 f8 h
9 f8 .n
9 f8 h
a fee h
a fee h
e f8 h
ND/ARG
ESNrT.

1111111111
1111131111

D

Y

U
E

1 1 1 1 1 1 1 1 1 1

U

JJJJJJJJJJ

E

Attire

Attractiveness

Ph sical

U

aaaaaaaaaa

aaaaaaaaaa

NFAS
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
FASH
ATT
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
.0 Cd
UNAT

HID
e f
e f
e f
e a...
e f
e f
e f
e f
e f
e f
ION
RAC
e f
e f
e f
e f
e f
e f
e f
e a...
e f
e f
TRA

N
A
B

8 8 8 8 8 8 8 8 8 8 T A 8 8 8 8 8 8 8 8 8 8
I B
h h h h h h h h h h V L h h h h h h h h h h

1 i i 1 1 1 1 1 1 1

JJJJJJJJJJ

C
T
I

L
E

1111111
11311111

1

E

V
E

j

J

58

 

Speech Fluency

 

Socioeconomic Status

(Apparent)
a a a a a a a a a a

8883838888

ELOQUENT

FLUID
.D Ode 4.8 h
b cde f8 h
b cde f8 h
b cde 18 h
b cde .18 h
b 6.0... T8 h
b ode f8 h
b cde .18 h
b cde 418 h
b Bud e f 8 h
HESITANT
HALTING

MORE UEALT
b eds 418 h
b 0.09 f8 h
b cde f8 h
b ncd e f 8 h
b cde .18 h
b nod e f 8 h
b Bed 8 f 8 h
b .cd 9 f 8 h
b cde f8 h
b Ode f8 h
LESS WEALT

1 1 1 1 1 1 1 1 1 1

JJJJJJJJJJ

H
Y

1 1 1 1 1 1 1 1 1 1

H
Y

JJJJJJJJJJ

 

Inter ersonal

Ideas

 

Expression of

a a a a a a a a a a
b b b b b b b b b b

a a a a a a a a a a
b b b b b b b b b b

N
O
V

CCCCCCCCCC

d d d d d d d d d d

E
X

p
R
E

I

CCCCCCCCCC

p
R
E

P

d d d d d d d d d d L S

L S

e e e e e e e e e e I
C
E

f f f f f f f f f f

E
R
T

L E
O N
G T

f f f f f f f f f f

O E
G N
I T

g h
8 h
8 h
8 h
g h
8 h
8 .n
g h
8 .D
co h
ATI
[CA
8 h
g h
g h
g h
g h
g .n
g h
8 h
8 h
8 h
ATI
CAL

1111111
1353311353

1

1

O
N

1 1 1 1 1 1 1 1 1 1 L

JJJJJJJJJJ

O
N

Persistence

Encircle

your name:

(a)

(b)

3833383883

 

 

p
E

b b b b b b b b b b

p

E
V R
E

(c)

(d)

CCCCCCCCCC

 

N S

d d d d d d d d d d

S

 

(e)
(f)

 

O
T
f f f f f f f f f f

R I
Y S
T

 

(g)

 

E
N

8 8 8 8 8 8 8 8 8 8

E
N

(h)

T

h h h h h h h h h h

T

 

(1)
(j)

i 1 1 1 i i

1 1 1 1

 

 

j

JJJJJJJJJ

Appendix C

Group Behavior Rating Scalaa for ARS and ARO

Instructions': On this minibooklet’s last page note that
all group members’ names have been listed. Encircle your
own name. Starting with the following page, encircle the
letter that best represents your personal impression of each
members’ actual behaviorppithinjall group sessiona up to
pap. These ratings will be most useful if you use the full
range of possible ratings for each scale.

Rate all group members, including self and leader(s).
Complete all ratings on aagh page befora turning ahead to
the next. Unlike other scales which address behaviors, the
Liked versus .Disliked scale solicits your personal

PESEOHSES .

Instructions“: Starting with the folowing page, encircle
the letter between the extremes that best represents how you
think aach member rates your behavior within the group up to
now. Please use the full range of possible ratings for each
scale. Do not rate yourself. All ratings will be kept
confidential. Complete all ratings on each page before
turning ahead to the next.

 

a a a a a a a a a a H a a a a a a a a a a S
b b b b b b b b b b D I b b b b b b b b b b H
L I D O
I c c c c c c c c c c S E c c c c c c c c c o W
K d d d d d d d d d d L S d d d d d d d d d d S
E e e e e e e e e e e I e e e e e e e e e e
D K F F
f f f f f f f f f f E E f f f f f f f f f f E
8 8 8 8 8 8 8 8 8 D E 8 8 8 8 8 8 8 8 8 E
h h h h h h h h h h I h h h h h h h h h h I
i i i i i i i i i i N i i i i i i i i i i N
JJJJJJJJJJ GJJJJJJJJJJS
S

59

60

aaaa
bbbb

aaaaaaaaaa
bbbbbbbbbb

E
cccccccccc X

G

C
O

cccccccccc
dddddddddd
eeeeeeeeee

W
A

p
R
E
S

d d d d d d d d d d
e e e e e e e e e e
f f f f f f f f f f

U
A
R
D

L
D
ffffffffff

R
M

51v
8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
ED

8 h
g h
8 h
8 h
8 h
8 h
8 h
8 h
8 h
8 h

E

1111111111

JJJJJJJJJJ

1111111111

JJJJJJJJJJ

aaaaaaaaaa
bbbbbbbbbb

aaaaaaaaaa H

H

p
A
S

CCCCCCCCCC

A
C

bbbbbbbbbb A
R
M

CCCCCCCCCC

E
L
p

dddddddddd

T

dddddddddd S

S

SIV
e f.
e f
e f
e f
e f
e f
e f
e f
e f
e f
IVE
0T.
9 f
e f
9 f
e f
e f
e f
O f
e f
e f
e .1
0T

E

g h
8 h
8 h.
g h
06 h
8 H...
g h
g h
8 h
an .0
HER
E h
8 .n
8 h
8 h.
g h
8 h
8 h
g h
8 h
8 h
HER

1111111111

11111111115

JJJJJJJJJJ

S

J

J

J

8888888888

8888888888

bbbbbbbbbb

S
U
B

bbbbbbbbbb

D
O

CCCCCCCCCC

G
E
N

CCCCCCCCCC

H .
A

dddddddddd M

M

dddddddddd

R

INA
e f
e f
e f
e f
e f
e f
e f
e f.
e f
e f
ISS
TLE
e f
e f
e f
e f
e f
e f.
e f
e f
e f
e .7.
SH

N
T

8888888888

I
V
E

8888888888

hhhhhhhhhh

h h h h h h h h h h
1 1 1 1 1 1 1 1 1 1
J J J J J J J J J J

1111111111
313351115)

61

Encircle
your name:

(a)

(b)
(c)

a a a a a a a a a a R

A

E
J
E

b b b b b b b b b b
c c c c c c c c c c
d d d d d d d d d d
e e e e e e e e e e

C
C
E

(d)

(e)

C
T

p
T

(f)

f f f f f f f f f f
8 8 8 8 8 8 8 8 8 8

(g)
(h)
(i)
(j)

0
T
h h h h h h h h h h H

1 1 1 1 1 1 1 1 1 1
JJJJJJJJJJ

O
T
H

E

E
R
S

R

‘Instructions for self and accorded ratings.

.Instructions for reflected ratings.

L I ST OF REFERENCES

LIST OF REFERENCES

Adams, H.B. (1984). ”Mental illness” or interpersonal
behavior? American Paychologist, £51 191-197.

Allen, C. (1984). On the validity of relative validity
studies of ”final say” measures of marital power.

Journal of Marriage and the Family, aa, 619-629.

Allport, G.W. (1961). Patterns and gropth in personality.
New York: Holt, Rinehart, and Winston.

Anastasi, A. (1982). Psychological taating. New York:

 

McMillan.
Argyris, C. (1962). Interpersonal competence and
organizational effgctiveness. Homewood II:

 

 

Irwin-Dorsey.

Ashkanasy, N. (1985). Rotter’s internal-external scale:

 

 

Confirmatory factor analysis and correlation with
social desirability for alternative scale formats.
Journal of Personality pand Social Ps cholo , 58(5),
1328-1341.

Bales, R.F. (1970). Personal and interpersonal behavior. New
York: Holt, Rinehart & Co.

Barnes, M., & Rosenthal, R. (1985). Interpersonal effects of

 

experimenter attractiveness, attire, and gender.
Journal of Personality and Social Psychology, p§(2),
435-446.

Benjamin, L. (1979). Structural analysis of differentiation
failure. Psychiatry, 5;, 2-23.

Benson, P., Karabenick, S., & Lerner, R. (1976). Pretty
please: The effects of physical attractiveness, race,
and sex on receiving help. Jourpala of Experimental
Social Psychology, pg, 409-415.

Berne, E. (1966). Principlea of group treatment. New York:
Grove, Press.

Bierman. R. (1969). Dimensions of interpersonal facilitation

in psychotherapy and child development. Psychological
Bulletin, 2;, 338-352.

62

83

Bord, R. (1975). Toward a social-psychological theory of
charismatic social influence processes. Social Forcaa,
22(3), 485-497.

Carson, R. (1969). Interaction concepts of ,pegaonality.
Chicago: Aldine.

Cash, T., Begley, P., McCown, D., & Wise, B. (1975). When
counselors are heard but not seen: The initial impact

 

of physical attractiveness. Journal of Counaaling
Psychology, 22(4), 273-279.

Chance, E. (1954). Father’s perception of self and
first-born child. In L.M. Stotz (Ed.), Father
relationa_aof war-born childrap. Stanford: Stanford
University Press.

Cherulnik, P., Neely, W., Flanagan, M., & Zachau, M. (1978).
Social skill and visual interaction. Journal of SociaL
Psychology, 104, 263-270.

Collins, B. (1974). Four components of the Rotter
internal-external scale: Belief in a difficult world, a
just world, a predictable world, and a politically
responsive world. Journal, of Personality and SociaL
Psychology, 22, 381-391.

Conte, H., & Plutchik, R. (1981). A circumplex model for

 

interpersonal personality traits. Journal of
Personality and Social Psychology, pp, 701-711.

 

Cooley, C. (1902). Human nature and the social ordap. New
York: Schocken Books.

Cowan, G., Drinkard, J., & MacGavin, L. (1984). The effects
of target, age, and gender on use of power strategies.
Journal of Personality and Social Paychology, p1(6),
1391-1398.

Cromwell, V., & Cromwell, R. (1978). Perceived dominance in
decision-making and conflict resolution among anglo,
black, and chicano couples. Journal of Marriage and the
Family, pp, 749-759.

Dahl, R. (1957). The concept of power. Behavioral 5918908,
2, 201-215.

Dovido, J., & Ellyson, S. (1982). Decoding visual dominance:
Attributions of power based on relative percentages of
looking while speaking and looking while listening.
Social Ps cholo uarterl , 32(2), 108-113.

64

 

Egan, G. (1978). Interpersonal living: Agakillaicontract
approach to hupan relationa training. Monterey:

 

 

Brooks/Cole.

Erickson, 8., Lind, E., Johnson, 8., & O'Barr, W. (1978).
Speech style and impression formation in a court
setting: The effects of ”powerful" and "powerless"
speech. Journal_of Experimental Sogial Paychology, Li,

286-279.
Falbo, T. (1977). The multidimensional scaling of power
strategies. Journal of Personality and SociaL

 

 

Psychology, 22, 537-548.

Falbo, T., & Peplau, L. (1980). Power strategies in intimate
relationships. Journal of Personality and SociaL
asychologx. 38(4), 618-628.

Farina, A. (1980). Patterns of role dominance and conflict
in parents of schizophrenic patients. Journal of
Abnormal and Social Paychology, 6;, 31-38.

Fast, J. (1977). The Body Language of Sax. PowerL and
Aggression (pp. 91-123). New York: M. Evans & Co.

Foa, U. (1981). Convergences in the analysis of the
structure of interpersonal behavior. PsychologicaL
Review, 22, 341-353.

Freedman, M.B.. Leary, T.F., Ossorio, A.G., & Coffey, H.S.
(1951). The interpersonal dimension of personality.

Journal of Personality, 29, 143-181.

 

 

 

 

Gavrilides, G. (1980). The relatiopahip of personal power
functionai to general happiness, intarpersonal risk,
interpersonally inducad anxiatpr, and security

 

operations. Unpublished doctoral dissertation, Michigan
State University.

Gerstenhaber, L. (1975). Acceptance versus rejection of
others and self in personality self-reports. (Doctoral
Dissertation, M.S.U., 1974). Dissertation Abstracts
InternagiopaL, 22, 338-352. .

Gibb. J.R. (1964). Climate for trust formation. In L.P.
Bradford, J.R. Gibb, and K.D. Benne (Eds.), T-group
theoryaand laboratory method (pp. 279-309). New York:
Wiley.

Gray-Little. B. (1982). Marital quality and power processes
among black couples.‘ Journal of Marriage and tha
Family, pp, 633-648.

85

Gray-Little, B., & Burks, N. (1983). Power and satisfaction
in marriage: A review and critique. anchological
Bulletin, 22(3), 513-538.

Greenwald, A., & Pratkanis, A. (1984). The self. In R. Wyer
& T. Srull (Eds.), Handbook of aocial cggnition (Vol.
3). Hillsdale, NJ: Erlbaum.

Heider, F. (1958). The p ychology of interpersonal

s
relations. New York: Wiley.

Hunter, J. (1977, March). Cluster analysis: Reliability,
construct validity. and the multiple indicatora.

approach to measurement. Paper presented at a workshop

of the United States Civil Service commission.

Hurley, J. R. (1976a). Helpful behavior in groups of mental
health professionals and undergraduates. International

Journal of Group Psychotherapy, 22, 173-189.

Hurley, J. R. (1978b). Two prepotent interpersonal
dimensions and the effects of trainers of t-groups.

Small Grpup Behavior, 2, 77-98.

Hurley, J. R. (1978). Toward seeing ourselves as others see
us. In L.R. Wolberg, M.L. Aronson, & A.R. Wolberg

(Eds.), Group therapy 1978 (pp. 105-118). New York:
Stratton Medical Books.

Hurley, J. R. (1983). [Correlations among measures of the
principle dimensions of interpersonal behavior].

Unpublished raw data.

Hurley, J. R. (1986a). Interpersonal behavior, range of
ratings, and personal security. anchological Raporta,
22, 219-228

Hurley, J. R. (1986b). Leaders’ behavior and group members’
interpersonal gains. Group, 29(3), 161-176.

Hurley, J. R., & Marsh, V.A. (1986). Contrasting

 

interpersonal correlates of mutual eye-contact reports
by self and others. Perceptual and Motor Skilla, 82,
1287-1274.

Hurley, J. R., & Myers, A. (1987). Defensiveness: Another
interpersonal correlate of range of ratings.

Psychological Reporaa, 89, 187-190.

Hurley, J. R., & Rosenberg, D. (1986). Leader behaviors

associated with group members’ interpersonal gains.
Unpublished manuscript, Michigan State University.

 

88

Jacobson, W. (1972). Power and interpersonal relations.
Belmont, 0a.: Wadsworth Publishing Co., Inc.

 

Joe, V. (1971). Review of the internal-external control
construct as a personality variable. PsychologicaL
Reports, 22, 619-640.

Kaplowitz, S. (1978). Towards a systematic theory of power
attribution. Social Psychology, £2(2), 131-148.

Kiesler, D.L. (1983). The 1982 interpersonal circle: A
taxonomy for complementarity in human transactions.

Psychological Review, 22, 185-214.

Kinch, J. (1963). A formalized theory of the self-concept.
American Journal of Sociology, 22, 481-488.

LaForge, R., & Suczek, R. (1955). The interpersonal
dimension of personality: III. An interpersonal
checklist. Journal of Personality, 23, 94-112.

Lefcourt, H. (1966). Internal versus external control of
reinforcement. Psychological Bullatin, 22(4), 206-220.

Lorr, M., & McNair, D. (1965). Expansion of the
interpersonal behavior circle. Journal of Personality
and Social Paychology, 2, 823-830.

Markus, H. (1977). Self-schemata and processing information
about the self. Journal of Personality and SociaL

Psychology, 22, 83-78.

Markus, H., Crane, M., Berstein, 5., & Siladi, M. (1982).
Self-schemes and gender. Journal of Personality and
Social Psychology, 32, 38-50.

Markus, H., Moreland, R., & Smith, J. (1985). Role of the
Self-concept in the perception of others. Journal of
Personality and Social Paychology, 32(8), 1494-1512.

Maslow, A.M. (1970). Motivationgand personality. New York:
Harper & Row.

Mead, G. (1934). Mind; aalf and aociety. Chicago: University
of Chicage Press.

Mead, G. (1982). The individual and tha aocial ,aelf: Tha

unpublished works of George Harbart Mead. Chicago: The
University of Chicago Press.

Miller, A. (1970). The role of physical attractiveness in
impression formation. Psychon. Sci., 22(4), 241-243.

87

Minton, H. (1968). Power as a personality construct. In B.A.

Maher (Ed.), Progress in experimental personality
research (Vol.4). New York: Academic Press.

Mischler, E., & Waxler, N. (1968). Interaction in fapilieap
An experipantal atudy of fapily procesap gang
aghizophrenia. New York: Wiley & Sons.

Nagel, J. (1968). Some questions about the concept of power.
Behaviopal Sciance, 22, 129-137.

Olsen. D., & Rabunski, C. (1972). Validity of four measures

of family power. Journa} of Marriage and the Family,
is. 224-234.

O’Sullivan, M., Ekman, P., Friesen, W., & Scherer, K.
(1985). What you say and .how you say it: The
contribution of speech content and voice quality to
judgments of others. Journal of Personality:and SociaL

Psychology, 22(1), 54-62.

Parkes, K. (1984). Locus of control, cognitive appraisal,
and coping in stressful episodes. Jourpal of
Personality and Social Ps cholo , 32(3), 655-668.

Paulhus, D. (1983). Sphere-specific measures of perceived
control. Journal of Personality_and Social Psychology,
3316), 1253-1265.

Paulhus, D., Molin, J., & Schuchts, R. (1979). Control
profiles of football players, tennis players, and
nonathletes. Journal of Social Paychology, LQQ,
199-205.

Peterson. D.R. (1985). Scope and generality of verbally
defined personality factors. anch. Raviap, 12, 48-59.

Riecken, H. (1958). The effective talkativeness on ability
to influence group solutions of problems. Sociometry,
22, 390-421

Roe. A. (1957). Early determinants of vocational choice.
Journal of Coupaaling Paychology, a, 212-217.

Rogers. C.R. (1951). Client-cantered therapy. Boston:
Houghton-Mifflin.

Rogers. T., Kuiper, N., & Kirker, W. (1977). Self-reference
and the encoding of personal information. Journal of

Personality and Social Paychology, 22. 677-888.

88

Rosenberg, M.‘ (1981). The self-concept: Social product and
social force. In M. Rosenberg & R. Turner (Eds.),

Social psychology: Social perspectives (p.593-624). New
York: Basic Books.

Rotter, J. (1954). Social Iaarning and clinical paychology.
Englewood Cliffs, N.J.: Prentice-Hall.

Rotter, J. (1968). Generalized expectancies fo internal
versus external control of reinforcement. Psychological
Monographs, 22(1, no. 809).

Rotter, J. (1975). Some problems and misconceptions relating
to the constrcut of internal versus external control of
reinforcement. Journal of Consulting and CIinicaL
Psychology, 52, 56-67.

Sabatelli, R., Buck, R., & Dreyer, A. (1983). Locus of
control, interpersonal trust, and nonverbal
communication accuracy. Journal of Personality and
Social Psychology, 3312), 399-409.

Safilos-Rothschild, C. (1970). The study of family power

structure: A review 1960-1969. Journal of Marriage and
the Family, 22, 539-552.

Scanzoni, J. (1971). The black fapily in modern society.
Boston: Allyn and bacon.

 

Schaefer, E. (1959). A circumplex model for maternal
behavior. Jougnal of Abnormal and Social Paychology,
§_9_9 226-235.

Schaefer, 5.5. (1981). Converging conceptual models for
maternl and child behavior. In J.C. Glidwell (Ed.),
Parental attitudes and child bapayior (pp. 124-146).
New York: C. Thomas.

Schafer, R., & Keith, P. (1985). A causal model approach to
the symbolic interactionist view of the self-concept.
Journal of Personality and Social Ps cholo , p2(4),
963-989.

Schafer, R., & Keith, P., & Lorenz, F. (1984).
Equity/inequity and the self-concept: An interactionist

perspective. Social anchological Quarterly, 51(1),
42-49e

Schneider, D. (1974). Effects of dress on self-presentation.
Psychological Raporta, 22, 167-170.

89

Schwartz, 8., Tesser, A., & Powell, E. (1982). Dominance
cues in nonverbal behavior. Social ,anchology

Quarterly, 52(2), 114-120.

Shrauger, S., & Shoenemean, T. (1979). Symbolic
interactionist view of the self-concept: Through the
looking glass darkly. Ps cholo ical Bulletin, 22,
549-573.

Smith, J. (1979). Use of liked--d;aliked scale in
interpersonal ratinga. Unpublished master’s thesis,
Michigan State University.

 

Strickland,B. (1977). Internal-external control of

reinforcement. In T. Blass (Ed.), Personality variables
in social behavior. Hillside N.J.: Erlbaum.

Sullivan, H.S. (1953). The interpersonal thapry of
psychiatry. New York: Norton.

Symonds, P.N. (1939). The psychology of parent-child
relations. New York: Appleton Century.

Tedeschi, J. (Ed.). (1972). The social influence processes.
Chicago: Aldine-Atherton, Inc.

Tedeschi, J., Schlenker, B., & Bonoma, T. (1973). Conflict,
Power, and Games. Chicago: Aldine Publishing Co.

Warner, R. & Sugarman, D. (1988). Attributions of
personality based on physical appearance, speech, and
handwriting. Journal of Personality yyand SociaL

 

 

Psychology, 22(4), 792-700.

Wiggins, J.S. (1982). Circumplex models of interpersonal
behavior in clinical psychology. In P.C. Kendall &
J.N. Butcher (Eds.), flandbook of research methods in

clinical psychology (pp. 183-221). New York: Wiley.

 

 

GAN STATE U

“ill 1 | llmnlllllllllllll“

12930061