“is.

This is to certify that the
thesis entitled
A Cognitive Developmental Analysis of Children's
Understanding of Social Relationships:
, Friendship, Marriage, and Divorce
presented by

C laire Bridget Lowry

has been accepted towards fulfillment
of the requirements for

Doctoral degreein Psychology

Major professor

 

Datejuu.‘ 3 l? 80

0-7639

a“. 1 M m l @117 I!!! L

: yaw»? 21,, .1
ifl:;:n(:{h \ a

  

fink; 31241 $823523

. 4'" Uflt’m é"

 

 
 
 
 
  
 

OVERDUE FINES
\:
25¢ P" ‘0 Per its.
RETuMlNG LIBRARY “MTERIALS.
Place in bookre
die

to mo!
'99 from circulationrc

 

A COGNITIVE DEVELOPMENTAL ANALYSIS OF CHILDREN'S
UNDERSTANDING OF SOCIAL RELATIONSHIPS:
FRIENDSHIP, MARRIAGE, AND DIVORCE

By

Claire Bridget Lowry

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1980

ABSTRACT

A COGNITIVE DEVELOPMENTAL ANALYSIS OF CHILDREN'S
UNDERSTANDING OF SOCIAL RELATIONSHIPS:
FRIENDSHIP, MARRIAGE, AND DIVORCE

By
Claire Bridget Lowry

This study was an attempt to investigate the development of
children's understanding of social relationships. Two specific
issues were addressed: (l) children's perceptions of whom they
might choose to befriend, marry, or divorce; and (2) children's
perceptions of issues which might help to consolidate or threaten a
marital relationship. Impetus for the focus and design of this
research stemmed from the rising interest in social cognitive
developmental theory, a growing body of research on person perception,
and reports on children's perceptions of familial roles.

The sample consisted of 40 subjects, divided across four cells
according to age and sex. The mean age was 8 years 2 months for the
younger group and 13 years 10 months for the older group. All were
from intact middle class homes. Data were obtained by using an
individualized interview format. The questions examined children's
perceptions of a friend, possible spouse, and possible ex-spouse,
motivation for marriage and divorce, characterization of on-going

marital and post-marital process, personal attitudes toward marriage,

Claire Bridget Lowry

and the process of marital problem solving. The analysis was in

part structural, attempting to delineate the cognitive framework
underlying children's reasoning about relationships. Major variables
were Social Penetration, Personal Involvement, and Evaluative Direc-
tion. The analysis was also descriptive, focusing on the specific
material that children raised. A pool of seventeen content cate-
gories were used.

Children's inferential ability showed improvement with age for
two of the three social relationships examined. Psychological inti-
macy was cited by older children as being particularly important in
friendship, while acceptance, equality, and role enactment were
noted as major factors in divorce. In contrast, both age groups
tended to highlight superficial issues such as appearance when dis-
cussing marriage. Children's ability to construe others in non-
egocentric terms did not show an age effect. All subjects tended to
respond in a non-egocentric manner. An age effect was not found for
the Evaluative Direction scale, with all subjects tending to focus
on the presence of qualities rather than on their absence.

Boys and girls showed equal facility in egocentric/decentered
thinking. Similarly, no sex effect was found for the Evaluative
Direction scale. However, for the dimension of Social Penetration,
an interaction was found between sex and social relationship. Girls
were more likely to make inferential statements in regard to friend-
ship, but this did not carry over to their perception of a spouse

or ex-spouse.

Claire Bridget Lowry

Comparison was also made across social relationships. Children
gave the most insightful descriptions for friends, but described
spouses in the most decentered terms. An ex-spouse tended to be
described in terms of qualities that were lacking.

In discussing issues contributing to marital solidarity and
marital tension, children saw people marrying on the basis of affec-
tive involvement, but divorcing on the basis of behavioral incompati-
bility. Their expectations for marriage included the sharing of time,
parenthood, and financial responsibility. Their expectations for
divorce included physical separation of the partners, a change in
lifestyle and parental relationship for the children, and resumption
of responsibility for oneself. Children tended to perceive couples
as generally satisfied and expected to get married themselves. How-
ever, they were very sensitive to marital tension, especially that
precipitated by issues about affective expression, childrearing,
and finances.

These results were discussed in terms of cognitive-developmental
theory, with particular reference to the distinction between cogni-
tive skill and cognitive performance. Implications were also pre-
sented for discussion about the intercorrelation of cognitive skills
and generality of inference across different aspects of the social
domain. Finally, the significance for future research and therapeutic

intervention were noted.

ACKNOWLEDGMENTS

I would like to express my appreciation to the members of my
committee: Dr. Ellen Strommen, chairperson, Dr. Lucy Ferguson,
Dr. John Hunter, and Dr. John McKinney.

Special thanks are due to Dr. Strommen who was always enthusi—
astic about the questions raised in this study, ready with the right
suggestions, and supportive--expecially during those dark November
days when I was stuck on page five of my literature review. Her
contribution of insight, time, and help were essential in my finally
seeing this project through to completion. I am also indebted to
Ellen for looking beyond the bounds of this research to the researcher,
no matter how warily I sometimes approached that role.

I am grateful to Dr. Ferguson, not only for her ideas and help
in this project, but also for her model of developmental and clinical
interests. My only regret is that I did not manage to work with
her more. My thanks go also to Dr. Hunter who supported the design
of this study, contributed much to its conceptualization, and who
spent many sunny days indoors helping me plan the statistical analy-
ses. My debt to Dr. McKinney extends well beyond this project. John
has contributed much to my development both as a professional and as
a person. I am especially appreciative of the confidence and friend-

ship he has given me for so many years.

ii

I would like to express my thanks to the children and families
who so willingly participated in this study. I would also like to
thank my colleagues at the University of North Carolina for their
support as I finished up this dissertation while doing a Fellowship
there. Particular acknowledgment for statistical consultation is
due to Dr. Helen Brantley, Department of Psychiatry and the staff of
the Statistical Laboratory in the Department of Psychology.

1 would like to thank several friends, Linda Giacomo, Diane
Klein, David Rubin, and Nancy Nagner, with whom I shared all the
trials and frustrations, depressed and manic days through the evolu-
tion of this project. Particular thanks are due to Linda who coded
data while I unpacked boxes in North Carolina and who helped me edit
my discussion as I coped with a deadline for completion of this
dissertation and an impending colloquium.

I am grateful to my parents and family for their unquestioning
support, encouragement, and modeling of competence.

Finally, I am grateful to the Government of Quebec and the
Canadian Medical Research Council for their financial support of my

graduate education.

iii

TABLE OF CONTENTS

LIST OF TABLES

INTRODUCTION
REVIEW OF LITERATURE .

Development of Social Cognitive Understanding
Development of Person Perception Skills
Structural Analysis . . .
Content Analysis .
Development of Familial Role Perceptions

STATEMENT OF PURPOSE AND HYPOTHESES .
METHODOLOGY .

Subjects .
Interview Schedule
Procedure .

Data Coding .

RESULTS

Person Perception in Friendship, Marriage, and
Divorce . . . . . . . .
Structural Dimensions .

Content Dimensions . . .

Motivation for Marriage and Divorce . .

Characterization of Marital and Post- Marital Process

Personal Attitude Toward Marriage . . .

Marital Tension and Problem Solving .

Importance of Marital Agreement
Likelihood of Marital Tension
Likelihood of Tension Resolution
Means of Problem Resolution

iv

Page

vi

Page

DISCUSSION . . . . . . . . . . . . . . . . . 54

Cognitive Structure and Content: Marriage, Friendship,

and Divorce . . . . . . . . . . . . . . 54
Issues Contributing to Marital Solidarity-Marital

Tension . . . . . . . . . . . . . . . 6O
Implications for Future Research and Therapeutic

Intervention . . . . . . . . . . . . . . 62

APPENDICES . . . . . . . . . . . . . . . . . 65
A. Letter to Parents . . . . . . . . . . . . . 66
B. Interview Protocol . . . . . . . . . . . . 68
C. Data Coding Manual . . . . . . . . . . . . 73

BIBLIOGRAPHY . . . . . . . . . . . . . . . . 79

 

Table

l0.

ll.

12.

l3.

14.

LIST OF TABLES

Mean and Standard Deviation Scores for Personal
Involvement Scale . . . . . . . .

Analysis of Variance for Personal Involvement Scale

Mean and Standard Deviation Scores for Social Penetra-
tion Scale

Analysis of Variance for Social Penetration Scale

Evaluative Description Scale-~Percentage of Subjects
Using Affirmative Descriptions . . . . .

Content Analysis for Person Perception in Friendship,
Marriage and Divorce--Percentage of Subjects Using
Each Category . . . . . . . . .

Content Analysis for Motivation for Marriage and
Divorce--Percentage of Subjects Using Each Category

Content Analysis for Characterization of Marriage and
Divorce--Percentage of Subjects Using Each Category

Mean and Standard Deviation Scores for Rating of
Affective State in Marriage and Divorce

Analysis of Variance for Affective State in Marriage
and Divorce . . . . . .

Personal Attitude Toward Marriage--Percentage of
Subjects with Positive Expectations .

Mean and Standard Deviation Scores for Ratings of
Importance of Marital Agreement .

Analysis of Variance for Ratings of Importance of
Marital Agreement . . . . . . . .

Mean and Standard Deviation Scores for Ratings of
Likelihood of Marital Tension . . . . .

vi

Page

32
33

35
35

36

39

41

43

45

45

46

47

48

49

Table

15.

16.

17.

18.

Analysis of Variance for Ratings of Likelihood of

Marital Tension

Mean and Standard Deviation Scores for Ratings of
Likelihood of Tension Resolution . .

Analysis of Variance for Ratings of Likelihood of

Tension Resolution

Types of Problem Resolution--Percentage of Subjects

Using Each Category

vii

Page

50

51

52

53

INTRODUCTION

In modern society, children are continually exposed to informa-
tion, debate and controversy about marital relationships. Parents
present a primary model of marital behaviors and attitudes. The popu-
lar media, music, television and the press, are flooded with reference
to relationships and lifestyles. Educational curricula now frequently
encompass classes on marriage and family life. Dating has become
increasingly prevalent among young adolescents and it seems as if the
age at which children begin to show heterosexual interests is being
pushed steadily lower.

This study was an attempt to investigate the development of
children's understanding of social relationships, especially their
concepts of marriage and divorce. Two specific issues were addressed:
(1) children's perceptions of whom they might befriend, marry, or
divorce; and (2) children's perceptions of issues which might help
to consolidate or threaten a marital relationship. The anaysis was
in part structural. It attempted to delineate the cognitive frame-
work underlying children's reasoning and perceptions about relation-
ships. The analysis was also descriptive. It attempted to appreciate
the richness and variety of ideas and attitudes expressed by children.

The significance of this study lies in its potential contribu-

tion to primary and secondary intervention. Cognition plays an

important role in shaping an individual's behavior. Analysis of
children's cognition about marital relationships may facilitate
explanation and prediction of patterns in family interaction. Simi-
larly, the attitudes of adults have their roots in childhood beliefs.
Analysis of the ideas and attitudes exhibited by children may help
to reveal the developmental origins of adult notions about marital
relationships. The significance of this study lies also in its
potential contribution of cognitive developmental theory. Particu-
larly relevent are questions about the interrelationship of social
cognitive abilities and the generality of various types of social

inference.

REVIEW OF LITERATURE

Unfortunately, there is a limited context of literature specifi-
cally relating to children's understanding of marital dynamics in
which to embed the present study. However, impetus for the focus and
design of this research stemmed from three main sources: the rising
interest in social cognitive-developmental theory, a growing body of
research on person perception, and the ongoing investigations of

children's perceptions of familial roles.

Development of Social Cognitive Understanding

The theoretical and empirical literature exploring how children
learn about the physical world around them has become voluminous over
the past decade. For the most part, such work has been carried out
within an organismic paradigm. Stimulated by work of Piaget, the
focus of analysis has been on the internal cognitive processes used
in constructing knowledge about the world (Baldwin, 1969a; Flavell,
1963, 1970; Kuhn, 1978; Piaget, 1970; Wallach, 1963; Sigel, 1964).
Theoretical and empirical literature on social development is per-
haps equally voluminous. However, such work has been carried out
largely within a mechanistic paradigm (Kahn, 1978). The focus of
analysis has been on the child's overt social behavior and on the
external stimuli which influence its acquisition and expression. In

contrast, it is only in very recent years that interest in social

3

cognition has come to the fore. Although interpersonal behavior and
dynamics received some attention within Piaget's theoretical frame-
work, there are now repeated calls for further investigation of the
application and adaptation of his concepts to analysis of children's
social understanding and reasoning (Damon, 1977; Kohlberg, 1969;
Kuhn, 1978; Lee, 1975; Selman, 1976a; Shantz, 1975; Youniss, 1975).

A basic premise of cognitive developmental theory is that the child
interacts with the environment by extracting information and organizing
it into meaningful patterns and schemas. These concepts serve as fil-
ters through which subsequent experiences are evaluated and interpreted.
Development involves the gradual elaboration and modification of these
cognitive representations into more comprehensive and integrated systems
of understanding (Flavell, 1970). Hence, the goal has been to describe
a sequence of cognitively based stages or qualitatively distinct ways
of organizing and understanding the social world. Particular attention
has been addressed to identifying cognitive stages which may show an
invariant sequence of development, cohesion across different social
issues, and universality across subject groups.

Although the relationship between physical and social cognition
has been subject to intense theoretical debate (Damon, 1977; Kohlberg,
1969; Lee, 1975; Piaget, 1970; Selman, 1976a; Shantz, 1975; Youniss,
1975), much of this investigation has been based on the general
assumption that the course of development of social and physical con-
cepts is likely to be parallel or at least descriptively similar.

More specifically, children's understanding of the social world is

expected to reflect both a gradual decline in idiosyncratic, global

and egocentric thinking, and a rise in logical, causal, and abstract
reasoning. A search of the literature revealed a large and increas-
ing number of exploratory attempts to examine such developmental
patterns in children's organization of thoughts about social issues.
The preponderance of effort has been focused on moral development
(Kohlberg, 1969) and perspective taking skills (Flavell, 1968, 1974).
However, there are also data extant on a wide range of topics, includ-
ing the child's concept of death (Childers & Wimmer, 1971), defense
mechanisms (Chandler, Paget & Koch, 1978; Whiteman, 1967), emotional
disorders (Coie & Pennington, 1976; Hoffman, Mersden & Kalter, 1977;
Kalter & Marsden, 1977; Maas, Marecek & Travers, 1977; Marsden &
Kalter, 1976; Novak, 1975); gender identity, sexuality and reproduc-
tion (Bernstein & Cowan, 1975; Kahn, Nash & Brucken, 1978), maternal
attachment (Marvin & Greenberg, 1978), physical illness (Campbell,
1975; Neuhauser, Amsterdam, Hines & Steward, 1978); politics
(reviewed by Flavell, 1970); self-concept (Bannister & Agnew, 1977;
Lewis & Brooks—Gunn, 1979; Mohr, 1978), social institutions, customs,
and conventions (Furth, Baur & Smith, 1976; Turiel, 1975), and
religion (reviewed by Flavell, 1970). There have also been investi-
gations of somewhat more esoteric social issues, such as esthetics
(Spiegel, 1950), ambivalence (Harter, 1977), queues (Moessinger,
1977), excuses (Leahy, 1978), kindness (Baldwin & Baldwin, 1970)
and fantasy (Morison & Gardner, 1978; Shapiro, Sherman & Osowsky,
1980).

Despite steady expansion of the boundaries of social cognitive

investigation, there has been little attention given to the child's

understanding of marital relationships. However, theoretical specu-
lation and empirical data on related social phenomena underscore the
value of such research and provide ample support for hypothesizing
that the child's reasoning about marital dynamics should similarly

reflect a cognitive developmental progression.

Development of Person Perception Skills

 

A particularly pertinent area of social cognitive research is
that generally labeled person perception. Other rubrics include
attribution process, impression formation, and naive psychology. The
question addressed is how people are conceptualized as individuals--
that is, what properties are attributed to them and how is such
information organized.

Research on person perception has a long history. Initial
efforts emphasized the judgmental aspects of person perception and
focussed on factors correlated with an individual's empathic ability,
insight, or accuracy in appraising the responses of others (Bruner &
Tagiuri, 1954; Taft, 1955). However, this line of research declined
following major methodological criticism (e.g., Cronbach, 1955), and
was replaced by a more phenomenological approach. Beginning with
the work of Heider (1944, 1958) and attribution theory, Kelley (1955)
on personal construct theory, and the research of Asch (1946) and
Bieri (1955), increasingly refined attempts have been made to study
the cognitive processes involved in person perception. Emphasis has
been placed on examining the type of impressions an individual forms

about another and the conceptual sophistication of such evaluations.

Extensive investigation based on adult populations conclusively indi-
cate that person perception is affected by a vast array of factors
(Bieri, Atkins, Briar, Leaman, Miller & Tripodi, 1975; Crockett, 1965;
Gardner & Schoen, 1962; Livesley & Bromley, 1973). Among these
determinants are the type, valence and congruency of attributes
possessed by the other person, the relationship between the observer
and the person perceived, as well as a host of stable characteris-
tics and transitory psychological states on the part of the perceiver.

The development of person perception style and skill is still a
comparatively unworked area. There has been no systematic analysis
of children's concepts of human behavior and personality comparable
to the study done with adults. However, there is a small but growing
body of research on this topic, examining both the structure and

content of children's perceptions.

Structural Analysis

 

Structural analysis attempts to delineate stages in hgy_children
reason about social issues. Research has focused on several theor-

etical cognitive constructs.

Cognitive complexity. The cocept of cognitive complexity,

 

introduced by Bieri (1955, 1961), has received a number of different
definitions. In this context, it may be defined as the capacity to
construe social behavior in a multi-dimensional way. A cognitively
complex percept is based on a more highly differentiated system of

dimensions for describing others' behavior than a cognitively simple

percept (Bieri et a1., 1975; Crockett, 1965). This has been opera-
tionalized in terms of the number of dimensions children use in their
portrayal of another, and the degree to which they are differentially
applied to people.

Research has tended to show a correlation between age and cogni-
tive complexity. Kohn and Fiedler (1961) had high school and college
freshmen rate significant others on twenty bipolar adjective scales.
Younger subjects tended to apply similar ratings across stimulus
objects, whereas older subjects tended to develop a more unique pro-
file for each subject. Looking at children age 8 to 16, Signell
(1966) and Vacc and Greenleaf (1975) found a progressively increased
ability to give detailed description of roles (Adapted Role Repertory
Test). Similar results were also found in studies using a more
unstructured methodology. Bigner (1974) found that across preschool
to eighth grade, children's free descriptions of siblings became
longer and contained more descriptive elements. Scarlett, Press and
Crockett (1971) found this same trend in children's free descriptions
of peers. In addition Bigner (1974) and Scarlett et a1. (1971)
found a significant correlation between cognitive complexity and the
valence of the person being described. Figures who were liked by
the subject tended to be described in a more complex fashion than
those with negative valence. Unfortunately, data on cognitive com-
plexity as operationally defined in this context are difficult to
interpret because of a possible confounding interaction between age

and verbal fluency.

Social penetration. The concept of social penetration has been

 

discussed at length by Altman and Taylor (1973). In brief, it
refers to the development of social exchange, gradually progressing
from superficial, nonpersonal areas to more intimate and deeper
interchange. People tend to let themselves be known gradually, at
first revealing less intimate information and only later sharing
more personal aspects of themselves. Cognitive developmental theory
predicts a parallel progression in the development of skill in per-
ceiving others. Between the preoperational stage and the stage of
formal operations, children shift from largely concrete, superficial
thinking to more abstract reasoning (Piaget, 1970; Werner, 1948).

In an early study, Gollin (1958) scored the extent to which
children age 8 to 17 made inferences which accounted for diversity
of behavior. At all ages, girls were more likely than boys to
ascribe motives or introduce some underlying condition to explain
the observed behavior. However, for both sexes, inferences were more
commonly found among older children. The method used by Bigner
(1974) and Scarlett et a1. (1971) was discussed above. Both studies
rated the children's descriptions on a four-point scale in terms of
emphasis on concrete or abstract elements. Scarlett et a1. (1971)
found that by fifth grade, most subjects were describing peers in
abstract terms. The nature of the relationship between the subject
and person described (liked/disliked) was not found to have a sig-
nificant effect. Bigner (1974) found that it was not until eighth
grade that abstract dimensions predominated in descriptions of sib-

lings. Disliked siblings tended to be described ataahigher cognitive

10

level than liked siblings. Significant sex differences were also
found, with girls using more abstract terms than boys. In a widely
cited study, Yarrow and Campbell (1963) used a seven-point scale to
rate the organization of descriptions of roommates at summer camp
given by children age 8 to 13. There were no significant sex dif-
ferences, but age trends were apparent. Older subjects tended to
give well-integrated personality portraits involving implicit and
explicit inferences about behavior, while younger children tended to
give reports consisting of vague, global generalizations. Also using
a rather detailed scoring analysis, Peevers and Secord (1973) found
a similar developmental progression in their rating of children's
descriptions of peers. However, in this study, boys were found to
give somewhat more differentiating descriptions than girls. Peevers
and Secord (1973) also reported that disliked peers were described
in more dispositional terms than liked peers.

In sum, these studies give definite support to the argument
that with age children become more penetrating in their social per-
ceptions. However, the degree of penetration seems to depend on
the type of person being described and that person's relationship to
the observer. At this point, data on sex differences are still

ambiguous.

Evaluative consistengy. This cognitive construct refers to the

 

extent to which children are able to integrate discordant and incon-
gruent attributes shown by another person. This issue has not been

extensively investigated. The study by Gollin (1958) discussed above

11

makes indirect reference to it. However, Peevers and Secord (1973)
directly rated the extent to which children recognized and articu-
lated both desirable and undesirable qualities in liked and disliked
peers. As one might expect, very young subjects (kindergarten)
tended to give rather inconsistent responses, perhaps because they
did not perceive the inconsistencies or did not see a need to be
consistent. Older subjects (grade 11) also tended to give incon-
sistent descriptions, most likely because of a greater tolerance

for and ability to integrate disparate characteristics. In contrast,
early and late latency age subjects tended to give rather one-sided
evaluations. No differences in evaluative consistency were found

between liked and disliked peers.

Personal involvement. This concept refers to the social per-

 

spective, egocentric or decentered, that children use in person per-
ception. Interest in perspective taking skill grew out of the
classical studies on conservation. Piaget (1970) noted that children
gradually become able to focus on several dimensions of a situation
and to relate these dimensions. This topic has been a perennial
favorite for theoretical speculation and research. Selman (1976a)
goes so far as to argue that social intelligence depends upon the
individual's ability to take the role of or put himself in the place
of another. Tomes of research have been published on developmental
stages in children's ability to consider another's visual, inten-
tional and affective experience. Early data were reviewed by Flavell

(1968), and more recent work by Shantz (1975). Much effort has also

12

been given to examining the relationship between perspective taking
skills and a variety of social cognitions and behaviors, such as
moral judgment (Selman, 1971b), aggression (Dodge, 1980), and altru-
ism (Rushton & Wiener, 1975).

Several studies have also examined the relationship between
perspective taking skill and person perception. Despite use of a
rather crude two-point scale, Bigner (1974) and Scarlett et a1. (1971)
found that children's descriptions became less egocentric with age,
reflecting an increasing ability to distinguish between self and
other. Peevers and Secord (1973) used a more refined scale which
looked at whether the child used an egocentric, mutual, or decentered
frame of reference. A similar developmental trend was found. How-
ever, the effects of the object's valence are less clear. Bigner
(1974) found that liked siblings were described in egocentric terms.
In contrast, Peevers and Secord (1973) reported that egocentric terms
were more likely to be applied to disliked peers.

Perhaps the most detailed examination of the contribUtion of
perspective taking skills to person perception has been made by
Selman (1977, 1976a, b, 1971a; Selman, Jaquette & Lavin, 1977;

Selman & Byrne, 1974; Cooney & Selman, 1978). While Selman and his
colleagues have begun to examine how perspective taking functions as
the structural basis for reasoning about individual (Selman et a1.,
1977), group dynamics (Selman et a1., 1977) and even parent-child
relationships (Cooney, 1978), primary emphasis has been given to

analysis of friendship. As friendship might be considered a

13

prototype of dyadic relationships, this work is especially relevant
to the present study. Hence, it is reviewed in some detail.

Data were obtained by interviewing children about their reactions
to movie characters who were confronted by an interpersonal dilemma.
Although this methodology can be criticized for being somewhat
restricted in the range of information presented to and requested
from the child, the technique is within social cognitive tradition,
adapted from the classic design of Kohlberg (1969). Results indi-
cated a close association between the child's level of perspective
taking skill and his/her reasoning about friendship.

Selman elaborated the concept of perspective taking to include
four stages of development. At level 0, the children are egocentric,
typically unable to realize that others have a different perspective
from their own. This is reflected in children's conceptions of
friendship by their failure to recognize its subjective or evaluative
nature (e.g., that friendship might be based on enjoying play with
another). At level 1, the subjective stage (about age 6 to 8), aware-
ness of differences in perspectives emerges. However, the child
tends to focus on only one perspective rather than coordinate per-
spectives. In analysis of friendship, the notion of subjectivity
and reciprocity begin to appear, albeit of a primitive, uni-
directional nature. Friendship is seen as being established and
maintained on the basis of a positive subjective reaction to the
actions of another (e.g., a friend is someone who gives to you). At

level 2, the self-reflective level (about age 8 to 10), the child

14

tends to alternate between social perspectives, but cannot process
or integrate them simultaneously. Children begin to recognize the
need for mutual cooperation in friendship, but only so far as it
serves mutual self-interests. At stage 3, the mutual perspectives
level (age 10+), perspectives are coordinated, as if viewed from a
third party position. At this level, friendship is described as
involving mutual concern about well being, rather than the more ego-
centric notion of self-interest.

In related research, Youniss (1975; Youniss & Volpe, 1978)
focused specifically on affirmation, the process by which children
communicate about the initiation and establishment of friendship.
Using a structured technique, Youniss found similar developmental
shifts in children's reasoning. His data underscore the children's
growing understanding of the mutuality and reciprocity involved in
friendship and increasing recognition of the other person as distinct
from themselves.

In sum, this literature on personal involvement suggest that
perspective taking skill may be a powerful determinant of the cogni-
tive sophistication of children's perceptions of others. However,
firm conclusions cannot be drawn about sex differences or the effects

of the stimulus object's relationship with the observer.

Content Analysis,
Content analysis complements structural analysis by providing a
descriptive characterization of the terms in which the child expe-

riences his social world. Although content analysis is of high

15

interest to the clinician, it has not been favored by the researcher.
This may reflect the latter's pejorative notion that it yields so-
called "soft" descriptive data, not easily subjected to quantifica-
tion or statistical procedures. However, there are a few relevant
studies which have focused on content.

0n the basis of nonsystematic observations, Watts (1944) con-
cluded that 6 to 7 year old children tended to be most concerned with
superficial characteristics, especially appearance. Analyses of
personality were extremely crude, with liked peers being described
as kind, good or nice, and disliked peers characterized as unkind or
bad. Yarrow and Campbell (1963) reported that children age 8 to 13
were most descriptive of peers' interpersonal behavior. The most
frequently mentioned interaction categories were those dealing with
physical and verbal sociability, conformity, play, and affiliation.
Some reference was also made to appearance, but this and other non-
interactive categories were featured in only half the descriptions
given by subjects. Age trends in the use of categories did not occur
except for the dimensions of aggression and domination, which were
more likely to be used by older subjects.

Content analysis has also been made of children's concepts of
friendship. Bigelow (1977; Bigelow & LaGaipa, 1975) asked Anglo-
Canadian and Scottish children to write essays illustrating their
expectations for best friends. Developmental changes were found on
several dimensions. During early latency, friendship appeared to be
based on common activities,propinquity and a cost-reward balance

determined by overt behaviors. Later, friendship is seen as a

16

contractual relationship, violation of which may lead to its sever-
ance until compensation of some type has been exchanged. Finally,
friendship is considered as a long-term relationship based on uncon-
ditional positive regard, empathy, understanding and trust. Sex
differences were not found except that girls were more likely to
refer to altruistic behaviors. Although widely cited, these data
cannot be readily interpreted. The method did not take into account
the tremendous variance in writing skill that might be expected
across children age 6 to 14. However, there does seem to be some
validity to the developmental trend, as similar results have been
noted by Berndt (1978). Berndt's method only required subjects
(kindergarten to grade 6) to give verbal descriptions of friendship.
Children of all ages expected friends to share, help, cooperate
rather than quarrel, and to engage in common activities. However,
issues of intimacy, loyalty, trust and personality were more likely
to be mentioned by older subjects. Sex differences were found on

two categories. Girls stressed intimacy and faithfulness more than
boys. Remarkably congruent age patterns have also been reported by
Selman et a1. (1977) and Damon (1977), in studies based on children's
responses to hypothetical dilemmas. Unfortunately, these reports
did not include analysis of sex differences. Finally, Hayes (1978)
focussed on conceptions of friendship held by preschool age children.
Content analysis indicated that propinquity, common activities, play
skill, evaluation and possessions were all used as a basis for liking
friends. Dislike of another was typically based on his/her rule

violation, aggression, and aberrant behavior.

17

To a degree, content analysis might seem merely to reflect
patterns already articulated through structural analysis, especially
developmental trends in social penetration. However, the value of
content analysis is that it is multi-dimensional. Hence, it gives
credit to the richness of children's cognition. Furthermore, it takes
into account subtle shifts and variance within a developmental level
and across sex groups, although the issue of sex differences has not

been well investigated to date.

Development of Familial Role Perceptions

Considerable attention has been given to parental role function-
ing. Not only have familial experiences been recognized as important
influences on the child's socialization and personality style, but
it has also been argued that the perception of parental roles may be
an important initial step in the child's developing cognition about
social roles (Brim, 1957).

Research in this area has evolved through several stages of
focus. Most early work was concerned with the relationship between
child development and the way parents' perceived themselves as carry-
ing out their role. Many of these studies were retrospective.
Recognizing the methodological problems inherent in such a design,
however, investigators moved to making direct observations of
parental behaviors or at least having parents give reports on their
current functioning.

However, such research has been criticized on two counts (Dubin

& Dubin, 1965). Studies often failed to examine children's actual

18

responses to parental behavior. Children cannot be assumed to
respond in a uniform manner. Furthermore, investigators failed to
examine children's interpretation of their experiences. It cannot
be assumed that the child's perception of his/her parents is neces-
sarily congruent with parents' self-perceptions. There is now a
growing body of research exploring the extent to which and manner in
which children understand the parental behavior they observe and
experience. Much of this work has been reviewed by Dubin and Dubin
(1965), Goldin (1969), and by Livesley and Bromley (1973).

A great variety of descriptive variables have been used in
analysis of parental roles. However, recent factor analytic work
suggested that three orthogonal factors appear to constitute a stable
framework within which to investigate child-rearing behaviors. These
factors have been labeled by Schaefer (1965a, b) as Acceptance versus
Rejection, Psychological Autonomy versus Control, and Firm versus Lax
Limits. In a slightly different model, Seigelman (1965) labeled the
orthogonal factors as Loving, Demanding, and Punishment.

Research has indicated that children have differing percepitons
of maternal and paternal roles. There is remarkable consensus across
studies (e.g., Emmerich, 1959; Kagan, 1956; Kagan & Lemkin, 1960) that
fathers are seen as more punitive, more feared, more dominant, and
less nurturant than mothers. Perceptions of parents also vary by sex
of the child. In his review of literature, Goldin (1969) noted that
boys have a less positive perception of both parents than do girls.

For example, such studies have indicated that in comparison to girls,

19

boys see fathers asshowing less affection, less approval, less avail-
ability, and more aggression. Mothers are seen as more dominating
and more punishing.

Of particular relevance to the present study in research indi-
cating that the age of the child is a critical factor in his/her
perception of parents. Children are aware of parental behaviors from
a very early age. Hartley (1960) reported that even children as young
as 3 years of age were able to make distinctions between maternal
and paternal roles, albeit at a crude level. Dubin and Dubin (1965)
concluded that with age children's perceptions become more realistic,
that is, more congruent with observed parental behavior. Studies have
also shown that with age, children come to take more subtle dimen-
sions into account. For example, Emmerich (1961) reported that
children showed less stereotyping. Other studies (e.g., Emmerich,
1959; Hess & Torney, 1962) noted that older children become able to
characterize parents along dimensions such as relative power, or
dominance, rather than being restricted to behavioral descriptions.
The child's ability to differentiate between role characteristics
and role enactment has not been widely investigated, although Sowers
(1937) noted that adolescents could make such a distinction. How-
ever, Dubin and Dubin (1965) argued that there may be a relative
progression in the development of children's perceptions, moving from
focus on individual behavior, to perceptions of classes of behaviors
that are seen as characteristic of a role, and finally to an ability to

make generalizations about role functions.

20

While the above discussion has focused on perceptions about
parental roles, there is also a wealth of literature on perceptions
about marital roles. Although the latter work has dealt almost
exclusively with adult samples,it is relevant in perhaps providing a
basis for discussion about how "realistic" are children's perceptions
about marital issues, at least as measured in the present study.

The major focus of much of this literature has been on delineat-
ing factors which facilitate or impede marital satisfaction. In a
review of classic studies, Hicks and Platt (1970) noted that marital
happiness tends to be associated with a variety of demographic vari-
ables such as higher occupational income, and educational levels, and
especially husband-wife similarity in socioeconomic background, age,
and religion. However, with the development of role theory, more
recent investigations have examined the relationship between role
expectations/enactment and ratings of marital happiness.

Nye and Berardo (1973) developed a family paradigm including
seven roles: provider, housekeeper, sexual, recreational, kinship,
supportive, and parental. There is agreement across studies that
well-functioning marriages are characterized by spouses who have
similar role expectations, or at least are willing to compromise
(e.g., Hicks & Platt, 1970; Locke & Williamson, 1958; Scanzioni,
1968). Stuckert (1963) noted that the husband's role definitions
and expectations may be more important to the early success of a
marriage than the wife's. Investigators have also attempted to

evaluate the relative contribution of the various aspects of marital

21

process to marital satisfaction. In several surveys of middle-class
couples, companionship, defined broadly to encompass recreational,
supportive, and intimacy needs, is seen as most critical (e.g.,
Bouerman, 1957; Chadwick, Albrecht & Kunz, 1976; Locke & Williamson,
1958; 0rden & Bradburn, 1968). Housekeeping was credited with having
little impact on marital happiness. Research has also indicated that
while parenting is seen as an important aspect of marital role, it

is simultaneously described as a frequent source of marital tension

(Chadwick et a1., 1976; Hicks & Platt, 1970).

STATEMENT OF PURPOSE AND HYPOTHESES

A major element missing from social cognitive literature to date
is an examination of how children perceive and understand the adult
world around them. This issue is of paramount importance. Children
are constant observers of adult interaction, and the interpretations
they make are a primary factor in the socialization process. They
integrate such observations into their current functioning and the
effects of these early observations may certainly be manifested when
they too acquire adult status.

Hence, the present study investigated one aspect of children's
understanding of the adult world, their concept of marriage and
divorce. More specifically, it first examined children's perceptions
of relationship partners. This is the basic question of person per-
ception--what sort of person does the child see as attractive to marry,
or as likely to provoke divorce. To compare this to children's
understanding of their own social activities, examination was also
made of the type of person they are likely to befriend. By way of
further elaboration, this study also examined children's perceptions
of motivations for marriage and divorce, and their characterization
of on-going marital and post-marital process.

The second basic focus of this study was on children's expecta-

tions about marital role enactment. In particular, analysis focused

22

23

on couples' modes of handling differences in point of view about role
performance. This issue was selected because problem solving is
obviously a primary issue in relationship maintenance. Indeed, extant
data on friendship highlighted children's expectations for coopera-
tion and tension resolution. Finally, this study examined children's
personal attitudes toward marriage.

As this study was exploratory, hypotheses could only be formu-
lated in general terms.

1. An age effect was anticipated, reflecting developmental
trends in the child's reasoning abilities. In particular, it was
expected that children's perceptions of relationship partners would
show a decline in egocentrism and a rise in ability to make inferences
about underlying personality features.

2. The design also included examination of sex differences.
Although extant literature on social cognition makes frequent mention
of this variable, the direction of effect is not clear. Hence, no
specific hypotheses were formulated for the present study.

3. Social cognitive literature also includes mixed results
about effects for the stimulus object's valence (liked/disliked).
This variable was included in the design through comparison across
marriage and divorce. However, the direction of effect could not be
anticipated.

4. This study also examined differences between children's
understanding of relationships in which they are actively engaged

(friendship) and relationships of which they are only a third party

24

observers (marriage and divorce). Although this point has not been
discussed in this report, it is the obvious hypothesis that familiar
experiences are likely to be understood at a higher cognitive level

than more foreign experiences.

METHODOLOGY

This study used a 2 x 2 design. The first independent variable
was sex. The second independent variable was age, including sub-
jects 7 to 8 years old (Grade 2), and 13 to 14 years old (Grade 8).

The dependent variables were derived from the subjects' responses
regarding three types of social relationships: friendship, marriage
and divorce. This was considered a repeated measures factor. Addi-
tional dependent variables were derived from the subjects' responses
regarding four aspects of marital role: housekeeping, economic pro-
vision and expenditure, childrearing, and affective expression. This

was also considered a repeated measures factor.

Subjects

Subject criteria included socioeconomic status, family structure,
and age. To participate in the study, the subject had to be of a
middle class, intact family. An intact family was defined as one in
which both natural parents resided and which had no history of marital
separation. Furthermore, only subjects 7 to 8 and 13 to 14 were
included.

These age groups were chosen for investigation as they were
likely to correspond with two major Piagetian stages of development,
concrete and formal operations. Age 7 was selected as the lower limit

in order to ensure that the children had developed the requisite

25

26

verbal skills to complete the interview. Initial plans for this
study also included preschool children, 4 to 5 years of age, corre-
sponding to Piaget's preoperational stage of cognitive development.
However, pilot testing raised serious doubts about their verbal compe-
tence. Such children tended to give very brief and inarticulate
responses, especially in reference to marriage and divorce. While
this might indicate that their concepts of such social relationships
have not developed beyond at best a simplistic level, it might also
merely reflect their verbal limitations. Hence, this young age group
was dropped from investigation at this time. A less verbally demand-
ing methodology would be required for the data to be considered mean-
ingful.

The sample was drawn from communities in the environ of a large
university. Names of prospective subjects were obtained from school
class lists provided by several families in the area. The mothers
of all prospective subjects who met the selection criteria were con-
tacted by phone to seek permission for their child's participation.
The rate of refusal was very low and usually attributed to scheduling
difficulties. A copy of the explanatory letter given to parents at
the time of their child's participation is available in Appendix A.

The final sample consisted of 40 subjects, divided across the
four cells according to age and sex. The younger group ranged in
age from 7 years 8 months to 8 years 11 months, with a mean age of
8 years 2 months. The older group ranged in age from 13 years 0

months to 14 years 11 months, with a mean age of 13 years 10 months.

27

All children came from middle class backgrounds as judged by family

residence.

Interview Schedule

 

Data were obtained by using an individualized interview format.
Each child was asked a series of set questions, some open-ended, and
others restricted to a multiple-choice response. There was a total
of fourteen questions.

Five aspects of children's understanding and attitudes about
friendship, marriage and divorce were examined:

1. Perceptions of a friend, a possible spouse, and a

possible ex-spouse
2. Perceived motivation for marriage and divorce
3. Characterization of on-going marital and post-marital
process

4. Personal attitudes toward marriage

5. The process of marital problem solving
The last portion of the interview encompassed exploration of the per-
ceived contribution of various aspects of marital role to marital
tension, the likelihood of its resolution, and the range of possible
problem-solving techniques.

A copy of the interview questions is provided in Appendix B.
It will be noted that the question regarding friendship was placed
first in the interview schedule. This was done to help reduce the
subjects' anxiety. Friendship was likely to be a more familiar and

less threatening topic for children than marriage and divorce.

28

The interview format and final interview schedule were developed
on the basis of extensive pilot testing. Several other experimental
techniques were considered and tried. One pilot design presented
subjects with a descriptive narrative and situational dilemmas and
then assessed their reactions using an open—ended format. This is a
standard technique in cognitive developmental research, popularized
by the work of Kohlberg (1969) and more recently by Selman (e.g.,
1976a). Another pilot design called for subjects to be interviewed,
but included only a multiple-choice response format. This method is
also popular, and its advocates cite the reduced dependence on verbal
skill as a prime advantage. Methodological issues were extensively
discussed earlier in this report. One of the major criticisms of
techniques mentioned above is that they impose g priori restraints
on the type of data collected and presuppose that the material the
investigator presents and the response options he/she provides are
indeed the most relevant for the child. This caveat seemed particu-
larly pertinent to the present study. Because of the dearth of
literature on children's perceptions of marriage and divorce, the
present study represented an exploratory venture. The largely open-
ended interview format was adopted on the recommendation that it

best taps the potential richness of the child's ideas and expression.

Procedure
All subjects were tested individually at their homes. Typically,
one or both parents were also in the house. The study was explained

to them and written permission obtained for their child's

29

participation. Verbal agreement to participate was also obtained
from each subject. No subject refused to be involved in the study.
Indeed, the subjects were strikingly cooperative, attentive and
intrigued by the somewhat novel situation. All subjects were inter-
viewed in a quiet room in which the confidentiality of their responses
was ensured. A brief period was always spent establishing rapport.
Each interview took approximately thirty minutes. The study was
introduced by a short explanation that the investigator was inter-
ested in what children thought about different types of people and
the ways they behave toward each other. Every attempt was made to
ensure that the subjects understood the interview questions. Most
questions included several alternate phrasings. After receiving an
initial response, the investigator used a series of prompts to facili-
tate further responsiveness. The goal was to tap the subjects' most
elaborated or highest level of understanding. The prompts were
strictly nondirective and included the usual phrases such as "tell
me more" or "what else do you think?" It was not until subject had
apparently exhausted his range of responses that the next question in
the series was presented. Response were recorded verbatim. An
attempt had been made in pilot testing to have subjects write their
responses. However, this design was discarded because of readily

apparent discrepancies between their verbal and writing skills.

Data Coding

 

Subjects' responses on the open-ended interview quetions were

divided into units of information, each containing reference to a

3O

discrete behavior, characteristic, or evaluation. These units were
used as the basis for data analysis. A systematic coding system

was developed to achieve quantification or descriptive analysis of

the response elements. Some of the coding categories were adopted
from scoring protocols used in related studies. Alterations and addi-
tions were developed by the investigator on the basis of pilot results
and the specific issues addressed in the present study. Coding was
done independently by the investigator and an experienced graduate
student in clinical psychology. Definitions and examples of the cod-

ing categories are presented in Appendix C.

RESULTS

Inter-Coder Agreement
The consistency of the data coding was calculated according to
the most stringent procedure. For each category the number of units
on which the coders were in agreement was compared to the total number
of units being coded. Consistency was extremely high. Coefficients
ranged across categories from 80% to 100%, with 35 out of 42 cate-

gories showing consistency at 90% or higher.

Person Perception in Friendship, Marriage, and Divorce
Descriptions of a friend, a possible spouse, and a possible
ex-spouse were examined from both structural and content perspec-

tives.

Structural Dimensions

Personal involvement. This dimension refers to the frame of

 

reference used'h1describing others. It assesses the degree to which
the subject maintains himself/herself as the pivot focus. Three
levels of Personal Involvement were distinguished: (l) egocentric--
descriptions which are narcissistic and subjective; (2) mutual--
descriptions which make reference to the relationship between the
subject and others; and (3) other-oriented--descriptions which are
objective and decentered (Peevers & Secord, 1973). These categories

were assumed to lie along a developmental continuum.

31

32

Scores were assigned according to the most prominent mode of
response, the level containing the highest number of response units.
It was hypothesized that this would tap the child's most likely level
of day-to-day conceptualization. In cases of a tie in prominence
rankings, the score for the higher level in the three-point scale was
assigned.

Means and standard deviations for each subject group are reported

in Table l. The results were examined using a 2 x 2 x 3 analysis of

Table 1: Mean and Standard Deviation Scores for Personal Involvement

 

 

 

Scale
Friendship Marriage Divorce

Girls: 7' 2.4 2.8 2.0

Young SD .84 .63 .33
Boys: Y 1.8 3.0 2.8
SD 1.03 0 .96

Girls: 'Y 2.1 3.4 1.8

Old SD .89 .96 1.03
Boys: 1' 2.1 2.8 2.0

SD .74 .63 .33

 

variance for age, sex, and social relationship, with repeated meas-
ures on the last factor. A summary of the analysis is presented in
Table 2. There was a signifiant main effect for the type of social

relationship (F2 64 = 8.866, p < .001). However, main effects for

33

Table 2: Analysis of Variance for Personal Involvement Scale

 

 

 

Source df MS F
Age 1 12 1.125

Sex 1 .3 .281

Age . Sex 1 .3 .281

Error Between 36 1.067

Social Relationship 2 6.1 8.866*
Age . Social Relationship 2 .3 .436

Sex . Social Relationship 2 1.2 1.744

Age . Sex . Social Rela- 2 .4 .581

tionship
Error Within 64 .680
* p < .001

age and sex were not significant and there were no interaction
effects.

Post hoc analysis was done using the Scheffé technique for
multiple comparisons between means. Signficant differences were
found between the degree of Personal Involvement in friendship and

marriage (t = 3.525, p < .05) and between divorce and marriage

obs

(t bs = 3.715, p < .05). Subjects across age and sex groups tended

o
to describe marriage in more objective and decentered terms than they

described friendship and divorce.

34

Socialgpenetration. This dimension refers to the depth of the
child's insight. It assesses the degree to which the subject goes
beyond surface characteristics to make inference about internal quali-
ties. Three levels of Social Penetration were distinguished: (1)
external orientation--descriptions which focused on appearance or
possessions; (2) behavioral orientation--descriptions which focus on
action and activities, and (3) internal orientation-~descriptions
which focus on inferred traits and attitudes. These categories were
also assumed to lie on a developmental continuum, with the more pene-
trating levels assumed to require greater insight and integration.

Scores were assigned using the same procedure as for Personal
Involvement. Each subject was credited for his/her most prominent
mode of response, and in the case of tied frequencies, the higher
level in the three-point scale was scored.

Means and standard deviations for each subject group are given
in Table 3. A 2 x 2 x 3 analysis of variance was performed for age,
sex, and social relationship, with repeated measures on the last
factor. Table 4 presents a summary of these results. The main
effects for age and sex were significant. The main effect for
the type of social relationship was not significant. However,
there were significant interaction effects between type of social
relationships and sex (F2, 64 = 8.0, p §_.001), and among type of
social relationship, age and sex (F2 ’ 64 = 8.0, p §_.001). That
is, older children were more likely to make inferences about covert
features than younger children. Furthermore, girls showed greater

penetration than boys on friendship, but not on marriage and divorce.

35

Table 3: Mean and Standard Deviation Scores for Social Penetration

 

 

 

Scale
Friendship Marriage Divorce

Girls 'Y 2.6 2.1 2.3
Young SD .99 .48 .89
Boys 'N' 2.0 2.1 2.4

SD .81 .84 .66

Girls '7 3.0 2.7 2.4

01d SD .68 .62 .0
Boys 7' 1.8 2.6 2.6

SD .7 .62 .63

 

Table 4: Analysis of Variance for Social Penetration Scale

 

 

 

Source df MS F
Age 1 2.15 6.2*
Sex 1 2.13 6.14*
Age . 1 .28 .808
Error Between 36 .347
Social Relationships 2 .055 .159
Social Relationships . Age 2 3.065 1.0
Social Relationships . Sex 2 .605 8.0*
Social Relationships . Age 2 3.04 8.0*

. Sex
Error Within 64 .342

*p < .05

**p < .001

36

Evaluative direction. This dimension has not been previously

 

studied in terms of person perception. It refers to whether the
descriptions of others were given in affirmative or negative terms.
This does not imply reference to whether the evaluation was positive
or negative. Rather, it takes into account whether the subject
focused on the presence (Jr absence of a characteristic. Whether
children show a developmental trend in their use of affirmative or
negative structure has not been delineated in previous research.
Hence, this dimension was considered dichotomous in the present study,
with a score for each subject assigned according to the most promi-
nently used category.

The proportion of subjects using each category is reported in

Table 5. Separate chi-squares were computed for each type of social

Table 5: Evaluative Description Scale--Percentage of Subjects Using
Affirmative Descriptions '

 

 

 

Friendship Marriage Divorce
Girls 100 90 80
Young
Boys 100 100 60
Girls 100 90 60
Old

Boys 100 90 90

 

37

relationship to determine age and sex differences. None of these
analyses was significant. Hence, subject groups were combined to
examine differences in Evaluative Direction across friendship, mar-
riage, and divorce using Cochran's Q test for related proportions.
This analysis showed a significant effect (0 = 37.22, p < .001). The
chi-square analog of the Scheffé test was then used as the post hoc
comparison procedure for the Cochran test (Marascuilo & McSweeney,
1977). Results revealed significant differences between friendship
and divorce (p < .05), and between marriage and divorce (p < .05).
Subjects tended to describe divorce in largely negative terms,
whereas friendship and marriage were more likely to be described in

affirmative terms.

Content Dimensions

 

Subjects' descriptions of a friend, possible spouse, and possible
ex-spouse were each coded along twelve content dimensions. These
categories referred to: Possessions, Proximity, Appearance, Social
Behavior, Activities, Personality, Psychological Intimacy, Acceptance-
Equality, Affective Involvement, Compatibility, Parenthood, and Sexu-
ality. This coding system was adapted from the work of Bigelow (1977;
Bigelow & LaGaipa, 1975) and Hayes (1978). These categories were not
assumed to lie in a developmental continuum. Hence, each content
dimension was considered separately.

Analysis was based on a frequency count of the use/nonuse of each
category. Credit for usage was given regardless of the number of

descriptive units so categorized. For the most part, descriptions

38

were coded along several dimensions, with subjects rarely giving
emphasis to a particular category. Hence, the notion of a most promi-
nant mode of response for each subject was not applicable.

The proportion of subjects using each category are presented in
Table 6. The effects of age and sex on each dimension were examined
using chi-squares. Significant effects were found for several cate-
gories. With regard to friendship, girls were more likely than boys
to be concerned about Personality factors (X2 = 6.547, p < .05), and
the older subjects were more likely than younger subjects to be con-
cerned with Psychological Intimacy (x2 = 8.025, p < .01). With
regard to divorce, descriptions given by older subjects were more
likely than those of younger subjects to contain reference to issues
of Acceptance-Equality (X2 = 22.94, p < .001), and issues about appro-
priate Social Behavior (X2 = 5.013, p < .05). There were no age or
sex differences in descriptions given for a marriage partner.

Cochran's test for related proportions was used to compare across
descriptions of a friend, possible spouse, and possible ex-spouse.
For infrequently used categories, the Binomial test was substituted
according to Siegel's (1956) recommendation. Where there were no age
or sex differences, analysis was done on the total sample. Signifi-
cant effects were found for Appearance (0 = 12.846, p < .001), and
for Sexuality (p < .01). For girls there was a significant effect
for Personality (Q = 9.8, p < .01). For older subjects there were
significant effects for Acceptance-Equality (Q = 7.091, p < .05), and
for Psychological Intimacy (Q = 7.091, p < .05).

39

 

 

 

 

 

 

 

 

 

op cm on op o o o o -1 - 1- 1- apwpmzxmm
o o— op op op op op om 1- - 1- - noogucmcmm
om om om om om om op op oo oo oe om zuwpmnmumaeou
acm5m>po>cm
op op op om o op o om om op om om m>wuumww<
xpmpmacm
cm as op op op oa O or o o o oF -mocmuamou<
momewucH
ow o o o o_ on o o co om op o Pmuwmopoguxmm
om oe oe oe on om om om om o“ ow om xuwpmcomcmg
0 or o o o_ om op om ca oe om om mmmuw>wuo<
om oh om om cm on or om on on ow oe cow>msmm meuom
o o 2 o 8 on 8 oo on o 8 S 85:82
o om op op o op o o om om o op xuwspxoca
on o op o o cm on or op op om om mcowmmommoa
mace mpcww mxom mpcwo mxom mpcwu whom mpcwu maom mpcwo mzdm mpgwo
2o 9.2; 2° 953 2o 95.;
muco>vo mmmwccmz awcmucmwcu

 

agommpmu comm mcwm: muumnnam
we mmmucmucmm1-muco>wo new mmmwccmz .awgmucowcm c? copuamucma comcma com mvmzpmc< acmucou no «pack

40

The chi-square analog of the Scheffé test was again used as the
post hoc comparison procedure for the Cochran test (Marascuilo &
McSweeney, 1977). Appearance was more likely to be featured in
descriptions of a spouse than in descriptions of a friend (p < .05)
or an ex-spouse (p < .05). Sexuality was more at issue in divorce
than in marriage (p < .05). Taking age and sex differences into
consideration, Psychological Intimacy was more likely to be mentioned
by older subjects in reference to friendship than marriage (p < .05)
or divorce (p < .05). This relationship did not hold for the younger
subjects. There were significant differences for older subjects in
use of the category Acceptance-Equality between divorce and marriage
(p < .05) and between marriage and friendship (p < .05). Personality
characteristics were more likely to be cited by girls in the context
of friendship than divorce (p < .05). This pattern was not found

for boys.

Motivation for Marriage and Divorce

 

Children's perceptions about why adults marry and divorce were
each coded along ten content dimensions. These categories referred
to: Appearance, Possession, Activities, Compatibility, Social
Behavior, Independence,Depth-Commitment, Affective Involvement, Sex-
uality, and Parenthood. Analysis was based on a frequency count of
the use/nonuse of each category.

The proportion of subjects using each category are presented in
Table 7. Chi-square analyses revealed little effect for age and

sex on children's explanations of why people marry and divorce. The

41

 

 

 

 

 

 

 

 

op o o c op o o op coozpcmcma
om op om op o o op op zuwpmaxmm
ow om om om om cop om om ucmsm>Fo>2H m>wpommw<
op cm 0 0— cm om om o pcmspvssou-gunmo
OF or op op o o o op mocmncmamucm1mucmc=ucsz
op om op op op o o_ o cow>mnmm meuom
ow om om cm 0 op o o xurpwnmumneou
OP o o o om om op op momwcmucH-mmpuw>wuo<
o c or o 0 ON o o mcomemmmom
o o o o o 8 8 o 85:82
maom mpcwo whom mpcwo whom mpcwo whom mpcmw
upo mczo> u—o maze»
moio>wo mmmwccmz

 

accomumu comm acmmz
mpomnnsm to mumpchLma1-moco>wo vcm mmmwccmz Loo cowpw>wpoz coo mwmapmc< acmpcou "N mpnmh

42

only significant difference was between young girls and boys, with
the former group placing more emphasis on the issue of Affective
Involvement in divorce (X2 = 5.0, p < .05).

A comparison of the reasons given for marriage as opposed to
divorce was done using McNemar's test for two related proportions.
For categories which were infrequently used, the Binomial test was
”substituted according to Siegel's (1956) recommendation. Signifi-
cant differences were found on only two categories. Children were
likely to perceive marriage as resulting from Affective Involvement
(p < .05). In contrast, divorce was seen as a function of the degree

of Compatibility (p < .05).

Characterization of Marital and Post-Marital Process

 

This portion of the interview attempted to tap the child's per-
ception of on-going marital and post-marital process. This was first
examined by asking open-ended questions. The data were coded along
ten content dimensions. These categories encompassed: Proximity,
Possessions, Financial Responsibility, Responsibility for Self,
Parenthood, Sexuality, Compatibility, Independence, Affective Involve-
ment, Depth-Commitment. As in the content analysis delineated
above, analysis was based on a frequency count of the use/nonuse of
each category.

The proportion of subjects using each category are presented in
Table 8. Age and sex differences on each dimension were examined
using chi-square analyses, or in the case of low frequency usage,

using Fisher's Exact Probability Test (Siegel, 1956). Few age and

43

 

 

 

 

 

 

 

 

om o o o om om o op acmEquEoQ-nuamo
ON 04 cm on o op om om acmem>_o>=H m>woumtt<
om om o o op o o o mucmucmamuca
o om o o o_ om o o aywpwawquEou
o o~ op op op op o o— xuwpmsxom
op ow om om op o om om coonucmcmm
om o o as op o o_ o cpmm Let soapwnwmcoammm
op o_ o op on o om o zawpwnwmcoammm meucm=_d
om op o o_ on o o op meowmmmmmoa
om om om om ow oe on so xpwswxocm
maom mpcwo mxom mpcww mzom mpcwu whom mpcwu
u_o mesa» upo a==o>
moco>wo mmmwccmz

 

xgommpmu comm anew:

muomnnsm mo monocoocma1-moco>wo new mmmwccmz mo cowumNpcmuomcmso coo mwmapmc< ucwucou ”w mpnmp

44

sex differences were revealed. The only significant age difference
was that older subjects were more likely than younger subjects to
characterize marriage in terms of the Depth-Commitment involved in
the relationship (X2 = 9.643, P.fi.-01)' The only significant sex
difference was that young girls were more likely than young boys to
describe divorce as entailing Responsibility for Self (p < .05).

On-going marital and post-marital process were compared using
McNemar's test for related proportions or the Binomial test in cases
of low frequency category usage (Siegel, 1956). Marriage in contrast
to divorce tended to be characterized in terms of Affective Involve-
ment (X2 = 4.9, p < .05) and for older subjects, in terms of Depth-
Commitment (x2 = 4.9, p < .05).

Children's perceptions of on-going marital and post-marital
process were also examined using a more structured methodology. Par—
ticular interest was focused on children's perceptions of the affec-
tive state of people who were married or divorced. Data were obtained
using a four-point scale, ranging from very happy/satisfied to very
unhappy/dissatisfied.

Means and standard deviations for each subject group are given
in Table 9. A 2 x 2 x 2 analysis of variance was performed for age,
sex and social relationship, with repeated measures on the last
factor. Table 10 presents a summary of the analysis. There was a
main effect for social relationship (F1,36 = 34.93, p < .001).
Married persons were perceived to be happier than divorced persons.

Age and sex differences were not significant.

45

 

 

 

 

 

 

Table 9: Mean and Standard Deviation Scores for Rating of Affec-
tive State in Marriage and Divorce
Marriage Divorce
Girls 7' 4.0 1.6
SD .0 .7
Young __
Boys X 3.7 2.2
SD .68 1.23
Girls T 4.0 1.6
SD .0 .52
Old __
Boys X 3.8 1.7
SD .42 .68
Table 10: Analysis of Variance for Affective State in Marriage
and Divorce
Source df MS F
Age 1 .2 .318
Sex 1 .05 .079
Age . Sex 1 .2 .318
Error Between 24 .629
Social Relationship 1 88.2 34.93*
Social Relationship . Age 1 .45 .178
Social Relationship . Sex 1 1.8 .713
Social Relationship . Age
. Sex 1 .45 178
Error Within 36 2.525

 

*p < .001

46

Personal Attitude Toward Marriage

 

Each subject was also asked about his/her own future plans about
marriage. The data were coded into dichotomous categories. These
referred to: Positive Expectations or Negative Expectations (i.e.,
expectations of marrying/not marrying in the future).

The proportion of subjects using each category is reported in

Table 11. Chi-square analyses were performed to isolate age and

Table 11: Personal Attitude Toward Marriage--Percentage of Subjects
with Positive Expectations

 

 

Young Old
Girls 80 90
Boys 30 60

 

sex differences. As might be expected, girls were more likely than
boys to anticipate getting married in the future (X2 = 5.385, p<:.05).

No other differences were significant.

Marital Tension and Problem Solving
The final portion of the interview examined children's percep-
tions about the sources of marital tension and means for its resolu-
tion. Four major aspects of marital role were highlighted: house-
keeping, financial provision and expenditure, childrearing, and

affective expression.

47

Importance of Marital Agreement

 

Subjects were asked to rate the importance that husbands and
wives agree on each aspect of marital role. The four-point scale
range from not at all important to very important.

Means and standard deviations for each subject group are cited in
Table 12. The results were examined using a 2 x 2 x 4 analysis of

variance for age, sex and marital role, with repeated measures on the

Table 12: Mean and Standard Deviation Scores for Ratings of Impor-
tance of Marital Agreement

 

House- Child-

 

 

Financial keeping rearing Affect
Girls 7 4.0 3.7 4.0 3.9
SD .0 .48 .0 .32
Young __
Boys X 3.5 3.5 4.0 3.9
SD .71 .97 .0 .32
Girls 'X’ 3.7 3.5 3.9 4.0
SD .48 .53 .32 .0
Old _.
Boys X 3.8 3.3 4.0 3.9
SD . .42 .68 .O .32

 

last factor. A summary of the analysis is given in Table 13. There
was a significant main effect for marital role (F3 109 = 10.766,
p <.OOl). However, effects for age and sex did not approach signifi-

cance. Post hoc comparisons across aspects of marital role were

48

Table 13: Analysis of Variance for Ratings of Importance of Marital

 

 

 

Agreement

Source df MS F
Age 1 .1 .345
Sex 1 .4 1.38
Age . Sex 1 .225 .776
Error Between 36 .29
Marital Role 3 1.841 10.766*
Marital Role . Age 3 117.0 .684
Marital Role . Sex 3 .15 .877
Marital Role . Age

. Sex 3 .242 1.415
Error within 109 .71

*p < .001

performed using the Scheffé technique. Significant differences were
found between the perceived importance of marital agreement around
issues of housekeeping and that of childrearing (tobs = 5.307, p<:.05)
and affective expression (tobs = 4.749, p < .05). That is, consensus
about childrearing and affective expression was seen as more impor-

tant across all age and sex groups than was housekeeping.

Likelihood of Marital Tension

 

Subjects were requested 1x1 rate how often they thought couples
had difficulty reaching agreement on housekeeping, financial issues,
childrearing, and affective expression. The four-point scale ranged

from never to all the time.

49

Table 14 shows means and standard deviations of the ratings

given by each group of subjects. As above, a 2 x 2 x 4 analysis of

Table 14: Mean and Standard Deviation Scores for Ratings of Likeli-
hood of Marital Tension

 

 

 

. House- Child-
Flnances keeping rearing Affect
Girls 7 2.2 3.0 3.4 3.0
Young SD .79 .82 1.35 1.15
Boys '7 2.3 2.7 3.2 3.2
SD .48 .95 1.30 .92
Girls 7' 2.8 2.3 2.3 2.2
01d SD .42 .48 .68 .63
Boys 7 2.9 2.6 2.8 2.4
SD .56 .70 .79 .62

 

variance was used for age, sex and marital role, with repeated
measures on the last factor. The results indicated a significant
interaction between age and marital role (F3,109 = 9.936, p < .001).
Other interaction effects and main effects were not significant
(Table 15). Post hoc comparisons across marital role were performed
separately for the two age groups. Based on the Scheffé technique,
significant differences were found between ratings given by the
younger subjects for financial issues and those given for childrear-

ing (tobs = 3.601, p < .05) and affective expre551on (tobs = 2.915,

50

Table 15: Analysis of Variance for Ratings of Likelihood of Marital

 

 

 

Tension

Source df MS F
Age 1 4.557 3.759
Sex 1 .507 .418
Age . Sex 1 1.057 .872
Error Between 36 1.212
Marital Role 3 1.006 2.367
Marital Role . Age 3 4.223 9.936*
Marital Role . Sex 3 .073 .172
Marital Role . Age . Sex 3 .34 .8
Error within 109 .425

*p < .001

p < .05). Young children tended to view marital couples as having
relatively little difficulty reaching agreement on financial matters.
In comparison, issues of childrearing and affective expression tended
to be viewed as much more problematic for couples to agree upon.
Older subjects showed no significant differences in their ratings
across aspects of marital role, viewing them as relatively equal

contributors to marital tension.

Likelihood of Tension Resolution
In this section of the interview, subjects rated how easily they

thought marital problems could be resolved. A four-point scale was

51

used, ranging from problems being very difficult to very easy to
solve.
Table 16 shows the means and standard deviations of ratings given

by each group of subjects. A 2 x 2 x 4 analysis of variance was

Table 16: Mean and Standard Deviation Scores for Ratings of Like-
lihood of Tension Resolution

 

 

 

. House- Child-
Flnances keeping rearing Affect
Girls 'X 2.2 2.8 2.1 2.6
Young so .42 1.14 .99 .82
Boys 7' 2.6 2.5 2.2 3.2
SD .96 .71 .63 1.14
Girls 'X 2.4 2.8 2.6 2.8
Old 50 .48 .63 .96 .79
Boys 7' 2.0 2.2 1.9 2.0
SD .47 .63 .73 .82

 

performed for age, sex and marital role, with repeated measures on
the last factor. This analysis is summarized in Table 17. The data
yielded a main effect for marital role (F3,109 = 3.644, p < .05).
Dispute about finances and childrearing were seen as more difficult
to resolve than disputes over housekeeping or affect. There was also

a significant interaction effect between age and sex (F = 5.272,

1,36
p < .01). Boys became slightly more optimistic about disputes with

52

Table 17: Analysis of Variance for Ratings of Likelihood of Tension

 

 

 

Resolution

Source df MS F
Age 1 1.407 .87
Sex 1 1.087 .672
Age . Sex 1 8.525 5.272**
Error Between 36 1.617
Marital Role 3 2.19 3.644*
Marital Role . Age 3 .273 .454
Marital Role . Sex 3 .313 .521
Marital Role . Age . Sex 3 .267 .444
Error within 109 .601

*p < .05

**p < .01

age while girls became less optimistic. There were no other signifi-

cant effects.

Means of Problem Resolution

 

The final portion of the interview required subjects to suggest
ways couples could try to solve their marital problems. This mate-
rial was coded along four content dimensions. These included:
Apology, Denial, External Aid, and Mutual Discussion. Analysis was
based on a frequency count of the use/nonuse of each category.

The proportion of subjects making reference to each category are

presented in Table 18. Age and sex differences were investigated

53

Table 18: Types of Problem Resolution--Percentage of Subjects Using
Each Category

 

 

 

Apology Denial External Aid Discussion
Girls 10 4O 20 60
Young
Boys 10 50 10 40
Girls 10 20 4O 40
Old
Boys 0 30 50 50

 

using chi-square analysis. No signifiant age differences were found.
However, there was a significant effect for sex with regard to use of
discussion as a means of problem solving (X2 = 5.833, p < .05). Girls
were much more likely than boys to offer this as a means for reducing

marital conflict.

DISCUSSION

The following discussion reviews these findings and attempts to
interpret them in view of data and arguments in the literature and
in view of this study's hypotheses regarding the effects of age,
sex, object valence (liked/disliked), and familiarity of relationship
on social cognition. The data reported in this study highlight two
major issues: (1) the cognitive structure and content of children's
perceptions of whom they might befriend, marry, or divorce; and
(2) children's perceptions of issues which may help to consolidate
or threaten a marital relationship.

Cognitive Structure and Content:
Marriage, FriendshipLand Divorce

 

This study looked at three structural elements underlying
children's reasoning about social relationships: Personal Involve-
ment, Social Penetration, and Evaluative Direction. Cognitive devel-
opmental literature makes strong predictions about the effects of age
on the first two of these structural elements. Prior research shows
that in reasoning about the physical world and some aspects of the
social world, children become increasingly competent in inferential
and nonegocentric thought.

The results of this study are only partially consistent with

these predictions. Considering the Social Penetration scale, the

54

55

data are in the expected direction when children are describing
friends or a possible ex-spouse. For these social relationships,
older children showed greater ability to go beyond surface character-
istics and focus on more inferential aspects of the partner and
relationship than younger children. Content analysis illustrated

the types of covert factors to which children become sensitive. Psy-
chological intimacy was cited by older children as being particularly
important in friendship. When considering divorce, older children
noted the importance of feeling equal and accepted in a relationship,
and awareness of societal expectations about the marital role. All
of these dimensions require considerable social insight. However, an
age effect was not found for marriage. Older children did not
describe a marriage partner in more inferential terms than younger
children. One possible explanation is suggested by the content
analysis. There was a strong pull across age groups to describe a
marriage partner in terms of appearance, a very superficial dimen-
sion.

While children's inferential ability as measured by the Social
Penetration scale showed improvement with age for two of the three
social relationships examined (friendship and divorce), children's
ability to construe others in decentered terms did not show an age
effect. Across friendship, marriage, and divorce, younger and
older subjects did not differ in the extent to which they used an
egocentric/decentered frame of reference to describe others. Both

age groups tended to respond in a nonegocentric manner. This was

56

surprising as the development of skill in social penetration and
decentered reasoning are usually thought to be correlated.

In part this may reflect the ceiling of the scale used. How-
ever, another likely explanation for the lack of an age effect on
the Personal Involvement scale is that the wording of the interview
question tended to specifically foster decentered thinking from both
young and older children. Extant person perception studies which
report a developmental trend toward decentered thinking (e.g., Bigner,
1974) have phrased the interview question in terms of "what makes
someone your friend." Among young subjects, this phrasing has tended
to pull for descriptions of what friends give to you or what friends
share together. These types of responses fall toward the low end
of the egocentric/decentered continuum. In contrast, this study
asked subjects to describe the "sort of person" a friend, spouse, or
ex-spouse would be. This wording may make the task of decentering
easier for younger subjects by attuning them to describing the person
per se, rather than describing the person in relation to him/herself.
Hence, the young subjects in this study were able to match the cogni-
tive level of older subjects, giving responses falling toward the
more sophisticated end of the egocentric/decentered continuum.

Cognitive theory has not dealt with developmental trends in
children's use of affirmative or negative structure (Evaluative
Direction) in person perception. One might hypothesize that affirma-
tion is a more soohisticated structure in that the child is required

to have a richer repertoire of responses, rather than just relying

57

on negation. However, the results of this study did not indicate
any developmental trends.

In addition to age differences, this study also examined whether
there are sex differences in conceptualization of marriage, friend-
ship, and divorce. Boys and girls showed equal facility in their
egocentric/decentered thinking. Similarly, no sex effect was found
in terms of the Evaluative Direction scale. However, for the dimen-
sion of Social Penetration, an interaction was found between sex and
the way children characterized a friend, spouse, or ex-spouse. Girls
were found to be more likely to make inferential statements in regard
to friendship, but this did not carry over to their perceptions of
a spouse or ex-spouse. The direction of sex difference effects in
regard to friendship are consistent with previous research (e.g.,
Gollin, 1958; Bigner, 1974; Scarlett et a1., 1971). With few excep-
tions, girls have been found to use more "abstract terms" to describe
others than do boys. The content analysis included in the present
study allows one to be more specific. Girls were found to make
reference particularly to personality traits.

These results can be understood in light of research on child-
ren's friendship behaviors. Friendships among boys tend to be
casual and activity oriented. In comparison, friendships among
girls tend to be more enduring, more intensive, and used in more
ego-supportive ways. One might anticipate that the nature of the
friendship interchange among girls promotes, and in turn is promoted

by, their intuitive perceptual style.

58

Explanation for the absence of sex differences on the Social
Penetration scale for marriage and divorce is more difficult. The
results can perhaps best be interpreted in terms of Vygotsky's (1962)
distinction between "spontaneous" and "scientific" concepts. Accord-
ing to Vygotsky, the former are based on everyday experience, while
the latter are acquired through tuition. Children's notions about
friendship are most likely to be derived from their direct expe-
riences. As discussed above, this may differ for boys and girls,
leading to differences in their levels of conceptualization about the
relationship. However, the concepts of marriage and divorce are less
familiar to children and hence their thinking about these relation-
ships is more likely to be a function of tuition. Such learning
experiences are likely to be the same for both sexes.

In addition to examining age and sex effects within each of the
structural dimension, this study also compared these dimensions across
children's concepts of friendship, marriage and divorce. It had
been hypothesized that children would show greatest cognitive soohis-
tication on friendship because it is more familiar to them. As
discussed above, support for this hypothesis was found, at least for
girls, on the Social Penetration scale. However, on the Personal
Involvement scale, marriage was found to be described in more decen-
tered terms than were friendship and divorce.

This finding may be understood in terms of the cognitive sophis-
tication of the stereotypic way these relationships are conceptual-

ized. Description of a marriage partner tends to elicit the social

59

ideal endlessly popularized in the media (visions of the fair lady
or the knight in shining armor). For example, in the content analy-
sis, a spouse was more likely to be described in terms of appearance
than were a friend or an ex-spouse. Hence, for marriage, children
may use a social (i.e., decentered) rather than a personal (i.e.,
egocentric) frame of reference. In contrast, the stereotype for
divorce has a more narcissistic flavor (the partner who failed to
meet one's needs). Once again, the content analysis supports this
hypothesis. Children were most likely to make reference to issues
of personal acceptance and equality when describing an ex-spouse.
They also made more reference to sexuality in terms of divorce than
of marriage. These factors are likely to lead to a predominant use
of an egocentric frame of reference. For friendship, social stereo-
types may also have narcissistic overtones. Perhaps more influential
is their own experience in such relationships. The content analysis
highlighted children's frequent reference to mutual interchange.
Hence, rather than eliciting a more sophisticated level of concep-
tualization, familiarity may in fact promote use of egocentric frame
of reference to describe friends.

Marriage, friendship, and divorce were also compared in terms
of Evaluative Direction. Children were found to describe divorce
in more negative terms than they did friendship and marriage. This
finding is consistent with the point made above that divorce has a
narcissistic pull. By describing divorce partners in terms of quali-

ties which they lack, children emphasized the notion of failure to

6O

meet expectations. In contrast, marriage and friendship were
described in predominantly affirmative terms, emphasizing the pres-
ence of qualities and behaviors.

Issues Contributing to Marital
Solidarity-Marital Tension

 

 

To this point, the discussion has centered on the cognitive
level of children's understanding of friendship, marriage, and
divorce relationships. This study also analyzed what factors child-
ren perceived as facilitating or threatening marital accord.

All children described the decision to marry primarily as a
function of affective involvement, lbving the partner. Few other
sources of motivation were mentioned. As might be expected, older
children had a somewhat more sophisticated notion of love than did
younger children. Older children tended to elaborate that marital
love embodies a commitment to the relationship and that it differs
in depth from other types of caring (e.g., for friends or family).

As might also be expected, young girls tended to be more attuned to
affective issues than young boys. While marriage was seen as a
function of affective involvement, all children described the
decision to divorce as a function of compatibility, not getting along
with the partner. For the most part, the question of compatibility
was couched in behavioral rather than affective terms.

The expectations children held for marriage included spending
much time together, having children, and assuming financial responsi-

bility. The latter expectation is particularly interesting, in that

61

it may reflect children's perception of marriage as a symbol of
acquiring adult status. The expectations children held for divorce
included physical separation of the partners, a change in lifestyles
and parental relationship for the children, and resumption of
responsibility for oneself. This latter point was emphasized more
by young girls than young boys, perhaps reflecting the farmer's
adoption of stereotypes about a woman's greater dependency in a
marital relationship.

For the most part, children tended to be optimistic about mar-
riage. They perceived couples as typically being satisfied with
their relationships. They also saw themselves as likely to marry in
the future, although young boys were somewhat more ambivalent on this
point than young girls. However, the children were also clearly
sensitive to the possibility of marital tension.

This study examined children's perceptions about four possible
sources of marital tension: finances, childrearing, housekeeping,
and affective expression. All children recognized the importance
of marital agreement on affective expression and young children
especially mentioned this as a frequent source of tension. All
children also recognized the importance of marital agreement about
childrearing, and perceived disagreements on this point as particu-
larly hard to resolve. However, young children perceived childrear-
ing issues as a more frequent source of stress than did older
children. This perhaps reflects the greater egocentrism of younger

children, and the greater demands that young children make upon

62

parents. This finding is consistent with research on adults' per-
ceptions of children as a stress on the marital relationship. Child-
ren also perceived that financial issues might be hard to resolve.
However, older children were more likely than younger children to

see this as a frequent source of tension.

Finally, this study examined children's perceptions about how
couples might be able to resolve their differences. All children,
but especially girls, recommended discussion and attempts at problem
solving. There was a trend for older subjects to suggest seeking
help from an outside party.

Implications for Future Research and
Therapeutic Intervention

 

 

In sum, this study has interesting implications for cognitive
developmental theory. The inconsistency in developmental trends on
the Social Penetration scale and the Personal Involvement scale
suggests that these are two independent constructs and may not be as
highly correlated as one might assume. A child showing inferential
thinking style may not necessarily show a tendency to decenter. The
results of this study also suggest that children show considerable
divergence in the type of reasoning they apply to the various concepts
in the social world. This is consistent with documentation of the
decalage across understanding about different aspects of the physical
world.

These trends underscore that investigators must interpret data

with caution. There is clearly little justification given in this

63

study for the pervasive practice of generalizing from research on a
particular social facet to the child's overall understanding of social
cognition. By the same argument, there is still less justification
for hypothesizing that developmental trends in children's cognition
about the physical world will necessarily be congruent with the
development of social cognition. Additional research is needed in
order to explore the generality or interrelationship within and across
domains of cognition.

Furthermore, the decalage found in this and other studies ques-
tions the frequent practice of interpreting data in terms of the
development of cognitive “skill." As children were found to deal with
different social situations at varying degrees of cognitive saphis-
tication, it may be more appropriate to talk about developmental
trends in the child's cognitive 'performance." Additional research
is needed in order to determine what factors facilitate a higher or
lower level of cognitive reasoning about an issue. The present study
suggested that the degree and nature of a child's familiarity with
a social situation may be a significant factor. For example, it
would be of major value to examine the effects of divorce on child-
ren's cognition about social relationships. Finally, when consider-
ing cognitive "performance," there is also need for research using
a range of methodological procedures. The present study highlighted
the impact of even subtle differences in the wording of the inter-
view questions.

This study also has implications for therapeutic work with

children and families. The results highlighted how early the

64

developmental origins of adult notions about relationships may date,
suggesting the value of primary intervention. Processing about
issues such as marriage and divorce cannot be considered the sole
prerogative of adults or even adolescents. Young children are also
very much involved in such issues and intervention should be directed
at this age group. The results also highlighted how sensitive child-
ren are to marital tension. Of particular relevance is the child-
ren's perception of themselves as a source of such tension. It is
perhaps not surprising then that clinicians so frequently encounter
children who feel responsible for parental divorce. It is also sig-
nificant to note how the children generally perceive the expression
of affect, especially love and caring, as problematic. Once again,
it should be underscored that concerns on this count may be acquired
at a very early age. Perhaps the most encouraging note is that
children frequently cited mental health professionals as a possible
source of help. Whether such a positive attitude has been acquired
by the children through their own contact with counselors in the
school system, or acquired through socialization by parents and
others, it would appear that mental health professionals may have
come some way in countering prejudice against requrests for a pro-

vision of such services.

APPENDICES

65

APPENDIX A

LETTER TO PARENTS

66

Dear Parents:

In recent years, families have undergone many changes and a
variety of family life styles have developed. Because the family
is of major importance to the child, researchers are increasingly
involved in learning how families operate.

I am conducting a research project on children's perceptions
about how families operate, especially how parents relate to one
another. Families of course vary considerably and people actually
behave and feel toward others in a whole range of ways. However, I
am interested in learning what children notice about relationships
in families.

The project takes about 20 minutes and can be done at your home

at any convenient time. The child is asked a series of open-ended
questions about various apsects of families, especially parents.
For example, there are questions on family roles, family problem
solving, and what helps to make marriages happy or unhappy. All
questions deal with families in general and not with the child's
own family.

I am looking for children age 7 to 8 and 13 to 14 who might be
interested in participating in this study. I have found that child-
ren usually enjoy participating in research and are very willing to
share their ideas on the topics explored in this project. Naturally,
all materials will be kept confidential.

If you would like to participate in this study or would like

more information, please contact Claire Lowry at 332-8981 (home)
or at 355-9561 (department).

67

APPENDIX B

INTERVIEW PROTOCOL

68

Subject Number:
Date of Interview:
Date of Birth:
Age:

1. Do you have a best friend? What sort of person is he/she?
What is he/she like?

2. Suppose you were going to get married, what sort of person would
you want to marry? What would that person be like?

3. How is being married to someone different from just being friends
with someone?

4. Why do you think two people would decide to get married?

**Sometimes people feel that they don't want to be married to each
other any more and they decide to get a divorce.**

5. Suppose you were married, what sort of person might make you
feel unhappy? What sort of person might make you not want
to be married any more?

6. How is being divorced from someone different from being married
to that person? What changes when people stop being married?

7. Why do you think two people might decide to get a divorce? Why
might they decide not to stay married any more?

8. When people who are married have problems, what should they
do to help solve them? What could they try to do to make things
better?

69

70

**People who are married have to decide about a lot of things.**

9.

How important do you think it is for people who are married to

agree about:

(a) earning and spending money?

 

 

 

 

l 2 3 4
not at all mostly not sort of very
important important important important
(b) looking after the house/apartment?

l 2 3 4
not at all mostly not sort of very
important important important important
(c) looking after the children?

1 2 3 4
not at all mostly not sort of very
important important important important
(d) showing care and affection?

l 2 3 4
not at all mostly not sort of very
important importnat important important

10.

71
How often do you think people who are married have trouble
deciding about?

(a) earning and spending money?

l 2 3 4

 

never sometimes often all the time

(b) looking after the house/apartment?

1 2 3 4

 

never sometimes often all the time

(c) looking after the children?

1 2 3 4

 

never sometimes often all the time

(d) showing care and affection?

1 2 3 4

 

never sometimes often all the time

11.

12.

13.

14.

72

How hard do you think it is for people who are married to
solve problems about:

(a) earning and spending money?

1 2 3 4

 

very hard sort of hard sort of easy very easy

(b) looking after the house/apartment?

1 2 3 4

 

very hard sort of hard sort of easy very easy

(c) looking after the children?

1 2 3 4

 

very hard sort of hard sort of easy very easy

(d) showing care and affection?

l 2 3 4

 

very hard sort of hard sort of easy very easy

How do you think most people feel about being married?

1 2 3 4

 

very sad sort of sad sort of happy very happy

How do you think most people feel after they decide to stop
being married and get divorced?

l 2 3 4

 

very sad sort of sad sort of happy very happy

00 you think you will get married when you grow up? If no,
why not?

APPENDIX C

DATA CODING MANUAL

73

DEFINITIONS AND EXAMPLES OF CODING CATEGORIES

Structural Dimensions
These categories deal with the cognitive framework underlying

person perception in friendship, marriage, and divorce.

Personal Involvement

This dimension refers to the frame of references used in
describing others. It assesses the degree to which the subject
maintains himself/herself as the pivotal focus.

1. Egocentric--descriptions which are narcissistic and

 

subjective--"A friend makes me feel good"; "I'd divorce
someone who didn't give me enough money."

2. ‘Mgtgglg-descriptions which make reference to the rela-
tionship between the subject and others--"As friends,
we are really close"; "A friend is someone who does
things with me."

3. Other-oriented—-descriptions which are objective and

 

decentered--"I'd marry someone who had a good character";

"My friend likes bowling."

Social Penetration

This dimension refers to the depth of the child's insight.
It assesses the degree to which the subject goes beyond surface
characteristics to make inferences about internal qualities.

74

75

External 0rientation--descriptions which focus on appear-

 

ance or possessions-~"My friend brings over toys to play
with“; "When I marry, I'd have a nice house and a pool."

Behavioral Orientation--descriptions which focus on actions

 

and activities--"I'd divorce someone who didn't know how
to behave in public and always got drunk"; "A friend is
someone you do things with."

Internal Orientation--descriptions which focus on inferred

 

traits and attitudes--"He's generous and caring"; "He

wouldn't mean to ever hurt me."

Evaluation Direction

 

This dimension takes into account whether the subject focused

on the presence or absence of a characteristic.

1.

2.

Affirmative--presence of a characteristic--"I'd want the

 

person I marry to be fun-loving and respectable"; "My
friend has lots of toys."

Negative--absence of a characteristic--"I'd divorce someone
who didn't love me or wasn't kind"; "I wouldn't be friends

with someone who didn't take me places."

Content Dimensions

Content analysis attempts to appreciate the richness and variety

of ideas and attitudes expressed by children. Twelve of these were

used to score person perception in friendship, marriage, and divorce.

Ten were also used to score children's characterization of ongoing

76

marital and post-marital process. The final entry refers to the
four categories used to code children's problem-solving suggestions.
Possessions--"My friend has a dog I play with"; "I'd want to
marry a rich man."
Proximity--"I'm friends with Sally because she lives next
door"; "I'd divorce someone who was never around at home."
Appearance--"I'd marry a cute girl"; "If my wife got fat I'd
hate it."

Social Behavior--"I couldn't stand being married to someone who

 

was rude"; "It is nice to have friends who get along with each other."
Activities--"Jenny and I go to ballet class together"; "We do
lots of things together."

Personalit --"My friends are always really terrific"; "I'd

 

want my wife to be sensitive and perceptive."

Psychological Intimacy--"I think married people should be able

 

to talk to each other about anything";"1'd trust my friend with any
secret."

Acceptance-Equality--"I wouldn't want to be married to someone

 

who was always nagging me to be different"; "A person has to respect
you for it to work out."

Affective Involvement--"A friend is someone you really care

 

about"; "If you hate someone, there is no point staying married."

Compatibilit --"We got along well and liked the same things";

 

"Divorced people are always fighting and can never decide about

things."

77

Parenthood--"I'd want to make sure that who ever I married

 

wanted kids"; "I wouldn't stay with someone who was mean to the
children."

Sexuality--"You have to be attracted to someone to get married";
"I'd separate from someone who was running around with other men."

Independence--"If I was married, I'd still want to be able to

 

have my own life"; "People get married to get away from home."

Depth-Commitment--"A marriage means that you want to stay with

 

that person no matter what"; "It is a very special relationship
because you know that person better than anyone else does."

Financial Responsibilit --"When you get married, YOU have to

 

pay for a house and everything else"; "You need to have a job before
getting married."

Responsibility for Self--"When you get divorced, you have to

 

start taking care of yourself again"; "Being divorced is hard because

there is no one there to help you out."

Problem Solving

 

Apology--"You can buy her flowers or candy"; "Say you are
sorry.“

Dgnjglf-"Just forget about it and start over"; "Stop fussing
and ignore it."

External Aid--"You can talk to your minister or a psychiatrist";

 

"Go talk it over with someone neutral who can talk to both of you

without getting mad."

78

Discussion--"People have to talk about their problems together

 

to figure things out"; "Problems don't go away unless you work at

them."

BIBLIOGRAPHY

79

BIBLIOGRAPHY

Altman, 1., & Taylor, D. Socialgpenetration: The development of
interpersonal relationships. New York: Halt Rinehart &
Winston, 1973.

 

Asch, S. Forming impressions of personality. Journal of Abnormal
Social Psychology, 1946, 41, 258-290.

 

 

Baldwin, A. A cognitive theory of socialization. In D. Goslin
(Ed.), Handbook of socialization theory and research. Chicago:
McNally, 1969.

Baldwin, C., & Baldwin, S. Children's judgments of kindness. Child
Development, 1970, 41, 29-47.

 

Bannister, D., & Agnew, J. The child's constrewing of self. In
J. Cole (Ed.), Nebraska symposium on motivation, 1976 (Vol. 24).
Lincoln: University of Nebraska Press, 1977.

 

Berndt, T. Children's concepts of friendship and the behavior
expected of friends. Paper presented at the annual meeting of
the American Psychological Association, Toronto, 1978.

Bernstein, A., & Cowan, P. Children's concepts of how people get
babies. In E. M. Hetherington & R. Parke (Eds.), Contemporary
readings in childgpsychology. New York: McGraw-Hill, 1977.

 

 

Bieri, J. Cognitive complexity--simplicity and predictive behavior.
Journal of Abnormal and Social Psychology, 1955, 51, 263-268.

Bieri, J., Atkins, A., Briar, S., Leaman, R., Miller, H., & Tripodi, T.
Clinical and social judgment. New York: Krieger, 1975.

 

Bigelow, B. Children's friendship expectations: A cognitive develop-
mental study. Child Development, 1977, 48, 246-253.

 

Bigelow, B., & LaGaipa, J. Children's written descriptions of friend-
ship: A multidimensional analysis. Developmental Psychology,
1975, 11, 857-858.

Bigner, J. A Wernerian developmental analysis of children's descrip-
tions of siblings. Child Development, 1974, 45, 317-323.

80

81
Bowerman, C. Adjustment in marriage: Over-all and in specific
areas. Sociology and Social Research, 1957, 41, 257-263.

Brim, O. The parent-child relation as a social system: 1. Parent
and child roles. Child Development, 1957, 28, 343-364.

 

Bruner, J., & Tagiuri, R. The perception of people. In G. Lindzey
(Ed.), Handbook of Social Psychology. Boston: Addison-Wesley,
1954.

 

Campbell, J. Illness is a point of view: The development of
children's concepts of illness. Child Development, 1975, 46,
92-100.

 

Chadwick, B., Albrecht, S., & Kunz, A. Marital and family role sat-
isfaction. Journal of Marriage and the Family, 1976, 38, 431-
440.

 

Chandler, M., Paget, K., & Koch, D. The child's demystification of
psychological defense mechanisms: A structural and develop-
mental analysis. Developmental Psychology, 1978, 14, 197-205.

 

Childers, P., & Wimmer, M. The concept of death in early childhood.
Child Development, 1971, 42, 1299-1301.

 

Coie, J., & Pennington, 8. Children's perceptions of deviance and
disorder. Child Development, 1976, 41, 407-413.

 

Cooney, E. Social cognitive development: An experimental inter-
vention in the elementary grades (Doctoral dissertation, Harvard
University, 1978).

Cooney, E., & Selman, R. Children's use of social conceptions:
Toward a dynamic model of social cognition. New Directions for
Child Development, 1978, 1, 23-44.

 

Crockett, W. Cognitive complexity and impression formation. In
B. Macher (Ed.), Progress in gxperimental personality research
(Vol. 2). New York: Academic Press, 1965.

Cronbach, L. On the nonrational application of information measures
in psychology. In H. Quastler (Ed.), Information theory in
psychology. Chicago: Free Press, 1955.

 

Damon, W. The social world of the child. San Francisco: Jossey-
Bass, 1977.

Dodge, K. Social cognition and children's aggressive behavior.
Child Development, 1980, 51, 162-170.

 

82

Dubin, R., & Dubin, E. Children's social perceptions: A review of
research. Child Development, 1965, 33, 809-838.

 

Emmerich, W. Young children's discriminations of parent and child
roles. Child Development, 1959, 39, 403-419.

 

Emmerich, W. Family role concepts of children ages six to ten.
Child Development, 1961, 33, 609-624.

 

Flavell, J. The developmental psycholOQy of Jean Piaget. New York:
Van Nostrand, 1963.

 

Flavell, J. The development of role-taking and communication skills
in children. New York: Wiley, 1968.

 

Flavell, J. Concept development. In P. Mussen (Ed.), Carmichael's
manual of child psychology (3rd ed.) (Vol. 1). New York:
Wiley, 1970.

 

 

Furth, H., Baur, M., & Smith, J. Children's conceptions of social
institutions: A Piagetial framework. Human Development, 1976,
19, 351-374.

 

Gardner, R., & Schoen, R. Differentiation and abstraction in con-
cept formation. Psychological Monographs, 1962, No. 560.

 

Goldin, P. A review of children's reports of parents' behaviors.
ngchological Bulletin, 1969, 11, 222-236.

 

Gollin, E. Organizational characteristics of social judgment: A
developmental investigation. Journal of Personality, 1858,
26, 139-154.

Harter, S. A cognitive-developmental approach to children's expres-
sion of conflicting feelings and a technique to facilitate such
expression in play therapy. Journal of Consulting and Clinical
Psychology, 1977, 53, 417-432.

Hartley, R. Children's concepts of male and female roles. Merrill-
Palmer Quarterly, 1960, 9, 83-91.

 

Hayes, 0. Cognitive bases for liking and disliking among preschool
children. Child Development, 1978, 33, 906-909.

Heider, F. Social perception and phenomenal causality. Psychologi-

 

Heider, F. The psychology of interpersonal relations. New York:
Wiley, 1958.

83

Hess, R., & Torney, T. Religion, age, and sex in children's per-
;gptiogs of family authority. Child Development, 1962, 33,
-78 . __'

 

Hicks, M., & Platt,lL. Marital happiness and stability; A review
of the research in the sixties. Journal of Marriage and the
Family, 1970, 33, 553-574.

 

Hoffman, E., Mersden, G., & Kalter, N. Children's understanding of
their emotionally disturbed peers: A replication. Journal of
Clinical Psychology, 1977, 33, 949-953.

Kagan, J. The child's perception of the parent. Journal of Abnormal
and Social Psychology, 1956, 33, 257-258.

 

Kagan, T., & Lemkin, T. The child's differential perception of
parental attributes. Journal of Abnormal and Social Psychology,
1960, 31, 440-447.

Kahn, D., Nash, S., & Brucken, L. Sex role concept of two- and
three-year-olds. Child Development, 1978, 52, 445-451.

Kalter, N., & Marsden, G. Children's understanding of their emo-
tionally disturbed peers II. Etiological factors. Psychiatry,
1977, 39, 48-54.

Kelley, G. The psychology of personal constructs. New York: Norton,
1955.

Kohlberg, L. Stage and sequence: The cognitive developmental approach
to socialization. In D. Goslin (Ed.), Handbook of socialization
theory and research. Chicago: McNally, 1969.

Kohn, A., & Fiedler, F. Age and sex differences in the perception
of persons. Sociometry, 1961, 34, 257-264.

 

Kuhn, D. Mechanisms of cognitive and social development: One psy-
chology or two? Human Development, 1978, 31, 92-118.

Leahy, R. Children's judgments of excuses. Paper presented at the
Southeastern Conference on Human Development, Atlanta, Georgia,
1978.

Lewis, M., & Brooks-Gunn, J. Toward a theory of social cognition:
The development of self. New Directions for Child Development,
1979, 3, 1-19.

Livesley, W., & Bromley, 0. Person perception in childhood and
adolescence. London: Wiley, 1973.

 

 

84
Loche, H., & Williamson, R. Marital adjustment: A factor analysis
study. American Sociological Review, 1958, 33, 562-569.

Maas, E., Marecek, J., & Travers, J. Children's conceptions of
disordered behavior. Child Development, 1978, 33, 146-154.

 

Marascuilo, L., & McSweeney, M. Nonparametric and distribution free
methods for the social sciences. San Francisco: Brooks/
Cole, 977.

 

Marsden, G., & Kalter, N. Children's understanding of their emo-
tionally disturbed peers I. The concept of emotional dis-
turbance. Psyghiatry, 1976, 33, 227-238.

 

Marvin, R., & Greenberg, M. Preschooler's changing conceptions of
their mothers: A social-cognitive study of mother-child
attachment. Paper presented at the Southeastern Conference on
Human Development, Atlanta, Georgia, 1978.

Moessinger, P. Developmental study of the idea of lining up.
Journal of Psychology, 1977, 33, 173-178.

 

Mohr, 0. Development of attributes of personal identity. Develop-
mental Psychology1 1978, 15, 427-428.

 

Morison, P., & Gardner, H. Dragons and dinosaurs: The child's
capacity to differentiate fantasy from reality. Child Develop-
ment, 1978, 53, 642-648.

Neuhauser, C., Amsterdam, B., Hines, P., & Steward, M. Children's
concepts of healing: Cognitive development and locus of con-
trol. American Journal of Orthopsychiatry, 1978, 43, 335-341.

Novak, 0. Children's responses to imaginary peers labelled as
emotionally disturbed. Psychology in the Schools, 1975, 13,
103-106.

Nye, F., & Berardo, F. The family: Its structure and interaction.
New York: Macmillan, 1973.

Orden, S., & Bradburn, N. Dimensions of marriage happiness. Ameri-
can Journal of Sociology, 1968, 13, 715-731.

Peevers, B., & Secord, P. Developmental changes in attribution of
descriptive concepts to persons. Journal of Personality and
Social Psycholpgy, 1973, 31, 120-128.

Piaget, J. Piaget's theory. In P. Mussen (Ed.), Carmichael's Manual
of Child Psychology (Vol. 1). New York: Wiley, 1970.

85

Rushton, J., & Wiener, J. Altruism and cognitive development in
children. British Journal of Social and Clinical Psychology,
1975, 13, 341-349.

Scanzoni, J. A social system analysis of dissolved and existing
marriages. Journal of Marriaggand the Family, 1968, 39,
452-461.

Scarlett, H., Press, A., & Crockett, W. Children's descriptions
of peers: A Wernerian developmental analysis. Child Develop-
ment, 1971, 53, 439-453.

Schaefer, E. Children's reports of parental behavior: An inventory.
Child Development, 1965a, 33, 431-424.

 

Schaefer, E. A configurational analysis of children's reports of
parental behavior. Journal of Consulting Psycholpgy, 1965b,
29, 552-557.

Selman, R. Taking another's perspective: Role taking development in
early childhood. Child Development, 1971a, 53, 1721-1734.

Selman, R. The relation of role taking to the development of moral
judgment in children. Child Development, 1971, 53, 79-91.

 

Selman, R. Toward a structural analysis of developing interpersonal
relations concepts: Research with normal and disturbed pre-
adolescent boys. In A. Pick (Ed.), Annual Minnesota symposium
on child development X. St. Paul: University of Minnesota
Press,*l976a.

 

Selman, R. Social-cognitive understanding. In T. Lickona (Ed.),
Moral development and behavior. New York: Holt, Rinehart &
Winston, 1976b.

 

Selman, R. A structural-developmental model of social cognition:
Implication for intervention research. Counseling Psychologist,
1977, g, 3'6. .

Selman, R., & Byrne, D. A structural-developmental analysis of
levels of role taking in middle childhood. Child Development,
1974, 53, 803-806.

Selman, R., Jaquette, D., & Lavin, D. Interpersonal awareness in
children: Toward an integration of developmental and clinical
child psychology. American Journal of Orthopsychiatry, 1977,
47, 264-274.

Shantz, C. The development of social cognition. In E. M. Hethering-
ton (Ed.), Review of child development theory and research
(Vol. 5). Chicago: University of Chicago Press, 1975.

86

Shapiro, T., Sherman, M., 8 Osowsky, 1. Preschool children's con-
cepts of ghosts. Journal of Child Psychiatry, 1980, 13,
41-55.

Siegelman, M. Evaluation of Bronfenbrenner's questionnaire for
children concerning parental behavior. Child Development, 1965,
36, 164-174.

Siegel, S. Nonparametric statistics for the behavioral sciences.
New York: McGraw Hill, 1956.

Siegel, I. The attaniment of concepts. In M. Hoffman 8 L. Hoffman
(Eds.), Review of child development research. New York:
Russel Sage, 1964.

 

Signell, K. Cognitive complexity in person perception and nation
perception: A developmental approach. Journal of Personality,
1966, 33, 517-537.

Sowers, A. Parent-child relationships from the child's point of
view. Journal of Experimental Education, 1937, 3, 305-331.

Spiegel, L. The child's concept of beauty. A study in concept
formation. Journal of Genetic Psychology, 1950, 11, 11-23.

Stuckert, R. Role perception and marital satisfaction--A configura-
tional approach. Journal of Marriage and the Family, 1963, 33,
415-419.

Turiel, E. The development of social concepts: Mores, customs, and
conventions. In D. DePalma 8 J. Foley (Eds.), Moral Develop-
ment: Current Theory and Research. New York: Erlbaum,

1975.

Vacc, N., 8 Greenleaf, W. Sequential development of cognitive com—
plexity. Perceptual and Motor Skills, 1975, 51, 319-322.

Wallach, M. Research on children's thinking. In National Society
for the Study of Education, 62nd Yearbook, Child Psychology.
Chicago: University of Chicago Press, 1963.

Watts, A. The laoguage and mental development of children. London:
Harrap, 1944.

Werner, H. Comparative psyohology of mental development. New York:
Science Editions, 1948.

Whiteman, M. Children's conceptions of psychological causality.
Child Development, 1967, 33, 143-155.

 

 

87

Vygotsky, L. Thought and Language. Cambridge, Mass.: MIT Press,
962.

Yarrow, M., 8 Campbell, J. Person perception in children. Merrill-
PalmeryOuarterly, 1963, 3, 57-72.

 

Youniss, J., 8 Volpe, J. A relational analysis of children's friend-
ship. New Directions for Child Development, 1978, 1, 1-22.
Youniss, J. Another perspective on social cognition. In A. Pick
(Ed.), Minnesota symposium on childpsychology (Vol. 9).
St. Paul: University of Minnesota Press, 1975.