Ms -."'Iu‘ l‘ ' "
. o. 1." jive?! {)é'r II' "(\‘I ’anfi,
' ., . IEIII'IWII‘II " MI
' I'"" IRII’f-QfiI'Il‘w ""fII-I’
' ,II'IIfl ‘1 35" '
-:. 31 at

«

I
w IXI‘

I l\,' "‘3’ I‘lu :‘
g 1 , ., .‘ ,I if“) I
Mfif 'H. ' fut"... ., {:‘L‘: 5' '1?“
u 'I'I

:\:\:‘
Mv'”

'UI‘V‘ ’“’\;;.:|'":;‘ IL»:
I'-

I
‘I
II I}???

5'ut"'

.w‘ -

| ‘. . ..-'..:;.”|'.m

fi§y / £8135? 'LI'II
k-fiw II."

Il\$l1:ua‘..".ww ‘. ..
I" 2'? W :3 I

W Yt' 1; "I‘I'J :IS w,.
x “I. Iqa gr} .1353

k' H .
J"'_. 1;} . - It} xiv-“1'13”. .
I“ ”II?‘§¢§5

 

,I :tl‘fl'y If." '.
u.-J'.m ~ MI. I.

 

7 may
mm sum
Univerdty_

 

This is to certify that the

dissertation entitled

COGNITIVE STYLES
PSYCHOSOCIAL COMPETENCE

presented by

Sallie Annelle Norquist

has been accepted towards fulfillment
of the requirements for

Ph.D. degflxin Psycholoqy

gnaw/M

Major profeéor
John R. Hurley

 

Date 7 19 85

"cl”.--AL‘; - . -‘ In - n ,4 - 0.12771

 

IASLJ

LIBRARIES
”—

 

 

 

'1 Q5.I:I$.lqa"lvivl‘
5;“? 2 7 rag” 2

JZIowe>
JUN 1 1 E55

E i , m P,7
Dffifggs

 

BEIURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

I

f

 

WNITIVE STYLES AND
PSYCHOSGZIAL COMPETENCE

BY
Sallie Annelle Norquist

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology
I 985

ABSTRACT

C(BNITIVE STYLES AND
PSYCHOSOCIAL OOHPETENCE

BY
Sallie Annelle Norquist

This stud/ investigated relationships between psychosocial competence and three
aspectsofcognitive style. cognitive complexity, perwptualaccuracy, andfield independence.
Data were derived from 53 persons' ratings of self and peers' behavior within small groups of
college stucbnts that met for about 50 hours during a 10-week period. chhosocial
competence was separately assessed by two methods: (a) the product of peers' mean ratings of
each individual's within-group behaviors on measures of Acceptance versus Rejection of self
and also of others; and (b) participants direct interpersonal competence rank-orderings of
self and all other members of their small group.

These two measures of psychosocial competenm unexpectedly failed to correlate
simificantly, blurring the meaning of other outcomes Participants' perceptual accuracy
skills on the selected interpersonal variables proved difficult to assess conficbntly and the
derived measures did not link significantly to either psychosocial wmpetence indicator.
Comitive complexity did correlate significantly and positively with the ranking-bmed
psychosocial competence measure, as hypothesized, but not to the rating-based similar
measure. The latter, however, didlink significantly, as predicted, to moderately low field
intbpencbnce worers (Level II), while lower field independent persons (Level I) were
ratedthe least competent psychmially. Level III and IV field incbpendence scorers were

rated intermediately for psychosocial mmpetenm. Unexpectedly, the most field incbpendent
(Level IV) quarter of this sample tended to be perceived as more psychosocially competent
than the more moderately field independent (Level III) quarter, although this difference was
not significant.

Notable interactions were observed among some variables The most mnitively
simple persons were perceptually more accurate in rating similarly classified individuals, as
expected, but thwe more cognitively complex were not found significantly more wcurate than
others in their ratings of other cognitively complex persons. Also, field incbpendence
unexpmtedly failed to link signifimntly with degree of acceptance of either self or of others.
Albeit nonsignifioently, highly field inmpentbnt (Level IV) persons tentbd to score higher in
acceptance of both self and others than did their low field in¢pendent counterparts (Level I).
Several issues of definition and methodology were discussed, but the persistent difficulty in
achieving any reasonable consensus on a comprehensive and satisfactory definition for the

elusive construct of psychosocial competence emerged as the salient problem.

To John

H

ACKNOWLEDGEMENTS

I would like to express my warm appreciation to my committee members. In addition
to helping with the development, implementation, and final outcome of this project, each of
them has contributed substantially to my developing identity as a psychologist.

I would like to thank Dr. John Hurley, my chairperson, for being a dependable friend
throughout my graduate school years. His suggestions, support, editing skills, patience,
and faith in my potential as a researcher and a clinical psychologist were invaluable. I would
like to thank Dr. Robert Caldwell for the seriousness with which he takes his teaching
responsibilities, his concern with doing quality work, and his warm, subtly humorous,
always dependable manner of being with his students. Dr. Norman Abeles has provided me
with many opportunities for professional growth through my work with him in the
Psychological Clinic. These experiences have bolstered my confidence in leaving graduate
school and entering the field as a professional psychologist, and for this I would like to thank
him. From Dr. Elaine Donelson l have learned to respect myself and what I have to offer, and
have gained a much keener eye for sexual discrimination - in my own and in others' attitudes.

In adiition I have greatly appreciated her insightful research comments.

Special thanks also goes to John, for lovingly bearing my anxieties involving this
project, and giving timely help with editing and statisties, and to Kathleen, Lisa, and
Wendy, always willing to listen, support, andcare. Thanks also to Suzi Pavick, who made

it more possible to finish my dissertation away from the university.

iii

TABLE OF CONTENTS

Page

LIST OF TABLES ............................ vii
INTRODUCTION ............................ 2
REVIEW OF THE LITERATURE ....................... 4
A Definitions of Psychosocial Competence .......... . . . . 4

B. Personality Correlates of Psychosocial Competence ......... 6

C. Empathy and Psychosocial Competence ............... 8

D. Current Definition of Ps/chosocial Competence ............ 9
Cognitive Aspects of Psychosocial Behavior ................ l 2
A. Cognitive Complexity ...................... IS

B. Perceptual Accuracy ...................... 18

C. Field Independence .................. . . . . 21
Relationships Between Dependent Variables ............... 24

HYPOTHESES .............................

A. Field Independence and Psychosocial Competence ...... . . . . 27

B. Cognitive Complexity and Psychosocial Competenm .......... 29

C. Perceptual Accuracy and Psychosocial Competence .......... 30

D. Interactions between Variables .................. 30
METHOD ............................... 33
A Participants ......................... 33

B. Measures .......................... 34

1. Cognitive Complexity .................... 34

2. Field Independence ................ . . . . 34

3. Perceptual Accuracy .................... 36

4. Psychwociel Competence .................. 38

C. Procedure .......................... 40

D. Scoring of the Nemures ..................... 41

I. ARS-ARO ......................... 4l

Iv

2. Perceptual Accuracy .................... 42

3. CognitiveComplexity .................... 44

4. Fieldlndependence ................ . . . . 44
RESULTS .............................. 4S
I.Preliminary Statistical Investigations . . . . . . . . . . . . . . 45
A. CognitiveComplexity . . . . . . . . . . . . . . . . . 45
I. Interrater Reliability . . . . . . . . . . . 45

2. Distribution and Significant Correlations . . . . . . . . . 45

B. Fieldlndependence . . . . . . . . . . . . . . . 46
1. Distribution of OEFT Scores . . . . . . . . . . . . . 46

C. Perceptual Accuracy Measures . . . . . . . . . . . . . . 47
I. ICL Discrepancy. . . . . . . . 47

2. Interpersonal Scale Measure ofDominance and Warmth . . . . 48

3. Conceptual Accuracy . . . . . . . . . . . . 49

4. Composite Perceptual Accuracy Measure . . . . . . . . . 49

D. Psychosocial Competence Measures . . . . . . . . . . . . . 50
I. ARSARO . . . . . . . . . . . . . . . . . . . SO

2. Rank- Order Measure . . . . . . . . . . . . . . . .SI

3. Significant Correlatesof RANK . . . . . . . . . 52

4. Correlations Among Psychosocial Competence Measures . . . . . 53

II. Results of the Hypotheses . . . . . . . . . . . . . . . . . 54
A. Fieldlndependenceand Psychmial Competence . . . . . 54
I. Hypothesis Ia . . . . . . . . . . . . . . . . . . S4

2. Hypothesis lb . . . . . . . . . . . . . . . . . . 54

3. Hypothesis Ic . . . . . . .' . . . . . . . . . . . 55

8. Cognitive Complexityand Psychosocial Competence . . . . . . . . 56
I. Hypothesis II . . . . . . . . . . . . . . . . . . 56

2. Hypothesis III . . . . . . . . . . . . . . . . . . 56

C. Perceptual Accuracy and Psychosocial Competence . . . . . . . . 57
I. Hypothesis IV . . . . . . . . . . . . . . . . . . S7

D. Interactions Between Variables
l. Hypothesis V
2. Hypothesis VI
DISCUSSION
I. Variables Usedinthislnvestigation: Pertinent Factors
A. Psychmial Competence Measures

I. Insignificant Correlation between Measures
2. Difference between ROI and ROII

B. Group Embedded Figures Test Scores . . .

C. Relationship between Cognitive Complexity and Sex.

D. Perceptual Accuracy Measures . .
II. Discussion of the Findings

A. Psychosocial Competence and Field Independence
B. Psychosocial Competence and Cognitive Complexity

C. Psychosocial Competence and Perceptual Accuracy .

III. Summary

APPENDICES
A. Interpersonal Scales
8. Rank Ordering Exercise I
0. Rank Ordering Exercise II
D. Table of Raw Data

LIST OF REFERENCES .

vi

$8

58
59

.60
.60

6O

. 60
.65

. . 65
. 66
. 67

68
68

.70

71

72

75
76
77
78

80

LIST OF TABLES

I Interrater Reliability of Crockett's R00

2 Distribution of OEFT Across Cuartiles

3 Correlations Among Perceptual Accuracy Measures

4 ARS X ARO Mean, SD, and Range

5 Correlations Among Psychosocial Competence Measures

6 Mean Psychosocial Competence Score at Four Field Independence Levels
7 Correlations between Psychosocial Competence and Perceptual Accuracy
8 T-Tests, Level I versus Level IV of OEFT with regard to Acceptance

of Self and Others

vii

.45

47
so
51
53
55
57

59

INTRODUCTION

Interpersonal effectiveness appws essential for both the individual‘s psychological
well-being and for societal stability. Deficits in the critical psychosocial skills needed to
achieve personal male have been long recognized as contributors to psychological impairment.
Hospitalized psychiatric patients, for example, have been found impoverished in social
achievement as compared to nonhospitaliaed individuals from the same socioeconomic strata
(Zigler 8‘ Phillips, I960, I962). Furthermore, shorter institutionalization periods
distinguished psychiatric patients who had mmonstrated relatively mists premorbid social
functioning from those less adequate in premorbid social functioning (Zigler & Phillips,
I962). In hisessay "Senseof IntarpersmaIOompetence," White (I973) Wthatone
factor contributing to schizophrenia is "a bad start in eliciting response. from the human
environment" (p. 522).

Abba (l964) himlimted the int that the label ”mental illness” is arbitrarily
applied to certain types of malachptive interpersonal behavior, or “patterns of conduct with
'symptoms' of a psychosocial rather than a medical nature" (p. I94). As he noted, the most
effective ”cures“ for these illnesses are those which alter these maladaptive psychosocial acts
toward greater moderation, versatility, appropriateness, mdcompetence.

Despite the apparent importance of the construct of psychmocial competence for
understanding both arbptive functioning and psychopathology, little empirical work has ban
dine in this area The present study is an investimtion of the relationships between various
cognitive stylesandpsychosocial competence. AsShuretl981) noted, psychosocial

competence can be stressed from the perspective of cognition since how one thinks
(rematically affects one's actions (see also Ellis & Grieger, I977). Three major cognitive
style variables, cogiitive complexity, perceptual accuracy, and field incbpencbnce, have
been selected from social and personality research for use in the current study. These
variables offer promise for better understanding psycimial competence. This work will
explore their relationship to a rank-ordering measure of psychosocial competenm, and to a
pair of variables that have ban jointly equated with psychosocial competence: the ”as of
self-acceptance (ARS) andthedegreeofother-aoceptancc (ARO). Derived from observations
of individual's behavior airing the course of interpersonally oriented groups, each variable's

relative importance to psychosocial competence will be woertained.

REVIEW OF THE LITERATURE

DE'I' [El 'IC I

Wine ( I981) noted that competence models tend to take an optimistic view of human
nature, seeing humans as growing, changing, and leaning through interactions with the
environment. As compared with the traditional Infect—orierned medical model, competence
marble emphasize positive behaviors and capacities more than mficiencies and pathology.
Competence marble define the relationship between the individual and the environment
intercepentbnt, fluid, and charmable. Interest in competence-based coiiceptualizetions of
human functioning has increased as the circle of crities of the traditional “medical morbl"
approach (seefor example, Benjamin, I981) hasgrown.

One of the earliest known definitions of psychomial competence was Socrates' view of
competent individuals: “Those who manage well the circumstances they encounter (ally, and
who possess a jutbement which is accurate in meeting occasions as they arise aid rarely miss
the expedient course of action.” Doll (I953), in file Measurement of .Sxia/
ampere/roe, give a more succinct (bfinition: "the functional ability of the human organism
for exercising personal iflpencbnce and social responsibility" (p. 2).

Three different mproaches to the chfinition of competence were (blineated by Goldfried
and D'Zurilla (I969). The first was to Mine competence as the sum total of an indivimal's
social achievements and accomplishments in such areas as education, occupdion, marriage,
aid social activities. In line with this approach, Lanyon (I967) consicbred coll” stucbnts
as socially competent “to the extent that their backgrounds Uld present lives showed behaviors

or characteristies which indicated social participation, interpersonal competence,
achievement, and environmental mastery“ (p. 495). The difficulty of this approach is that
it is influenced by various external factors, such as availability of money, social prestige,
etc. It also leaves unanswered the question of how theee social accomplishments have been
achieved

A more operational approach to the conceptualization of competence was referred to by
Ooldfriad and D'Zurilla as the “behavioral-environment interactions associated with effective
functioning.“ In this remrd they tbscribad three aspects of competence that were reported by
OIwWIn (I967) at a National Institute of Mental leth (NIMH) conference (a) the mility
to reach a given (pal using a va‘iety of pathways. (b) the ability to move within and use the
resources offered by a variety of social systems, and (c) effective reality testing The latter
was mfined as ”not merely the lack of psychopathological impairment, but also a positive,
broad, and sophisticated unwrstanding of the world" (p. 32). Ooldfried and D'Zurilla
summarized the second mproach in terms of the individual's “effective response.“ An
effective response was further (bfined as “a pattern of responses to a problematic situation
which alters the situation so that it is no longer problematical, and at the same time produces a
maximum of other positive consequences and a minimum of nemtive ores” (p. 158).

The second wproach (i. e.. effective functioning) has been more recently Mined
Meiclunbaum, Butler, and Oruson (l98l) presenteda mocbl of psychosocial competence
which encompassed “the individual's meaning wstem, overt behavior, and cognitive proeeesee
in continuous interaction with one another and the social environment" (p. 55). The authors

viewed this entire chain of events as a summary of the construct of psychosocial competence.

 

The final approach to the mfinition of competence outlined by Goldfried and D'Zurilla,
and that which is most relevant for the current study, focused upon the attitudes, motives,
personality dynamiee, and traits which are the pruumed internal antecemnts of psychosocial
competence. Thus, Facts and Cottrell (I955) rewind autormy, creativity, empathy,
health, intellimnce, and judpment to be the essential elements of psychosocial competenm.
Eisler and F rubriksen (1980) cited early modelsof interpersonal functioning as postulating
traits such as empathy, emotional stability, maturity, and needs for affiliation or affection as
forces motivating positive interpersonal functioning Unsuceeesful social functioning by
contrast, was associated with traitssuchasintroversion, neuroticism, andibpencbncy.

In 1963, Oormn chveloped a 90-item measure of interpersonal value orientation,
called the W Survey of Interpol-mo] Values. In this measure he defined the
psychosocially competent as those who scored above the 60th percentile on Inthpencbnce,
Benevolence. and Leatbrship, but below the 40th percentile on Conformity, Recognition, and
Support. Wine (1981) asserted that psychosocial competence thfinitions should reflect
female subcultural valuee such as interpersonal sensitivity, altruism, and conesrn for
communal mod, thereby tdting Emma of women's waster Indivimal aid historical
experience in the interpersonal realm.

In a study of the antacatbnts of optimal psychological adjustment, Siemlman, Black,
Black, and Lippe (I970) consitbred the following Q-Sort items to be among the most
positively Mining has the cmacity for close interpersonal relationships, is socially
perceptive of a with ramp of interpersonal relationships, appears straightforward,
forthright, candid in chalings with others, whevesinasympatheticorconsithrate manner,

is compassionate, has warmth, and (hes not have a lot of hostility toward others. The
psychosocial nature of the content of those items sxmasts a positive relationship between
psychwocial competence and optimal psychological adjustment.

One's personal orientation may affect the capability to be psychosocially competent,
wcording to Weinstein (1969). Rigidity or rule-bouncemees was one such orientation.
Rigidity interferes with psychosocial competence by resulting in a reluctance to supercede the
bounchries of the roles one had learned earlier when these roles are no longer effective.
Rigidity in certain upsets of one's self-conespt may also interfere with competence, for
example, when one refuses to ask for assistance because s/he can't bear to be in the position of
needing something from someone else. Ironically, Machiavellianism is another orientation
sugmsted as associated with psychosocial competence, according to empirical evidence
(Singer, 1964). There are many situations where the willingness to use whatever tacties the
situation may require to meet short rm male serves the Machiavellian well in achieving her
or his pe‘sonal ends through social effectiveness. However, this (hes not mean that in the long
run a Machiavellian would be judmd as psychosocially competent.

Broader dispositional factors may also influence motivation for psychosocial
competence. An internal locus of control would theoretically lead to him motivation for
psychosocial competence. in that one would see oneself as capable of having an effect upon the
environment Motivation for achievement through psychmial competence is refined to the
extent that one is alienated, has an external locus of control, and is oriented toward avoiding
failure rather than maximizing suceeea.

Bergin and Garfield (1971) listed thefollowing skills as DBflBflCIBI to any human

interaction: an mcurate and sensitive awareness of another person's feelings, aspirations,

values, beliefs, and perceptions, a (hep concern for the other person's welfare (without
being (laminating), and an open, noncbfensive, mnuine being According to Weinstein, the
ability to take the role of the other accurately, that is, being able to ""correctly predict the

impact that various lines of action will have on alter's definition of the situation“ (p. 757),

is pivotal in psychosocial competence. in other words, psychosocial competence requires
empathy.

 

What are the charmteristim of the effective empathizer? Katz (1963) saw the
qualities of the mod empathinr as similar in many respects to the qualities of the gifted artist.
In both there is a “need to involve the emotions, to relax conscious controls, and to permit
oneself to be projected onto other objects“ (p. 135). He described good empathizers as having
an easy aid natural mprecietion for nuances of feelings in others, being capable of disenmging
from their identifications with others, and having flexible ego bouncbries. The latter allow
the empathizer to alternate between empathic experience and reflective analysis of these
experiences. Exercising the above abilities requires a sense of inner security on the part of
the effective empathizer. This is thrived from a sense of self—acceptance, and allows for
feelings of warmth and acceptance towards others. Indeed, acceptance and respect for the other
was consimred by Katzto be the mark ofthngodempathizer.

Rothenberg (1970) noted that a positive relationship has often been found between
social sensitivity and social adjustment, sugpsting that “an uncbrstanding of others' feelings
towvds oneself as well as the reasons for these feelings, are crutial for the development of
goodinterpersonal relations” (p. 336). Smith's (I973) book on training individualsto be

“sensitive“ to others defined sensitivity as “the empirical uncerstanding of other people as
measured by predictive wcuracy" (pp. 9-10). Based partially on others' analyses of the
components of sensitivity, Smith icbntified four constituent abilities. ‘The ability to look at
and listen to another person and remember what s/he looked like and said“ (p. 24) was termed
Observational Senstivity. Theoretical Sensitivity was mfined as the mility to 'selmt did use
thwies to make more accurate predictions about others' (p. 24). Smith pointed out that
ansitivity trainees often used theoretical concepts to interpret a situation to the point of
neglecting to carefully observe the facts. The third sensitivity component outlined by Smith
was sensitivity to the milled other (Nomothetic Sensitivity). This involved the mility to
'leern about the typical members of a group and to use the knowledge in making more accurate
predictions about individuals in that group" (p. 25). ldiographic Sensitivity was the final
component and required being able to make increasingly accurate predictions abwt an
individual as one's exposure to and information about the target person increased. Less is
known about this component than any other, as few researchers in this area have exposed their
subjects to the tarmt person for even one hour. The propm study will seek individuals‘
judgments about group colleagues after interacting with them for over 40 hours.

Weinstein 0969) listed several capacities underlying empathy. These were cue
sensitivity, perceptual vigilance, and role-taking accuracy. Role-taking accuracy involved
the ability to entertain multiple perspectives simultaneously, and required abstract skills that

have mnerally bssn considered a central component of intellimnce.

 

if the preceding discussion leaves the reatbr unmrtain of the nature of psychosocial

lO

competence, it is for the simple reason that tMre is little consensus as to exactly how to thfine
it. After is years of experience utilizing the social competence construct, Zigler
acknowledpd, “it is with the knowlm that has mcrued from all these efforts, aid with a
recognition of their inconsistencies, that we say that we know of no rimrous or even mildly
satisfying infinition of the construct of the term 'social competence" (Zigler at Trickett,
l978, p. 794).

Gibb (l964) asserted that ”a person learns to wow through his increasing mptance
of himself aid others' (p.279). The importmce of acssptance of self and other has been noted
by others as well. In fact, several investimtors ( Achms, T964; Argyris, l962; foe,
l96| ; Hurley, l976a) have ancestor: that psychosocial competence is born of the acceptance
of both mlf and others. This is the tbfinition that will be used in the current study. As
Argyris (l962) noted, an unmrstmding of psychosocial competence requires examining the
nature of interpersonal relationships. in systematically presenting this viewpoint, he
provitbd the following postulates:

i) "A basic need of man is to increase his sense of self-aweptance and acceptance of others'
(p. 20).

2) "Acceptame is intimately related to swarm, because... we will not tend to perceive that
behavior that threatens our self” (p. 19).

3) Therefore, “Awareness and acceptance of self and others are hard to separate“ (p. 20).
4) Psychosocial competence tends to increase as individuals “are aware of their impact upon
others, andothers' impact upon them" (p. 26).

Thus, Argyris suggested that competence stems from a sense of self-acceptance

ll

aid acceptance of others. Relatedly, he stated that human growth toward more
meaningful relationships is increased to the extent that self-awareness and acceptance of
others takes place in any interpersonal relationship.

Other researchers in this field have offered similar conceptualizations. Tyler's
three-faceted hierarchical competence configuration morbl of psychosocial functioning
consisted of self-attituths, world attituas, aid relevant behavioral attributes (Tyler &
(hi2, I977). The self-attitucb component was similar to acceptance of self, as it involved
maintaining a favorable self-evaluation and a "sense that one is causally important in one's
own life“ (p. 442). Constructive interaction with others involved sustaining a world
attitude of basic optimistic trust, which seems akin to acceptance of others. The third feast of
Tyler's moml was less relevant. it involved behavioral attributes 'that serve as ormnizing,
implementing, andfulfilling components" (p. 442).

The mpropriateness of self-acceptance and other-acceptance as measures of
psychosocial competence was also sugpsted by Foe (l961). He observed that an
"interpersonal at is an attempt to establish the emotional relationship of the actor toward
himself and toward the other, as well as to establish the social relationship of the self and the
other with respect to a larmr reference woup" (p. 350) and "the same act statesthe
positionoftheactortowu‘dtheselfandtoward the other“ (p. 350).

The centrality of wceptance of self and acceptance of other to both healthy and
malathpted psychosocial behavior has been indicated in a number of studies. Foe (1961)
summarized these studies with the statement that the findings 'suggest a circumplex around
thetwoorthomnalaxesofDominance-SubmissionandAffaction-Hostility" (p. 352). Adams
(l964) auto that 'The Dominance-Submissim axis (Mines the use of acceptance or

12

rejection of self, while the Affection-Hostility axis mfinm the agree of acceptance or
rejection of the other.” Thus, "an interpersonal set may be regrcbd as the Cartesian product
of these two sets of values” (p. 195).

The weptance/ rejection of self (ARS) and acssptance/ rejection of others (ARO)
scales used in the current investigtion were gveloped ow upon the above evignce
(Hurley, l976a). Psychosocial competence, or wceptance/ rejection of self and others will
be wed in the current investigation within the context of small ungrgraduate
interpersonally oriented g‘oups. This context is well suited for the current investigtion in
thatme members rateemh other onARSandAROafter morethan 4O hoursofmtual
interaction with each other in a group atmosphere which has as one of its primary gals the

further development of psychosocial skills.

 

Given the viewpoint that psychosocial mmpetence can be defined g acceptance of self
and other, what are the salient cognitive espmts of psychosocial behavior? In consigring
the structure of cognitive abilities, Feffer (I966) extencbd Piaget‘s concept of
'Mntering". His interpretation of this abrupt in the psychmial (binain la‘i to the
hypothgis that “effective social interaction is a function of each participating individual's
mility to consigr (her/his) behavior from more that one perspective simultaneously" (p.
415). He found empirical support for this view using the Role Taking Task (RTT). This
seems related to the role-taking capacity earlier mentioned as underlying empathy.

As pointed out by Van Maanen (I979), psychosocial cmacity refers to an individual‘s
daility ”to engg in social intercourse and, therefore, to potentially share meaning with

13

others" (p. 25). He glineated three primary mental capacities similar to structural
properties of the mind, which allow one to perform effectively in interpersonal interactions.
The first was the capacity to produce temporal frameworks. This involved the abilities to
recall the pm, to be aware of the present, and to plan for the future. The ability to
caiegrize experience was the second underlying mental capacity. This was an as necessary
for recognizing, clsssifying, and ungrstanding one's experiences. The last set of
psychosocial capmities inscribed by Van Maenen is that of interpretive procedures, referring
to the mind's ability to construct, interpret, and understand on-ging social interaction.

As noted by Zajonk (1968), in social psychology it is the cognitive representation of
social stimuli, rather that their objective properties that are assumed to be most important.
The chracteristic ways that individuals cormptually orgnize the environment have been
referred to by Goldstein and Blackman (1978) as cognitive styles. From this point of view
at individual's environment takes on psychological meaning (spending on how s/he filters and
prsssssss the available objective stimuli. An emphasis on the structure, rather than upon
the content of thought, is common to the numerous theories and research engavors on
comitive style. One of the most prominent of these theories has been Harvey, Hunt, and
county's (1961 ) theory of Integrative Qimplexity.

Oonsspts are of prime importance in the theory of lntewetive Complexity. A concept
is defined as a “schema for evaluating impinging stimulus objects or events“ (p. 10), and as
such proviths a medium through which the individual relates to the surrounding world The
development aid functioning of the self is assumed to be inseparable from the development and
functioning of a concept. Concepts are seen as on a continuum ranging from concrete to
abstract Indivickials can be ordered on this continuum according to their ability to

14

differentiate and integrate information. Those that function more concretely are more
absolutistic and stereotypical in their cognitive processes and responses. miceptualizing for
them is more in terms of black and white or, at most, a minimum of alternatives. As
development proceeds toward the abstract end of the continuum, the individual's psychological
functioning moves away from concrete functioning toward a strongr differentiation between
the self and the social environment, to the development of empathy, and, ultimately, to a
balmce of affiliative and incbpetmnt tentbncias (i.e., mptance of self and others).

Personal construct psychology, founded by Kelly (1955), is similar to Harvey,
Hunt, and Schrogr's theory in that both regrd the indivimal's conceptual orgnization as a
means of ungrstanding personality. These theories differ in that Kelly emphasized the
importanceof the write/7! (Zajonc, 1968) of construal systems, whereas Harvey, Hunt,
aid Schrotbr emphasind the sires-rural features. Kelly viewed individuals as creators of
meaning throng taking an active role in the perception, construal, and interpretation of the
world around them. His fundamental postulate was that "a person's procmes are
psychologically channelizad by the ways in which he anticipates events" (1955, p. 46).
Three principles derived from this postulate have been pertinent to psychosocial relations.
The Individiality Corollary asserted that 'persons differ from each other in their
constructions of events" (1955, p. 12). The Commonality Corollary stated ”to the extent
that one person employs a construction of experience which is similar to that employed by
another, his psychological processes are similar to those of the other person" (Kelly, 1970,
p. 20). The third psychosocially relevant principle has been the Sociality mallary,
stipulating that ”to the extent that one person construes the construction processes of

mother, he may play a role in a social process involving the other person" (Kelly, 1970, p.

15

22). Research on these corollaries has inclugd investigations of how individuals make
inferences about other's constructs as a basis for effective communication and mutual
ungrstanding, of ways in which we orgnize information about our social environment, and
of the role of similarities between persons (in terms of the content of their personal
construct systems) in the gvelopment of interpersonal relationships.

From the standpoint of personal construct theory, social gvelopment entails
systematic change in the structure and content of an individual's construct system. This
involves progressive increases in the number of constructs that an individual uses to ascribe
others (i.e., an increase in complexity) and "is accompanied by a rebel shift of emphasis
from primary concern with mpeerance, social roles, and behavior, to a precbminant
interest in personality" (Amine-Webber, 1979, p. 200). This seemsto reflectagneral
procession from seeing others mainly in terms of stereotypes, to a more differentiated and
individiated view of them as persons (Duck, cited in Agms-Webber, 1979).

A similar view was proposed by Warner (1957). He viewed cognitive development
as a presses of alvencing "from a state of relative globality and lack of differentiation to a
siateof increasing differentiation, articulation, and hierarchic integration" (p. 127).
Thus, maturation was (timed as a prog‘eesion from simple to complex. In terms of coping
with the social environment, one's success seems to be in a large part gtermined by the
differentiatim (or complexity) of one's cognitive representation of that environment

(Zajonc, l968).

1.) il' 13 l 'l
The concept of cogtitive complexity-simplicity was introduced by Bieri (1955)

16

shortly after the publication of Kelly's Me Psycho/ow of Perms! Chasm/ct:
(1955). Bieri defined complexity in terms of the relative number of constructs (i.e.,
differentiation) in an individual's construct system. A more recent cbfinition referred to
cogiitive complexity as “the number and interrelatednees of rules or schana used for
discriminating, encoding, and retrieving information about the social environment" (Bruch,
Heisler, 8t Conroy, 1981). Bieri's methodof appraising ggree of complexity was thrived
from Kelly's Role Construct Repertory Test Essentially, Bieri's method consisted of a
matrix or aid on which the subject was gked to nane specific persons in her/his life and to
rate these individuals on va‘ious subjectively gnerated construct dimensions ( eg,
pinging-shy, adjusted-maladjusted, etc). Many alternate measures of cognitive
complexity have been proposed since Bieri's initial conceptualization. These have included
Crockett's Role Categry Questionnaire (1965), ~Carr's Interpersonal Discrimination Task
(1980), mdSmith and Leach's (1972) hierarchical clustering technique.

Subseqient research has found cognitive complexity to be an important variable in
sophisticated psychological functioning This seems logical, consigring that the more
cognitively differentiated individual is assumed to have a geater number of dimensions with
which to construe the behavior of others. Carr (1980) found that cogiitively complex
jumps made finer interpersonal discriminations. He also reported several studies relating
success of social intermtion to the m of compatibility in level of comitive complexity.
Bruch, Heisler, andConray (1981) foundthatwomen him in comitive complexitywere
more assertive in difficult situations and also expressed consideration of the needs of others
more often. Sechrist aid Jmkson (1961) rworted a .54 correlation between measuresof

cogiitive complexity and social intellignce.

17

copiitive cornplexity's position within the nornological network has been repeatedly
maestioned It has also been questioned as to whether it is relatively stable over time and
different stimulus domains (i.e., whether it is strait). Crockett (1965) marted that
complexity is specific to a prticular cognitive rbmain, while Bieri (1956) provirbd some
evicbnce of its gnarality, using both personal and nonpersonal stimuli. Little universality
wg found by Valnoy (1965) in a stud/ consitbring 15 potential comitive complexity
measures. More recently, after a review of relevant studies, Goldstein and Blackmail
(1979) conclutbd that comitive complexity scores seemed to be similar mress different
stimulus mmains, but that environmental factors may modify complexity.

One of the major difficulties in comitive complexity research has been the leak of
simificant linkage among different complexity measures, and umtainties about what
measures are the most meaningful. In addressing this issue, O'Keefe and Sypher (1981)
reviewed research relevant to the evaluation of various complexity measures on the basis of
five criteria; high test- retest reliability, association with chronological age across childlood
aldublescenca, inrbpenrbnce from verbal abiliies and intellignce, associations with other
social-cogitive measures, and associations with other measures of communicative
functioning The two complexity measures used most often have been Bieri's modification of
the Rap Test and Crockett's Role Catemrv Questionnaire (RCO). Research to date indicates
that Crockett‘s RCO has been satisfactory on the five criteria named above. Biari's measure,
on the other hand, related nonlinearly to rue, often had a low test-retest reliability,
associated inconsistently with other measures of social-cognitive functioning, and had not
been positively associated with high cornunicative functioning

in the present study cognitive complexity will be measured by Crockett's R00.

18

Reckman (1977) found it positively related to social competence as measured by frequency of
involvement in social activities, competence in mating interpersoanl action plans, and
comfort aid satisfaction with peer relationships. In the same study Bieri's measure did not

correlate with social competence.

BMW

Perceptual wcurwy, presently Mined as the ability to jm others' attributes in
relation to others' self-report, is the next variable selected for the study. It should be
emphasizedherethat raters wereaskedto rateotherswcordlngto how the raterssawthem,
rather then moording to the rater's prediction of how the ratae would rate her/himself. The
criterion for judging mracy was the was of ag‘aement between the rating given by the
raters, andthe ratee's self-ratings.

The relationship between perceptual accuray and psychosocial competence seems
unexplored Although person perception was heavily investigated for some time in social
psychology, disparate methodologies and artifactual influences prodmed equivocal, often
conflicting, findings (see Filak, 1982 for a review). Nevertheless, perceptual aesuracy
has emermd as an important psychosocial variable in non—analogue research on extenmd
relationships. For instmce, Lucky (1960) found wuence of spouses' perceptions, or
perceptual mrmy between spouses to be related to marital satisfaction. Relatedly,
therapist's perceptual accuracy of various client personality and symptom indiess at the
conclusionofpsychothermy was foundbyseveral researcherstoberelatedtothesuccessof
treatment (Cartwright 8s Lerner, 1963; Filak & Abeles, 1983; Schrier, 1953).

Based on previous research, it mpears likely that perceptual accuracy is not

19

exclusively a cognitive variable. Rather, the agree of perceptual wcuraoy in an
interpersonal relationship is likely to reflect the qualfly of that relationship (Filak &
Abeles, 1983; Lucky, 1960). In other words, a person's relative agree of accuracy in
judging others is likely to reflect that person's sensitivity and attunement to these others, aid
hencethe qualityof involvement with others. Perceptual accuracy, therefore, is likelytobe
useful in ssssssing participant's adjustment and attestation to an interpersonally-oriented
amp. It is hypothesized that participant‘s perceptual accuracy scores will be positively
related to their rated pwchosocial competence scores. Interactions betwssn psychosocial
competence aid perceptual accuracy seems likely, since psychosocial competence is likely to

lead to waster perceptual accuracy, but also, waster perceptual wcuracy would likely lead
to positive feediack from others. This could lead to increased self-esteem and hence increases

in self aid others' views of one's psychosocial competence.

Methodological consitbrations are critical in the investigtion of perceptual accuracy
(Fildt 8s Abeles, 1983). Thetype ofaccuracytask, the natureof thetask, the typeofthe
person being judged, and the criteria for judging accuracy are all important variables in
perceptual wcuracy research. Other variables, such as the nature of the relationship and its
wration before the accurezy task is given are also influential.

Conceptual versus predictive accuracy have been the two major types of mproaches to
person perception research. These approaches reflect distinct comitive operations
(chher, 1967; Taft, 1955). In thecoimptualization mpromh, theindivimal is ukedto
subjectively formulate her/ his View of another. The predictive apprfmh, on the other hand,
requires that a person try to predict how mother would behave. perform, or jump him or
herself. The relative importance of each of these variables to psyclmocial contexts is not

20

known, however, the conceptualization approach has been posited to be the more meaningful
exercise (Filak, 1982). It is freer of methodological artifacts found in predictor tasks. such
astha three of similarity batwwn the perceiver and perceived (Cronbach, 1955). In the
current study, the conceptualization approach is utilized “Perceptual accuracy” is the term
used to refer to scores related to this mproach, since the perceiver uses her or his own
senses in appraising others.

The nature of the selected task is the judging of parmnality attributes of others, and
the sample involves “normal" college students. Both the perceived individual's self— ratings
and her/his ratings of how s/he thinks s/he is seen by others will be the criteria for judging
accuracy. Both types of self-ratings are included to explore the relationship between then.
Validity for the use of individual's self-report of personality for the criteria for judging
accuracyhasbeenfoundbyseveral researchers (Oiedt, I958; Gottman 8e Markman, 1978;
ria‘kowsky, 1979; Hischel, 1972). Finally, accuracy at the beginning of a relationship
has been found to be unimportant to psychotherapy outcome, whereas accuracy in the later
pa‘tof therwy relationships has been found relevant (Cartwrimt 8t Lerner, I963; Filak
at Abeles, 1983; Kurtz & Grummon, 1972). Accordingly, theperceptual accuracy tasks in
this study will be given near the end of the 40-hour group experience.

Some research has been done on the relationship between cognitive complexity and the
muracy of perception , but the results seem inconclusive (see Goldstein 8s Blackman,
1978, for a review). Theoretical predictions hm on this research would be difficult
beseuse of the different measures of perceptual accuracy and cognitive complexity used in
these studies. After a consideration of the final cognitive variable. field capenmnce, a

21

model will be proposed to anticipate the relationship between all of the variables in the
current study.

W

The concept of field incbpencence arose from studies of perception of the upright
citing back to 1942 (Witkin 8s Gootbnwgh, 1981). On the basis of large individual
differences in perception, Witkin differentiated between permns (bpancbnt on the structure
of the prevailing visual field (i.e., field (bpemmt) and those who overcame the organization
of the field by thaling with it analytically (i.e., field incepenmnt). In time, these individual
tentblicies toward field depenmnce and field inmpenmnce were found related to intellectual
activities, specifically, a global versus analytical (or articulated) cognitive style. The
mmpticn of field incbpmdence as a cognitive style has recently been questioned by McKenna
(1984), also by Widimr, Knumon, aid Rorer (1980). Using factor analysis, the latter
resea‘chers obtained results best interpreted as indicating that the present field
(bpancbnce-intbpentbnce measures appraise ability rather than cognitive style. As yet,
field (hombres-imam measures are still commonly cited as indicators of cognitive
style.

The concept of psychological differentiation was introduced after global versus
cticulated differences were found to be related to individual differences in how concept, in
the nature of the self, and in the controls and defenses used most often. More differentiated
ormnizations were postulated to show ”greater self-m" segremtion, signifying definite
boundaries between an inner core of attributes, feelings, and needs imntified the self on
the one hmd, and the outer world, on the other hand, particularly other people“ (Witkin,

22

Ooomiiough 5s Oltman, 1979, p. 1127). A greater connectedness between self and others
was charmtaristic of a less differentiated system.

Persons with a more differentiated, or field immnt, mesh of cogiitive
functioning have been found more autonomous in social-interpersonal situations, to have a
well obveloped sense of personal identity, to use specialized mfenses (such as isolation,
intellectualization, and projection) and to show weater regulation of affective discharm and
motor ctivity. Field mpancbnt individuals, on the other hand, rely more on external means
for (bfining themselves, have a more open, interwsonal orientation, use less specialized
defenses (such as repression and rental), and show more diffuse anxiety and impulsivity
(Witkin, Goocbnough, & Oltman, 1979).

Many researchers have probed relationships of field depancbnce to interpersonal or
psychosocial behavior. Reviewingthssefindings, Witkin and Gootbnough (1977) notedthat
there are marked psychosocial behavior differences between field thpendent and field
intbpencbnt persons. Field (bpeiicbnt people are more attentive to social cues, have a more
interpersonal orientation, prefer closeness to others, self-disclsss more, have greater eye-
contmt with others, are more open emotionally, and tend to evaluate others more positively.
In short, field mpencbnt indivichials show a variety of characteristiss that mate it likely that
they will mt along better with others. In contrast to the strong interest in others shown by
the field mpancbnt, field inmpentbnt individuals have an impersonal orientation. They are
more autonomous in social relations, show more physical «id psychological distancing from
others, and prefer nonsociel situations. Characteristics ascribed to field Wt
individuals incluw being warm. affectionate. tactful, accommodating. nonevaluative,

mcapting, and not likely to express hostility directly. Field incbpancbnt persons, in

23

contrast, have been ascribed such charmteristics as chmaflding, inconsitbrate,
manipulative (to achieve their own ends), cold. and distant in relating. These traits seem
antithetical to competent psychosocial behavior within a gem of peers.

In relation to psychosocial competence, pmlbtnb nurses who were fund most
competent were found to be field menthnt, while competent surgical nurses were
relatively field inchpantbnt (Quinlan & Blatt, 1972). Also, groups with more field
(bpancbnt members tended to be relatively more effective in conflict rssslution (Oltman,
Occlhmugh, Freednan 8t Friednan, 1975; Shulman, 1975). Witkin aid Goocbnougi
(1981) Stimulated that the field inmpencbnt re more limited in psychosocial competence and
that their social skills “are likely to represent the application of their restructuring skills to
the social cbmain rather than investment in relations with others' (p. 45). The poster
psychosocial skills shown by the field dependmt may well reflect a stronger tentbnw to rely
on external referents.

Individials generally tend to be either relatively high in cognitive restructuring
skills (i.e., field intbpencbnt) or. relatively him in psychosocial competencies (i.e., field
(reactant). Despite this tendency, it is reasonable to mums that, given appropriate life
aid edicational experiences, some persons may develop high level skills in both areas.
Witkin designated those who have acsees to both psychosocial m cognitive restructuring
skills as ”mobile“. The instrument used in the present study to ....-.o field

 

mpetmnce-intbpentbnce (the GEFT, described in the Methm section) was chimed to
measure the relative presence or disence of cognitive restructuring skills. As such, field
mm is not directly measured It is possible that any persons in the present study who

test as field inmpentbnt on the GEFT , yet also show psychosocial competence in their group

24

(as measured by ARS-ARO), are mobile with respmt to field chpembnce—incbpendenm.
Consequently, in this investigation, those who score low on the GEFT will be referred to as
being low in field inmpeimice, rather than being field chpancbnt

Ganble aid Ginsberg (1981) disagreed with Witkin's notion of psychological
differentiation, which postulated high intercorrelationsof perceptive, cognitive, andsocial
differentiation. Instead, they asserted that differentiation in each of these three minains is
mpatmnt upon the ecological moms of the culture. Thus, it would be quite possible for
individuals to be dole to discriminate socially, but not perceptually, or cognitively.

Relatively field (hpenthnt and field inrhpenmnt indivimals tend to be primarily
invested in different domains, resulting in psychological development along different
pathways. The field intbpenmnts' more autonomous functioning is likely to foster the
(L-velcpment of weater cognitive restructuring capacities but not psychosocial competence.
The reverse is true for field fbpentbnt individuals. However, Witkin, mm, and
Oltman (1979) have proposed that differentiation may mvelop multilinearly; mnuine
mvelopmant taking pleas along both the psychosocial competence and the comitive
restructuring pathways of lesser and greater differentiation. Therefore, it seems reasonable
to postulate that those who are more differentiated along the psyclmial competence pathway

could also show relatively high perceptual mraoy skills.

811' |' E! D llll'll
No prior work has investimted the relationships among these selected variables.
Based on the previously reviewed research, it can be hypothesized that the three thpentbnt

widiles, cognitive complexity, perceptual muracy, andfieldintbpentbnce will relate

25

memingfully to the intbpencbnt variable, psychosocial competence. Thererereasonable
crowds for expecting that both cognitive complexity and perceptual accuracy will have a
straig'itforward positive, linear relationship with psychomial competence. In contrast,
neither hiqi nor extremely low field inthpentbnt scores are expected to relate montcnically to
psychosocial competence. Rather, midranm field iMencbnce scores, especially thom
falling in the second quartile of the obtained GEFT scores, are expected to be positively
associated with psychosocial competence.

The inclusion of the three (bpentbnt variables in this study allows for an exploratory
investimtion of their interrelationships. For instanm, cmhitive complexity may be an
important variable in psychmocial competence, but only when perssptual accuracy is
pramnt. While no systematic research can be relied on to generate specific hypotheses
(hence this works' exploratory nature) familiarity with the conssptual underpinnings of
these varidiles allows for some mneral speculations.

The relationship between cognitive complexity and perwptual accuracy is difficult to
ascertain. While a mural positive miation seems likely, puzzles remain. For instance,
will the cognitively complex be equally accurate when assessing both other comitively
complex indivimals md ccmitlvely simple Iridivimals, or will the cupitively simple be
more accurate when appraising cognitively simple individuals? The relationship between
field intbpencbnce and perceptual accuracy seems similarly perplexing For instmce, low
field indepentbnt individuals, having been hypothesized to be more attuned to others, may be
more waged in this regard, but may be disadvantaged on the perceptual mourn! task.
This is die to the fact that they may be less likely to possess the cognitive skills necessary for
mrata imminent. In adiition, both the comitively simple and the field inibpancbnt may

26

not be able to differentiate sufficiently between self and other to perceive self and other
accurately.

Perceptual accuracy may mediate between both other cbpendent variables (cognitive
complexityand field inmpentbnce) and the independent variable, psychosocial competence.
For instance, the comitively complex may be more psychosocially competent only when they
re perceptually attuned to others. Conversely, cognitive complexity may lead to
psychosocial incompetence when the cognitive constructs are too far removed from others'

actual behaviors.

27

HYPOTHESES

The hypotheses in this stud/ concern the interrelationships between the three

cogiitive style variables; field intbpendence, cognitive complexity, andperceptual wcuraw,
and the WM variable, psychosocial competence. Hypotheses are grouped according to
incbpeth variables.

 

Willa. There will be a significant difference in psychosocial competence
between subjects scoring in the different quartiles of field

independence.

This hypothesis is based on the prediction discussed earlier that, in general, those
individuals who we low on the GEFT should show more psychmocial competence than those
who do well on the GEFT. This hypothwis predicts that psychosocial competence will vary
with level of new inoeperoenoe. Hypotheses lb and 1c further speciiy the variation across
levels that is predicted. To test this hypothesis, the scores obtained in the present sample on
field indapancbme were divicbd into four levels ranging from low field inchpencbnce (Level I)
to high field incbpendence (Level IV). A division into quartiles of the room of scores obtainw
by the present sample was utilind to obtain the four levels. This hypothesis was tested by

using analysis of variance.

28

Hypothesis 1b. Level 11 field independence scorers will score higher on
psyclmociel competence than either Level I or Level IV field

incbpanibnce were.

The rationale is that individuals whose OEFT scores fell at Level II should be able to
make more of a separation between themselves and others than Level I persons, and yet be
more attuned to othws than the highly field indepenmnt morars (Level IV). A one-tailed I

-test will be used to test this hypothesis, due to the direction of predicted differences.

Hypothesis Io. Level 111 field inmpendence worers who also score above
averqe on perceptual accuracy will show higher psychomial
competence scores than either Level I or Level IV worers.

The underlying assumption is that Level 111 field independent scorers are theoretically
less psychosocially oriented and less attunw to others than Level I and II scorers. If they also
score high on perceptual accuracy, however, their conceptual skills should enable them to
interim more competently with others, we to their more accurate conceptualizations of
others. This should be mially true in the present interpersonally-oriented coups,
which were geared toward mining a clearer understanding of how self is perceived by others.
This hypothesis will be tested with a one-tailed I -test.

29

 

Hypothesis II. There will be a significant positive relationship between cognitive

complexity and psychosocial competence.

This hypothesis stems from Reckman's (1977) results, and the assumption that
more comitively complex individuals will have access to a raster number of dimensions for
construing others. Therefore, they should be able to make finer discriminations in coming to
at uncerstending of an individual. I am assuming here that a greater understanding of an
individual is likely to be manifested by constructive interactions with that person. Given the
inmrpersonally-oriented nature of the cognitive complexity measure used in this study, it
also seems reasonable to assume that individuals among the current sample who score high in
comitive complexity have more experience in relating interpersonally, and that this
experience will correlate positively with psychosocial competence. Pearson's r will be used

to test this hypothesis.

Hypothesis III. Among high cognitive complexity scorers, those who are more
perssptually accurate will have higher psychosocial competenm

scores.

Him cognitive complexity scorers use a greater number of constructs in tbscribing
others. Having a larger number of constructs to draw from in wnceptualizing others has been

hypothesindto been important factor for psychosocial competence (see Hypotheses II).

30

Hypothesis Ill predicts that the perceptual accuracy of the constructs used in conssptualizing
about, or interacting with, another isanadditional important factor influencing one's ability
to be psychosocially competent. Specifically it is hypothesized that those individuals who
score him in both mnitive complexity and perceptual accuracy will show greater
psychosocial competence skills than those who score high in cognitive complexity alone.
“High“ scores are defined as those that are above the mean scores obtained by the present
sanple. Two-way analysisof variancetests will servetotest Hypothesis III.

 

Hypothesis 1V. There will be a significant positive relationship between perceptual

assuracy and psychosocial competence.

This hypothesis was thrived partially from the information presented in the
literature review, along with the rationale that perms regarcbd as most psychosocially
competent by woup peers should--over the courseof 40 or more hours of working togther
within these small groups--have come to a clearer understanding of other group members.
Pearson's r will be used to test this hypothesis.

1 I l' I l l! . I I
In Edition to the intermtions among the ihtbpenmm did thpenfbnt va‘idiles listed
above (Hypotheses lo and III), the following interactional patterns are expected

31

Hypothesis V. There will be a simple interaction between cognitive complexity of
raters and ratees (above average vs. below average), with regard to

perwptual accuracy.

In particular, it is predicted that the cognitively complex will be more perceptually
wrote in rating cognitively complex individials, whereas the cognitively simple will be
more perceptually mcurate in rating other cognitively simple puticipants. The rationale is
that cognitively complex persons theoretically will perceive the cognitively simple in a more
complex and differentiated manner that the comitively simple will perceive themselves. The
reverse should be true of the cognitively simple individuals’ perceptions of the mitively
complex. If this is true, the above combinations of complexity scores should result in lower

perceptual accuracy scores. This hypothesis will be tested using I -tssts.

Hypothesis VI. Level 1 field intbpendence scorers will obtain significantly higher
scores on Acceptance/Rejection of Others than will Level IV scorers,
while Level IV field incbpencbnce scorers will obtain significantly
himer scores on Acceptance/Rejection of Self then will Level I

SCG‘BI‘S.

This hypothesis is trawn from the theoretical umbrpinnings of the concept of field
incependence. The tenchncy of low OEFT scorers to rely more on external soursss for
self-(bfinition. to experience a greater sense of connectemess between self and others, to

prefer closeness to others, and to evaluate others more positively seems to sugpst that

32

they place a very high priority on acceptance of others. Him OEFT worers, on the other
hand, have beencharmterizedes cbmanding, inconsiderate, manipulative, cold, anddistant
in relating These traits clearly ob not portray someone who would be seen as higily
accepting of others. Rather they aim that such an individual would be more concerned with
pursuing her/his own needs and gals over accommotbting to others; an orientation which
augusts more acceptance of self than of others. Hypothesis VI will be mpreised by
one-tailed I-tests, duetothespecificityof thedirection predicted.

METHOD

E I' . 1

Students in an upper level uncbrg‘aduate psychology course aimed at experiential
learning about interpersonal prosessss served as subjects for the current investigation. The
structure of the course was such that students were assigmd to small groups of abwt five to
aidit members, including one or two facilitators, to meet for two 90-minute sessions per
week throughout the 10-week term, plus two 12-hour extenmd sessions (after
mproximately the third and the seventh wssks). The purpose of these groups was to explore
onming interpersonal processes as an aid in mveloping better interpesonal skills. Required
textbook readings and entries in a personal log after each group session were used to help meet
the stated purpose. The group process emphasized mtting to know each other through
here-md-now interactions, rather than bringing outsicb information about themselves into
the coup. In forming the groups efforts were more to balance the proportions of male and
female members. and friends were not allowed to be in the same group. Facilitators or
leacbrs were former g‘oup members selected from volunteers for further training. Date
from all group members, including facilitators, from two sucssesive terms (Fall 1983 and
Winter 1984) were utilized A total of 53 individuals participated (29 females and 2‘1
males), thirteen of when served as facilitators. Information was gathered from a total of ten
g‘oups. All members supplied all measures in three groups. Information was mthuad from
all but one g~oup member in six more groups, and all members but two returned the
measures in the final group. Oonsicering only those persons who returned all measures, the
size of thesecroups ranmdfrom four to seven (3 with 4 members, 4 with 5 members.

33

34

2 with 6 members, and I with 7 members). About three-fourths of the participants were
between the ans of 20 and 23, although their ems ranged from 19 to 39. About two—thirds
(66%) were psychology majors, followed by majors in the physical health professions

(approximately i558), andcommunications (mproximately 85%).

Was

WIN. Comitiva complexity was measured with Crockett's Role
Outflow Questionnaire (R00). For it, participmts were given five minutes each to write
an impression of persons fitting two role Inscriptions. Two Inscriptions were asked for; one
of someone they know and like, and one of someone they know and dislike. Instructions
require subjects to Wiba the person as fully as possible, paying attention to particular
habits, attituchs, beliefs, mannerisms, and waysof treating others. The scoring procedure
for this measure, as outlined by Crockett, Press, Delia and Kenny (I974), consists of
counting the number of constructs used to cbscribe emh person. Aspects of the other's
personality and behavior are counted, while physical charmtaristies are not The sum of the
constructs prodimd in both situations constituted the final score. Thus, the higher the score,
the more comitiveiy complex the subject is believed to be. Interrater reliability
ax'relatims for theRCCI commonly exceed .90 (O‘Keefe & Sypher. I981). Ithesbeenfound
positively mecciated with perspective-taking ability (Hale & Delia, I976), measures of
communicative efficiency and effectiveness (Hale, I980) and level of permive strategy
(Delia, Kline, 8t Burlson, I979).

Hakim Field intbpendenca was measured by the Group Embodied Figures Test

35

(IIFT). The OEFT was modeled after the original indivimally ministered Embattled Figures
Test (EFT). Both tests require that the participant find a simple figure which has been
incorporated into a more complex one, but obscured perceptually by line patterns. The
participmt is prevented from seeing both the simple and the complex figure simultaneously,
but is allowed to examine the stimulus as often as neared This is accomplished by having all
the complex figures on the right hand side of the booklet paws and the simple figures on the
outsim back cover. Seventeen of the GEFT's I8 complex figures were taken from the EFT. To
minister the GEFT, puticipmts are each given a test booklet including two practice
problems, a first section, a second section, and a third section. Two minutes are allowed for
the first section, and 5 minutes each for the second and third sections, after which the test
booklets we collected Instructions asked the participants to trace certain simple forms
within the more complex figures provitbd Scoring is the total number of simple forms
treaed correctly in the second and third sections. The higher the more, the more field
intbpenmnt the subject. An internal mnsistency of .82 has been found for both males and
females taking the OEFT, using the Spearman-Brown prophecy formula. Witkin at a1.
(1971) report correlations between the GEFT and the EFT that seem reasondily high (.82 for
men did .63 for women). The GEFT has also been evaluated in terms of its relationship to
another measure of psychological differentiation, the agree of articulation of the body
concept (the ABC scale). Correlations betwmn the ABC scale aid the OEFT yielcbd .71 and .55
for males and females, respectively, stlgpsting a reasonable amount of overlap between these
two measures of psychological differentiation. Hen show a slimt but significant (n< .005)
Wicytooutperform women on theOEFT (Witkin, Oltman, Raskin. & Karp, I971).

Lusk aid Wrimt (1981), however, noted that when differences in curricula are taken into

36

aacount, performance differentials between men and women disappeared They also
Monstrated that learning occurs airing the GEFT, in that scores on the second half of the test
are mnerally higher than scores on thefirst half. A study by Carter and Loo (I980) aimed
at providing further data on the norms for the GEFT found both collew males and females were
more field intbpencbnt than those mrlier reported in the MFT manual. They 81mm that
these differences may reflect differenses in educational settings and cultural charms in the
pest (more.

Wm Three measures of perceptual accuracy were obtained The first was
obtained from the use of the Interpersonal Check-List (LaForm & Sumek, 1955). a
self-atninishred mjectiva checklist consisting of 128 items. It wu cbvised as a set of
interpersonal variables listed in a circumplex of .16 or eight segments, each representing
characteristic ways of relating to others. Two orthogonal dimensions undergird the ICL;
Dominance-Submission (DOI‘I) on thevertical axis, endLove-Hate (LOV) onthe horizontal
axis. In minismring the ICL, amh participant was given a booklet and was asked to tbscribe
self and specified others by filling in the circle in front of each item as to whether it was or
was not Inscriptive of the person that was being ascribed Emh item was to be imntified as
either true or false for each target Participants were encourqu to m with their first
impressions and to work quickly. Results of the ICL can be plotted on the diagnostic g‘id
consisting of the eight segnents. Support for the validity of the ICL‘s octant constellations and
underlying interpersonal dimensions was proviced by Lange's (1970) study. LeForm's
manual W112i. (1973) we reliability test-retest correlations ranging from .64 to
.77 and cited several studies showing evicbnce relating to the validity of the ICL. Wimins

37

(I982) provimd a thorough review of the ICL, along with other major two-dimensional
representations of interpersonal behavior. To assess the perceptual accuracy of woup
members, the discrepancy between the DOM aid LOV score given a particular g‘oup member
and the DOM and L01! score that particular woup member we her/himself on the ICL was
(btermiried As noted earlier, (388 pa 20) both the DOM and LOV score resulting from the
perceived individual's ratings of how s/he sees her or himself and how s/he thinks s/he is
seen by others were explored as criteria for judging accuracy. Both types of self-ratings
were inclum for the purpm of exploring the relationship between the two criteria.
Pa‘ticipants were asked to use only each member's in-group verbelizations and behavior as a
reference for ascribing that individual. In each mrceptual accuracy task, participants were
mked to rate emh individual ascording to how the rater saw that particular individual, not
moording to their prediction of how the ratae saw her/himself.

A second measure of perceptual accuracy was more straightforward and less lengthy
than the ICL's. This consisted of mking participants to rate each member of their group
(including themselves) on four 9-point scales; Dawes of Dominance, Deg‘ae of
Submissivenees, Degree of Warmth and Caring, and Degree of Hostility (see Appendix A).
This second task (the “Interpersonal Scales") was a more direct conceptual-perceptual
accuracy task, as oppmed to the more verbal and molecular nature of the ICL task. Although
the final ICL tabulation yields scores on the axes of LOV, and DOM, these summary scores
are the result of the rater's responses to 128 trait adjectives that potentially ascribe the
ratae. The Interpersonal Scales required a more global assessment of where the ratae stands
with remrd to these four dimensions.

The final perceptual muracy measure waslbtermined from the scores on both of

38

the prior perceptual accuracy measures, the Interpersonal Checklist and the Interpersonal
Scales. It consisted of the absolute value of the discrepancy between the rater's ratings of the
rates on the Interpersonal Scales (Dominance, Submissiveness, Warmth and Caring, Uld
Hostility) minustheratee's self ratingson thefour dimensionsof the ICL (DOM, LOV, SUB,
aid HATE). For example, if Mary mve Debbie a 7 on the Dominance subscale of the
Interpersonal Scales, and Debbie's self-rated ICL DOM score was a 12, then May's
perceptual accuracy score would be S; simifying that she was discrepant by 5 points from
Debbie's own rating of herself. This final perceptual accuracy measure assessed the
discrepancy octween the rater's global conceptualization of the ratae along the four
dimensions of the Interpersonal Scale measure as compared to the self-rating of the ratae
resulting from a tabulation of her/his responses to the 128 cbscriptive adjectives of the ICL.
All three measures were inclum in an attempt to disesrn which mmure or what

combination of these three measures would be the most valid measure of perwptual accuracy.

WW Three measures of psyclmocial competence were utilized in the
present study. The main measure was Acceptmce/Rejection of Others (ARO) and

Amptmoe/Rejectmce of Self (ARS) (Hurley, 1976a). (See pm 9- 12 for justification
of the use of these measures to assess psychosocial competence.) These scales were developed
based on the eviwice that a broad variety of interpersonal behavior can be represented by two
principal inthpenthnt dimensions (see Hurley, l976a). Because of the mneral independence
of the ARS and ARO dimensions, they can be mpicted orthoginally producing four quadrants.
Thus, as Hurley (I978) notes, the theoretical construct of psychosocial competence runs
along the diml from the low ARS, low ARO quarant to the high ARS, high ARO quairant.

39

To assess the ARS aid ARO dimensions, interpersonal perceptions were gathered on nine
semantic differential scales which were presented to the participants in a mini-booklet form.
The four scales used to assess ARS we. Shows feelings-mime feelings,
Expressive-Mend, Active-Passive, andDuninant--Submissive. Warm--Cold, Helps
others-"Harms others, Gentle-Harsh, andAccepts others--Rejects others were the scales
messing the ARC dimension. The ninth scale, which was presented first, was
Like-Dislike. It was included in an attempt to minimize the influence of liking or disliking
on the eight subsequent scales. Each member was asked to rate self and all other group
members on each scale, which ranges from 0 (negative) to 9 (positive). Thus, 36 was the
highest obtainable score on ARS or ARO. Group members were asked to use the widist possible
rm of ratings and to consider only behavior they had directly obmrvad within gsup
sessions as the criterion for rating Evidence supporting the construct validity aid
reliability of ARS md ARO has been presented by Hurley (l976b, in particular see Table 3).
Two rank-orderings by group members as to the level of psychosocial competenm
displayed by each termt mrson within their woup also served as measures of psyclmial
competence. The first (Rank Ordering Exercise 1; See Appendix 8) consisted of a shaetof
peer with a (bfinition of psychosocial competence followed by enough blank spaces to
accommwete every group member. The instructions asked the group member to
hierarchically orw the members of their group (including self) from most to least
psychologically competent in terms of this definition. If they themed two group members were
mout equal in their likeness to the definition they still were required to rmk them, but
allowed to indicate a tied rank. However, no more than two members could be portrayed as

tied by my participant.

40

The second rank-orcbring measure (Rank Ordering Exercise II; See Appendix C)
was similar to the first with remrd to the format and instructions, except members were
asked to rank-orchr everyone in their g‘oup (including self) according to their personal
definition of interpersonal competence as written out at that pays bottom. Both measures
were inclum to ascertain if there was an mpreciable relationship between ratings based on

an externally defined versus a more internal and subjectively tbtermined refinition of
psychosocial competence.

Me

As a part of the regular class curriculum, group members were given theARO-ARS
booklet of scales to cheeribe self and all other participants in their group twice (bring the
term; after about 22 hours and 43 hours of g‘oup participation. About one week after each
ministration, every group member received mmplete feedback about these ratings in the
form of a copy of the precise ratings given her/him by each other member plus a graphic
display of the discrepancies between self-ratings and those assigned each person by each other
g'oup member. Members were encouramd to discuss these discrepancies in the hopes of
better understanding these perceptual differences and enhancing both the group process and
their own psychosocial skills. Depending in part on the mowing behaviors and persistence of
the facilitator( 3), these discussions were math more intensive and rich in some mp3 tInen
in others. Only those ARS-ARO ratings given near the term's end (i.e., after mproximately
43 hours of smell g‘oup interaction) were utilized in the current investigation.

Both the RC1) and the DEFT were ministered to participants wring their weekly
diuretic class meeting near the midile of the 10-week term. The Interpersonal Competence

4i

Scales and the two rank-uttering measures of psychosocial competence were given to the
participants wring class time near term's and to mecimize the amount of prior group
experience with each otMr. During this same class meeting the ICL was given out to take
home and return the following week. Group facilitators were given all of the measures during
their weekly supervision sessions on approximately the same (by that group members
received theirs in class. Most members and facilitators who were absent the (by these

measures were given out completed them soon afterwards in the experimenter's office.

Wanna

MD. A psychosocial competence score was determined from each irndividual's
ARS-ARO scores by multiplying their mean ARS more by their mean ARO score. Only the
scores given by others were included in theARS andARO means. Scoring the measure in this
way we equal weight to the importance of self-acceptance (ARS) and other-mptance
(ARO) with regard to psychosocial competence. Thus, someone who scored 20 on
other-mptance and IQ on self-acceptance was scored at the same level of psychosocial
competence as someone who scored 10 on other-acceptance and 20 on self-wceptance.
Although these two individuals might differ markedly in their behavioral style of intermting
with others, their functional level of competence within the interpersornal group setting
would be the same. The individual who is low in acceptance of others and hign in acceptance of
self might be more obviously incompetent psychosocially as compared with the mirror
opposite, as s/he would be more apt to overtly offend others, be lacking in uncerstanding,
empathy and warmth towards others, and tend to disregard othws' viewpoints. The mirror

opposite individual (hign in other-acceptanceandlow in self-mceptance), however, mignt

42

be emally psyclmially dysfunctional within an interpersonal row. The latter person
would be apt to have trouble initiating, expressing her or his own opinions, confronting,
expressing engr, showing self-mpreciation, and interacting in a mutual give-and-take
manner with others. In a woup whose explicit purpose is to learn about self arnd others
through intermting with them, the individual who is low in self-mceptance would not likely
be seen as highly competent psychwocially.

The rank-orcbr psychosocial competence measures were scored by calculating the
sum of the ranks given each group member (includirng self-ratings), then converting the
scores obtained for emh g‘oup into quantiles. This was dine to assure uniformity in scores

between the groups cbspite the varying sins of the groups.

2W Perceptual accuracy as measured by the 101. ratings was
determined first by scorirng the two principal undergirding dimensions (Lov and Don)
moording LaForge's (I973) formula. Perceptual accuracy was then determined by the
mscrepancy between an irndividual's ratings of another on the two axes of the ICL, and that

other person's self ratings. This was assessed by utilizing the stanmrd formula for

 

demrmining Euclidian distance: R: -/(Dom1-Dom2)2 + (Lev, + Log): Lerorge

(I973) has noted that this procedure is relatively free from the disturbing effects of
elevation, scatter, and depentbnce. There is an almost complete lmk of correlation between
thetwo dimensionsof Dominanceand Love (LaForw, I973).

The second measure of perceptual wcuracy involved direct ratings of each group
member (irncluding self) on emh of four scales: Dominance, Submission, Love, and

Aim-Hostility. Ratings were on a scale of 1 to 9 , in increments of 0.5. Perceptual

43

wcurmy scores here were determined by subtracting self-ratings from the man ratings
received from others. Discrepancy scores consisted of the magnitude of the difference
betwwn self-rating and mean ratings received by others. No attention was paid to the
direction of the difference. For example, if person X rated herself as 4.5 on the Dominance
scale, and person Y rated her as 7, the discrepancy would be 2.5. To (intermine person Y's
level of perwptual accuracy in rating group members the absolute value of all of This
discrepancy scores would be summed then divicbd by the number of members in the group
(excludirng himself).

The scoring of the third perceptual accuracy measure was jointly mtermined from
both of the prior perceptual accuracy measures, the ICL, and the Interpersonal Scales. The
process for the scoring of this measure, labeled Conceptual Accuracy was nblineated on pam
38.

Because the conceptual variables of psychosocial competence and perceptual mouracy
have not been fully valicbted, the author followed the principle of Multiple Operationalism
(Crano 8e Brewer, I973). This prirnciple was based on the likelihood of an imperfect
correlation between the conesptual variable and the observed manifestation. It, therefore,
involved recognizing that no single observation provims enough information to define a
tlnaoretical concept (Crano at Brewer, I973). To increase the probability of obtaining a
valid operationalization of constructs of psychosocial competence and perceptual accuracy in
the current investigption, more than one measure was proposed Based on the statistical
properties and face validity of these measures, the author constructed each construct's best

operationalization.

W, Cognitive complexity was scored according to the
specifications in the manual (Crockett et al., 1974). All cognitive complexity mta were
scored by two raters, after which interrater reliability was interminw. Final cognitive
complexity scores averaged the scores amigned to each participant by each rater. The final
scores were diviced into quartiles for use in testing the hypotheses, as no known norms were

available.

W. Scoring of field independence was according to the procedures
outlined in the manual (Witkin et al., 1971). Collem norms obtained on this measure by
both Witkin et al (I971 ) and Carter and Lao (I980) were not utilized to establish the GEFT
mutilas. This was due to an unbalanced distribution of the scores obtained by the present
sample when they were divicbd into un'tiles according to either of the above-mentioned
norms. Instead, the quartiles were established using the range of GEFT scores obtained by

the prwent sample.

RESULTS

1. Preliminary Statistical Analyses
A. Woman

1. Interrater Reliability:

Two raters, the present author and an undarwaduate work-study student rated
all of the cognitive complexity (eta provided by the Role Category Questionnaire (R00)
wording to Crockett et al.'s (1974) manual. Each rater had approximately five hours of
practice in rating R005 provimd on a different sample before rating the present (eta. The
results of these practice ratings were then discussed jointly and discrepancies were accounted
for in an effort to clarify differences in ratings. Interrater ROQ reliability for the current
study is presented in Table 1 and was typical of interrater reliability correlations found in

other R00 works (O'Keefe & Sypher, 1981).

TABLE I

Interrater Reliability for Crockett's RCO (N = 55)

BM 3W film W m
l 7- 60 21.1 742
.90
2 6- 58 22.8 773

2. Distribution and Significant Correlations of the R00:
Rm scores were obtained for 51 of the 53 participants in the study. The

45

46

remaining two were absent from class the day it was given and did not respond to attempts to
arrange another administration. The mean score for cognitive complexity was 22.8, with a
stancbrd deviation of 8.8 and a range of 6.5 - 59. As no previous Rm norms could be found,
the present scores were divided into quartiles. Cognitive complexity (raw score) was found
significantly correlated with sex ( r(51)= .42, p < .001 ): women ( N: 25.9) scored well

above their men peers ( N = 18.7).

a. modem:
1. Distribution:

General college norms were not used to divide this sample into quartiles because
thater en unmual distribution of scores. Instew, quartiles were mterminad bmed
on the obtained distribution of scores. As compared to Witkin et. al.'s (I971) college norms,
the present sample lacked a sufficient number of person's scoring in the second (less field
independent) quartile, and wording to Wter 8r Leo's (1980) later set of college norms
this sample lacked scores in the fourth (or most field independent) quartile. The mean,
mowiation, and rangofthe obtained GEFT scoreswere 12.3, 4.3, and 2- 18,
respectively. The distribution of the scores across quartiles determined by this sample's
cute, as well as the established college norms, is presented in Table 2. GEFT scores were
obtained from 50 of the 53 participants. The remaining three pe‘ticipants were absent from
class the (by it was given and did not respond to attempts to arratm enotlner administration.

Man and women did not differ significantly in their GEFT scores in this sample.

47

TABLE 2
Distribution of Group Embattled Figures Test
Across Quartiles Determined by College arid Sample Norms

131 2nd 3nd 31b
Sample Norms:
RanmofScores 2-9 10-13 14-15 16-18
N 11 12 11 16
Witkin Norms(N= 397):
RangeofScores(Male) 0-9 10-12 13-15 16-18
(Female) 0-9 10-11 12-14 15-18
N of CurrentSample 15 7 10 18
Carter&Loo (N = 266):
Rang:ofScores(Male) 0-11 12-15 16-17 18
(Female) 0-9 10-13 14-16 17-18
N of CurrentSampIe 17 12 I4 7

The following were the final four perceptual accuracy measures used. The first three
were thrived from the earlier specified perceptual accuracy measures, while the fourth was a
composite of these three.
i. 101. Discrepancy (lCL-DISCR):
This measure W how accurately a person rated others in her/his group
bad on the 128-item ICI. qnsstionnaire. Both rater and ratae filled out the mestionnaire,
andthe accuracyscore wasbasedonthe discrepancybetween each's LOV andDOM factor score

as was more fully described on peg 42. As noted on peg 37, two types ofself-rating were

48

used: (a) the ICL more resulting from the target individual's self-ratings (es inscribed
move) and (b) another resulting from how the target individual thought others viewed
her/him. Both self-ratings were collected to explore their relationship. An investimtion of
the differences between these measures on the ICL DOM and LOV axes produced nonsignifimnt
findings. Due to a substantial correlation between the two ratings ( r(53) = .72, p < .001 )
and a need to reduce the number of perceptual accuracy measures, the latter ratings were

cropped There were no significant sex differences on the retained measure.

2. Interpersonal Scale Measure of Dominance and Warmth and Caring (DOM-WC):

This mono measure of perceptual accuracy involved participants' direct ratings of
self and their group members on the following 9-point scales: Dominance, Submission,
Warmth and Caring, and Hostility. Accuracy scores were determined by subtracting each
ratee's self- ratings from those given by each other group member. These different ratings
across the four scales lacked consistency and included negative intercorrelations between some
of the subscales (Hostility versus Dominance; r(49)= -.22, ns, Hostility versus Warmth
and Caring; r ( 49) - -.18, n3), resulting in participants averaging approximately the same
scores. This was resolved by using the mean discrepancy on each subscale for determining the
final score, rather than the average discrepancy across all four subscales Discrepancies on
only two of the subscales were significantly positively correlated (Hostility versus
Submissivensss; r (49)- .42, p < .001), and neither of these correlated positively with
the other perceptual accuracy measures. After reviewing these correlations, it was (bcitbd
to combine subscales Dominance with Warmth and Caring for use as the second measure of

49

perceptual accuracy. This was due to their positive but very wwk association (r= .16, ns)
aid especially their significant, although modest, correlationswith the other perceptual
accuracy measures used in this study ( see Table 3, p. 50). Both woman and more cognitively
complex subjects in the present sample scored as significantly less accurate on this measure
(r(49)=.27, p<.05 and M48) = .32, p< .01, respectively). The fmt that Dothwef‘e
simificant may be related to the earlier reported correlation between sex and cognitive
complexity (r (51) = .42, p < .001). Post-hoc examination revealed that sex was
significantly related only to accurate ratings of the Dominance subscale (r(49) - .26, M
.05). Cognitive complexity, however, was significantly related to perceptual accuracy on
both the Dominance and the Warmth and Caring subscales considered separately ( r (48) =
.23, [M .05; r(48) = .26, p< .05, respectively).

3. Conceptual Accurmy (CONCEP-AC):

Labeled conceptual accuracy, a third measure assessed the discrepancy between the
rater's global conceptualization of the ratae along the four dimensions of the interpersonal
Scale measure as compared with the ratee's self-ratings yielded by a tabulation of her/his
responses to the ICL'S 128 thscriptive adjectives. There were no significant sex differenees
on this measure, but a week although significant positive correlation between conceptual
accuracy and GEFT (r(47)= .23, p< .05) indicated that more field-inchpencbnt persons

were slightly less conceptually accurate in this sample.

4. Composite Perceptual Accuracy Measure (COMB-PA):

The final perceptual accuracy measure aggreth the three measures just

50

ascribed; ICL Discrepancy (iCL-DISCR). Interpersonal Scales Dominance plus Warmth and
Caring (DOM-WC), and Conesptual Accuracy (CONCEP-AC). in light ofthe weakly positive
intermaasure correlations, these three were combined This reprwentad an attempt to
increase the probability of obtaining a valid operationalization of the construct of perwptual
accuracy (see the principle of Nulitiple Operationalism, Crano at Brewer, 1973). This
was momplished by first converting each component to a z-more, then summing across the
measures.

Table 3 shows the intercorrelations among all four perceptual accuracy measures. As
with the earlier reported measures, several instances of missing data reduced the sample IV
for this composite.

TABLE 3

Correlations Among Perceptual Accuracy Meeeures (N- 46)

ICL-DISCR 29* .19 .69***
DON-WC - .30‘ 34““
CDNCEP-AC - .69‘"m

* p( .05, ii“ [H .001

0. Was:
The two psychosocial competence measures used were:
i. ARS x A120

51

This measure of psychosocial competence was drawn from the acceptance/rejeCtion of
self (ARS) and wceptance/rejection of others (ARO) scales developed by Hurley (1976a).
As discussed on page 41, ARS x ARO scores were determined by multiplying the mean ARS and
ARO scores given each participant by her/his small group members. Mean, standard deviation,

and ram for ARS and ARO as given by others, and ARS X ARO are presented in Table 4.

TABLE 4
ARS, ARO, and ARSxARO

Mean, 5.0., UldRarm (N- 53)

mob]: Elfin .5.D, bl Boom
ARS (otharS) 27.0 3.1 53 20.4-33.8
ARO (others) 28.8 2.6 53 24.2-34.6
ARS x ARO 780.1 129.6 53 493.7- 1 169.5

2. Rank-Order Measure of Psychosocial Competence (RANK):
Participants were asked to complete two rank-ormr measures of psychmial

competence for this study; the first (R01) asked them to rank self and all other group
members according to a supplied cbfinition of psychosocial competence, while the second
(R011) wked that they rank-order all according to their personal definitions of psychosocial

competeme (more information on these measures was presented on pegs 39-40). Scores for

52

these measures were determined by calculating the sum of the rank-ortbr scores given each
group member, then converting each group's set of summed scores into quartiles. This was
cbne to assure uniformity in scores between the groups tbspita their size differences. Final
rank-order wores were reverse-scored (i.e., so that the fourth quartile was the highest
pwchosocial competence more and the first quartile was the lowest) for ease in
' interpretation. Results showed these measures correlated substantially (r(53) =.74, p <
.001), therefore the two measures were combined into a composite ”RANK“ measure.
Post-hoe analyses calculated in an attempt to explain the unaccountable covariance
between these measures yielcbd differences in each's correlations with sex, cognitive
complexity, and other-given ICL DOM scores. Specifically, women and the cognitively
complex were ranked significantly higher on R01 (r(53) = .26, p < .05, and r(51) = .31,
p< .01 , respectively), while them ranked highly on ROII tended to be given higher ICL

Dominance scores (r(47) = .28, p< .05).

3. Simificant Correlates of RANK:

RANK had several interesting correlates. Group facilitators were seen as
significantly more psychosocially competent (r (53) = .55, p < .001) than group I
members. This correlation proviths some concurrent validity of the RANK measure, in that
facilitators were selected for this role partially because of their apparently mod
interpersonal skills. They were also more experienced than first-time members in the group
setting Another interesting correlation was that of RANK and self -rated self-acceptance
( r (53) = .40, p < .002). These results sugmsted that individuals who were ranked highly
on psychosocial competence also tancbd to rate themselves higher on self-acceptance, which is

53

partially supportive of the postulate presented earlier; that high psychosocial competence
consisted of a high agree of self-acceptance (i.e., high ARS) and a high degree of acceptance
of other (i.e., high ARO). Self-rated ARO and ratings of ARS or ARO eesimad by others
(those scores used in the ARS X ARO measure), however, did not correlate signifimntly with
RANK.

4. Correlations Among Psyclmocial Competence Measures:
Male 5 presents the correlations among most of the above-mentioned
psychosocial competence measures. The lmk of a significant correlation between the two final

measures of psychmial competence, ARS x ARO and RANK, is noteworthy and it's meaning
will be explored in the Discussion section.

TABLE 5
Correlations Among Psychosocial Competence Measures (N = 53)

802 A1315;+ ABD” ABSLABD. BANK
R01 14*" -.oo .oo .01 33*“
R02 - -.19 .12 -.05 33*“
ARS - .52.“ .90 mi .10
A110 - an" - .07
ARSxARO - .03

*Scores given by others p< .001

54

II. Results of Hypotheses

 

1. Hypothesis is: There will be a significant difference in psychosocial
competence between subjects scoring in the different
quartiles of field independence.

Hypothesis is predicted that psychwocial competence would vary across the quartiles
of field intbpenthnce. Analysis of variance was used to test this hypothesis. Results showed
that there was a slight statistical trend for the ARS x ARD memure to be mtimt on
subjects‘ field independence scores, but neither ARS x ARO nor RANK was significantly linked
tofield intbpencbnce (F(3) = 2.21 , p < .10, ns; F(3) = 0.47, ns, respectively).

2. Hypothesis 1b: Level 11 field incbpendence scorers will score
higher on psychosocial competence than either Levels 1 or 11!
field indepencbnce scorers.

In comparison to the move hypothesis, this hypothesis made a more specific
prediction 081an field inmpencbnoe level and psychosocial competence. The assumption here
was that persons scoring in Level 11 of the GEFT would be more psychwocially oriented then
the others, andstill havea greater capacity for understanding other's viewpoints. Due to
the specificity of direction that was predicted, the one-tailed [-test was used here. There
were no significant results using the RANK measure of psychosocial competence. 0n the ARS x
ARO mpendent measure there was partial support for the hypothesis. Level 11 field

incbpencbnce scorers were significantly higher than Level I scorers ( I (21) = 2.62, p <

5'5

.01), but they did not more significantly higher than Level 111 scorers ( ((26) = .69, ns)
(see Table 6). While not part of the hypothesis, Level II scorers, as would be expected, also
wored significantly higher on ARS x ARO than Level 111 scorers ( ((21) = 1.95, p< .05)

(see Table 6).
TABLE 6
Mean Psychmial Competence Scores
at Four Field Indepencbnce Levels

ABSXABQ .N. Mean .SD.
1 l 1 708.81 95.72
LEVEL: 11 12 832.63 128.78
111 1 1 728.49 127.36
iv 16 795.37 158.25

BMK

1 1 1 4.82 2.27
LEVEL: I I 12 5.25 2.45
111 1 1 5.73 1.95
W 16 4.75 2.38

3. Hypothmis 1c: Level 111 field independence worers who also scored
above average on perceptual accuracy will show higher psycho-
social competence scores than either Level I or iv scorers.
The unwrlying assumption here was that although Level 111 field indepeimtt scorers
should theoretically be less psychosocially competent than those less field independent,

56

unusually high perceptual accuracy skills might enable some Level III scorers to be more
competent in their interactions with others. The criteria groups selected were Levels I at
11!, since level 11 was still expected to have the highest psychosocial competence scores. The
hypothesis was not supported by the data. The results showed mean ARS x ARO scores
received by those Level 111 GEFT scorers who were more perceptually accurate on ICL-DISCR
(N = 673.6) were actually lower than Level 111 field independence participants who were
less parwptually accurate on ICL-DISCR ( N: 759.8), although to a nonsignificant cbgrae( I

(9)= 1.26, ns).

8.0'1'01'1121'101'

l. Hypothesis II: There will be a significant positive relationship between

cognitive complexity and psychosocial competence.

The assumption here was that more cognitively complex individuals should have
the potential for a greater understanding of others, and that this understanding could be
manifested in more constructive interactions with others. This hypothesis was weakly but
statistically supported for RANK (r(51)= .261, p< .05), but not for ARSx ARO (r
(51)= -.056, ns). As mentioned earlier, there was a significant correlation between
cognitive complexity and sex (with females marina higher), but sex linked to neither RANK
or ARS x ARO, indicating that sex was not a confounding variable in this finding.

2. Hypothesis 111: Among high cognitive complexity scorers, those who
are more perceptually accurate will have higher

psychosocial competence wares.

57
Totest this hypothesis, cognitive complexity scores and the scores of each of the
perceptual accuracy measures were dichatomized at the mean. Eight ANOVAS were com-
puted (both of the psychosocial competence measures and R00 scores by each of the four
perceptual accuracy measures). This yielded an N of 9 to 13 participants who scored high on
the R00 but low in perceptual accuracy and an IV of 10 to 14 participants who scored high on
both the R00 and perceptual accuracy (mpending on the perceptual accuracy measure).
None of the results supported this hypothesis.

 

1. Hypothesis iv: There will be a significant positive relationship between
perceptual accuracy mid psychosocial competence.

One of the underlying assumptions for hypothesis IV was that persons viewed as the
most psychosocially competent by other group members would have relatively more accurate
perceptions of these peers. This hypothesis was not supported, as is apparent in Table 7
below.

TABLE 7

Correlations between Psychosocial Competence

and Perceptual accuracy
BANK ARSXABQ
LDISCR -.05 (N - 46) -.04
DOM-WC .12 (N = 49) .06
CONCEP-AC -.07 (N = 48) -.03

COMB-PA -.03 (N = 46) -.01

S8

DLptenmtionsbatweenXacinblea

1. Hypothesis V: There will be a simple interaction effect between the cognitive
complexity of raters and ratees (above average versus below
average) with remrd to perceptual accuracy.

More particularly, it was predicted that those scoring highest in cognitive complexity
would be more perceptually accurate in rating cognitively complex individuals, whereas the
cognitively simple would be more perceptually accurate in rating other cognitively simple
participants. Of the four perceptual accuraw variables, significant results were obtained
only with CDNCEP-AC, where the cognitively simple were significantly more accurate in
rating other cognitively simple participants (versus rating cognitively complex participants)
(1(81) = -2.4, p< .05).

However, the coglitively complex were not more peresptually accurate in rating
other cognitively complex participants. instead, the cognitively complex were significantly
more perceptually accurate in rating cognitively simple than cognitively complex persons ( t
(96): 2.53, p < .01). The mnitively complex were also significantly more accurate in
rating the cognitively simple than the cognitively simple were in rating the cognitively
wmplex (I (96) = -2.79, p< .01 ). A post-hoc t-test computedto ascertain whether
there was any overall significant difference in accuracy of rating among the comitively
complex versus the cognitively simple was insignificant ( I ( 179): -.61 , ns). Sex did not
correlate significantly with CONCEP-AC. (It should be noted here that the term chgreee of
freemms are due to the fact that each group member's scores were paired with every other
group member, resulting in a much larger number of comparisons then the total number of

participants. )

59

2. Hypothesis V1: Those individuals who score in Level I of field
independence will obtain significantly higher
scores on Acceptance/Rejection of Others (ARO) than will
those who wore in Level 1V, while Level 1V participants will
obtain significantly higher scores on ARS than will Level 1
participants.

Hypothesis VI was drawn from the theoretical uncbrpinnings of the concept of field
independence. Low GEFT scorers were postulated to be more other-accepting, and high GEFT
more self-wcepting. One-tailed 1' -tests were used to test this hypothesis, due to the
specificity of the direction that was predicted. Neither hypothesis was supported, although
there were trends, as expected on ARS and contrary to expectations on ARO, for Level iV

scorers to be higher than Level 1 worers on both measures (see Table 8).

TABLE 8
t-tests, Level 1 Versus Level IV of GEFT with Regard to

AweptanceofSelfandOthers(Levell IV- 11, Level 1V N- 16)

ARO
LEVEL 1 27.75 2.2 - 1.41 .09
LEVEL w 29. l 4 2.9

ARS
LEVEL 1 25.50 3.0 - 1.31 .10

LEVEL IV 27.10 3.0

DISCUSSION

The present study investigated the relationships between several mewures of
cognitive style and psychosocial competence. The purpose was to determine whether
hypothesized linkages among these constructs could be confirmed. Three major cognitive style
variables, field independenm, cognitive complexity, and perceptual accuracy, were
investigated. Psychosocial competence was measured by participants' rank-orderings of each
other along this dimension and by a separate measure that combined peers' ratings of
self-acceptance and other-wcaptahce. Overall the findings were mixed and complicated by
the absence of a significant correlation between the two psychosocial competence indicators.

The discussion is divided into three sections. The first discusses features of the
variables used in the study that may significantly (bar on the results. This is followed by a

discussion of the central findings and finally, asummary of the overall findings in this study.
1. Variables Used in this Investigation: Pertinent Factors

A. Psychosocial Competence Meeeures:
1. insignificant Correlation between Measures.
Two very different measures of psychomial competence were utilized. The
ARS x ARO measure was less direct in that it represented a presumed manifestation of
psychosocial competence: the degree of self- and other- acceptance. The rank-order
measure (RANK). on the other hand, asked the rater todirectiy assess the participants'

60

61

psychosocial competence. It was presumed that these two psychosocial competence measures
would correlate positively and significantly. Earlier research on the ARS x ARO (Hurley,
l976a, l976b, T978) index had found it correlated with variables associated with
psychmial competence. Yet the ARS x ARO incbx and RANK failed to correlate significantly.

Neither measure fared worse than the other in relating to the three cognitive style
indicators. RANK correlated significantly but weakly with ROG-based cognitive complexity,
while ARS x ARO did not. ARS x ARO related to the GEFT scores weakly, but more strongly
than did RANK. As a check of the reliability of the psychosocial competency measures,
participants' ability to significantly differentiate between group members in their ratings of
each other was evaluated. This was calculated for ARS x ARO by analysis of variances for
correlated groups, and for the rank-ordering measures with Kentbll's coefficient of
concordance (W), a measure of the association of ranks (see Kerlinger, I973, p. 292). A
conversion formula to the F-ratio was then used to evaluate W's significance. Both the
rank-orcbr meaures and ARS x ARO fared well in this reliability check. Eight of the 10
groups demonstrated significant reliability on the ARS x ARO measure (7 at p <. 01 and l at p
< .05), accounting foranaverageof 40.43 of the variance (range: 115-728 ). Significant
results were obtained with 7 of the 10 amps on the first rank-order measure (6 at [K .01 ,
and l at p< .05), which had asked participants to rank all group members according to an
explicit (bfinition of psychmial competence taken from the constructs of self- and
other-meptance. The second rank-ortbr measure, which asked participants to rank each
other according to their personal definition of psychosocial competence, also did well. Here
all 10 of the groups obtained significant results (8 at p< .01, 2 at m .05).

The psychosocial competence variables were also investigated for their linkage to

62

liking, in an attempt woertain if either was confounded with liking for others. A substantial
linkage was found between ARS x ARO and a scale ”Liked — Disliked" ( r= .Sl, p< .00l ).
while the correlation between RANK and liking was nonsignificant ( r = .l l, ns). These
findings suggest that the more liked persons were more likely to be rated highly on self- and
other-acceptance. Thus, ARS x ARO scores were linked more to regree of liking for the other
than were RANK scores.

There are several other noteworthy differences between the two pwchosocial
competence measures that may have contributed to their lack of correlation. These concern
differences in their confidentiality, different use of self-scores, and the use of a floating
reference point with respect to groups for the ARS x ARO scores.

With regard to differences in confidentiality, it will be recalled that ARS x ARO
ratings were shared with group members as a regular part of the class format, whereas the
instructions for the RANK measures specified that thefi rankings would be kept wnfitbntiel.
As a consequence, ARS x ARO scores may have bmn wnfounded with concerns about how
particular group members would react to the ratings given, whereas this was not an iesue for
RANK. This fact may partially account for ARS x ARO’s association with liking.

Another difference between the meeeures concerns the use of self-ratings. These
were explicitly exclutbd in the ARS x ARO tabulations, under the assumption that they would
contribute irrelevant variance to the final scores (i.e., that raters might have a tendency to
under- or over-rate themselves due to level of self-esteem or other factors). The RANK
mores, on the other hand, included self-rankings. Post-hos analyses revealed a significant
correlation between self-based ARS scores and RANK ( r(53) = .40, p< .002). This was not

the with self-based ARO. However, the correlation between ARS x ARO and RANK was

63

so low (r = .03), that adding these self-ratings to the ARS x ARO measure would still have
yielded a nonsignificant relationship between the two measures. Nonetheless, the significant
mrrelation between self-based ARS and RANK suggests that the inclusion of self-ratings on the
ARS x ARO measure may enhance its value, especially as a proposed indicator of psychosocial
competence.

A third important difference concerned the use of a floating reference point for the
ARS x ARO measure. RANK, it will be recalled, was determinm by dividing the mores given
within each group into quartiles, whereas ARS x ARO scores were not anchored within each
amp, but rather, were based on a floating reference point with respect to groups. To assees
whether this floating reference point for the ARS x ARO scores affected the results, pmt-hoc
calculations were done using z-scores determined separately for emh of the 10 groups. All
of these post-hoe comparisons were nonsignificant, sumting that the measure was not
importantly affected by the floating reference point.

in review of post-hot: investigations of the lack of a significant relationship between
the two psychosocial competence variables the following can be concluded:

l ) Neither fared better in terms of the conmptualized hypotheses.

2) Both measures appeared to be reliable.

3) There were several important and potentially influencial differences between the

measures, including:
a) ARS x ARO scores were more strongly associated with liking.
b) ARS x ARO mores may have been influenced by the fact that these ratings
were fully shared with ratees within each group.
c) Self-ratings were excluded in the ARS x ARO scores, whereas they were

64

inclum in the RANK scores.
d) ARS x ARO was owed on a floating reference point, however, post-hoe

analyses suggested that this difference was unimportant.

In this study's findings the lack of a relationship between the ARS x ARO measure «M a
direct measure of psychosocial competence makes it difficult to defend either index as a
satisfactory measureof psychosocial competence. It is importanttonote, however, that the
lack of a significant correlation between these two measures does not mean that one or both did
not measure psychosocial competence. RANK assessed participants' wncepwellzan‘ans of
psychwocial competence. The fact that these conceptions of psychwocial competence did not
align with the ratings of self and other acceptance (ARS x ARO) does not invalidate the latter's
pertinence to the skills of interacting with others in a competent manner. Had a third
measure of psychosocial competence been inclumd in the study, one perhaps better
mcumented as a valid measure of the construct, the question of the validity of these two
measures may have been more easily resolved. The difficulty here is that there is no well-
accepted definition of psychosocial competence, let alone a recognized measure of it. It was
the problem of the lack of a clear and rwognized conception of what is involved in psychmial
competence that this study was designm to explore.

Another equally plausable explanation for the lack of significant results in this
investigation is that self-selection and 40 hours of training in an interpersonal woup setting
with the goal of mveloping better interpersonal skills may have resulted in a restricted range
of psychosocial competence scores, thus eliminating the possibility of an moate test of the
hypotheses.

6S

2. Differences between ROI and ROII

As reported earlier (p. 52), post-hos analyses revealed several differences
between ROI and ROII that could account for some of the unaccounted covariance between the
two measures. Women and more cognitively complex participants were ranked more highly
on ROI while those ranked more highly on ROII tented to have been given higher scores on the
ICL Dom axis. The instructions for ROI asked that the participants rank-order according to
the individual in their group who "evaluates her/himself most favorably, gets along best with
others, and is the most understanding, caring, and aware of her/his impact on others.”
Traits such as being understanding, being caring, getting along well with others, and being
socially aware are suggestive of relationships with others, which has recently been specified
as an important defining aspect of women's development (see Gilligan, I982). It is possible
that the definition given for ROI pulled for more sex-biased responses from participants, who
then ranked women more highly then men. The high ranking of the cognitively complex on
ROI may also be related to the fostering in women in this culture of interests and skills in
interpersonal relationships. When participants were asked to rank wording to their
personal definition of psychosocial competence for ROII, however, the aspect of tbminance

became more important.

8. Group Embedded Figures Test Scores:

As noted in the Results section, general college norms for the GEFT were not used due
to the present sample's uneven distribution of GEFT scores. It is possible that the shifting
across the four quadrants through the use of improvised norms may have washed-out real

differences between actual scorers in each level. Had it been possible to use mnaral college

66

norms, the GEFT results may have been different. As compared with the most recent college
norms for the GEFT (Carter 8‘ Loo, I980), the present sample -- perhaps due to its
preponderance (about 742) of social science majors -- lacked scorers in the most field
incbpendent quartile, Level IV. Field inmpendent skills tend to be more mociated with
majors in mathematics and physical sciences, which were under-represented (about ISZ)

here. Future research could expect similar samples if subjects are drawn principally from

social science classes.

0. Relationship between Cognitive Complexity and Sex:

As reported earlier, cognitive complexity was found significantly correlated with sex
(r(Sl )= .42, p< .001). Women scored significantly higher on R00 cognitive complexity
then did men. For any finding related to cognitive complexity, an alternative explanation
would have been that the sex variable was a confounding factor. However, sex did not
correlate significantly with any of the variables with which significant findings were found
with cognitive complexity. This leads the researcher to conclude that cognitive complexity's
other linkages were not confounded with sex.

Prior research reporting RCQ sex differences was not found, so this finding was
unexpected. It would be interesting to test its stability in additional research; to ascertain if
women generally score higher than men on RCQ cognitive complexity. It seems likely that this
task's psychomial character of describing others in as much cbpth as possible within a
limited amount of time may have own less appealing to men than to women in this sample.
Sex—role stereotyping for women in this culture has generally tented to foster interests and

skills in interpersonal relationships (seeeillimn, I982).

67

D. Perceptual Accuracy Heeeeeur:

As noted earlier, there is no well-recognized and accepted measure of perceptual
accuracy. Therefore, in an attempt to incrm the probability of obtaining a valid
operationalization of the construct of perceptual accuracy in the current investigation a
composite perceptual accuracy measure was improvised. The significant correlations among
its components, although mnerally quite small (median r = .29), supported the general
construct of perceptual accuracy (see Table 3, p. 50). The initial statistical difficulties
with the Interpersonal Scales measure are noteworthy, however (see p. 48-49). It may be
recalled that each of these scales involved messing others on a unipolar subscale (i. e.,
Dominanm, Submission, etc). Scores on several of these subscales correlated negatively
with each other, however, and averaging across them resulted in participants obtaining
approximately the same scores. This suggests that these perceptual accuraw tasks may have
been overly specific and that unipolar scales may not provide adequate measures of the
wnstruct. Specific personality features of the rater, rather than a broarbr trait of
perceptual accuracy, may have been related to amh task. For example, person A may have
been more perceptually accurate in ratings others on the dimension of Dominance but not on
Warmth and Caring, whereas the opposite may have held for person 8, etc. The findings
suggest that, similar to the present investimtion of the psychosocial competence construct,
the construct of perceptual accuracy should not be gauwd on the ratings of single dimensions.

An attempt was made to explore the relationship between the two types of self- ratings
on the ICL-DISCR measure (see p. 37) by investimting whether they would differ on the ICL's
DOM and LOV axes. This also yielded nonsignificant findings. It is likely, then, that error

variance and nonspecific factors mounted for the 482 of each measure's variance not

68

included in their significant correlation ( r(53) = .72, p< .001).

Both women and the more cognitively complex in the present sample scored
significantly less accurateon DOM-WC (r(49) = .27, p< .05, and r(48) = .32, p< .01,
respectively). In a post-hoc analysis it was found that women were significantly lees
amurate in rating only on the Dominance subscale. When the post-hos analyses were applied
to cognitive complexity scores, it fared better. Cognitive complexity was significantly
related to each scale considered separately. However, neither the cognitive complexity nor
the sex finding should be weighed heavily besause neither correlated across other perceptual

accuracy measures.

ll. Discussion of the Findings

A. Psychmial Competence and field Independence:

The predicted relationships between the cognitive style variable field intbpendence
and psychosocial competence were only partially supported. It was predicted that Level II
GEFT scorers would be highest on ARS x ARO and RANK, followed by Level III scorers who
scored high on perceptual accuracy, Level IV scorers, and finally, those scoring In Level I.
Level II participants did score significantly higher than Level I scorers by the ARS x ARO
memure, as predicted, but not by the RANK measure. The prediction that Level II
participants would score significantly higher than Level IV scorers on ARS x ARO and RANK
was not supported. Perhaps Level IV inclucbd some participants who were “mobile“ with
respecttothis variable (i.e., manifesting high skills in both field-intbpentbnce and field-

depen¢nce).

69

The findings also failed to support the hypothesis that Level III participants with
better perceptual accuracy scores would obtain higher mores on ARS x ARO and RANK than
Level I or Level IV participants. In part this was due to the lack of a significant relationship
between perceptual accuracy and psychosocial competence. There was even a trend with
ICL-DISCR towards the contrary finding; namely, that the least perceptually accurate Level
III scorers were rated more highly on ARS x ARO than Level III participants with above
average perceptual accuracy scores. It is not clear to this researcher why thm moderately
high field independent participants who scored lower in perceptual accuracy scored higher on
self and other acceptance then did moderately high field inmpencbnt participants who scored
higher on perceptual accuracy. Perhaps the Level III participants who scored low in
perceptual accuracy were more similar to their field chpencbnt counterparts, who
thwretically may rim to be more accepting of self and others out of dependence on their
surroundings (or field) for the cognitive structuring abilities that they lack.

Hypothwes We and Vlb predicted that Level IV field independent scorers would obtain
significantly higher scores on ARS than Level I scorers, while these latter would obtain the
higher ARO scores. Making a prediction of differences between thm groups appeared sound,
although the results did not support the hypothesis. Instew there was a trend for Level IV
scorers to obtain higher scores on both ARS and ARO than those obtained by Level I scorers.
Level IV scorers in this investigation fared better than predicted on ARS x ARO, obtaining the
second highest scores on this measure--scores that were not significantly lower than the
scores obtained by Level II participants. It appeared that the cognitive abilities of the himly
field independent served them well enough in the interpersonal group setting to be seen as

more accepting of self and others than Level I and Level III participants, and not significantly

70

lower than Level II persons on self and other acceptance. The fact that Level IV participants
scored higher than Level I participants appeared reasonable, but this author cannot account
for their slightly superior performance over Level III participants, who would be more
similar than Level IV to Level II scorers. Although these results are somewhat perplexing,
the findings themselves point to the value of investigating this cognitive variable in social

interactions.

8. Psychosocial Competence and Cognitive Complexity:

The predicted significant positive relationship between cognitive complexity and
psychosocial competence was found with the RANK measure of psychosocial competence, but
not the ARS x ARO measure. This provides some support for the hypothesized link between
cognitive complexity and psychosocial competence. In line with previous conceptions related
to cognitive complexity, participants with higher cognitive complexity mores were wen as
interacting more competently in an interpersonal group setting, as compared with their less
cognitively complex counterparts.

The prediction that the combination of a high thgree of cognitive complexity and
comparably higher scores on the study's perceptual accuracy measures would increase
psychosocial competence abilities (as wmpared to those who score high in cognitive
complexity alone) was not supported. Again, these findings were apparently linked to the
lack of a significant relationship between participants perceptual accuracy and psychosocial

competenm SCOPBS.

71

C. Psychomcial Competence and Perceptual Accuracy:

As mentioned above, the prediction that perceptual accuracy and psychosocial
competence would be significantly and positively related was unsupported. These results
suggested that accuracy in perceiving or construing others was not related to how accepting
they were of self and others, or how competently the participants in this study interacted, as
measured by RANK. The lack of support of this hypothesis on all four of the perceptual
accuracy variables strengthens these findings.

Significant findings were obtained with the CONCEP-AC measure of perceptual
accuracy, however, in predictions about the relationship between perceptual accurmy and
cognitive complexity. It was predicted that those smring above average on cognitively
complexity would be significantly more accurate in rating other cognitively complex
participants, whereas the comitively simple (those scoring below average on cognitive
complexity) would rate other cognitively simple participants significantly more accurately.
Findings supported the prediction that the cognitively simple would rate other cognitively
simple participants significantly more accurately, but not the prediction that the cognitively
complex would rate other cognitively complex participants more accurately. In fact, the
pattern that emermd from the data was that both the cognitively simple and the omnitively
complex were significantly accurate in rating the cognitively simple, whereas neither the
cognitively simple nor the cognitively complex rated the cognitively complex well.

The unexpected low perceptual accuraw scores found for the group of mnitively
complex participants in rating other cognitively complex participants may have been due to a
tendenw to perceive others with too much complexity and differentiation. Or perhaps the

cognitively complex were more difficult in mneral to rate due to their greater agree of

72

differentiation. In spite of the assumed greater complexity and differentiation of the
cognitively complex, however, they were able to accurately rate the cognitively simple
participants to a significant agree; in fact, as well as the cognitively simple participants.
It may be that being requested to rate their perceptions of the cognitively simple participants
only on the four dimensions of the Interpersonal Scales, rather than to rate their perceptions
of them in a more global and less specifically definw manner, helped restrict distortions that
the cognitively complex participants might have made in perceiving the cognitively simple
participants.

Whatever the possible reasons, the relationship between levels of cognitive
complexity and perceptual accuracy is an interesting area standing in need of further
investigation, according to the present data. One possible area for investigation is the effect
of the manner of assessment of perceptual accuracy. For example, had the raters been asked
to rate their group members according to how they thought each of the members would rate
her/himself (as opposed to according to how the raters themselves saw the group members,
as was done here), a different, and perhaps more valid measurement of perceptual accuracy

might have resulted.

III. Summary:

The most serious limitation in the present investigtion was the lack of a significant

correlation between the two psychosocial competence measures. This rencbred conflicting
raults, such as those obtained on Hypothesis II, where one of the psychosocial competence

73

measures was significantly related to cognitively complexity (as predicted), but the other
was not. This limits strong conclusions about the relationship between the cognitive style
variables used in this study and psychosocial competence. As mentioned in the literature
review, part of the difficulty with an investigation such as this is the lack of established
mreement and instrumentation on the psychosocial competence construct.

Despite this important reservation, the present work contributed several
interesting results to the literature of research attempting to delineate cognitive factors
contributingtopsychosocial competence. Thus, it was found that (l) asexpected, cognitive
complexity was significantly, albeit it weakly, related to psychosocial competence ( r= .26),
as assesmd by RANK; (2) accuracy in perceiving others, as compared to how those others
perceive themselves, proved difficult to assess and the resulting index was unrelated to how
competently the present participants interacted with each other. An additional finding with
regard to perceptual accuracy was that cognitively simple participants were rated
significantly more accurately by both cognitively simple and cognitively complex individuals,
while, in contrast, neither the cognitively simple nor the cognitively complex were accurate
in rating the cognitively complex; (3) in this sample, asexpected, the moderately low field
indepentent individuals (Level II) were the most psychosocially competent, while the
extremely low field independent individuals (Level I) were the least competent psychosocially.
Level III and IV field incbpencbnt participants scored between these two extremes on
psychosocial competence. Unexpectedly, the highly field independent (Level IV) scored as
more psychosocially competent than the moderately field-indepencbnt (Level III). An
adiitional finding was that there was no relationship between level of field indepean and

cbgree of acceptance of either self or others. Highly field independent participants (Level IV)

74

in this investigation scored slightly, but not significantly, higher on both self and other

acceptance compared to their low field independent counterparts (Level I).

APPENDICES

APPENDIX A

INTERPERSONAL SCALES

APPENDIX A
Name:
Facilitators:
Date:

 

 

 

W

Below are four scales, each assessing a different personality trait salient to group
interaction. dust above each of these scales is a list of each individual in your group, with a
letter beside each name. Rate each group member, including yourself, on each of the four
scales by writing the appropriate letter on each of the scales to cbsignate how you see each
individual in relation to these traits. Make your decision based on each member's
verbalizations and behaviors to date in your group. All answers will be kept confidential.

 

 

 

 

 

 

 

 

 

Names of Group Members: Example: a b b

e I 6| I I

h b d

c C 8

L c d b

e d c c

f d a

9

h

i
9 -- -- -- -- 9
8 -- -- -- -- 8
7 -- -- -- -- 7
6 -- -- -- -- 6
5 -- -- -- -- 5
4 -- -- -- -- 4
3 -- -- —- -- 3
2 -- -- -- -- 2
l -- -- -- -- l

Degree of Degree of Degree of Degree of
Dominance Submissiveness Warmth and Hmtility
Caring

75

APPENDIX B

RANK ORDERING EXERCISE I

APPENDIXB
B ID I . E . I

Pleeee rank-order all the members of your group (including yourself) from “most
like" to “least like“ the definition stated below. In other words, if you thought “Mary“
(hypothetical name) in your group was the person who evaluated herself most favorably, got
along best with others, and was the most uncbrstanding, caring, and aware of her impact on
others, as compared to all other members in your group (including yourself), then write
“Mary" in the blank by number I. It is pmible that no one in your group has all of these
characteristics, or else has some more than others. It is up to you to rank-orchr group
members as to who best fits the thscription overall. If you think two group members are
. about equal in their likeness to the definition, you must still rank them (i.e., putoneabove
the other on the sheet), but you may indicate theyou see them as tied by placing a "T“ after
each of their names and drawing a line to connect the ”T's." For example: 3. Mary T_

4. Debbie T-
No more than two members may be portrayed as tied with each other and there can be no more
than two groups of ties among all the group members. This task should take approximately
l0- IS minutes. All answers will be kept condidential.

it"l' E E Hill":

The individual in your group who evaluates her/himself most favorably, ms along
best with others, and is the most understanding, caring, and aware of her/his impact on
others.

Most Like Definition
Your Name:

 

l.
Facilitator(s):

 

2.
Date:

 

9.
I 0.
Least Like Definition
76

APPENDIX C

RANK ORDERING EXERCISE II

APPENDIXC
E I'D I . E . II

Please rank—order all the members of your group (including yourself) from most
interpersonally competent to least interpersonal ly competent. If you think two group
members are about equal in thier level of interpersonal competence, you must still rank them
(i.e., put one above the other on the sheet), butyou may indicate thatyou see them as tied by
placing a ”T“ after each of their names and drawing a line to connect the “T's.“ For example:
3. Mary T-

4. Debbie T_

No more than two members may be portrayed as tied with each other, and there can be no
more than two groups of ties among all the group members. This task should take
approximately lO- I5 minutes. All answers will be kept confitbntial.

Most Interpersonally Competent
Your Name:

 

l.
Facilitator(s):

 

2.
Date:

 

8.
9.

Least Interpersonal ly Competent

What does interpersonal competence mean to you? (i.e., what is your definition of
interpersonal competence?)

 

 

 

 

77

APPENDIX D

RAW DATA TABLE

APPENDIX D

We“

Client Code '

Group "

Sex (O=man, l=womenl
GEFT Score

RCO Score

lCL-DISCR Score
DOM-WC Score
CWCEP-AC Score
ARS (given by others)
ARO (given by others)
ARS x ARO

RANK

Client Code '

Group '

Sex (O=man, lswomen)
GEFT Score

RCO Score

ICL-DISCR Score
DOM-WC Score
CONCEP-AC Score
ARS (given by others)
ARO (given by others)
ARS x ARO

RANK

Client Coda '

Group '

Sex (O=man, l=women)
GEFT Score

RCQ Score

ICL-DISCR Score
DOM-WC Score
CCNCEP-AC Score
ARS (given by others)
ARC (given by others)
ARS x ARO

RANK

01020305050602060911JJJZ

01

0
16
16 .0
07.2
i .7

1 1.2
30.2
32.4
976
2

03

14
15.5
11.3
1.0
9.23
25.3
30.5
771

25

05

1
07
26.0
1 0.4
2.9
1 0.3
27 .6
31 .3
670
6

01

0

16
16.0
13.3
1.7
9.34
25.4
25.2
640
3

03

16
30.5
06.4

1 0.9

23.6
29.0
690

26

05

0

16
11.5
13.1
2.0
12.3
22.5
27.0
607
2

0 1

1

O9

1 6 .5
07.6
2.2

1 0.0
24.6
29.4
674
3

03

15
21 .5
04.5
2.2
14.6
26.0
26.5
741

22
05

16.0
3.0

32.5
27.6
903
6

01

1

12
24.5
06.5
3.3
9.65
29.6
31 .2
923
4

03

11
19.5
10.4
4.2
13.6
27.3
29.3
799

05

09
27 .0
1 7 .4
4.5

1 0.6
31 .5
33.0
1039
6

01

1

07
26.5
07.3
4.0

1 1.3
25.6
29.4
752
6

30.3
31 .5
954

06

17
13.0
12.3
2.7
12.0
29.2
31.5
919
3

76

01 02 02
0 0 l
10 - 09
19.5 21.5 21.5
06.3 10.0 06.3
2.1 3.1 2.6
12.5 10.2 11.7
26.4 30.2 29.4
26.6 31.6 30.6
755 960 905
6 2 3

04 04
07 17
14.0 24.5
12.9 -
3.7 -
10.7 -
20.4 26.6
24.2 30.0
493 656

16

17.5
09.0
1.2

10.0
33.6
34.6
1 169

3.0 31 32
06 06 06
0 1 1

15 11 15
20.5 30.0 26.5
12.2 15.6 12.6
3.9 3.6 3.3
11.3 11.7 11.1
29.0 26.2 30.5
27.2 26.6 31.5
766 754 960
3 3 4

02 02 02 02
1 1 0 1

06 13 17 16
40.5 15.0 17.0 45.0
14.7 -- 10.6 17.1
2.6 - 2.9 3.6
11.6 - 12.6 13.4
26.4 24.6 29.0 23.6
26.4 26.6 27.6 25.6
677 659 600 609

4 7 6 7

2.1 22 .23 2A
04 04 04 05
1 1 0 1
09 13 04 06
26.0 26.0 19.0 14.5
- 19.5 12.9 19.7
- 2.2 2.5 2.3
-- 12.0 12.2 12.7
30.0 23.0 26.6 23.0
30.0 30.2 27.4 26.0
900 694 769 644

6 6 7 2

33 3A .35 36

O6 06 06 07
1 1 0 1

11 17 12 05
27.5 59.0 23.0 21.5
- 09.4 06.5 14.0
3.3 3.3 3.3 4.1

11.6 11.6 12.6 06.6
31.6 24.2 26.5 23.2
32.7 27.6 24.5 26.3

1039 672 649 656

6 6 3 6

Client Code ‘

Group '

Sex (O=man, i=women)
GEFT Score

RCO Score

lCL-DISCR Score
DOM-WC Score
CONCEP-AC Score
ARS (given by others)
ARO (given by others)
ARS x ARO

RANK

Client Code '

Group '

Sex (O=man, l=women)
GEFT Score

RCO Score

ICL-DISCR Score
DOM-WC Score
CONCEP-AC Score
ARS (given by others)
ARO (given by others)
ARS x ARO

RANK

313632593152535555963156

07

0
16
34.5
10.1
3.9

1 1.4
27 .6
24.7
666
6

$9

09

l
l 3
1 9 .4
4.3
1 2 .2
29 .6
29 .4
670
6

07

1

14
17 .5
13.7
3.6
9.6
26 .2
29.0
760
6

50

10

l

06
26.0
14.1
4.4
10.3
25.3
27.5

07

O

16
21.0
13.2
4.4
12.3
29.7
31.5
935
6

5.1

10

0

14
15.0
13.4
1.6
5.01
22.5
26.0

07

0

1 5
26.0
09.5
2.7

1 1.2
25.6
26 .5
663
6

52

1 0

0

06
26.5
2.4
6.40
27 .6
27 .5

695 565 764

2

7

3

06

1

16
29.5
19 .0
3.9

1 2 .4
25.0
26 .3
707
6

53

10

0

06
13.5
10.3
1.5
10.4
27.3
26.3
716
6

06

0
16
16.5
24.5
4.3
15.6
26 .5
31 .0
663
2

06

1

17
23.0
16.6
2.9
14.7
26.6
32 .5
671
4

06

0
14
17.5
1 1.7
2.6
17.4
20 .5
24.3
498
4

09

0

02
19.0
15.4
3.0
10.7
29.2
25.4
741
3

09 09 09
0 0 1
12 ll 03
06.5 16.5 27.0
14.3 09.7 11.3
2.1 2.7 2.9
11.6 12.2 10.1
30.6 25.6 21.6
25.4 30.2 31.0
777 773 669
7 2 4

 

' Gala-PA is not included as it consisted of a summation of the 2- scores for ICL-DISCR, DOM-WC.

end CONCEP-AC.

LIST OF REFERENCES

LIST OF REFERENCES

Adams, H8. (1964). Mental illness or interpersonal behavior. American
Psycnolog/Lst , l9, l9] - I97.

Adams-Webber, 0. (I979). Construing persons in social contexts. In P. Stringer at D.
Bannister (Eds), Constructs of wolality ano' Individuality (pp. 195- 219).
London: Academic Press.

Argyris, C. (1962). Interpersonal competence and organizational effectiveness.
Chicago: The Dorsey Press, Inc.

Benjamin, LS. (I98l). Apsychosocial competence claesification system. In 8.0. Wine &
M.D. Smye (Eds), Social Competence (DD. l89-23l). NewYork: The
Guilford Press.

Bierl, 8. (I955). Cognitive compiexity-simplicityand predictive behavior. Journal of
Abnormal and Social Psycnology, 51, 263-268.

Bieri, 0., 8K Blacker, E. (1956). The generalityof cognitive complexity in the perception
of people and inkblots. Journal of Aonormal and Social Psychology, 5.3,
lI2-ll‘l. '

Block, J. (1971). lives tnrougn time. Berkeley: Bancroft Books.

Bronfenbrenner,U. , Harding, J., & Gellwey, M. (1958). The measurement of skill in
social perception. In McClelland, D. C. ( Eds), Talent and axlaty (pp.
29-11 I). NewYork: D. Van Nostrand Company, Inc.

Bruch, M.A., Heisler, 8.0., & Conroy, C.G. (I981). Effects of conceptual complexity on
assertive behavior. Journal of Counseling Psycno/ogy, 2:9, 377-388.

Carr, 8.1-2. (I980). Personal construct theory and psychotherapy research. InA.W.
Landfield & L.M. Leitner (Eds), Personal construct psychology,
psychotnarapy and personality (pp. 233-270). New York: John Wiley and
Sons.

Carter, H., at Loo, R. (I980). Group embedded figures test: Psychometric data.
Perceptual and Motor Skills, 50, 32—34.

Cartwright, R.D., at Lerner, B. (I963). Empathy, needto change, andimprovement with
psychotherapy. Journal of consulting Psycnology, 27, I38- I44.

Crane, W. D., at Brewer, M. B. (1973). Princrplas' ofrasaarcn in social
psychology NewYork: McGraw-Hill Book Company.

80

81

Crockett, W.H. (1965). Cognitive oomplexityandimpreesion formation. In BA Maher
(Ed), Progress in experimental personality research (Vol. 2) (Pp.
47-90). NewYork: AcademicPress.

Crockett, W.H., Press, A.N., Delia, JG, 8‘ Kenny, CT. (1974). Structural Analysis
of the organization of written impressions. Unpublished manuscript,
University of Kansas.

Delia, J.G., Kline, S.L., & Burleson, BR. (1979). The developmentof persuasive
communication strategies in kincbrwrteners through twelfth grecbrs.
Communication Monographs, 46, 241 - 256.

Doll, EA. (1953). Measurement of social competence. U.S.A.: Educational Test
Bureau.

Eisler, RM, & Frederiksen, UN. (1980). Perfecting social skills. NewYork:
Plenum Press.

Ellis, A 8K Grieger, R. (1977). fiano'hoot of Rational [motive fherapy. New
York: Springer Publishing Co.

Fancher, RE. (1967). Accuracy versus validity in personal perception. Journal of
Consulting Psychology, 3l, 264-269.

Feffer, M., at Suchotliff, L. (1966). Decenteringimplicationsofsocial interactions.
Journal of Personality and Social Psycho/own 4, 415-422.

Filak, J. (1982). congruence of perception on client symptoms and therapy
outcome Unpublished master's thesis, Michigan State University, East Lansing,
Mi.

Filek, J., at Abelee, NA. (1984). Posttherapy congruenceonclient symptoms and therapy
outcome. Professional Psychology: Research and Practice , l 5 ,
846-855.

Foe, U.G. (1961). Convergenoes in the analysis of the structure of interpersonal whavior.
Psychological Review, 68, 341 -353.

Foote, N., at Cottrell, LS. (1955). Identity and interpersonal competence.
Chicago: University of Chicago Press.

Gamble, T.J., & Ginsberg, PE. (1981). Differentiation, cognition, andsocial evolution.
Journal of Cross cultural Psychology, l2, 445-459.

Garfield, S. (1971). Research on client variables in psychotherapy. In A. Bergin 8K 5.
Garfield (Eds. ), fiandhoot of psychotherapy and hehavior change. New
York: John Wiley and Sons.

82

Gibb, J. R. (1964). Climate for trust formation. In L. P. Bradford, J. R. Gibb, andK. D.
Benne (Eds), f-Croup lheory anulaooratory Nathan, (pp. 279-309).
NewYork: Wiley.

Giadt, EH. (1958). Cuesassociated with accurate and inaccurate interview impressions.
Psychiatry. Zl, 405-409.

Gilligan, C. (1982). In ao'ifferent voice. Cambridge: Harvard University Press.
Gledwin, T. (1967). Social competence and clinical practice. Psychiatry, 30, 30-43.

Goffman, E. (1956). Presentation of self in everyday life. Edinburgh: University
of Edinburgh Social Science Reeeerch Centre.

Goldfriend, MR, 8: D'Zurille, T.J. (1969). Behavioral-analytic model for aseeesing
competence. In 0D. Spielberger (Ed), Current topics in clinical and
community psychology (Vol. I, pp. 151-196). NewYork: Academic Press.

Goldstein, K.M., & Bleckmen, S. (1978). Cognitive Style. NewYork: John Wiley and
Sons.

Gordan, LV. (1976). Revised manual for survey of interpersonal values.
Chicago: Science Reeeerch Association.

Gottman, E., & Markman, H.J. (1978). Experimental chsigns in psychotherapy rmrch.
In S.L. Garfield 8: AF. Bergin (Eds), fiandooot of psychotherapy and
behavior change: An empirical analysis (PP. 23-62). New York: John
Wiley and Sons.

Hale, C.L., & Dalia, JG. (1976). Cognitive complexityandsocial perspective-taking.
Communication l‘tonographs, 43, 195-203.

Hale, CL. (1980). Cognitive complexity-simplicityasacbterminent of comunicetion
effectiveness. communication Nonographs, 47, 304-311.

Harvey, 0.J., Hunt, D.E., 8t Schroder, H.M. (1961). Conceptual systems and
personality organization New York: John Wiley and Sons, Inc.

Hurley, JR. (1976). Two prepotent interpersonal dimensionsendthe effects of trainers on
T-groups. Small Croup Cehavior, 7, 77-98.

Hurley, JR. (1976). Helpful behaviors in groups of mental health professionals and
undergraduates. International Journal of Croup Psychotherapy , 26,
173- 189.

Hurley, JR. (1978). Toward seeing ourselvesesothersseeus. In LR. Wolberg & M.L.
Aronson (Eds), Croup I’herapy (pp. 105-118). NewYork: Stratton Books.

83

Katz, R.L. (1963). impethy, its nature and uses. London: Collier-MacMillen
Limited.

Kelly, GA. (1955). fhe psychology of personal constructs New York: Norton.

Kelly, GA. (1970). Abrief introduction to personal construct theory. In D. Bannister
(Ed. ), Perspectives in personal construct theory. New York: Academic
Press.

Kerlinger, F.N. (1973). Founoations of behavioral research. New York: Holt,
Rinehart 8t Winston, Inc.

Kurtz, R.R., & Grumman, CL. (1972). Different appromhestothe measurement of
therapist empathy and their relationship to therapy outcome. Journal of
Consulting and Clinical Psychology, 3.9, 106-115.

LaForge, R., 8: Suczek, (1955). The interpersonal dimensionofpersonality: III. An
interpersonal check list. Journal of Personality , 24, 94-112.

LaForge, R. (1973). Using the ICL An interim revision of Research Use of the lCt,
83 Homestead Blvd, Mill Valley, CA.

Lm, DE. (1970). Validation of the orthogonal dimensions underlying the lCL and the
octant constellations assumed to be their measure. Journal of Projective
rechnioues and Personality Assessment , 54, 519- 527.

Lanyon, RI. (1967). Meesurementofsociel competence in college males. Journal of
consulting Psychology, Jl, 495-498.

Lucky, EB. (1960). Marital satisfaction and its association with congruence of perception.
Marriage and fam ily t iving, 22, 49-55.

Lusk, E., 8‘ Wright, H. (1981). Differences lnsexend curricula on learning the group
embedded figures test. Perceptual and Motor Skills, 53, 8-10.

Markowsky, A. (1979). Direct assessment of psychopathology Unpublished
mctoral dissertation, University of Pennsylvania.

McKenna, F.P. (1984). Measuresof field dependence: Cognitive styleoroognitive ability?
Journal of Personality and Social Psychology, 47, 593-603.

Meichenbaum, 0., Butler, L., & Gruson, L. (1981). Towardamnceptual model of social
competence. In J.D. Wine & M.D. Smye (Eds), Social competence (pp.
36-60). New York: The Guilford Press.

84

Mischel, W. (1972). Direct versus indirect personality assessment: Evidence and
implications. Journal of Consulting and Clinical Psychology, 43,
319-324.

O'Keefe, DJ, 8! Sypher, HE. (1981). Cognitive wmplexity measures and the
relationship of cognitive complexity to communication. Human Communication
Research, 8, 72-92.

Oltman, P.K., Goodanough, D.R., Witkin, H.A., Freedman, N., & Friedman, F. (1975).
Psychological differentiation as a factor in conflict resolution. Journal of
Personality and Social Psychology, 32, 730- 736.

Quinlan, D.M., 6t Blatt, SJ. (1972). Field articulation and performance under stress:
Differential predictions in surgical and psychiatric nursing training. Journal of
Consulting and Clinical Psychology, 39, 517.

Reckman, RF. (1977). Social competence and construing peers. Unpublished
doctoral dissertation, University of Cincinnati.

Rothenburg, BB. (1970). Children's social sensitivity and the reletionshipto
interpersonal competence, intrapersonal comfort, and intellectual level.
Developmental Psychology, 2, 335- 350.

Schrier, H. (1953). The significance of identification in therapy. American Journal of
Crthopsych/atry , 2.3, 585-604.

Sachrest, L., & Jackson, ON. (1961). Social intellignceandaccuracyof interpersonal
predictions. Journal of Personality, 29, 167-182.

Seigelman, E., Block, J., Block, J., at Lippe, A. (1970). Antecedentsof optimal
psychological adj ustment. Journal of Consulting and Clinical Psychology,
35, 283-289.

Shulman, E. (1975). Conformity in a modified Asch-type situation. Unpublished
doctoral dissertation, City University of New York.

Shure, M. (1981). Social competenceasaproblem solving skill. In J.D. Wine & MD.
Smye (Eds), Social competence (pp. 158-185). NewYork: TheGuilford
Press.

Singer, J.E. (1964). Theuseofmanipulative strategies: Machiavellianismand
attractiveness. Sociometry. 27, 120-150.

Smith, HC (1973). Sensitivity fraining. NewYork: McGrew-Hill.

85

Smith, 8., 8 Leach, C. (1972). Ahierarchical measureof cognitive complexity.
British Journal of Psycho/ow, 63, 561-568.

Taft, R. (1955). The abilitytojudge people. Psychological Bulletin, 52, 1-23.

Tyler, F.B., 8t Gatz, M. (1977). Development of individual psychosocial competence ina
high school setting. Journal of Consulting and Clinical Psychology, 45,
441-449.

Yen Maenen, J. (1979). On the understanding of interpersonal relations. in W. Bannis,
JV. Maenen, E.H. Schein, & F.l. Steele (Eds), [ssays in interpersonal
dynamics (pp. 13-42). Chicam: Dorsey Press.

Yannoy, J.S. (1965). Generality of cognitive complexity-simplicityasapersonality
construct. Journal of Personality and Social Psychology, 3, 385-396.

Weinstein, EA. (1969). The development of interpersonal competence. In D.A. Goslin
(Ed), Handbook of socialization theory and research (pp. 753-775).
Chicago: Rand McNaI ly and Company.

Warner, H. (1957). The concept of development from a comparative and organismic view.
In 08. Harris (Ed), The concept of development (pp. 125- 148).
Minneapolis: University of Minnesota Press.

White, R.W. (I973). Senseof interpersonal competence. In W.G. Bennis, D.T. Berlew,
EH. Schein, & F.l. Steele (Eds), Interpersonal oynamics (pp. 518-523).
Chicago: Dorsey Press.

Widiger, T.A., Knudsen, RM, 8( Rorer, LG. (1980). Convergent and discriminant
validity of mamuresof cognitive styles and abilities. Journal of Personality and
Social Psychology, 39, 116-129.

Wiggins, JS. (1982). Circumplex models of interpersonal behavior in clinical psychology.
In P. C. Kendall &J. N. Butcher (Eds), Handbook of research methods in
clinical psychology (pp. 183-221), NewYork: Wiley.

Wine, JD. (1981). From infecttocompetence models. In J.D. Wine & M.D. Smye
(Eds), Social competence (pp. 3-35). NewYork: Academic Press.

Witkin, HA (1978). Cognitive styles in personal and cultural adaptation.
Worcester: Clark University.

Witkin, H.A., Dyk, R.B., Faterson, H.F., Goolbnough, D.R., & Karp, SA. (1962).
Psychological differentiation . New York: Wiley.

86

Witkin, H.A., & Goodanough, DR. (1977). Field dependence andinterpersonal behavior.
Psychological Bulletin, 64, 661-689.

Witkin, H.A., & Goodanough, DR. (1981). Cognitive styles.- [ssence and origins.
NewYork: International Universities Press, Inc.

Witkin, H.A., Goodanough, D.R., & Oltman, PK. (1979). Psychological differentiation:
Current status. Journal of Personality and Social Psychology, 3 7,
1127-1145.

Witkin, H.A., Oltman, P.K., Raskin, E., & Karp, SA. (1971). A manual for the
embeooed figures test. Palo Alto, CA: Consulting Psychologists Press.

Zajonc, RB. (1968). Cognitivetheoriesin social psychology. In G. Lindzey & E. Aronson
(Eds). Handbook arsaciai psycho/00y (rev. ed.) (pp. 320-411). Reading,
MA: Addison-Wesley.

Zigler, E., & Phillips, L. (1960). Social effectiveness and symptomatic behaviors.
Journal of Abnormal and Social Psychology, 6 l, 231 -328.

Zigler, E., at Phillips, L. (1962). Social competence and the process reactive distinction
in psychotherapy. Journal of Abnormal and Social Psychology, 65,
215-222.

Zigler, E., & Trickett, PK. (I978). 10, social competence, andevaluation of early
childhood intervention programs. American Psychologist, 53, 789-798.

"‘liliiiililiiliiill“