‘ . Q I D ‘ I. .
lull" \§U¢" Ila v H
p:“‘h|l.].l¥.. ..l
I 03.1 tVIlHMV‘o Irlv In n09? LII.“ - . .
I ‘tlufi‘t ‘ nu
IIILI 1" "hhnn v.0..‘w- ..
_ l\l‘ll§ ll- ‘gl‘ I!
.llh'v'. ! C
II.-. 1“].1‘:|
..-.}Or‘

.

. ‘ .. .LAA
p . I'.|
' I 501% . A.-
‘ I y 5'.
. bar»! yr'l.~ll‘\.

'Il': . ‘1': \ I.‘ P: '6‘..l“l.wl.. . unn-
.ll‘.l|r.\lt\l|\‘l-~t|. II a I... 0.:‘l "
31...}...3c :rnpuflfttru 31). to.-.“ . 1.1!. 1

..‘I.I.r '.|\1l..‘v' 24.- II ‘1 no. III! 10.
I. 1.031. I l'-A\ll.>n v0: .7. 'u.’,£‘

.O.|Il El. 1... RO‘RI; I. '. S10 algal.

o- u ' l.‘ 5' 0
l O ’ntA, 1
III rsv! ‘ q

I z.~
.‘l-t“)-

c. 1.0:...92
. A
.‘l

U I
.V‘Ou

t .
>VJ. 'w\v0| A

I l < v. .
J". .y..l “Jpn.
‘J .5. .4 ha

‘ A. II I )u'il 9’. to II: IIIIII‘OO.
9'; III. Lt lo ‘ \'I~I h if . v ollt '1'!-
MI; ntdfi'tVI" l" :1 I. I? .u b .
"n \ . 16 .51 111‘,‘ 1
[VI llll‘ '1‘ .IIA' I léhl‘vl It .
ya . .I Y 1» .1‘alll2‘o“.
' A..." I ll ( "h ..)‘||..i'-Il|lnll‘..tv..
.I all 04\ -‘l \ ..

‘(qu‘b‘ JUI ‘l‘l

u ilultll {II-351

. . 1|! 1 a II It
.I. I .

 

 

.1" 1. --o--- 9,1... ‘- —--

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled
FAMILY PROCESS , PARENT-CHILD
SUBSYSTEMS AND SELF-CONCEPT
IN STEPCHILDREN

presented by

Peggy A. Spiegel

has been accepted towards fulfillment
of the requirements for

Ph D. degreeinPSYChOlOgY

 

 

Major professo

y/géé
// /

“Elli-n.- Alli—‘1 A - 11- ll‘

1 ' ' ' ' 042771

 

MSU

LIBRARIES
.—:—.

 

 

 

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

 

 

-9? O ‘9?"“F‘r-
e ”3‘ fJCJq

., . I,“ "I I -
- N‘ I} .1. i 7"

w"""=
‘93 a; .. - :1

 

 

FAMILY PROCESS,
PARENT-CHILD SUBSYSTEMS
AND SELF-CONCEPT

IN STEPCHILDREN

BY

Peggy A. Spiegel

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1986

ABSTRACT

FAMILY PROCESS,
PARENT-CHILD SUBSYSTEMS
AND SELF-CONCEPT
IN STEPCHILDREN

BY
Peggy A. Spiegel

Stepfamilies have become more prevalent in recent
years, and the relationship between stepfamily living and
child adjustment is important to understand. Previous
research has attempted to demonstrate that self-concept in
children is related to family structure. This work contends
that children in nuclear families will have the highest
self-esteem; other family structures (e.g., blended
families) will result in psychological deficits for the
children. This study hypothesized that it is not the
structure of the family but perceptions of family
functioning that will predict a child's self—esteem.
Accordingly, it was predicted that there would be no
difference in self-concept between nuclear family and
stepfamily children, but, in both nuclear families and
reconstituted families, self-concept would be correlated
with family system and subsystem functioning.

Thirty stepchildren and 30 nuclear family children
between 6 and 12 years old completed the Family Adaptability

and Cohesion Evaluation Scales (FACES-II), the Parent

Perception Inventory (PPI) and the Piers-Harris Children's
Self-Concept Scale. The FACES-II, measuring system
functioning, yields Adaptability and Cohesion scores. The
PPI, the measure of parent-child subsystem functioning,
yields a Positive and Negative score for each parent. As
predicted, there was no difference in self-concept between
nuclear family and stepfamily children. For the
stepchildren, the Piers-Harris was significantly correlated
with Cohesion in Family with Mother, Adaptability in Family
with Mother, Mother Positive and Mother Negative, and was
marginally correlated with Stepmother Positive and
Stepfather Negative. For the nuclear family children, the
Piers-Harris only correlated with the subsystem
variables--Mother Positive, Mother Negative, Father Positive
and Father Negative. It was also found that remarried
families have a greater tendency toward distress than
nuclear families.

It was concluded that family process variables are
related to self-esteem but the pattern of these systemic and
subsystemic processes differs in stepfamilies and nuclear
families. In spite of this, there was no associated
difference in self-concept between stepchildren and nuclear
family children. Methodological issues and directions for

future research were also discussed.

ACKNOWLEDGMENTS

Gary Stollak, the chairman of my dissertation
committee, has been helpful and encouraging throughout this
study. He was particularly helpful with the initial
conceptualization of the project and with the trials of
getting subjects. I would also like to thank the other
members of my committee. Larry Messé’helped enormously with.
the data analysis. I appreciated Ellen Strommen's
encouragement and interest in the elementary age group.

Anne Soderman assisted in gaining access to the Waverly
School District, and her thoughtful consideration of the
stepfamily literature was appreciated.

I also want to thank Ted Sample of Holt School District
and Dr. Warren Starr of Waverly School District. I'd like
to extend my appreciation to the principals and teachers at
the Haslett, Holt and Waverly elementary and middle schools,
and to John Borgert, the principal of Mason Middle School.
The study could not have been completed without their
willingness to allow letters to go home with the children in
these four school districts. Also, the members of the
Lansing Chapter of the Stepfamily Association of America
were kind enough to allow me to present the study to their

group.

iv

Kim Draeger, Donna Guido, Demarie Gundle, Judy
Fleissner, Michael Henk and Wendy Steinhacker, my research
assistants, collected the data, scored it and put in on the
computer. They worked hard, and their interest in the
project was gratifying.

Carl Chenkin, Louise Finlayson, Nick Ialongo, Wendy
Sabbath, Cheryl Smith-Winberry and Ann Wagner have been
friends and colleagues throughout. They gave practical help
and emotional support, and I am most appreciative. As
always, Nick was very generous in helping with the computer
work, and Cheryl's statistical assistance was crucial.

Finally, I am most grateful to the sixty children and
families who gave generously of their time to participate in

this study. I have learned much from them.

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . .

LIST OF FIGURES O O O O O O C O O O 0

INTRODUCTION . . . . . . . . . . .
Previous Studies on the Self-Concept of
Stepchildren . . . . . . .
Evaluation of the Literature on Self- -Concept
of Stepchildren . . . . . . .
Family Process Variables and Child
Adjustment . . . . . . . . . .

Evaluation of Literature on Family Process
and Child Adjustment . . . . . . .

The Present Study . . . . . . . . .

Hypotheses . . . . . . . . . . .

METHOD 0 O O O O O O O O O O O O 0

Subject Recruitment. . . . . . . . .
Instruments . . . . . . . . . . .
Procedure 0 O O O O O O O O O O 0

RESULTS. 0 O O O O O O O O O O O 0

Description of Families . . . . . . .
Description of Stepfamilies . . . . . .
FACES-II: Family Types and Distance from

Center Scores . . . . . . . . .
Means and Standard Deviations for Cohesion

and Adaptability. . . . . . . . .
Family Types: Balanced, Mid-Range or Extreme
Distance from Center (DFC) Scores . . . .
Multivariate Analyses . . . . . . . .
Results of Tests of Hypotheses . . . .
Further Analysis Utilizing a FACES- II

Linear Model . . . . . . . . . .
Multiple Regression Analyses. . . . . .

vi

Page
viii

ix

12
15

20
21
22

24

24
26
31

32

32
33

36

37
38
40
41
42

45
45

DISCUSSION

Methodolog
Considerat
Conclusion

ical Issues . . . . . . . .
ions for Future Research. . . .

LIST OF REFERENCES 0 C O I O O O I O 0

APPENDIX A:
Public Sch

APPENDIX B:

Letter to Parents and Stepparents:
0018 O O O O O O O O O 0

Letter to Parents and Stepparents:

Stepfamily Association of America . . . .

APPENDIX C:
APPENDIX D:

APPENDIX E:
Scale .

APPENDIX E:

APPENDIX C:

FACES“II o o o o o o o o a
Parent Perception Inventory . .

Piers-Harris Children's Self-Concept

Marriage and Family Questionnaire.

Consent Form. . . . . . . .

vii

50
59
60
63

64

69

71
72

76

79
82
86

LIST OF TABLES

Table 1. Distribution of Subjects by

Recruitment Locale and Type of Family . . . .
Table 2. Distribution of Subjects by Age,

Sex and Type of Child . . . . . . . . .
Table 3. Means and Standard Deviations for

Cohesion and Adaptability . . . . . . . .
Table 4. Frequency and Percentage of Balanced,

Mid-Range and Extreme Families for

Stepchildren's Family with Mother, Stepchildren's

Family with Father and Nuclear Family Children .
Table 5. Pearson Correlations of FACES-II

and PPI scores with Piers-Harris Scores . . .
Table 6. Multiple Regression of Family

Variables on Piers-Harris Scores,

Stepchildren . . . . . . . . . . . .
Table 7. Multiple Regression of Family

Variables on Piers-Harris Scores, Stepchildren

with Four Parents. . . . . . . . . . .
Table 8. Multiple Regression of Family

Variables on Piers-Harris Scores, Stepchildren

with Three Parents . . . . . . . . . .
Table 9. Multiple Regression of Family

Variables on Piers-Harris Scores, Nuclear

Family Children. . . . . . . . . . . .

viii

Page

25
26

37

39

44

47

47

48

49

LIST OF FIGURES

Page
Figure l. The Circumplex Model. . . . . . . . 19

ix

I NTRODUCTION

Stepfamilies, also referred to in the literature as
blended, reconstituted and remarried families, are becoming
increasingly prevalent. In 1978, 10.2% of all children
under 18 lived with one biological parent and one
stepparent. This represents about seven million children.
The proportion of two-parent children living with a
stepparent and a natural parent is predicted to reach 15% by
1990 (Click, 1979). Thus, as more and more children will be
living in stepfamilies for at least part of their lives, the
effects of stepfamily living on children's growth are
important to understand. In response to this need, the
question of whether membership in a reconstituted family
relates to poorer psychological adjustment for children has
been studied.

Research has been dominated by the deficit comparison
model, which assumes that stepchildren will show
psychological deficits in comparison to children from
nuclear families. However, at least with regard to the
variable of self-concept, research generally has not
supported this contention (Ganong & Coleman, 1984).
Consistent with this literature, the present study addresses

the hypothesis that it is not the type of family structure

(i.e., nuclear or blended) but the perceived quality of
family system and subsystem functioning which is related to
the child's self—concept.

The balance of this introduction reviews and critiques
the literatures on self-concept in stepchildren and on the
relationship between family process variables and child
adjustment. The rationale and hypotheses of the present

investigation are then presented.

Previous Studies on the Self-Concept of Stepchildren

 

Parish and Taylor (1979) studied the self-concept of
406 grade school and junior high school students from
nuclear (N = 347), divorced (N = 44) and stepfather (N = 15)
families. The children were administered the Personal
Attribute Inventory for Children (PAIC). The results
indicated that there was no difference in self-concept
between children from intactfll and stepfather families.
However, children from divorced families had lower
self-concept than those from intact families.

Children (N = 738) in grades five through eight, again
from intact, divorced or blended families, were investigated
by Parish and Dostal (1980). The subjects completed the
PAIC three times, indicating which attributes they
considered descriptive of themselves, their mothers and
their fathers. For self-concept, the results were similar

to those of the previous study. The self-concept of

 

GlThe terms "nuclear" and "intact" will be used synonymously.

children from intact families was not significantly
different from the self-concept of stepchildren. However,
intact family children did have significantly higher
self-concept than single-parent family children. The
single-parent family children and the stepchildren did not
significantly differ in self-concept. In addition, there
were no significant sex differences found.

Results also indicated that intact family children
evaluated their mothers significantly more favorably than
those from either blended families or divorced families.

The latter two groups did not significantly differ.

Further, girls and boys did not rate their mothers
significantly differently. For fathers, there were
significant differences among the family types. Evaluations
of fathers from intact families were the most positive,
followed by those of fathers from blended families, with
those of fathers from divorced families being the least
favorable.

Sex of subject did significantly mediate the
relationship between ratings of fathers and type of family.
Divorced boys rated their fathers significantly more
negatively than any other group. The father ratings of boys
and girls from intact families varied significantly from the
ratings by girls from divorced and remarried families. None
of the other paired comparisons reached statistical

significance.

Again utilizing the PAIC, Parish and Nunn (1981)
examined the relationship between 132 fifth through eighth
grade children's self—concept and their evaluations of their
single parent or stepfather families as a function of four
conditions: father loss through divorce, father loss
through death, "happy" family, and "unhappy" family. The
results were mixed. In the "death of father" condition,
self—concept was not significantly correlated with child's
evaluation of mother, father or stepfather whereas with
divorce, self-concept was not significantly correlated with.
the child's evaluation of the parents. Self-concepts of
children from "happy" families were not found to be
significantly correlated with these children's evaluations
of their parents, but self-concepts in "unhappy" families
were related to their perceptions of their parents.

The authors had hypothesized that the correlations
between children's self-concepts and their evaluations of
their parents would be greatest in those circumstances
(unhappy families and divorced families) where their needs
are thought to be threatened and, as a result, where their
growth in autonomy is thought to be most likely impaired.
They concluded that their hypothesis was supported.
However, this conclusion seems unwarranted by the data. The
description of their subjects is very unclear. Some of the
"divorced" children had stepfathers, and the divorced
children and stepchildren were not distinguished in their

statistical analyses. Moreover, death of the father is a

disruptive and threatening event for children; this issue is
not addressed by Parish and Nunn (1981). On the whole, this
study was poorly conceptualized and executed. It is
noteworthy, however, in that it does consider how the level
of child-perceived family happiness may be related to the
child's self-concept.

Another study (Parish & Parish, 1983) of fifth through
eighth grade students compared children from intact families
(N = 397), divorced families (N = 19), and divorced
remarried families with a stepfather (N = 55). The children
completed the PAIC twice, once describing themselves and i
once describing their families. It was found that
children's self-concepts were correlated with evaluations of
their families in both nuclear and remarried families; there
did not prove to be a significant correlation for
single-parent families.

Santrock, Warshak, Lindbergh and Meadows (1982)
compared children's and parents' observed social behavior in
nuclear, divorced and stepfather families. Twelve families
from each of these three family types were studied. Half of
the 36 children were boys and half were girls, all between 6
and 11 years old. Families were matched on age at onset of
divorce, and children had been living in the stepfather
family for a minimum of 18 months. The data consisted of
videotaped observations of parent-child interaction; the
parent's behavior and the child's behavior were coded

separately. Parent behaviors included the following:

control, encourages emotional independence, engages in
intellectually meaningful verbal interaction, attentive to
child, authoritarian, permissive, authoritative and
maturity. Child behaviors included the following: warmth,
self-esteem, anxiety, anger, demandingness, maturity,
sociability and independence.

The most consistent findings suggested that boys in
stepfather families showed more competent social behavior
than boys in intact families, a pattern which corresponded
with the more competent parenting behavior that was also
displayed in those stepfather families. Specifically, male
stepchildren demonstrated higher self-esteem, more warmth,
less anxiety and less anger than intact family boys. Girls
in stepfather families showed more anxiety than girls in
intact families. Boys showed more warmth toward their
stepfathers than did girls. Few differences were found
between divorced and stepfather children and divorced and
nuclear family children and parents. Mothers of boys in
stepfather families made more meaningful statements to their
children than the divorced mothers of boys. The authors
conclude that the social behavior of children is not
necessarily less competent in stepfather families than in
intact or divorced families. Parenting behavior, sex of
child and marital conflict in any type of family structure
are implicated as more likely explanations of the child's

social behavior (Santrock et al., 1982).

The variable of self—esteem in particular is most
relevant to the aims of the present study. There were no
differences across family types and sex in self-esteem
except that male stepchildren had higher self-esteem than
boys from intact families (Santrock et al., 1982). This
study, then, also does not support the contention that
stepchildren show deficits in self-esteem.

Raschke and Raschke (1979) studied third, sixth and
eighth graders regarding self-concept, family structure and

family conflict. The children were from intact (N = 173),

single-parent (N 32), and reconstituted (stepfather = 29,
stepmother = 3) families. They completed the Piers-Harris
Children's Self-Concept Scale and a questionnaire measuring
three types of family conflict (fighting in the family,
adults fighting and perceived happiness of adults). The
results indicated that there were no significant
relationships between self-concept and type of family
structure. However, in both nuclear and single-parent
families, self-concept scores were signficantly lower for
children who reported higher levels of family conflict.
There were too few subjects, however, to determine whether
family conflict and self-concept were related in
reconstituted families. It is important to note that the
measure of family conflict was comprised of three questions,
one each about fighting in the family, parental fighting and

parental happiness. With so few items, the reliability and

validity of the measure is questionable.

Data from two national surveys were re-analyzed to
compare those from nuclear families and stepfather families
on social and social-psychological characteristics (Wilson,
Zurcher, McAdams, & Curtis, 1975). These characteristics
were divided into eight substantive categories:
demographic, religious, stratification, political, crime and
delinquency, interpersonal relationships (general),
interpersonal relationships (marriage and family), and
personal evaluation.

The first set of data was comprised of 68 questionnaire
items from the 1973 General Social Survey conducted by the
National Opinion Research Center (NORC) of the University of
Chicago. The subjects were adults; 1174 had lived in intact
families and 43 in stepfather families. Of these 68
variables, no statistical differences were found on 58 of
them. In particular, no differences between intact family
children and stepchildren were found for four personal
evaluations: estimate of general happiness, general
assessment of life, assessment of general health and
satisfaction with health. There were, however, some
differences between the groups: stepchildren were older,
were more often reared as Protestant and currently claimed
to be Protestant, less often thought that pornography led to
moral breakdown, had stepfathers whose level of educational
attainment was higher, themselves had lower educational
attainment, had lower total family income, were less

inclined to feel that most people were helpful and that most

people were fair, and were less satisified with their
conjugal family life.

The 1973 Youth in Transition survey conducted by the
Institute for Social Research of the University of Michigan
furnished data on tenth graders, 1573 intact family
adolescents and 79 teenagers from stepfather families. Only
one of the 39 variable revealed significant differences
between the groups. As in the first sample, nuclear family
children had more formal education than stepchildren. This
result disappeared, however, when both mental ability and
parental socioeconomic status were controlled for
statistically. No difference in self-esteem between the
groups of adolescents was found (Wilson et al., 1975).

In another study (Parish, 1981), 349 college students
evaluated themselves, their mothers, their fathers, and
their families on the Personal Attribute Inventory (PAI).
Five types of families were represented: nuclear; divorced,
not remarried; divorced, remarried; death, not remarried;
and death, remarried. The father was always the deceased
parent, and the fathers did not have custody in the divorced
families. Self-concept did vary as a function of family
happiness. It was lower for subjects who had rated their
families as unhappy versus those who had rated their
families as happy. Ratings of mothers did not vary as a
function of type of family or family happiness. Evaluations
of fathers did differ by family type. Students from

divorced nonremarried families and divorced remarried

10

families rated their biological fathers significantly more
negatively than students from intact families and
nonremarried families and remarried families where the
father had died. There was no significant effect as a
function of the happiness-unhappiness rating of the family.
There was an interaction effect (type of family x happiness
of family). Biological fathers were generally rated most
favorably by students from intact happy families, remarried
happy families (father died) and nonremarried unhappy
families (father died).

To this point, most of the studies reviewed have found
no systematic differences in self-concept between intact
family and stepfamily children, with a few finding mixed
results. Two investigations have found lower self-image in
stepchildren (Kaplan & Pokorny, 1971; Rosenberg, 1965).
Rosenberg (1965) studied 5,024 eleventh and twelfth graders,
of which 262 were stepchildren from both stepmother and
stepfather families. As measured by the Rosenberg
Self-Esteem Scale, stepchildren had lower self-esteem than
divorced children whose parents had not remarried, and even
lower self-esteem than children from intact families. In
the case of children who had lost a parent through death,
their self-esteem was lower than nuclear family children if
the surviving parent remarried, whereas self-esteem for
children whose parents did not remarry was equal to that of
children from intact families. These results must be viewed

cautiously, as no statistical analyses were reported.

11

Rather, the percent of subjects having low, medium and high
self-esteem in each family type were compared without formal
statistical testing of significant differences.

Kaplan and Pokorny (1971) conducted a careful analysis
of 500 adults, 54 of whom had lived in stepmother or
stepfather families as children. These subjects were
interviewed and completed the Rosenberg Self—Esteem Scale.
The data were analyzed to determine the particular
characteristics (living arrangements, death or divorce of
parents(s), age at divorce or death, presence or absence of
parental remarriage, age at parental remarriage) which were
associated with high self-derogation (low self-esteem)
relative to subjects from intact families. Results were
reported for the total sample and for subgroups
differentiatied by the race, sex and social class of the
subject.

The results indicated that, for the whole population,
there were no differences in self-derogation dependent upon
intact or broken home status. When the data were further
analyzed, the majority of the 190 chi square tests were not
significant. However, when there were differences, high
self-derogation was a function of a complex interaction of
broken home and subject characteristics. Results indicated
that if 1) the father died, 2) the home was disrupted by
socially stigmatizing events such as illegitimacy or mental
hospitalization of a parent, 3) rejecting parental attitudes

were present, 4) the subject was above the age of 12 at the

12

time of the broken home, or 5) the subject was eight years
of age or older at the time the parent remarried, then
subjects from broken homes were significantly more likely to
display low self—esteem when compared with subjects from
intact families (Kaplan & Pokorny, 1971).

Parish and Philip (1982) conducted an intervention
study with 376 third through eighth graders. There were 326
children from intact families, 38 from divorced families and
12 from divorced remarried families. In a single inservice,
these students‘ teachers were each given information
regarding Maslow's Hierarchy of Needs . The teachers were
asked to assess and provide for those needs "to the very
best of their ability" (Parish & Philip, 1982, p. 61). One
week prior and one month post training the students were
adminstered the PAIC. It was found that the self-concept
improved significantly for intact family children but did
not change for divorced or divorced remaried subjects.
However, no comparison of the self-concepts of the children
from each type of family was reported. Furthermore,
teacher's expectations about children from different family
structures were ignored as a confounding variable (Ganong &

Coleman, 1984).

Evaluation of the Literature on Self-Concept of Stepchildren
This small literature is difficult to evaluate as a

whole due to differing ages of subjects and different

methods of measuring self-image. Subjects from grade school

age through adulthood have been studied. To measure

13

self—concept, the Parish studies utilized the PAI or the
PAIC, Rosenberg (1965) and Kaplan and Pokorny (1971) used
the Rosenberg Self-Esteem Scale, Raschke and Raschke (1979)
utilized the Piers-Harris Children's Self-Concept Scale,
Santrock et al. (1982) used a behavioral assessment and
Wilson et al. (1975) utilized a series of questions of their
own devising. Further, most of the research reporting no
relationship between stepchild status and self-image
utilized non-probability samples of school children, with
the exception of Wilson et a1. (1975). In the series of
Parish studies, either the number of stepchildren in the
entire sample was small (Parish & Philip, 1982; Parish &
Taylor, 1979) or not reported (Parish, 1981; Parish & Nunn,
1981).

Age may be one variable which explains the studies
which did find lower self-concept in stepchildren. Ganong
and Coleman (1984) argue that investigations which sampled
adolescents (Rosenberg, 1965) or adults (Kaplan & Pokorny,
1971) found differences in self-concept whereas the studies
that sampled grade school and middle school aged children
generally found no differences. However, the Wilson et al.
(1975) study sampled adolescents and adults and found no
differences. Additionally, Ganong and Coleman (1984) point
out that the two earliest studies found lower self-concept
in stepchildren while more recent studies found no
differences between the self-image of stepchildren and

nuclear family children. Perhaps this is a cohort effect;

14

it may reflect a growing acceptance of stepchildren in our
culture.

As a whole, the literature does not demonstrate that
the mere fact of residence in a remarried family is linked
with self-concept deficits. ’Stepfamily researchers may need
to rethink their emphasis on the deficit comparison model,
which assumes that the family type (nuclear or remarried
family) is crucial (Ganong & Coleman, 1984). Similarly,
others have commented that "it is more fruitful to consider
the social interactional dynamics that lead to a given
outcome than to focus on the form of the family as the
critical independent variable" (Marotz-Baden, Adams, Bueche,
Munro, & Munro, 1979, p. 5). In fact, two of the studies
reviewed above addressed family process variables, finding
that it was not the family structure which predicted
self-concept; rather, it was the amount of child-perceived
family harmony. Family conflict predicted low self-concept
in one study, but only for nuclear and divorced families
(Raschke & Raschke, 1979); family "unhappiness" was related
to low self-concept in the other study (Parish, 1981).

Since family process is hypothesized to relate to
self-concept, the literature relevant to this issue will be
reviewed next. Unfortunately, the literature tends to deal
with psychological adjustment broadly defined, rather than
self-concept in particular; thus, both the material specific
to self-concept as well as the work considering adjustment

in general are discusssed.

15

Family Process Variables and Child Adjustment

 

One body of literature which addresses the relationship
between family process variables and child adjustment is
largely theoretical. Several major models of normal family
functioning have been articulated by family therapists. In
these theories, normal family processes result in healthy
adjustment of family members, especially children. These
models will be reviewed below.

The structural model (Aponte & Van Deusen, 1981;
Minuchin, 1974; Minuchin, Montalvo, Guerney, Rosman &
Schumer, 1967; Walsh, 1982) centers on the concepts of
subsystems and boundaries. The family differentiates and
carries out its functions through subsystems, formed by
generation, sex, interest or function. The boundaries of
the subsystem--the rules defining who participates and
how--function to protect the differentiation of the family
system. In healthy families, these boundaries are clear and
firm. Furthermore, the importance of an appropriate
hierarchy is stressed. The spouse subsystem must protect
its boundary, and the parental subsystem must carry out
child rearing functions in a coordinated manner. In
addition, the child learns appropriate social behavior
through interaction with each parent and by observing the
parents' interaction.

This model also posits that optimal functioning is
effected through the flexibility of the system. The system

must facilitate autonomy and interdependence of its members,

16

individual growth and system maintenance, and continuity and
change. Change occurs in response to developmental changes
within the family as well as environmental demands.

The strategic therapists (Bodin, 1981; Stanton, 1981;
Walsh, 1982) view normal families as extremely flexible,
using a large repertoire of behaviors to cope with
inevitable problems. Haley (1980) in particular stresses
that well-functioning families are able to adjust to life
cycle transitions (e.g., children's entry into school or
adolescence). Additionally, normal families are organized
hierarchically. Clear rules govern power and status
differentials among the generations. Parents exercise
appropriate controls and guidance with regard to their
children.

Satir (1964) describes four components of a healthy
family. The sense of self-worth in family members is high.
Communication is direct, clear, specific and honest. Family
system rules for members are flexible, human, appropriate
and subject to alteration. In addition, the family linkage
to society is open and hopeful. Lidz (1976) argues that for
healthy development a child requires two competent parents:
a same-sex parent to identify with and model after; and an
opposite-sex parent as a love-object, whose love provides
the child with a sense of self-worth. Thus, healthy
parent-child subsystems are important for the child's

optimal development.

17

More recently, empirically based family models have
been developed, including the Circumplex Model of Marital
and Family Systems developed by David Olson and his
colleagues (Olson, Russell, & Sprenkle, 1980; Olson,
Russell, & Sprenkle, 1983; Olson, Sprenkle, & Russell,
1979), which will be utilized in this study. The Circumplex
Model encompasses three dimensions of family behavior, which
are cohesion, adaptability and communication. Reviewing
concepts from the family therapy literature described above,
as well as from other social science fields such as
sociology, they concluded that many theorists used differing.
terms to describe the same concepts. Olson and his
colleagues chose the terms "cohesion" and "adaptability" to
describe essential processes that characterize all families.
Cohesion is defined as the "emotional bonding that family
members have toward one another." Adaptability is defined
as ”the ability of a marital or family system to change its
power structure, role relationships, and relationship rules
in response to situational and developmental stress" (Olson,
Russell, 5 Sprenkle, 1983, p. 70). The third dimension,
communication, encompasses such positive skills as empathy,
reflective listening and supportive comments, as well as
negative communication patterns such as criticism, double
messages and double binds.

There are four levels of cohesion: disengaged (very
low), separated (low to moderate), connected (moderate to

high) and enmeshed (very high). Moderate levels of cohesion

18

(separated and connected) are associated with optimal family
functioning, whereas the two extremes (disengaged and
enmeshed) are problematic. The ideal level of cohesion is
closeness among family members accompanied by a sense that
each individual is capable of functioning independently of
the family. As with the cohesion dimension, there are four
levels of adaptability: rigid (very low), structured (low
to moderate), flexible (moderate to high) and chaotic (very
high). The two moderate levels of adaptability (structured
and flexible) reflect the healthiest family functioning,
whereas poorly functioning families are characterized by the
extremes of adaptability (rigid and chaotic). The ideal
level of adaptability is a balance between stability and
change. Well-functioning families need to be stable as well
as to grow and change at appropriate times throught the
family life cycle.

If the four levels of each of these two dimensions are
combined with each other, 16 different types of marital and
family systems result. Figure l (p. 19) presents this
Circumplex Model (Olson et al., 1982, p. 7). The
communication dimension is not depicted in this figure
because it is considered a facilitating dimension. That is,
it facilitates couples' and families' movement along the
two, more basic, dimensions. Note that the 16 types of
families fall into three basic groups. One group falls at
the two central levels on both dimensions (four balanced

types), the second group is extreme on both dimensions (four

T N...

High
I

FLEXIBLE

§-<-+—r-—w>-w>o>

 

<

Figure 1.

smucnmo ‘1 smucrumuv

19

<———Low—c0HESIow

DISENGAGED SEPWIED
\\\\\““"
CHAOIICALLY CHADIICALLY

DISENGAGED EP

, sternum
,- onseucuoto

////
////

SEPARATE!)

STRUCTURAL“

GAGED SEPAIIMED

\ENGAGEO \\\\“ »
III. mm

The Circumplex Model.

 

    

High—9

 

CONNECTED mutsnso
“\\ V
monu\ CHAOTICA
comma enmesuso
//// \
// \\
amen FLEXIBLY \ .
counsmo enmesueo \

//// ’
///// ,.

muctUnALu I

mummuv
enmesutn ‘

 

Mummies '

1
NE!)

°_...‘

Z

 

 

20

extreme types), and the third group is extreme on only one
dimension (eight mid-range types). These three groups
represent a range in family functioning from optimal to
distressed. The Circumplex Model is thus a curvilinear
model; it is the balanced families, those with moderate
levels of cohesion and adaptability, which are postulated to
be the healthiest. Those families on the low or high end of
cohesion and adaptability will be relatively less healthy.
Olson and his colleagues have developed an instrument
to measure the concepts of cohesion and adaptability. This
instrument, the Family Adaptability and Cohesion Evaluation
Scales (FACES), has been utilized in several studies. It
was used by Portner (1980) in comparing 55 families (with an
adolescent) in family therapy with a control group of 117
nonproblem families. Nonclinic families were more likely to
fall in the balanced area of the Circumplex model than were
the families in treatment. Similarly, 33 families with
runaways were compared to these same 117 nonproblem families
(Bell, 1982). According to mothers and adolescents, but not
fathers, more control families than clinic families were in
the balanced range. Conversely, more runaway than control

families were found to be mid-range and extreme.

Evaluation of Literature on Family Process and Child

Adjustment

 

As can be seen from the above review, neither theory
nor research has explicitly dealt with family forms other

than nuclear families. In this literature, there is no

21

research on the impact of family process variables on
children's adjustment in remarried families.

Additionally, the child adjustment variables tend to be
poorly articulated. Global health or pathology is generally
studied; there is little attention to particular aspects of
functioning. Self—concept, an important indicator of
emotional health, is mentioned in the theoretical
formulations but has not been studied systematically.

Moreover, in general this literature lacks a solid
research foundation. Those models derived from the family
therapy field (e.g., Minuchin, Haley, Satir and Lidz) are
almost completely theoretical. The more empirical efforts
of Olson, while an improvement, have not been thoroughly
replicated and do not yet have strong quantitative support.
Along this line, Riskin (1982) has commented: "... the
field is noticeably weak in concepts and hypotheses that can
be operationalized and subjected to empirical tests" (p.
69).

Finally, the research cited has tended to concentrate
on adolescents. Given that a significant percentage of
couples divorce when children are preschoolers and remarry
before or when they are elementary school aged, it seems

necessary to study this age group more intensively.

The Present Study

 

The current study focused on four areas. First,
similar to past studies, it aimed to show that family type

would not predict child self-esteem. Secondly, it aimed to

22

demonstrate that, for stepchildren, family system and
subsystem functioning would predict self-esteem. Thirdly,
it explored the stepchild's perceptions of his/her Egg
family systems and how they contributed to self-esteem. The
children reported on the functioning of each of their
families as well as the relationships with their mothers,
fathers, stepmothers and stepfathers, as applicable.

Lastly, this investigation aimed to demonstrate that the
relationship between family process variables and
self-esteem would be the same in reconstituted families and

nuclear families.

Hypotheses

 

The specific hypotheses of this study are enumerated
below.

1. Consistent with the previous research on family
type and self-concept (Parish & Dostal, 1980; Parish &
Taylor, 1979; Raschke & Raschke, 1979; Santrock et al.,
1982; Wilson et al., 1975), there will be no significant
difference in self-esteem between nuclear family children
and stepchildren.

Hypotheses 2 through 6 are based upon the family theory
and therapy literature reviewed earlier (e.g., Minuchin,
1974; Bodin, 1981; Haley, 1980; Satir, 1964; Lidz, 1976).

2. For stepchildren, their self-concept will be
positively correlated with their perceptions of family
system functioning, such that self-esteem will be highest in

the healthiest families. This will be the case for the

23

child's ratings of both his/her family with mother and
his/her family with father.

3. Self-esteem will be related to the stepchild's
ratings of the parents' behavior such that his/her
self-concept will be positively correlated with his/her
perceptions of good parent-child subsystem functioning and
negatively correlated with his/her perceptions of poor
parent-child subsystem functioning. This will be true for
the child's perceptions of mother and father.

4. Self-esteem will be related to the stepchild's
ratings of the stepparents' behavior such that his/her
self-concept will be positively correlated with perceptions
of good stepparent-stepchild subsystem functioning and
negatively correlated with perceptions of poor
stepparent-stepchild subsystem functioning. This will be
true for the child's perceptions of stepmother and
stepfather.

5. For nuclear family children, self-esteem will be
positively correlated with their perceptions of family
system functioning.

6. For nuclear family children, their self-esteem will
be positively correlated with their perceptions of good
parent-child subsystem functioning and negatively correlated
with their perceptions of poor parent-child subsystem
functioning. This will be true for the child's perceptions

of mother and father.

METHOD

Subject Recruitment

 

Thirty stepchildren and 30 intact family children
between 6 and 12 years old were recruited for this study.
Introductory letters (see Appendix A) explaining the project
were sent home with students in various Ingham County,
Michigan communities. First through sixth graders in
Haslett, Holt and Waverly School Districts and the sixth
graders in Mason School District received letters. This
included four schools in Haslett, six schools in Holt, five
schools in Waverly and one school in Mason. Letters (see
Appendix B) were also distributed at a meeting of the
Stepfamily Association of America, Lansing Chapter (SAA),
and a presentation was made to a subgroup of the SAA
members.

Fifty of the subjects were recruited through two of the
school districts, Haslett and Holt. The remaining ten
children were recruited through a variety of sources. Four
were recruited through the other two school districts, three
children from Waverly and one child from Mason. Two
subjects were friends of people participating in the study,
one family was known to one of the research assistants, and

another family was known to a professor associated with the

24

25

project. An additional family was recruited through the
SAA. The final family was recruited through an announcement
of the study in a university newspaper. A breakdown of the
subjects by recruitment locale and type of family is
presented in Table 1.

Table 1. Distribution of Subjects by Recruitment Locale and
Type of Family.

Recruitment Locale

Family Type Haslett Holt Other*
Stepfamily 8 12 10
Nuclear Family 13 17

 

*This includes recruitment through Waverly, Mason, friends,
SAA and newspaper.

Each family was then contacted by phone to ensure
eligibility, answer questions and make arrangements for data
collection. A stepchild was eligible for the study if his
or her custodial parent had been remarried for at least six
months. Additionally, the stepfamily had to be a result of
divorce.

More nuclear families than stepfamilies responded to
the recruitment efforts. Those nuclear families that best
matched the 30 stepfamilies on age and sex of target child
were asked to participate. The final sample consisted of 31
females (16 step, 15 nuclear) and 29 males (14 step, 15
nuclear). The distribution of subjects by age, sex and type

of child is presented in Table 2.

26

Instruments

 

A. Family Adaptability and Cohesion Evaluation Scales

 

(FACES-II): The FACES-II (Olson, Bell, & Portner, 1982) is
a 30-item self-report scale. The 30 items are statements
about family life, such as "Our family does things together"
and ”Discipline is fair in our family." The subjects were
read each statement and responded on a 5-point scale,
ranging from "almost never" (1) to "almost always" (5). The
instrument yields two scores, total cohesion and total
adaptability.
gablg 2. Distribution of Subjects by Age, Sex and Type of

1 .

Age

Stepchildren
Female 2 1 4 3 2 3 1 16
Male 0 5 l 3 3 l 1 14
Total 2 6 5 6 5 4 2 30

Nuclear Family

Children
Female 1 2 4 3 2 2 l 15
Male 1 4 1 3 3 2 1 15
Total 2 6 5 6 5 4 2 30

 

An analysis of the internal consistency of the FACES-II
by Olson et al. (1982) revealed an alpha coefficient of .90
for the total scale, .87 for cohesion and .78 for
adaptability. The test-retest reliability for an earlier,

50-item version of the FACES-II was determined with a four

27

to five week interval between testings. The Pearson
correlation for the full scale was .84; the correlations
were .83 and .80 for cohesion and adaptability,
respectively. Construct validity was investigated utilizing
factor analysis, and it was found that the 16 cohesion items
loaded on Factor 1, and the adaptability items loaded
primarily on Factor 2 (Olson et al., 1982).

The FACES-II was slightly modified for this
investigation. Usually an individual responds twice to each
question, once describing the family currently and the
second time indicating his or her ideal view of the family.
Since the second response is not relevant to this research,
it was deleted to save administration time. Secondly, since
each stepchild had to complete two FACES-II, one for each
family, two additional versions of the instrument were
utilized. As applicable, the phrase "When with my mother
and stepfather" or "When with my mother" preceded all
questions on the second version; similarly, "When with my
father and stepmother" or "When with my father" preceded all
questions on the third version.

Additionally, the FACES-II was designed for children of
age 12 and older. For younger children, a 5-point response
scale can be difficult to understand. Thus, the procedure
that Hazzard and Christensen (1983) have incorporated into
the administration of the PPI was also utilized for the
FACES-II. The child saw a piece of paper with five

thermometers graphically representing the five responses

28

ranging from "almost never" to "almost always." Thus, an
empty thermometer meant "almost never" and a full
thermometer indicated "almost always."

Lastly, some words in the FACES-II were difficult for
younger subjects to understand. Definitions of those words
were written directly on the test and were used by the
interviewers for explanatory purposes when necessary. A
copy of the FACES-II and the response sheet is in Appendix
C.

B. Parent Perception Inventory (PPI): The PPI
(Hazzard, Christensen, & Margolin, 1983) is a measure of
children's reports of critical positive and negative
parental behaviors. This 36-item instrument yields four
9-item subscales: Mother Positive, Mother Negative, Father
Positive and Father Negative. Nine positive behavior
classes (positive reinforcement, comfort, talk time,
involvement in decision-making, time together, positive
evaluation, allowing independence, assistance, and nonverbal
affection) and nine negative behavior classes (privilege
removal, criticisms, command, physical punishment, yelling,
threatening, time-out, nagging, and ignoring) are assessed
for both parents. The child was read descriptions and
examples of each behavior class until he or she understood
the concept. For example, "How often does your father take
away things when you misbehave (like not letting you watch
TV or ride your bike or stay up late or eat dessert)?"

assesses "privilege removal." The child indicated a phrase

29

on a 5-point scale (never, a little, sometimes, pretty much,
a lot). For this sample, stepchildren were administered the
same questions regarding their stepmothers and stepfathers,
as applicable. A copy of the PPI (mother version) and the
response sheet is in Appendix D.

Hazzard et a1. (1983) reported a Cronbach's alpha of
.84 for Mother Positive, .78 for Mother Negative, .88 for
Father Positive, and .80 for Father Negative. As another
check of internal consistency, item-total correlations were
computed for each item and the subscale to which the item
contributed. All positive and negative items were
significantly correlated with the appropriate subscales.

The correlations ranged from .40 to .83 for the positive
items and from .34 to .72 for the negative items.

Convergent validity was assessed by Hazzard et al.
(1983) by correlating the four PPI subscales with scores on
both the Piers-Harris Children's Self-Concept Scale and a
parent-completed measure of child conduct disorder (the
Externalizing Scale of Achenbach's Child Behavior
Checklist). Six of the eight predicted correlations were
significant in the expected direction. In order to test the
discriminant validity of the PPI, correlations were again
calculated for the four subscales and two measures which
were not expected to be highly related to the PPI. These
were a child-completed achievement test (the Wide Range

Achievement Test) and the parent-completed Becker

30

Intellectual Inadequacy Scale. Again, six of the eight
correlations were non-significant as expected.

C. Piers-Harris Children's Self—Concept Scale

 

(Piers-Harris): The Piers-Harris (Piers, 1984) is a
self-report questionnaire designed to assess how children
and adolescents feel about themselves. It contains 80
statements such as "I am smart" and "I often get into
trouble." The subject responds with either "yes" or "no" to
each statement. The measure also yields six factor
analytically derived scales: Behavior, Intellectual and
School Status, Physical Appearance and Attributes, Anxiety,
Popularity, and Happiness and Satisfaction. Appendix E
contains a copy of the Piers-Harris.

A review of the reliability studies of the Piers-Harris
found that the test-retest reliability coefficients ranged
from .42 to .96, with a median coefficient of .73. Studies
of internal consistency revealed correlations of .88 to .93.
Numerous studies have also investigated the content,
criterion-related, construct, convergent and discriminant
validity of this instrument, and a review of these studies
reveals that the Piers-Harris is generally a well-validated
measure of self-concept in children (Piers, 1984).

D. Marriage and Familyvguestionnaire: This

 

questionnaire includes 17 questions focusing on demographic
(e.g., age and educational level of parents/stepparents),
marital history (e.g., current marriage and previous

marriages), family structure (e.g., age and sex of children)

31

characteristics, and visitation patterns. This instrument,
created for this study, is based on several unpublished
questionnaires utilized in other stepfamily research
(Clingempeel, personal communication; Clingempeel, Brand, &

Ievoli, 1984), and a copy is in Appendix F.

Procedure

 

Initially, three families and children were used as
pilot subjects. The wording of the FACES-II was clarified,
as were some of the items on the Marriage and Family
Questionnaire. The data were then collected by six trained
undergraduate research assistants. A one to two hour home
visit was conducted. The first portion of each visit was
spent establishing rapport, answering questions and
obtaining informed consent (see Appendix G). In separate
rooms of the home, the spouses then collaborated on the
demographic questionnaire, and the interviewer worked with
the child. The purpose of the study, confidentiality and
consent were explained to the child. If uncomfortable or
confused, the child was told to inform the interviewer. The
assistant then administered the FACES-II, the PPI and the
Piers-Harris to the child with the order of the measures
counterbalanced across children. Each item was read aloud,

and the response was recorded.

RESULTS

A summary of the demographic characteristics of this
sample will first be presented. Next, the data reduction
procedures for the FACES-II are explained. Third,
multivariate analyses to determine any differences as a
function of the school district are presented. Finally, the

results of the hypothesis testing are reported.

Description of Families

 

Unless otherwise noted, all demographic data on the
stepfamilies refer to the 30 custodial stepfamilies.

Means and standard deviations were obtained on the

following characteristics: (a) wife's age (stepfamily, M

33.5 years, SD = 3.7; nuclear, M = 35.9, SD = 4.5); (b)

husband's age (stepfamily, M = 34.8, SD = 4.7; nuclear, M
37.1, SD = 4.7); (c) wife's education (stepfamily, M = 13.8
years, SD = 2.3; nuclear, M = 14.5, SD = 2.4); and (d)
husband's education (stepfamily, M = 14.7, SD = 2.6;
nuclear, M = 15.4, SD = 3.3). A series of t-tests revealed,
with the exception of the wife's age, no significant
differences as a function of family type. Nuclear family
wives were older than stepfamily wives (t(58) = -2.26, p =

.03).

32

33

The total annual income of the 60 families ranged from
less than $15,000 for 4 families (stepfamily = 2, nuclear =
2): $20,000 to $29,999 for 9 families (stepfamily = 3,
nuclear = 6): $30,000 to $39,999 for 16 families (stepfamily
= 5, nuclear = 11); $40,000 to $49,999 for 15 families
(stepfamily = 9, nuclear = 6); and more than $50,000 for 16
families (stepfamily = 11, nuclear = 5). A 2 X 5

cross-tabulation with type of family by income group yielded

a non-significant chi-square ( x2 (4) 6.1, p = .19).

As for race, all the husbands were white with the
exception of one nuclear family husband who was black. One
stepfamily wife was oriental, and one nuclear family wife

was black; all other wives were white.

Description of Stepfamilies

 

In terms of the 30 stepchildren, the mean age at
divorce was 2.8 years (SD = 1.7). The mean age at mother's

2.0), and the mean age at father's

remarriage was 6.0 (SD

remarriage was 4.9 (SD 1.6). Fifteen of the stepchildren
had four parents (two parents and two stepparents). Fifteen
had three parents; of these, 13 had a mother, father and
stepfather, and two children had a father, mother and
stepmother.

Seventeen of the children saw their noncustodial parent

on a regular basis. The average number of days of

2.8); the average number

visitation per month was 5.4 (SD

of nights was 4.6 (SD = 2.5). The remaining 13 children did

34

not see the noncustodial parent on a regular basis; on the
average, this parent had seen them 6.4 days (SD = 7) in the
last six months.

There were eight structural types of remarried families
represented in this sample. These types are determined as a
function of a) whether one or both parents have children
from previous marriages, b) whether the mother, father or
both have custody of the children, and c) whether the
spouses have children from the current marriage. A simple
stepfamily is one in which only one parent has children from y
a previous marriage; a complex stepfamily is one in which
both parents have children from previous marriages. A
stepfather family is one in which only the wife has custody
of children; in a stepmother family, the husband has custody
of children. In a stepmother/stepfather family, both
spouses have custody of children. As an example, the
complex stepfather family, a common type of stepfamily,
contains a wife with custody of children from a previous
marriage and a husband with children of whom he does not
have custody.

In this sample, the frequency of the different types of
stepfamilies was: 1) simple stepfather (N = 7); 2) complex
stepfather (N = 7); 3) simple stepfather, with children from
current marriage (N = 4); 4) complex stepfather, with
children from current marriage (N = 2); 5) simple
stepmother, with children from current marriage (N = 4); 6)

complex stepmother/stepfather (N = 4); 7) complex

35

stepmother/stepfather, with children from current marriage
(N = 1); and 8) other (N = 1). This latter family was an
unusual simple stepmother family. The spouses had been
married to each other, had a daughter and then divorced,
with the mother retaining custody of the daughter. The
father remarried and had two children. He then divorced his
second wife, retained custody of the two children and
remarried his first wife. One of the father's children by
his second marriage was in this study.

The mean length of the current remarriage was 38.4
months (SD = 22.8). Fifteen of the 30 husbands had been
married and divorced once before, five had been married and
divorced twice previously and ten were in their first
marriage. The mean length of the husband's first marriage

was 81.3 months (SD

43.2) and, for the second marriage,

was 76.2 months (SD 69.3). The husband was single an

average of 26.7 months (SD = 20.5) between his first and
second marriages and 48.4 months (SD = 53.8) between his
second and third marriages.

Twenty four of the 30 wives had been married once
previously, four had been married twice previously, and two
had never been married previously. All of the previous
marriages had ended in divorce except one of the wives'
first marriages which ended with the death of her husband.

The mean length of the wife's first marriage was 83.9 months

(SD 35.6) and, for her second marriage, was 26.8 months

(SD 6.9). The wife was single an average of 36.9 months

36

(SD = 24) between her first and second marriages and 37

months (SD = 16.1) between her second and third marriages.

FACES-II: Family Types and Distance from Center Scores

 

The FACES-II cohesion and adaptability scores for each
family were combined in two ways. As discussed in the
introduction, each family was first categorized within the
Circumplex Model, depending on both its cohesion and
adaptability score. These two scores determined which of 16
types a particular family was. Since the 16 types were
grouped into three larger categories (balanced, mid-range
and extreme), each family then fell into one of these larger
groups as well. In addition, the cohesion and adaptability
scores were combined to derive a linear score, called
Distance from Center of Circumplex (DFC). This DFC score
indicates the distance of a family's cohesion and
adaptability score from the center of the model. The lower
the DFC score, the closer a family is to the center. Those
families closest to the center are the most balanced and the
most healthy.

Before the results of the analyses pertaining to the
FACES-II are reported, the procedure for obtaining the
cohesion and adaptability means and standard deviations will
be explained. These means and standard deviations are
important because they are the criteria for identifying the
16 family types and because they are utilized in calculating

DFC scores.

37

Means and Standard Deviations for Cohesion and Adaptability

The computation of the means for the cohesion and
adaptability scores was complicated by the fact that while
each nuclear family child had one cohesion score and one
adaptability score, each stepchild had two cohesion scores
and two adaptability scores, one set for each of the child's
two families. In order to examine potential differences in
mean scores depending upon the procedure by which they were
derived, two means for each of the two dimensions (cohesion
and adaptability) were initially computed. The cohesion
scores of the 30 nuclear family children, and the 30
cohesion scores for the stepchild's family with mother were
averaged for one mean; the same 30 cohesion scores for the
nuclear family children and the 30 cohesion scores for the
stepchild's family with father were averaged for the other
mean. The same procedure was utilized with the adaptability
dimension. The means and standard deviations resulting from
this procedure are reported in Table 3.

Table 3. Means and Standard Deviations for Cohesion and
Adaptability.

 

Cohesion Adaptability
M SD M SD
Nuclear-
Family with Mother 60.32 8.75 43.12 8.09
(N = 60)
Nuclear-
Family with Father 60.07 10.41 42.62 7.58

(N = 60)

 

38

The means and standard deviations for each of the two
dimensions were compared, and they were not found to differ
significantly. As another check, it was determined whether
the two sets of adaptability and cohesion scores would yield
differences in the classification of each family, since both
the means and the standard deviations are utilized in
categorizing a family as one of the 16 types. In fact, only
two of the thirty nuclear families were classified
differently by using the two sets of means and standard
deviations. The region (balanced, mid-range or extreme) was
not affected. Therefore, it was determined that either mean.
could be used in the DFC calculations. As a precaution,
however, all nuclear DFC scores were calculated twice where
relevant, and both DFC calculations were used in the
analyses. As expected, there was no difference in the
results as a function of the two different DFC calculations;
thus, for simplicity's sake, only the analyses utilizing the

Nuclear-Family with Mother mean (see Table 3) are reported.

Family Types: Balanced, Mid-Range or Extreme

Olson's procedure (Olson, personal communication)
utilizing standard deviations as the criterion values for
classifying families was adopted. On the cohesion
dimension, those with scores greater than one standard
deviation below the mean were labeled disengaged families,
those who scored within one standard deviation below the
mean were labeled separated families, those families scoring

within one standard deviation above the mean were labeled

39

connected, and those families scoring greater than one
standard deviation above the mean were labeled enmeshed. On
the adaptability dimenSion, rigid, structured, flexible and
chaotic families, respectively, were classified similarly.
Each family in the study was then placed in one of the 16
family types based on a combination of its cohesion and
adaptability scores.

Once all the families were classified, they were placed
into the three main groups: balanced, mid-range and extreme
(as explained in the introduction). Table 4 presents the
frequencies and percentages for each of these groups for
stepchildren's mother's family and father‘s family, and
nuclear children's families.

Table 4. Frequency and Percentage of Balanced, Mid-Range
and Extreme Families for Stepchildren's Family with Mother,

Stepchildren's Family with Father and Nuclear Family
Children.

 

 

Family with Family with Nuclear

Mother Father Family

N % N % N %

Balanced 9 30 16 53 23 77
Mid-Range 12 40 7 23 6 20
Extreme 9 30 7 23 1 3

 

An overall chi-square was performed to determine
whether the distribution of balanced, mid-range and extreme
families differed depending upon the type of family, and the
chi-square was found to be significant ( x2 (4) = 14.72, p <

.01). Two additional chi-squares where then completed. The

40

first compared nuclear families with stepchildren's family
with mother. These two family categories were found to have
significantly different distributions ( x2 (2) = 14.53, p <
.001). As can be seen from Table 4, family with mother has
a much higher proportion of mid-range and extreme families
than does nuclear families. In particular, it is noteworthy
that while only 3% of nuclear families were extreme, 30% of
stepchildren's mother's family were extreme. Another
chi-square, comparing family with father and nuclear
families, revealed a trend toward significance ( x2 (2) =
5.83, p < .10) and was close to significance at the .05
level. The pattern in this case was the same as in the
previous comparison; a higher proportion of extreme families
were found in family with father (23%) than in nuclear
families (3%).

Distance from Center (DFC) Scores

 

Two DFC scores were calculated for each stepchild, one
for each of his or her families, and one DFC score was
calculated for each nuclear family child. These DFC scores
were then utilized for the analyses of the hypotheses
concerning the FACES-II. The scores were correlated with
the Piers-Harris scores as well as utilized in multiple
regression analysis to determine their predictive power.

The formula for determining each DFC score (Olson,

Portner & Lavee, 1985) is:

41

 

(subject's cohesion score - XC)2
- 2

(subject's adaptability score - XA)

+

mean of cohesion

where: EC
mean of adaptability

XA

Multivariate Analyses

 

Prior to the major analyses, the data were analyzed to
determine whether there were differences as a function of
the school district (or other location) from which the
subject was recruited. As mentioned earlier, the sample of
stepchildren was recruited from three areas (Haslett, Holt
and Other), whereas the sample of nuclear family children
was recruited from two areas (Haslett and Holt). Because
there were different numbers of recruitment locales for the
stepchildren and nuclear family children, two MANOVAs were
performed for each clustered set of variables. The first
was a 2 X 2 (type of child X recruitment locale) MANOVA,
with Haslett and Holt as the two locales. The second MANOVA
dealt only with the 30 stepchildren and utilized one factor,
recruitment locale, with the three areas mentioned above.
The results of these two MANOVAs were then considered
together.

Three sets of MANOVAs were completed. A set was
performed on four demographic variables, husband's and
wife's age and husband's and wife's education. Another set

of MANOVAs was performed on the FACES-II variables cohesion,

42

adaptability and DFC. MANOVAs were also completed on the
four PPI scores, Mother Positive, Mother Negative, Father
Positive and Father Negative. In all three cases, the
overall multivariate F's did not reveal a recruitment main
effect or a recruitment-type of child interaction effect.

A 2 X 2 ANOVA (type of child X recruitment locale) and
a one way ANOVA for the three stepfamily recruitment locales
were performed with the Piers-Harris as the dependent
variable. Neither analysis generated significant effects;
therefore, it appears that the Piers-Harris scores were not

affected by the source of recruitment.

Results of Tests of Hypotheses

Hypothesis 1 predicted that there would be no
difference in self-concept between nuclear family children
and stepchildren. As predicted, the Piers-Harris scores for
the two groups of children were not significantly different
(t (59) = -.46, p = .64).

The FACES-II DFC and the PPI scores were correlated
with the Piers-Harris scores. These correlations were
computed separately for the stepchildren and the nuclear
family children. For the stepchildren, hypothesis 2
predicted that both DFC scores would be negatively
correlated with the Piers-Harris scores. It was also
hypothesized that Mother Positive, Father Positive,
Stepmother Positive and Stepfather Positive scores on the
PPI would be positively correlated with the Piers-Harris

scores (hypothesis 3). Mother Negative, Father Negative,

43

Stepmother Negative and Stepfather Negative scores were
expected to be negatively correlated with the Piers-Harris
scores (hypothesis 4).

The Pearson correlations are presented in Table 5.
Contrary to prediction, these data reveal that neither DFC
score for mother's family or the DFC score for father's
family was significantly correlated with the Piers-Harris,
but both correlations were in the predicted direction. In
terms of the PPI, the hypotheses were partially supported.
Two of the four parent PPI scores were significantly
correlated in the predicted direction with self-concept.
Mother Positive was correlated .38 (p = .02) and Mother
Negative was correlated -.30 (p = .05). Although the
correlations were not significant, both Father Positive and
Father Negative were correlated with the Piers-Harris in the
predicted direction. While none was significant, three of
the four stepparent scores (Stepmother Positive, Stepfather
Positive and Stepfather Negative) were correlated with
self-concept in the predicted direction, and two of these
correlations approached significance. Stepmother Positive
was correlated .37 (p = .07), and Stepfather Negative was
correlated -.25 (p = .10). Overall, two of the ten
predicted correlations were significant, two were marginally
significant and nine were in the expected direction.

For the nuclear family children, it had been predicted
that the Piers-Harris scores would be negatively correlated

with DFC (hypothesis 5), Mother Negative and Father Negative

44

scores, and positively correlated with Mother Positive and
Father Positive scores (hypothesis 6). These correlations
are also presented in Table 5. Contrary to prediction,
there was no significant relationship between the DFC score
and the Piers-Harris score, although, again, the correlation
was in the predicted direction. In contrast, all four PPI
hypotheses were confirmed. Self-concept was correlated .37
(p = .02) with Mother Positive, -.44 (p = .01) with Mother
Negative, .40 (p = .02) with Father Positive, and -.38 (p =
.02) with Father Negative.

Table 5. Pearson Correlations of FACES-II and PPI scores
with Piers-Harris Scores.

Variable Coefficient P Value*
Stepchildren
DFC-Mother -.10 .59
DFC-Father -.08 .66
Cohesion-Mother .44 .02
Adaptability-Mother .36 .05
Cohesion-Father .25 .18
Adaptability-Father .10 .59
Mother Positive .38 .02
Mother Negative -.30 .05
Father Positive .08 .34
Father Negative -.20 .15
Stepmother Positive .37 .07
Stepmother Negative .11 .34
Stepfather Positive .21 .15
Stepfather Negative -.25 .10
Nuclear Family Children
DFC -.04 .85
Cohesion .26 .16
Adaptability .11 .57
Mother Positive .37 .02
Mother Negative -.44 .01
Father Positive .40 .02
Father Negative -.38 .02

 

*Tests of significance are two-tailed for all FACES-II
scores and one-tailed for all PPI scores.

45

Further Analysis Utilizing a FACES-II Linear Model

 

Since the FACES-II DFC correlations did not yield
significant results, it was decided to examine the
alternative hypothesis that the relationship between
adaptability and cohesion on the one hand and self-concept
on the other hand was linear, rather than curvilinear, as
the Circumplex Model purports. Thus, for both stepchildren
and nuclear family children, cohesion and adaptability were
predicted to be positively correlated with the Piers-Harris.
The two scales were then separately correlated with the
Piers-Harris (see Table 5). For the stepchildren, both
cohesion (r = .44, p = .02) and adaptability (r = .36, p =
.05) in family with mother were significantly correlated
with the Piers-Harris. Neither dimension for family with
father was significantly correlated with the Piers-Harris,
although both correlations were in the predicted direction.
For the nuclear family children, both adaptability and
cohesion were correlated with self-concept in the predicted

direction, but not significantly so.

Multiple Regression Analyses

Multiple regression analyses were then performed in
order to evaluate the relative contribution of different
systemic and subsystemic variables to the variance in
self-concept. These analyses should best be considered
exploratory. It was speculated that the family system
variables would be the most powerful predictors, while the

parent-child subsystem variables would have less predictive

46

power. In addition, the pattern of variables contributing
to self-concept was expected to be the same in remarried
families and nuclear families.

The FACES-II adaptability and cohesion scores and the
PPI scales were the predictor variables, and the
Piers-Harris was the criterion variable. An overall
regression for the stepchildren was completed first; two
further regressions on subsamples of this group were then
performed. A regression analysis was also completed for the
nuclear family children.

The regression analyses of the stepchildren data must
be interpreted quite cautiously, given the small number of
subjects and the relatively large number of variables that
were entered into the regression equation. The results of
the overall regression for the stepchildren are reported in
Table 6. 100% (p = .003) of the variance in self—concept
was accounted for by six variables: Mother Cohesion,
Stepmother Negative, Stepmother Positive, Stepfather
Positive, Mother Positive and Father Negative. Mother
Positive accounted for the highest percentage of the
variance. Thus, aside from cohesion in family with mother,
the family system variables were not found to be important
predictors of self-concept.

Table 7 presents the regression results for one subset
of stepchildren (N = 15), those with four parents (mother,
stepfather, father and stepmother). In this case, 95% (p =

.04) of the variance in self-concept was accounted for by

47

 

Table 6. Multiple Regression of Family Variables on
Piers-Harris Scores, Stepchildren.

2 R2

Variable r R R Change
Cohesion-Mother's Family .44* .44 .19 .19
Stepmother Negative .ll .51 .26 .07
Stepmother Positive .37 .67 .45 .19
Stepfather Positive .21 .78 .61 .15
Mother Positive .38* .91* .83* .22**
Father Negative -.20 l.00** 1.00** .17
*p g .05

**p g .01
Table 7. Multiple Regression of Family Variables on
Piers-Harris Scores, Stepchildren with Four Parents.

2 R2

Variable r R R Change

Cohesion-Mother's Family .62* .62* .39* .39*
Stepmother Negative .11 .72* .51* .12
Stepmother Positive .38 .83** 70** 18*
Father Positive -.17 .90*** 81*** .12*
Mother Negative -.54* .93*** .87*** .05
Stepfather Negative -.32 .96*** 92*** .06*
Adaptability-Mother's Family .51* .97*** 94*** .01
Mother Positive .58* .97** 94** .003
Father Negative .14 .97* .94* .01
Adaptability-Father's Family .03 .97* .95* .002
Cohesion-Father's Family .23 .97 .95 .002

 

*p g .05
**p g .01
***p 5 .001

all except two of the predictor variables, cohesion in

father's family and Stepfather Positive.

Cohesion in

mother's family accounted for the greatest percentage of the

variance, followed by Stepmother Positive.

As expected, a

48

family system variable was the best predictor of
self-concept, although only one family system variable was
entered in the regression equation.

A second subsample of stepchildren (N = 13), those with
three parents (mother, stepfather and father), was also
analyzed by multiple regression. These results are
presented in Table 8. A trend approaching significance (p =
.08) indicated that 84% of the variance in self-concept was
accounted for by seven variables: Father Negative,
adaptability in mother's family, Stepfather Positive, Mother
Positive, Father Positive, Mother Negative and Stepfather
Negative. Individually, however, none of these variables
significantly accounted for any variance when entered into
the regression equation.

Table 8. Multiple Regression of Family Variables on
Piers-Harris Scores, Stepchildren with Three Parents.

2
. 2 R
Var1able r R R Change

Father Negative -.47 .47 .22 .22
Adaptability-Mother's Family .17 .53 .29 .06
Stepfather Positive .06 .62 .39 .10
Mother Positive .14 .68 .46 .07
Father Positive .27 .73 .53 .07
Mother Negative -.06 .82 .68 .15
Stepfather Negative -.20 .92 .84 .16
Cohesion-Father's Family .30 .93 .87 .03
Adaptability-Father's Family .23 .93 .87 .01

 

The results of the regression for the nuclear family
children can be treated more confidently, as the number of

variables entered into the equation was half the number for

49

the stepchildren. The results of this regression are
presented in Table 9. 34% (p = .03) of the variance in
self-concept was accounted for by Mother Negative, Father
Positive, adaptability and Mother Positive. Mother Negative
accounted for the greatest percentage of the variance,
followed by Father Positive. Again, contrary to
expectation, the subsystem variables, rather than the family
system variables, were the most powerful predictors of
self-concept.

There was some similarity between the combined
regression for stepchildren and nuclear family children
regression in that mother-child subsystem functioning was
most important in contributing to self-concept. Otherwise,
there was no similar pattern in terms of which variables
accounted for the most variance. In addition, the pattern
for the stepchildren with three parents was noticeably
different from that for those children with four.

Table 9. Multiple Regression of Family Variables on
Piers-Harris Scores, Nuclear Family Children.

2
2 R
Variable r R R Change

Mother Negative -.44* .44* .l9* .19*
Father Positive .40* .57** .32** .13*
Adaptability .11 .58* .33* .01
Mother Positive .37* .59* .34* .01
Cohesion .26 .59 .34 .001
Father Negative -.38* .59 .34 .001

 

* g .05
**p g .01

DISCUSSION

There was no significant difference in self-concept
between stepchildren and nuclear family children. This
result does not support the deficit comparison model. It is
consistent with several previous studies (Parish & Dostal,
1980; Parish & Taylor, 1979; Raschke & Raschke, 1979;
Santrock et al., 1982; Wilson et al., 1975). Furthermore, a
recent study (published after this one was conducted), also
found no difference in the self-concept of six to eight year
old nuclear family and stepfamily children (Bray, 1986).

For the stepchildren, their perceptions of healthy
family system functioning in the family with mother and
positive mother-child subsystem functioning EEES related to
high self-esteem. Healthy family functioning was
characterized by high levels of closeness and flexibility.

A negative mother-child relationship was associated with
lower self-esteem. In addition, a negative
stepfather-stepchild relationship was marginally associated
with lowered self-concept. A positive mother-child
relationship was the best predictor of self-concept. The
above results should be interpreted with consideration of
the fact that the family with mother was the custodial
family in most cases, and the family always had a stepfather

when the mother had custody. System functioning in the

ROI

51

stepchild's family with father, which in most cases was the
noncustodial family, was not related to the child's
self-concept, although positive stepmother-stepchild
subsystem functioning was marginally related to higher
self-concept.

The fact that family functioning is related to child
adjustment is consistent with the family therapy literature
and empirically based family models (e.g., Epstein 8 Bishop,
1982; Lewis, Beavers, Gossett, 8 Phillips, 1976; Minuchin,
1974; Olson, Sprenkle, 8 Russell, 1979; Satir, 1964).
However, it appears that the family with which the child has
day-to-day contact, in this case the custodial
mother/stepfather family, seems to be the major family
system that affects the child's self-image. Interestingly,
the association between the stepfather—stepchild
relationship and self-concept is inconsistent with a
recently published study by Clingempeel and Segal (1986).
With 40 custodial mother/stepfather families, they found no
correlation between self-concept and either the
stepfather-stepdaughter relationship or the
stepfather-stepson relationship. Their results may be
different because their data analysis took into account the
sex of the child and the structural type of stepfamily.

The fact that the noncustodial family and the
relationship with the father did not impact self-concept may
be explained by a study by Wallerstein and Kelly (1980).

They observed that even the most caring fathers in their

52

sample found it quite difficult to maintain relationships
with their noncustodial children over time. Five years post
divorce, there had been a gradual but steady decrease in the
amount of visitation contact. Only 30% of the father-child
relationships were found to be emotionally supportive; the
remaining 70% had ongoing visitation, but the relationship
was often a formal arrangement with little emotional
vitality. If a similar preponderance of involved but formal
fathers was typical of this sample as well, the lack of an
effect on their children's self-concept would be
understandable. In the families where the father was
remarried, the marginal association between the
stepmother-stepchild relationship and self-concept may
reflect her greater involvement with the child's rearing.
Although not the biological parent, the stepmother is the
"mother figure" in the family and is likely to do more of
the childrearing.

The pattern of family system and subsystem variables
' accounting for the variance in self-concept was different
for stepchildren with four parents and those with three
parents. This raises the possibility that there are
differences in the relative importance of these variables in
different types of family circumstances. Because of a low
number of subjects, however, these results must be
considered highly speculative and await further
investigation with larger samples. Nevertheless, it may be

instructive to consider how differences in family

53

circumstances could mediate self-esteem in different ways.
The child with four parents is part of two stepfamilies; the
child with three parents is part of one stepfamily (mother
and stepfather, in this study) and one divorced family
(father, in this study). A child who is part of two
stepfamilies has two mothers and two fathers; hence, this
child may have more loyalty conflicts than the child who has
either two mothers or two fathers. Further, with four
parents there may be more adult conflict than with three.

On the other hand, there could be less conflict, since both
ex-spouses are remarried and would be less likely to be
jealous of the ex-spouse's new mate.

In the situation with three parents, perhaps the single
father is overly involved with the child because he lacks
other significant relationships. Another possibility,
however, is that the single father is less involved with his
child because he feels he cannot provide a family situation
equivalent to the one provided by the mother. Whatever the
case may be, three or four parent situations could
differentially affect family functioning, the child's
adjustment, and presumably, his or her self-concept.

In contrast to the stepchildren, the nuclear family
children manifested a different set of relationships between
system and subsystem variables and self-concept.
Surprisingly, there was no relationship between family
functioning and self-concept. On the other hand, positive

parent-child subsystem functioning was associated with

54

higher self-concept, and negative subsystem functioning was
related to lower self-concept. A positive mother-child
relationship was the best predictor of self-concept in
children. As with the stepchildren data, it makes intuitive
sense that the mother-child relationship would be the best
predictor, as the mother is still likely to be the most
involved parent, even given our changing sex role attitudes.
The lack of a relationship between children's
perceptions of family functioning and self-esteem in nuclear
families is puzzling, as it is inconsistent with basic
tenets in family theory, as well as previous research by
Olson, McCubbin, Barnes, Larsen, Muxen and Wilson (1983).
In a study of 1000 intact, normal families, the FACES-II was
correlated with a variety of outcome measures. The
inconsistency between the two studies may be explained by
sampling differences. The percentages of balanced,
mid-range and extreme families in the national sample and in
the nuclear families in the current sample were not
equivalent. In the national sample, 55%, 30% and 15% of the
families were balanced, mid-range and extreme, respectively.
In this sample, the percentages were 77%, 20% and 3%,
respectively, suggesting a "healthier" sample of families
volunteered for this study as compared to the national
study. Further, in the present study, there was a very
small range in FACES-II scores; an increase in statistical
power through a greater number of subjects might have

increased the range and included more extreme scores.

55

It was also found that remarried families were more
likely to be distressed than nuclear families. While the
majority of nuclear families and stepchildren's family with
father were balanced, the majority of stepchildren's family
with mother were mid-range. Comparing the nuclear families
and the family with mother, there was a greater percentage
of mid-range and extreme families in the stepfamily sample.
Similarly, family with father had a greater percentage of
extreme families, but this finding was marginal. The higher
percentage of mid-range and extreme families approximates
the pattern Olson, McCubbin et al. (1983) have reported in
clinical samples. Distressed families were more likely to
be classified as mid-range or extreme.

The fact that blended families are more likely to be
functioning at less than-optimal-levels is consonant with
the clinical literature documenting the many stresses
associated with stepfamily living (Visher 8 Visher, 1979).
Family cohesion is likely to be low at the time of
stepfamily formation, and families are likely to be
chaotically organized until rules and structure can evolve
over time. However, since several researchers (e.g., Ganong
8 Coleman, 1984) have noted there is a lack of data
concerning the strengths and healthy aspects of blended
families, it is also important to point out that these
results do not imply that all stepfamilies are distressed.

In this sample, 53% of father's family and 30% of mother's

56

family were functioning in the range considered most healthy
by Olson et al. (1979).

In spite of the fact that different systemic and
subsystemic processes govern the relationship of family
process variables and self-concept in stepfamilies and
nuclear families, and stepfamilies have a greater tendency
toward distress than nuclear families, it should be stressed
that the outcome for children, at least in terms of
self-concept, does not differ. Membership in a remarried
family is different than membership in a nuclear family, but.
it does not necessarily predict poorer adjustment for the
child.

Olson's curvilinear model is not supported by the
results of this study. This is consistent with some recent
research, especially work that has been published since the
present investigation was undertaken. The curvilinear model
has recently been critiqued or modified by other
theoreticians, by empirical studies utilizing the model, and
by new work by Olson and his colleagues. The
Beavers-Timberlawn Model of family competence, another
theoretical model which has become a standard in the family
theory field, portrays family life as existing on an
infinite linear continuum of competence and emphasizes the
limitless potential for growth. Beavers and Voeller (1983)
and Olson, Russell et al. (1983), while agreeing more or
less on the salient dimensions of family life, have debated

the linear/curvilinear issue. Recently, Green, Kolevzon and

57

Vosler (1985) and the Beavers group (Beavers, Hampson, 8
Hulgus, 1985) have directly compared the models. The former
study did not find evidence of curvilinearity for either
cohesion or adaptability. The latter study found strong
linear correlations between an instrument based on the
Beavers model and the two FACES-II scales.

Other research utilizing the FACES or FACES-II has
reported results along the same line. An investigation by
Sigafoos, Reiss, Rich and Douglas (1985) found a linear
association between adaptability and a behavioral index of
family functioning. A study by Bray, Howard, Magee and
Brady (1985) is particularly relevant here because its
subjects were school-aged children from intact, divorced and
remarried families. Among other results, both cohesion and
adaptability were correlated with behavior problems. Higher
levels of cohesion and adaptability were associated with
fewer behavior problems in the children.

Olson has modified his original model as a result of
the national survey (Olson, McCubbin et al., 1983). For the
1000 normal families, higher levels of cohesion and
adaptability were associated with better family functioning,
as measured by communication and satisfaction. These
results were attributed to the fact that these normal
families were different than the clinical populations upon
which the curvilinear model was developed and tested. The
normal families only represented a narrow spectrum of the

range of behavior on these two dimensions. As a result,

58

there were very few of these families that legitimately fell
into the "extreme" category. They were only categorized as
such because they were extreme relative to that sample's
mean rather than "truly" extreme and dysfunctional (Olson et
al., 1985).

Another study of a subgroup of this national sample
found support for both the curvilinear and linear models. In
an investigation of parent-adolescent communication, the
adolescents indicated a linear relationship between
communication, the Circumplex dimensions and family
satisfaction. However, the parents' reports supported the
curvilinear Circumplex Model (Barnes 8 Olson, 1985).

It is instructive to compare the percentages of
balanced, mid-range and extreme families in the national
sample and current sample of stepchildren's family with
mother. In the national group of intact, normal families,
55% , 30% and 15% of the families were balanced, mid-range
and extreme, respectively. For family with mother, they
were 30%, 40% and 30%, respectively. This suggests that the
latter group of families is on the border between normal and
clinical families. These stepfamilies are "normal" enough
to manifest the linear correlations expected from a
non-clinical sample, even though the higher percentage of
mid-range and extreme families is more typical of distressed

populations.

59

Methodological Issues

 

This study lacked power as a result of a small number
of children. The small number particularly affected the
regression analyses. Additionally, several correlations
approached significance and would likely have attained
significance had there been more subjects.

Furthermore, several variables were confounded in this
study. For the correlational analyses, children with one
stepfamily (three parents) were confounded with child living
in two stepfamilies (four parents). These two groups of
children were analyzed separately in the regression
analyses, but, again, a small number of subjects precluded
any definitive conclusions. Moreover, custodial stepmother
and stepfather families were confounded in the data
analysis. In the time since this study was begun, data have
accumulated to demonstrate that the structural type of
stepfamily and the type of stepfamily-sex of stepchild
interaction are important in the study of stepfamily
relationships (Clingempeel et al., 1984; Clingempeel,
Ievoli, 8 Brand, 1984; Clingempeel 8 Segal, 1986). There
were not enough stepmother families recruited to analyze
them as a separate group.

Measures of social desirability and defensiveness would
also have been of benefit. It appeared that some children
may have reported "defensively" high self-esteem and may

have also felt "compelled" to report positive attributes

60

about their parents and their families even though they were
guaranteed confidentiality.

In addition, parent and stepparent reports of family
functioning would have been valuable. Several researchers
have pointed out that one family member's observations
cannot be considered representative (Barnes 8 Olson, 1985;
Esses 8 Campbell, 1984). As reported above, Barnes and
Olson (1985) found differing results as a function of the
family member rating the family. Olson and colleagues
(Olson, McCubbin et al., 1983, Olson et al., 1985) have
developed procedures for combining family members' scores to.
create family cohesion and adaptability scores.

Finally, the effect of volunteer bias is unknown.
Perhaps the families which participated in the study
differed in some way from families which did not volunteer.
The study may have attracted some stepfamilies experiencing

problems and hoping to receive help through the project.

Considerations for Future Research

From this investigation, it appears that a fruitful
line of research would more specifically delineate the
processes that occur in stepfamilies and how they relate to
outcomes for the children. This project has examined some
system and parent-child subsystem processes, but other
system and subsystem variables could be studied as well.
For example, family theory has long maintained the
centrality of the marital subsystem, at least in intact

families. This may not necessarily be the case in

61

stepfamilies, however. For example, Crosbie-Burnett (1984)
found that mutually suitable stepparent-stepchild
relationships were more highly associated with family
happiness than was the marital relationship.

In addition, the child's noncustodial family, parent
and stepparent should continue to be studied. While this
project did not find strong associations between these
variables and self-concept, there were some suggestive
results that are worthy of further elaboration. Moreover,
the healthy functioning of the biological parents'
coparental subsystem has been theoretically linked with the -
positive adjustment of children whose parents have divorced.
Whether this is also the case with stepchildren remains to
be seen. In a situation where one or both of the parents
has remarried, the kind of coordination entailed in a
smoothly functioning coparental system may be threatening to
the new spouse(s) and disrupt the new marital subsystem(s).
Systems theory would predict that this would adversely
affect the children.

Furthermore, the sibling constellation in the
reconstituted family is another aspect of this system that
will affect the children's adjustment. A stepchild who has
stepsiblings may have more difficulty than one who does not.
In addition, a stepchild may have half siblings as well as
stepsiblings. The relationship between each of these sets
of circumstances and child adjustment is worthy of future

research.

62

Cross-sectional research such as this project cannot
answer the question of causality. Longitudinal work can
address how family process variables determine child
adjustment over time. Stepfamilies encounter different
developmental phases and tasks which change over time. In
fact, several longitudinal studies of remarriage, remarried
families and stepchildren are currently underway (Ahrons,
1979; Bray, 1986; Esses 8 Campbell, 1984; Hetherington 8
Clingempeel, 1985).

Longitudinal research could not only deal with the
developmental issues specific to stepfamilies, but those
changes that occur in any family across the life cycle. The
results from this sample of school-aged children are not
generalizable to any other age groups, particularly with
reference to the FACES-II data. Olson, McCubbin et al.
(1983) have reported that cohesion and adaptability scores
are highest in newly married couples, while cohesion is the
lowest when the oldest children are adolescents.
Conceivably, too much cohesion in families with adolescents
could be detrimental to self-concept, as it would not
encourage the developmental tasks of separation and identity
formation. Thus, future research needs to address family
process issues as they relate to children of different ages.

The FACES-III (Olson et al., 1985) is recommended for
future research. It was published subsequent to the present
data collection and could not be utilized for this study.

It appears to be psychometrically superior to the FACES-II,

63

and it contains items that are relevant for a variety of
family forms, including blended families. Further, several
behavioral measures of family functioning are available
(e.g., Forgatch 8 Wieder, 1981) and could be utilized for
multimethod investigation.

Finally, subsequent research should use larger,
randomly selected samples of reconstituted families.
Different structural types of stepfamilies should be
distinguished and studied separately. This is particularly

true for father/stepmother families.

Conclusion

 

Stepfamilies and stepchildren are a fact of life in
this country and will remain so in the forseeable future.
Until recently, there has been much myth but little research
on the many relevant psychological and sociological issues
for these children and families. This research has been an
attempt to dispel a few of these myths. In fact, research
is beginning to accumulate; a number of significant studies
were published during the course of this investigation.
Children of divorce and remarriage are being included in
risk research, and the treatment literature is expanding.
It is hoped that, in spite of the theoretical and
methodological complexities, the study of stepfamilies will

become even more prevalent.

L IST OF REFERENCES

LI ST OF REFERENCES

Ahrons, C. R. (1979). The binuclear family: Two
households, one family. Alternative Lifestyles, 2,
499-515.

 

Aponte, H., 8 Van Deusen, J. (1981). Structural family
therapy. In A. S. Gurman 8 D. P. Kniskern (Eds.),
Handbook of family therapy (pp. 310-360). New York:
Brunner/Mazel.

 

Barnes, H. L., 8 Olson, D. H. (1985). Parent-adolescent
communication and the circumplex model. Child
Development, 85, 438-447.

 

Beavers, W. R., Hampson, R. B., 8 Hulgus, Y. F. (1985).
Commentary: The Beavers systems approach to family
assessment. Family Process, 34, 398-405.

 

Beavers, W. R., 8 Voeller, M. N. (1983). Family models:
Comparing and contrasting the Olson circumplex model
with the Beavers systems model. Family Process, 32,
85-98.

Bell, R. (1982). Parent-adolescent interaction in runaway
families. Unpublished doctoral dissertation,
University of Minnesota.

Bodin, A. M. (1981). The interactional view: Family
therapy approaches of the Mental Research Institute.
In A. S. Gurman 8 D. P. Kniskern (Eds.), Handbook of
family therapy (pp. 267-309). New York:
Brunner7Mazel.

 

 

Bray, J. H. (in press). The effects of early remarriage on
children's development: Preliminary analyses of the
developmental issues in stepfamilies research project.
In E. M. Hetherington 8 J. Arasteh (Eds.), The impact
of divorce, single-parenting and step-parenting on
children. Washington, DC: National Institutes of
Health.

 

Bray, J. H., Howard, V., Magee, J. T., 8 Brady, C. P.
(1985). Children with behavior problems from intact,
divorced and remarried families. Paper presented at

64

65

the meeting of the American Psychological Association, Los
Angeles, CA.

Clingempeel, W. G., Brand, E., 8 Ievoli, R. (1984).
Stepparent-stepchild relationships in stepmother and
stepfather families: A multimethod study. Family
Relations, 55, 465-473.

 

Clingempeel, W. G., Ievoli, R., 8 Brand, E. (1984).
Structural complexity and the quality of
stepfather-stepchild relationships. Family Process,
35, 547-560.

 

Clingempeel, W. G., 8 Segal, S. (1986).
Stepparent-stepchild relationships and the
psychological adjustment of children in stepmother and
stepfather families. Child Development, 51, 474-484.

 

Crosbie-Burnett, M. (1984). The centrality of the step
relationship: A challenge to family theory and
practice. Family Relations, 55, 459-464.

 

Esses, L., 8 Campbell, R. (1984). Challenges in
researching the remarried. Family Relations, 55,
415-424.

 

Forgatch, M. S., 8 Wieder, G. (1981). Family problem
solving system. Unpublished manuscript.

 

Ganong, L. H., 8 Coleman, M. (1984). The effects of
remarriage on children: A review of the empirical
literature. Family Relations, 55, 389-406.

 

Glick, P. C. (1979). Children of divorced parents in
demographic perspective. Journal of Social Issues, 55,
170-182.

 

Green, R. G., Kolevzon, M. S., 8 Vosler, N. R. (1985). The
Beavers-Timberlawn model of family competence and the
circumplex model of family adaptability and cohesion:
Separate! but equal? Family Process, 35, 385-398.

Haley, J. (1980). Leaving Home. New York: McGraw Hill.

 

Hazzard, A., Christensen, A., 8 Margolin, G. (1983).
Children's perceptions of parental behaviors. Journal
of Abnormal Child Psychology, ll, 49-60.

Hetherington, E. M., 8 Clingempeel, W. G. (1985).
Adaptation to remarriage. Study in progess, University
of Virginia and Temple University.

66

Kaplan, H. B., 8 Pokorny, A. D. (1971). Self-derogations
and childhood broken home. Journal of Marriage and the

Lidz, T. (1976). The person (Rev. ed.). New York: Basic
Books.

 

Marotz-Baden, R., Adams, G., Bueche, N., Munro, B., 8 Munro,
G. (1979). Family form or family process?:
Reconsidering the deficit family model appraoch. 253
Family Coordinator, 25, 5-14.

McGoldrick, M., 8 Carter, E. A. (1980). Forming a
remarried family. In E. A. Carter 8 M. McGoldrick
(Eds.), The family life cycle: A framework for family
therapy (PP. 265-294). New York: Gardner Press.

Minuchin, S. (1974). Families and family therapy.
Cambridge, MA: Harvard University Press.

Minuchin, S., Montalvo, B., Guerney, G., Rosman, B., 8
Schumer, F. (1967). Families of the slums. New York:
Basic Books.

Olson, D. H., Bell, R., 8 Portner, J. (1982). FACES-II.
St. Paul, Minnesota.

Olson, D. H., McCubbin, H. I., Barnes, H., Larsen, A.,
Muxen, M., 8 Wilson, M. (1983). Families: What makes
them work. London: Sage Publications.

 

Olson, D. H., Portner, J., 8 Lavee, Y. (1985). FACES-III.
St. Paul, Minnesota.

Olson, D. H., Russell, C. S., 8 Sprenkle, D. H. (1980).
Circumplex model of marital and family systems II:
Empirical studies and clinical intervention. In J.
Vincent (Ed.), Advances in family interventionL
assessment and theory (Vol 1, pp. 129-179). Greenwich,
CT: JAI Press Inc.

 

Olson, D. H., Russell, C. S., 8 Sprenkle, D. H. (1983).
Circumplex model of marital and family systems: VI.
Theoretical update. Family Process, 33, 69-83.

Olson, D. H., Sprenkle, D. H., 8 Russell, C. S. (1979).
Circumplex model of marital and family systems: I.
Cohesion and adaptability dimensions, family types, and
clinical applications. Family Process, l5, 3-27.

Parish, T. (1981). Young adults' evaluations of themselves
and their parents as a function of family structure and
disposition. Journal of Youth and Adolescence, l5,
173-178.

 

67

Parish, T., 8 Dostal, J. (1980). Evaluations of self and
parent figures by children from intact, divorced, and
reconstituted families. Journal of Youth and
Adolescence, 5, 347-351.

 

Parish T. S., 8 Nunn, G. D. (1981). Children's
self-concepts and evaluations of parents as a function
of family structure and process. The Journal of
Psychology, 107, 105-108.

 

 

Parish, T., 8 Parish, J. (1983). Relationship between
evaluations of one's self and one's family by children
from intact, reconstituted, and single-parent families.
Journal of Genetic Psychology, 143, 293-294.

Parish, T., 8 Philip, M. (1982). The self-concepts of
children from intact and divorced families: Can they
be affected in school settings? Education, 103, 60-63.

 

Parish, T., 8 Taylor, J. C. (1979). The impact of divorce
and subsequent father absence on children's and
adolescents' self-concepts. Journal of Youth and
Adolescence, 5, 427-432.

 

 

Piers, E. V. (1984). Piers-Harris children's self-concept
scale: Revised manual 1984. Los Angeles, CA: Western
Psychological Services.

 

Portner, J. (1980). Parent-adolescent interaction of
families in treatment. Unpublished doctoral
dissertation, University of Minnesota.

 

 

Raschke, H. J., 8 Raschke, V. J. (1979). Family conflict
and children's self-concepts: A comparison of intact
and single-parent families. Journal of Marriage and
the Family, 3;, 367-374.

 

Rosenberg, M. (1965). Society and the adolescent
self-image. Princeton, NJ: Princeton University
Press.

 

 

Santrock, J. W., Warshak, R., Lindbergh, C., 8 Meadows, L.
(1982). Children's and parents' observed social
behavior in stepfather families. Child Development,
55, 472-480.

 

Satir. V. (1964). Conjoint family therapy. Palo Alto, CA:
Science and Behavior Books.

Sigafoos, A., Reiss, D., Rich, J., 8 Douglas, E. (1985).
Pragmatics in the measurement of family functioning.
Family Process, 35, 189-203.

 

68

Stanton, M. D. (1981). Strategic approaches to family
therapy. In A. S. Gurman 8 D. P. Kniskern (Eds.),
Handbook of family therapy (pp. 361-402). New York:
Brunner/Mazel.

Visher, E., 8 Visher, J. (1979). Stepfamilies: A_guide to
working with stepparents and stepchildren. New York:
Brunner/Mazel.

Wald, E. (1981). The remarried family: Challenge and
promise. New York: Brunner/Mazel.

Wallerstein, J., 8 Kelly, J. (1980). Surviving the
break-up. New York: Basic Books.

Walsh, F. (1982). Conceptualizations of normal family
functioning. In F. Walsh (Ed.), Normal family
processes, (pp. 3-42). New York: Guilford Press.

 

Wilson, K. L., Zurcher, L. A., McAdams, D. C., 8 Curtis, R.-
L. (1975). Stepfather and stepchildren: An
exploratory analysis from two national surveys.

Journal of Marriage and the Family, 51, 526-536.

APPENDICES

APPENDIX A

LETTER TO PARENTS AND STEPPARENTS

PUBLIC SCHOOLS

APPENDIX A

MICHIGAN STATE UNIVERSITY
Department of PsyChology East Lansing, Michigan 48824
Psychology Research Building

Dear Parents and Stepparents:

We are social scientists at Michigan State University very interested in what
makes families work well and how this relates to children's optimal development.
We do not know as much as we need to about how children see their families
working and how they see themselves. In addition, children now live in many
different kinds of family arrangements, and we also need to know more about

how the children in these many circumstances understand their families and
themselves. One family arrangement that we especially lack information about

is stepfamilies. Thus, we are looking to enlist the help of families in which
the children live with 1) both biological parents, 2) their mother and stepfather,
or 3) their father and stepmother, to help us gain more information.

We anticipate this information will be useful in two ways. First, knowledge of
the processes in all families that relate to the highest functioning in children
will help parents and professionals know what is important to encourage in
family life. Secondly, because the number of stepchildren is increasing quite ,
rapidly and so little is known about them, the demand for information has
heightened. This study can further educate parents, professionals, and the
public about what stepchildren in particular experience.

In order to participate, your child(ren) or stepchild(ren) must be in first
through sixth grade. For stepfamilies, the spouses must be remarried fbr at
least the past six months. If you would like to consider volunteering, we
want you to know what will be asked of your family. We will have your child
complete a few questionnaires. These would take about l5 hours to do. We
will also ask you to complete a short questionnaire giving us some background
information about your family. This would take about 15 minutes. The
questionnaires would be personally administered at your home at a time that
is convenient for you and your children. The information you provide would
be held in the strictest confidence. No names will be written on the
questionnaires. You and your child(ren) would be free to withdraw from the
project at any time. If desired, we will fbrward to you a summary of the
results of the project.

If you are interested in participating in this project, or want more
information before deciding whether or not to participate, please complete
the reply form on the back of this letter and have your child return it to
his or her teacher as soon as possible. He will contact you to answer your
questions and/or arrange for the completion of the questionnaires.

We very much need and would appreciate your help.

       

//

Gary Stollak, Ph.D.
Professor of Psychology

 

69

 

 

 

 

70

We would like to learn more about or would like to

participate in the project on families and children.

Your name(s)

 

Your child or children's name(s) and age(s)

Namefs) Agegs)

Your telephone number(s)

 

APPENDIX B

LETTER TO PARENTS AND STEPPARENTS

STEPFAMILY ASSOCIATION OF AMERICA

APPENDIX B
MICHIGAN STATE UNIVERSITY

 

DEPARTMENT OF PSYCHOLOGY EAST LANSING - MICHIGAN - 48824-1!”
PSYCHOLOGY RESEARCH BUILDING

Dear Parent and Stepparent:

We are social scientists at Michigan State University interested in stepfamilies
and stepchildren. As you probably know all too well, very little is known about
stepfamilies. In particular, we need to learn more about how stepchildren perceive
their families and themselves. We anticipate that the kind of information this
project gathers will be useful in increasing the understanding of stepfamilies by
both professionals and parents.

We are looking for families to help us increase our knowledge of the life of
stepchildren. In order to be eligible for this project, your child or children
must be in first through fifth grade, and you must be remarried for at least the
past six months. If you would like to consider volunteering, we wish you to know
what will be asked of your family. We will have your child complete a few
questionnaires. These would take about lg hours to do. We will also ask you to
complete a short questionnaire giving us some background information on your family.
This would take about 15 minutes. The questionnaires would be personally
administered at your home at a time that is convenient for you. The information
you provide would be held in the strictest confidence. No names will be written
on the questionnaires. You and your child(ren) would be free to withdraw from the
project at any time. If desired, we will forward you a summary of the results of
the project.

If you are interested in participating in this project, or want more information
before deciding whether or not to participate, please check the box indicating
this below and complete the reply form. We will contact you as soon as possible
to answer your questions and/or arrange for the completion of the questionnaires.
If you do not want to participate, please check the box indicating that below. In
either case, please detach the reply form and return it in the enclosed self-

addressed envelope as soon as possible.

Gary S 011 flak PhD
Professor of Psychology

Your consideration of this project is most appreciated.

     

 

We would like to participate in the project on families and children.

 

 

We would like more information.

 

 

 

Your name(s)

 

Your telephone number

 

 

We do not wish to participate in the project.

 

 

 

71

APPENDIX C

FACES-II

APPENDIX C

FACES‘II

Read the child the following directions:

I would like to know how much certain things happen in your family [two
families]. [If stepchild, add "I will ask you about how things are when
you are at your mom and stepdad's house and when you are at your dad and
stepmom's house."] I will flgp_talk to your parents about what you tell
me, so please say what you really think.

The research assistant explains the procedure and does a practice question
with the child before reading each item.

I. Family members are supportive of each other during difficult times.
(supportive - listen to each other, help each other out when there
are problems)

2. It is easy for everyone in the family to express his/her opinion.
(express - say what he/she thinks, give his/her ideas)

3. It is easier to discuss problems with people outside the family than
with other family members. (discuss-—talk over, talk about; outside
the family - friends, neighbors, teachers)

4. Each family member has input in major family decisions. (has input -
has a say, everyone decides what to do; major - big)

5. Our family gathers together in the same room. (gathers - gets
together, spends time together in same room)

Children have a say in their discipline. (discipline - punishment)
Our family does things together.

8. Family members discuss problems and feel good about the solutions.
(solutions - the way you solved it, what you figured out to do)

9. Everyone goes his/her own way. (goes own way - does things alone)

l0. Ne shift household responsibilities. (shift - change; responsibilities
- chores (garbage, dishes, pets)

ll. Family members know each others close friends.
12. It is hard to know what the rules are.

l3. Family members consult each other on decisions. (consult - ask each
others' ideas)

l4. Family members say what they want.

15. Our family has difficulty thinking of things to do.

l6. The children's suggestions are followed. (suggestions - ideas)
17. Family members feel very close to each other.

l8. Discipline is fair in our family.

19. Family members feel closer to people outside the family than to other
family members.

72

20.

2l.
22.
23.
24.
25.
26.

27.
28.
29.

30.

73

Our family tries new ways to deal with problems. (deal with -
work out)

Family members go along with what the family decides to do.
Everyone shares responsibilities.

Family members like to spend their free time with each other.

It is difficult to get a rule changed. (difficult - hard)

Family members avoid each other at home. (avoid - stay away from)

If problems arise, we compromise. (arise - come up; compromise -
everyone gives in some)

Ne approve of each other's friends. (approve - like)
Family members are afraid to say what's on their minds.

Family members pair up rather than do things as a total family.
(total family - all together)

Family members share interests and hobbies. (interests and hobbies -
like to do the same things (sports, reading, music, video games))

—"U

PSQNONU'IDOJN

_a
O

ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST
ALMOST

NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER

 

ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE
ONCE

IN A
IN A
IN A

H
2
>>>>>>>>>>>>>>

H
2
>>>>

FACES - II

WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE
WHILE

WHILE

SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES

74

PRETTYW
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY

CODE
HOUSEHOLD

MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH

MUCH

PRETTY MUCH

PRETTY

MUCH

NH

ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS
ALMOST ALWAYS

ALMOST ALWAYS

75

5336

ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS

 

ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS
ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS
ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS
ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS
ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS
ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS
ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS
ALMOST NEVER ONCE IN A WHILE SOMETIMES PRETTY MUCH ALMOST ALWAYS

APPENDIX D

PARENT PERCEPTION INVENTORY

APPENDIX D

PARENT PERCEPTION INVENTORY
(MY PARENTS AT HOME)

Read the child the following directions:

WE WOULD LIKE TO KNOW HOW MUCH YOU THINK YOUR MOM, YOUR DAD (YOUR
STEPMOM AND YOUR STEPDAD) DO CERTAIN THINGS AT HOME. WE WILL NOT TALK
TO YOUR PARENTS ABOUT WHAT YOU TELL US, SO PLEASE TELL US WHAT—YOU
REALLY THINK.

LET'S TRY A PRACTICE QUESTION:
HOW OFTEN DOES YOUR MOM CLEAN HOUSE?
DOES SHE CLEAN IT NEVER, A LITTLE, SOMETIMES, PRETTY MUCH, OR A LOT?

(Point to each word as you say it. For younger children, use the
thermometers to give further explanation if necessary. For example,
"THESE THERMOMETERS SHOW HOW OFTEN YOUR MOM DOES SOMETHING. IF SHE
NEVER DOES SOMETHING, THE THERMOMETER IS EMPTY. IF SHE DOES SOMETHING
SOMETIMES, THE THERMOMETER IS HALF FULL. IF SHE DOES SOMETHING A LOT.
THE THERMOMETER 15 FULL.")

CIRCLE THE WORDS (for younger children, "UNDER THE THERMOMETER”) WHICH
TELL HOW OFTEN YOUR MOM CLEANS THE HOUSE.

(After the child has circled his/her answer, check to make sure that
he/she understands the task.)
SO YOUR MOM CLEANS THE HOUSE (child's answer)?

NOW WE'LL START.

(For each concept,

a. State the item number.

b. Ask "HOW OFTEN DOES YOUR MOM. . ."

c. Give examples until the child understands the concept.
For starred items, repeat the response choices (e.g. DOES SHE
NEVER, A LITTLE. SOMETIMES, PRETTY MUCH, OR A LOT“) as you point to
each response.

* l. (Positive reinforcement)
THANK YOU FOR DOING THINGS. TELL YOU WHEN SHE LIKES WHAT YOU DID.
GIVE YOU SOMETHING OR LET YOU DO SOMETHING SPECIAL WHEN YOU'RE

GOOD

* 2. (Privilege removal)
TAKE AWAY THINGS WHEN YOU MISBEHAVE (LIKE NOT LETTING YOU WATCH
TV OR RIDE YOUR BIKE OR STAY UP LATE OR EAT DESSERT)

3. (Comfort)
TALK TO YOU WHEN YOU FEEL BAD AND HELP YOU TO FEEL BETTER, HELP

YOU WITH YOUR PROBLEMS, COMFORT YOU

76

10.

ll.

12.

l3.

l4.

l5.

l6.

l7.

l8.

77

(Criticism)
TELL YOU YOU'RE GOOD. TELL YOU THAT YOU MESSED UP OR DIDN'T DO
SOMETHING RIGHT, CRITICIZE YOU

(Talk time)
TALK TO YOU, LISTEN TO YOU, HAVE A GOOD CONVERSATION WITH YOU

(Command)
ORDER YOU AROUND, TELL YOU WHAT TO DO, GIVE COMMANDS

(Involvement in decision-making)
LET YOU HELP DECIDE WHAT TO DO, LET YOU HELP FIGURE OUT HOW TO
SOLVE PROBLEMS ,

(Physical punishment)
SPANK YOU, SLAP YOU, HIT YOU

(Time together)
PLAY WITH YOU, SPEND TIME WITH YOU, DO THINGS WITH YOU WHICH YOU

LIKE

(Yelling)
GET MAD AT YOU, YELL AT YOU, HOLLER AT YOU, SCREAM AT YOU, SHOUT
AT YOU

(Positive evaluation)
SAY NICE THINGS TO YOU, TELL YOU THAT YOU'RE A GOOD BOY/GIRL,
COMPLIMENT YOU

(Threatening)
THREATEN YOU. TELL YOU THAT YOU'LL GET INTO TROUBLE IF YOU DO

SOMETHING WRONG, WARN YOU

(Allowing independence)
LET YOU DO WHAT OTHER KIDS YOUR AGE DO, LET YOU DO THINGS ON YOUR
OWN

(Time-out)
SEND YOU TO A ROOM OR CORNER WHEN YOU DO SOMETHING WRONG

(Assistance)
HELP YOU WHEN YOU NEED IT (WITH A HARD JOB, WITH HOMEWORK, WHEN

YOU CAN'T DO SOMETHING BY YOURSELF)

(Nagging)
NAG YOU, TELL YOU WHAT TO 00 OVER AND OVER AGAIN, KEEP AFTER YOU

TO DO THINGS

(Non-verbal affection)
HUG YOU, KISS YOU, TICKLE YOU, SMILE AT YOU

(Ignoring)
IGNORE YOU, NOT PAY ANY ATTENTION TO YOU, NOT TALK TO YOU OR LOOK
AT YOU

{56335

NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER
NEVER

NEVER

A
A
A
A
A
A
A
A
A
A
A
A

>>>>>>>

LITTLE
LITTLE
LITTLE
LITTLE
LITTLE
LITTLE
LTTLE

LITTLE
LITTLE
LITTLE
LITTLE
LITTLE
LITTLE
LITTLE
LITTLE
LITTLE
LITTLE
LITTLE

LITTLE

SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES
SOMETIMES

78

PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY
PRETTY

PRETTY

\AJI)‘;

PARENT

MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH
MUCH

A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT
A LOT

APPENDIX E

PIERS-HARRIS CHILDREN'S SELF-CONCEPT SCALE

APPENDIX E

PIERS—HARRIS CHILDREN'S SELF—CONCEPT SCALE

Read the child the following directions:

Here are a set of statements that tell how some people feel about

I will read each statement and you decide whether or not it
If it is true or mostly true
If it is false or mostly false for you, say "no."
Do not say both
Remember that there are no right
Only you can tell us how you feel about yourself, so

themselves.
describes the way you feel about yourself.
for you, say “yes."
Answer every question, even if some are hard to decide.
"yes" and "no" for the same statement.
or wrong answers.

we hope you will mark the way you reall feel inside.

l. My classmates make fun of me ......................... yes
2. I am a happy person .................................. yes
3. It is hard for me to make friends .................... yes
4. I am often sad ....................................... yes
5. I am smart ........................................... yes
6. I am shy ............................................. yes
7. I get nervous when the teacher calls on me ........... yes
8. My looks bother me ................................... yes
9. When I grow up, I will be an important person ........ yes
lo. I get worried when we have tests in school ........... yes
ll. I am unpopular ....................................... yes
12. I am well behaved in school .......................... yes
13. It is usually my fault when something goes wrong ..... yes
l4. I cause trouble to my family ......................... yes
l5. I am strong .......................................... yes
l6. I have good ideas .................................... yes
l7. I am an important member of my family ................ yes
l8. I usually want my own way ............................ yes
l9. I am good at making things with my hands ............. yes
20. I give up easily .................................... yes
Zl. I am good in my school work .......................... yes
22. I do many bad things ................................. yes
23. I can draw well ...................................... yes
24. I am good in music ................................... yes
25. I behave badly at home ............................... yes
26. I am slow in finishing my school work ................ YES

79

no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no

27.
28.
29.
30.
3T.
32.
33.
34.
35.
36.
37.
38.
39.
40.
4T.
42.
43.
44.
45.
46.
47.
48.
49.
SO.
Sl.
52.
53.
54.
55.
56.
57.
58.
59.
60.
6I.

80

I am an important member of my class ................. yes
I am nervous ......................................... yes
I have pretty eyes ................................... yes
I can give a good report in front of the class ....... yes
In school I am a dreamer ............................. yes
I pick on my brother(s) and sister(s) ................ yes
My friends like my ideas ............................. yes
I often get into trouble ............................. yes
I am obedient at home ................................ yes
I am lucky ........................................... yes
I worry a lot ........................................ yes
My parents expect too much of me ..................... yes
I like being the way I am ............................ yes
I feel left out of things ............................ yes
I have nice hair ..................................... yes
I often volunteer in school .......................... yes
I wish I were different .............................. yes
I sleep well at night ................................ yes
I hate school ........................................ yes
I am among the last to be chosen for games ........... yes
I am sick a lot ...................................... yes
I am often mean to other people ...................... yes
My classmates in school think I have good ideas ------ YES
I am unhappy ......................................... yes
I have many friends .................................. yes
I am cheerful ........................................ yes
I am dumb about most things .......................... yes
I am good-looking .................................... yes
I have lots of pep ................................... yes
I get into a lot of fights ........................... yes
I am popular with boys ............................... yes
People pick on me .................................... yes
My family is disappointed in me ...................... yes
I have a pleasant face ............................... yes
When I try to Rake something, everything seems to

90 wrong ............................................. yes

no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
00
no
no
no
no
no
no

no

62.
63.
64.
65.
66.
67.
68.
69.
70.
7T.
72.
73.
74.
75.
76.
77.
78.
79.
80.

81

I am picked on at home ................................ yes
I am a leader in games and sports ..................... yes
I am clumsy ........................................... yes
In games and sports, I watch instead of play .......... yes
I forget what I learn ................................. yes
I am easy to get along with ........................... yes
I lose my temper easily ............................... yes
I am popular with girls ............................... yes
I am a good reader .................................... yes
I would rather work alone than with a group ........... yes
I like my brother (sister) ............................ yes
I have a good figure .................................. yes
I am often afraid ..................................... yes
I am always dropping or breaking things ............... yes
I can be trusted ...................................... yes
I am different from other people ...................... yes
I think bad thoughts .................................. yes
I cry easily .......................................... yes
I am a good person .................................... yes

no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no

APPENDIX F

MARRIAGE AND FAMILY QUESTIONNAIRE

APPENDIX F

 

 

Code
MARRIAGE 5 FAMILY QUESTIONNAIRE
a. Husband's age Husband's birth date
b. Wife's age Wife's birth date

 

a. Husband's occupation

 

b. Wife's occupation

 

Circle the highest level of formal education each of you has completed.

 

 

High Graduate or

Grade School School College Professional

a. Husband l 2 3.4 5 6 7 8 l 2 3 4 l 2 3 4 1 2 3 4 5+
b. Wife 1 2 3 4 5 6 7 8 l 2 3 4 l 2 3 4 l 2 3 4 5+

Circle the number which comes closest to your family's total yearly
income before taxes.

a. less than $15,000 d. $30,000 - $39,999
b. $15,000 - $19,999 e. $40,000 - $49,999
c. $20,000 - $29,999 f. over $50,000

Circle the number corresponding to your race.

'flhigg ‘glagg Hispanic Oriental Other (specify)
a. Husband
b. Wife

Length of current marriage (in months)

 

82

83

Please answer questions 7-9 if you are a previously married man'and questions
10-12 if you are a previously married woman. Otherwise, go to question 13.

HUSBAND

7. Number of previous marriages

 

8. First marriage

a.

b.

Co

(check one) ended in divorce or death of spouse

length of marriage (in months, from date of marriage to physical
separation or death of spouse)

 

length of time between separation or death and remarriage
(in months)

 

9. If applicable, answer questions a-c above for subsequent marriages.

WIFE

10. Number of previous marriages

Second Marriagg. Third Marriage

 

 

 

 

 

 

 

 

11. First marriage

12. If

(check one) ended in divorce or death of spouse

length of marriage (in months, from date of marriage to physical
separation of death of spouse)

 

length of time between separation or death and remarriage
(in months)

 

applicable, answer questions a-c above for subsequent marriages.
Second Marriagg' Third Marriage

 

 

 

 

 

 

13.

84

Please indicate the first name, sex, date of birth (DOB) and age of all
children from your current marriage and, if applicable, from.previous
marriages. Also, please indicate with whom each child lives.

Which marriage Who does child
First Name Sex DOB is child from? live with?

 

 

 

 

 

Please answer questions 14-17 if you have a stepchild or stepchildren.

14.

Your child (or children) participating in this study is (are) part of one
or two households. For each participating child, list the first name, ‘

check which adults are present and indicate the total number of children

for each household that child lives in.

Mother's Household

 

 

 

 

 

 

 

First Name Mother Stepfather # of children
1.
2.
3. ‘

Father's Household

First Name ’ Father Stepmother # of children
1.
2.
3.

 

 

 

15.

85

For each child participating in the study, indicate the following

information, as applicable:

Length of time (in

 

 

 

Age at months) between
death or separation and
separation Age at remarriage remarriage of
First Name of parents of motherjof father ofggptherlof father
I I I
1. ' I
I
I
2. ' l
I
3. l l
l » l
i l

 

Questions 16 and 17 concern visitation patterns.

16.

17.

If there has been a scheduled and regular pattern of visitation during the
last six months, how many days did the child's other parent see the child
in a typical month? How many nights did the child sleep over at his or

her other parent's home?

First Name Days of visitation/month Overnight visits/month

 

 

 

If there has not been scheduled and regular visitations, please estimate the
total number of days in the last six months each child had a visit from

the non-custodial parent.

First Name

 

Days of visitation/6 months

APPENDIX G

CONSENT FORM

 

APPENDIX G
CONSENT FORM

Please read the following statements regarding the study. If you agree with
the statements please sign below.

1. I agree that I am participating voluntarily in a project to study children's
perceptions of their families and themselves.

2. My participation will involve filling out a questionnaire.

3. I understand that, at my request, I can receive additional explanation of
the study after my participation is completed.

4. I understand that I am free to decline to participate and may terminate my
participation at any time without any consequence to me.

5. I am willing to allow the child(ren) in my family to be administered
questionnaires by project personnel, and I have had the contents of these
questionnaires explained to me.

6. I understand that my child(ren) is (are) free to decline to participate and
may terminate participation at any time without any consequence.

7. I understand that my child(ren) will receive an explanation of the project,
the participation involved and confidentiality.

8. I understand that any information collected will be used for scientific and
educational purposes only and in ways that will preserve anonymity. Your
names will not appear on any document reviewed by present or future project
personnel nor in any public reports or documents. The project will protect
the confidentiality of all family members.

Name

 

Date

 

I would like a summary of the results.

Name:

 

Address:

 

 

 

86