MARITAL ADJUSTMENT AND INTERACTION, _
RELATED To INDIVIDUAL ADJUSTMENT ‘
0F SPOUSES IN CLINIC AND ‘
NON - CLINIC FAMILIES

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
  KEES C. HOFMAN
1969

    
 
    

LIBRARY

THESIS Michigan State
L, University

This is to certify that the

thesis entitled

MARITAL ADJUSTMENT AND INTERACTION,
RELATED TO INDIVIDUAL ADJUSTMENT
OF SPOUSES IN CLINIC AND NON-CLINIC

FAMI LI E 8
presented by

Kees C. Hofman

has been accepted towards fulfillment
of the requirements for

_BD_._D_.._ degree in _Es¥chnlogy

 

 

Date %//7//70

0-169

 

 

ABSTRACT

MARITAL ADJUSTMENT AND INTERACTION,
RELATED TO INDIVIDUAL ADJUSTMENT
OF SPOUSES IN CLINIC AND
NON-CLINIC FAMILIES

By

Kees C. Hofman

NRarriage assessment primarily consisted of self-
repcurt :instruments, as represented by the Locke-Wallace
ScaJxa <of‘ Marital Adjustment (LW) and the Family Concept
Inverytcxry (FCI), until the recent innovative methodology
of otnsetfving spouses interact under laboratory conditions.
The Eyrefisent project used both approaches in an investiga_
tion (1f Inarital adjustment and its concomitants.

Inter-

acticnq rneasures used were: (a) spontaneous agreement

(SA); (t)) choice fulfillment (CF); (c) dominance (DOM);
(d) Checjxsion time (DT); and (e) interpersonal competence
(IC). 131 addition, the Tennessee Self Concept Scale
(TSCfiS) teas used to assess individual adjustment.

1X clinic (C) sample of 15 couples was constituted
of palwants of children who had been referred to an out-
patierfl: clinic with emotional difficulties. A same-sized
non—tfiLinic (NC) sample was recruited from couples desig-

nated.zn3 "well-adjusted" in their marriages by local

Kees C. Hofman

ministers. Selection criteria were that all couples be
literate, caucasian, married at least A years, and have at
least 1 child. In addition, no member of NC families was
to have been referred at any time for mental health care.

All spouses (N = 60) separately completed the Lw,
FCI, and T808, and ranked solutions to 3 hypothetical
family situations dealing with finances (task 1), parent-
child relations (task 2), and birth control measures (task
3). Upon completion of these tasks, spouses were requested
to resolve their differences on tasks 1, 2, and 3. The
ensuing discussions were tape-recorded and independently
rated by two judges to obtain IC scores (average agreement
was 81.2%). SA was defined as the rank—order correlation
between spouses' initial rankings. CF was the rank-order
correlation between spouses' initial ranking and the joint
ranking. DOM was derived from CF and SA. DT was time
needed for joint completion of tasks 1, 2, and 3.

It was hypothesized that: (a) individual adjustment
of spouses is highly positively correlated; (b) individual
adjustment is positively related to marital adjustment as
measured by the LW, FCI, and interaction indices (posi-
tively with SA, CF, and IC, but negatively with DT); (c)
Lw, FCI, SA, CF, and IC are positively inter-related, and
negatively correlated with DT; (d) the TSCS, Lw, FCI, and
interaction measures differentiate the C and NC samples

at statistically significant levels.

Kees C. Hofman

Results showed a statistically significant positive
correlation between spouses' TSCS adjustment as well as
maladjustment scores. Individual adjustment was signi-
ficantly and positively correlated with Lw, FCI, and, to
a small degree, with SA. Other interaction indices were
minimally and variably related to individual adjustment
scores. More pronounced correlations obtained between
interaction measures (especially SA and DT) and TSCS
response-style measures. Husband's (a) tendency to acquiesce
with positive statements, (b) degree of conflict in, and
(c) certainty of, self-perception were related to lower
SA and high DT scores. These relationships were less
pronounced between wives' response—style and interaction
measures. Lw and FCI correlated highly, but did not con-
sistently correlate as predicted with interaction indices.
Male Lw scores were positively correlated with SA on task
3, while female FCI scores were negatively related to DT
scores, also on task 3.

The C and NC samples differed significantly on most
measures, except the interaction indices. The NC sample
scored significantly higher on 9 of 10 TSCS adjustment
scales (and significantly lower on 5 of 6 maladjustment
scales) as well as on the Lw and FCI. The FCI discrimi-
nated more clearly, with less overlap, than did the Lw.

NC mean SA scores were significantly higher for

task 3 and for all tasks combined, but not for tasks 1

Kees C. Hofman

and 2. An analysis of variance yielded significant Task
and Task X Group effects. Neither the CF nor the DT mea-
sure differentiated between the samples. Similarly, there
was no difference on the DOM measure, although all couples
tended to show husband dominance, relative to wife domi—
nance or compromise, on task 1. When summing across tasks
there was no preferred dominance pattern. A very consist-
ent finding was that tasks 1, 2, and 3 were qualitatively
different and that couples' interaction varied significantly
from task to task. There were no sample differences on IC.
However, the C couples were assigned significantly more
feeling scores, relative to idea scores, than were assigned
to NC couples. This was interpreted to suggest that NC
couples' discussion remained more task-oriented than did

C couples' discussion.

A significant (p < .01) inverse relationship was
found between FCI and length of marriage. Most pronounced
when discussing birth control measures (task 3) was a
rather stable positive correlation between number of
children and dominance of husband.

Conclusions drawn were:

(a) the results supported the homogamy theory; i.e.,
spouses tend to be equally well- or mal-adjusted.

(b) people who were individually well adjusted also tend
to be better adjusted in, and more satisfied with,
their marriage as compared to less well adjusted

people.

Kees C. Hofman

(c) questionnaire methods, e.g., the FCI and to a lesser
degree the Lw, more accurately distinguish between
adaptive and maladaptive marriages, than do those
interaction indices used in this project.

(d) interpersonal processes, as measured by the present
interaction indices, are to a significant degree
dependent on task content.

Suggestions for further research were discussed.

MARITAL ADJUSTMENT AND INTERACTION,
RELATED TO INDIVIDUAL ADJUSTMENT
OF SPOUSES IN CLINIC AND

NON-CLINIC FAMILIES

By

Kees CC Hofman

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Psychology

1969

64.17 22,

7—7-70

TO MY IMMEDIATE FAMILY

PA, MOE, HANS, JOHN, WILL, NELLY, and MARTHA

ii

ACKNOWLEDGMENTS

The author wishes to express his sincere gratitude
and respect to Prof. John Hurley, thesis and guidance
committee chairman, for his interest, help, friendship,
and gentle prodding toward autonomy.

Appreciation is also extended to Professors Dozier
Thornton, Lucy Ferguson, and Jay Artis, members of the
thesis and guidance committee, for their interest and
constructive efforts.

To the staff of the Genesee County Community Mental
Health Clinic, Rev. A. Hoksbergen and Rev. P. Birdwell,
thank you for your aid in making this project possible.

A special word of thanks is due Miss Lana Ribble
for her aid in various phases of this project, especially
the preparation of the typescripts and general moral

support.

iii

TABLE OF CONTENTS

Page

DEDICATION . . . . . . . . . . . . . ii

ACKNOWLEDGMENTS . . . . . . . . . . . iii

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

LIST OF APPENDICES . . . . . . . . . . ix

Chapter

I. INTRODUCTION . . . . . . . . . . 1

Relevant Literature. . . . . . . 3

Individual Adjustment of Spouses. . 3

Marital Happiness. . . . . . . 7

Interaction Studies . . . . . 12

Summary of Literature Review . . . 22

Purpose. . . . . . . . . . . 23

II. METHOD . . . . . . . . . . . . 2A

Instruments . . . . . . 24

Tennessee Self Concept Scale . . . 2A
Locke-Wallace Scale of Marital

Adjustment . . . . . 26

Family Concept Inventory . . . . 28

Revealed Differences Task . . . . 29

Decision Efficiency Measures . . 31

Discussion Ratings. ‘. . . . . 32

Subjects . . . . . . . . . . 3A

Procedure . . . . . . . . . . 36

III. HYPOTHESES. . . . . . . . . . . 37

IV. RESULTS. . . . . . . . . . . . 39

Demographic Variables . . . . . . 39

Marital Adjustment Scales. . . . . 39

Tennessee Self Concept Scale. . . A3

Interaction Scores . . . . . . . A7

Spontaneous Agreement . . . . . A7

Choice Fulfillment . . . . . . 51

Decision Time . . . . . . . . 52

Dominance . . . . . . . . . 53

iv

Chapter

Discussion Ratings
Interpersonal Competency
Interspouse Correlations . .
Relationships among Variables . .
Demographic Variables and Marital
Adjustment Scores
Demographic Variables and Tennessee
Self Concept Scale.
Demographic Variables and Interaction
Scores. . . .
Marital Adjustment and Tennessee Self
Concept Scale .
Marital Adjustment and Interaction
Scores. . .
Tennessee Self Concept Scale and
Interaction Scores.

V. DISCUSSION . . . . . . . . . . .

Selection of Subjects
Marital Adjustment . . . .
Tennessee Self Concept Scale. . . . .
Interaction Scores . . . . . . . .
Spontaneous Agreement
Choice Fulfillment . . . .
Decision Time . . . . . . . . .
,Dominance . . . . .
Interpersonal Competency .
Intercorrelation of Interaction Scores.
Interspouse Correlations . . . . .
Relationships among Variables . .
Demographic Variables and Marital
Adjustment .
Demographic Variables and Interaction
Scores. .
Marital Adjustment and Tennessee Self
Concept Scale
Marital Adjustment and Interaction

Scores. . .
Tennessee Self Concept Scale and Inter-
action Scores . . . . . . . .
VI. CONCLUSIONS . . . . . .

Suggestions for Further Research
REFERENCES . . . . . . . . . . . . .
APPENDICES . . . . . .

103
105
109
117

Table’

10.

ll.

l2.

13.
1A.

LIST OF TABLES

Means, Standard Deviations and Reliability
of TSCS Scores for Three Groups. . . .

Correlations between Husband's and Wife's
Marital Satisfaction Scores and between
the Locke—Wallace and Family Concept
Instruments . . . . . . . .

Categories of Behavior Related to Inter-
personal Competency. . . . . . .

Demographic Variables, Means and Standard
Deviations. . . . . .

Marital Adjustment Scores, Means and
Standard Deviations. . . . . . .

Analysis of Covariance, Family Concept
Scores with Number of Years Married as
Covariate . . . . . . . . . .

Correlations between Locke-Wallace Scale
and Family Concept Inventory. . .

Tennessee Self Concept Scale, Means and
Standard Deviations, All Subjects

Tennessee Self Concept Scale, Means and
Standard Deviations, Males . .

Tennessee Self Concept Scale, Means and
Standard Deviations, Females. . . .

Correlations among Tennessee Self Concept
Scale Scores; All Subjects . . . . .

Spontaneous Agreement, Means and Standard
Deviations. . . . . . . . . .

Spontaneous Agreement, Analysis of Variance
Choice Fulfillment, Means and Standard

Deviations. . . . . . . . . .

vi

Page

27

3O

33

A0

A1

A2

43

AA

“5

A6

A8

A9
50

51

Table Page

15. Choice Fulfillment Scores, Analysis of
Variance. . . . . . . . . . 52

16. Decision Time, Means and Standard

Deviations . . . . . . . . . . . 53
17. Decision Time, Analysis of Variance . . . 53
18. Frequencies of Dominance Patterns . . . . 55
19. Dominance Scores, Analysis of Variance . . 55

20. Interpersonal Competency, Means and
Standard Deviations . . . . . . . . 57

21. Interpersonal Competency, Analysis of
Variance. . . . . . . . . . . . 57

22. Frequencies of Interpersonal Competency
Ratings . . . . . . . . . . . . 58

23. Correlations among Interaction Total
Scores . . . . . . . . . . . . 59

2A. Correlations among Interaction Scores for

each Task . . . . . . . . . . . 6O
25. Pattern of Couples' Number of Deviant

Signs Scores . . . . . . . . . . 62
26. InterSpouse Correlations . . . . . . . 62

27. Similar versus Dissimilar Couples in Both
Groups . . . . . . . . . . . 63

28. Correlations between Demographic and
Marital Adjustment Measures . . . . . 65

29. Correlations between Demographic Variables
and Tennessee Self Concept Scale, All
Subjects. . . . . . . . . . . . 66

30. Correlations between Demographic Variables
and Interaction Scores . . . . . . . 68

31. Correlations between Marital Adjustment
and Tennessee Self Concept Scale. . . . 69

32. Correlations between Marital Adjustment

and Tennessee Self Concept Scale,
(Continued). . . . . . . . . . . 7O

vii

Table Page

33. Correlations between Marital Adjustment
and Interaction Scores. . . . . . . 72

3A. Correlations between Individuals' Marital
Adjustment and Interaction Scores . . . 73

35. Correlations between selected Male TSCS
Interaction Total Scores . . . . . . 7A

36. Correlations between Selected Male TSCS
Interaction Scores . . . . . . . . 75

37. Correlations between Selected Female TSCS
Interaction Total Scores . . . . . . 76

38. Correlations between Selected Female TSCS
Interaction Scores . . . . . . . . 77

39. Item Rank Frequency . . . . . . . . 85

viii

LIST OF APPENDICES

Appendix

A. Instructions and Tasks Administered to
Subjects.

Revealed Differences Tasks . . .
Locke-Wallace Scale
Family Concept Inventory.

B. Scores of All Subjects.

Individual Scores
Couple Scores

ix

Page

118
120
12A
126
I29

130
133

CHAPTER I

INTRODUCTION

Although marriage is one of the most important human
relationships (in terms of time and emotional involvement)
it has been relatively ignored by psychology. This is not
true of sociology and anthropology, which have a long his-
tory of investigating the marital relationship. One of the
major factors of interest in sociological research on
marriage has been mate selection. The increasing divorce
rate stimulated the investigation of marriage, among these
most notably the ambitious project of Terman in the 1930's
(See also Burgess and Wallin, 195A), and necessitated some
means of assessing the quality of the marriage. An early
construct designed to meet this need was "marital happi-
ness." Davis (1929) asked subjects: "Is your marriage
happy or unhappy." This early example was followed by
several self—report measures which purportedly measure
marital "happiness," "adjustment," or "success." Tharp
(1963) identified four additional approaches to the
assessment of marriage. These are: a. interpersonal
perception (i.e. marital adjustment depends on degree of

congruency between perception of ideal and actual

behavior of spouse); b. identification (i.e. adjustment
depends on congruency between spouse's actual behavior and
the idealized opposite-sex parent image); c. complementar-
ity of psychic needs (i.e. adjustment depends on the "fit"
of the psychological needs between spouses); and d. role
theory (i.e. adjustment depends on similarity of percep-
tion of spousal roles). In recent years, another approach
to the investigation of marriage has emerged in the form
of interaction studies. This approach appears to be a
manifestation of the increasing awareness of the limita-
tions of the intrapsychic point of view in social science
and the attempt to conceptualize interpersonal behavior as
system. Framo (1966) called this approach the "transper-
sonal view," which he defined as ". . . people really do
have an effect upon one another when they are in close
relationship, a telling effect which is more than the
resultant of two interacting intrapsychic systems (p. A51)."
This approach emphasizes the desirability of observing
.actual behavior between spouses for the evaluation of a
marriage.

The purpose of the present study is to combine
several of these approaches in the comparison of the marri-
ages of clinic and non-clinic subjects by: (a) investigat-
ing the similarity between spouses on an overall measure of
mental health; (b) analyzing the reported marital happiness

or adjustment; (c) observing and analyzing marital

behavior in response to a problem-solving task; and (d)

investigating the relationships among these variables.

Relevant Literature

Individual Adjustment of Spouses

 

The thesis of the present study is that spouses tend
to occupy similar positions on a global measure of mental
health; namely, self-esteem. Previous research both
supports and questions this proposition. Burgess & Wallin
(195A) state: "There is a strong tendency for persons
with neurotic traits to be engaged to others like them-
selves. Correspondingly, non—neurotics unite with non-
neurotics (p. 115)." Tharp (1963) commented on this early
research by stating that the correlations, although posi-
tive, were of low order. Fisher & Fisher (1967) after an
extensive review of the literature and analysis of their
own results in a sample of 119 couples found only moder-
ate support for the homogomy theory (i.e. "likes marry
likes"), and commented: "Analysis of the family data
collected revealed a surprisingly variable pattern of
correlations between spouses (p. 119)."

The research of Winch and his associates (summarized
in Winch, 1958) was designed to disprove the homogomy of
psychic needs theory. Tharp (1963), after critically
evaluating their research methods and results, discredited

the complementary need theory of Winch and summarized the

research by stating that mate selection ". . . is a func-
tion of unknown psychological variables (p. 115)."
Similarly, Murstein (1961) stated: "The heterogeneity-
homogomy dichotomy seems a gross simplification of the
actual marital situation. The assumption seems more
plausible thus, that for adequate marital adjustment some
needs require complementary needs in the marital partners,
while others necessitate homogomous patterns (p. 196)."

It should be noted that much of the homogomy-heter-
ogomy of psychic needs controversy is in terms of person-
ality traits or needs. It is the purpose of this study to
deal with a global measure of mental health rather than
traits or personality characteristics. Traditionally, the
"healthy" person has been described as a flexible, pro-
blemrsolving individual who does not occupy an extreme
position on any dimension of most personality inventories
since most traits, when carried to an extreme, become
maladaptive. The less well—adjusted individual, on the
other hand, is usually described as a person with age-
inappropriate "needs" or rigid behavioral patterns which
are‘used at the exclusion of others and therefore result
in GXtrenm scores on one or more scales of personality
inVentcmdes. Most of the research has focused on the
traits cm*persona1ity characteristics of spouses. The
preset”: study tends to focus on an overall evaluation of

the "Huaturity," "adjustment" or "mental health" of the

spouses with the hypothesis that spouses tend to occupy
similar positions on an overall measure of health.

This position is also advanced by Satir (196A) who
posits that a person with low self-esteem seeks to bolster
it through approval from a marital partner; who, in turn,
needs a "weak" partner in order to build his/her own self-
esteem. Similarly, Bowen (1960) stated that partners with
equal levels of immaturity tend to marry, even though they
frequently rely on opposite defensive systems. Lidz,
Cornelison, Fleck & Terry (1957) advanced a similar posi—
tion after investigating the marriages of parents of
schizophrenics. They described a variety of patterns of
marriage; however, all were marked by similar degrees of
maturity of spouses.

Research evidence which supports the proposition of
homogomy in terms of level of adjustment or mental health
is found in studies by Eschleman (1965), Dean (1966) and
Murstein (1967). Eschleman (1965) reported a modest posi-
tive relationship between mental health and marital
adjustment in a sample of 82 couples. Mental health was
Hwasured by the Adult Form AA of the California Test of
Per'SOnality and a symptom check list. Marital adjustment
wasrmeasured by interpersonal checklist perception dis-
cmepancy scores (after LaForge & Suczek, 1955) and seven
queStiJJns concerning satisfaction with the marriage (e.g.

would Sflou marry the same person, have you regretted your

marriage, etc.). Correlations among these variables were
all significant at beyond the .01 level, ranging from -19
to .57 in the predicted direction.

Dean (1966) also presented supportive data by
reporting positive interspouse correlations (3 is .A2) for
an emotional maturity scale (derived from the Dean Emo—
tional Maturity Scale) and marital adjustment scores
(3 is .59, derived from the Nye Scale of Marital Adjust-
ment). The emotional maturity scores and marital adjust-
ment scores of individuals were also modestly positively
correlated (r is .28 for males, 3 is .35 for females).

Murstein (1967) investigated courtship progress and
individual mental health scores in a sample of 99 couples.
Six months after the original interview and tests (pri-
marily the Minnesota Multiphasic Personality Inventory) a
follow—up interview assessed whether the relationship had
progressed in intensity or deteriorated. Mental health
was measured by the neurotic triad score, Welsh's
Repression and Anxiety Scales and Barron's Ego Strength
Scale (all based on the MMPI) as well as an overall rating
0f the MMPI profile. Murstein reported that those indi-
Viduals judged to be normal were more likely to be engaged
to another normal partner (p < .03). However, correlation
coef-‘ficients between partners for the previously mentioned
scales; were not significantly different from zero (3

rangirug from ,0A2 to .156). He also reported "only

moderate support" for the hypothesis that couples similar
in mental health would progress further in courtship than
dissimilar couples. An intriguing finding was that the
mental health of the male was related to a higher degree
with courtship progress than that of the female.

To sum up: the research intended to show that
spouses tend to be either homogamous or heterogamous in
psychological variables has yielded a "surprisingly
variable pattern" of results. The present author believes
that this may have been partly due to the instruments
used and also the failure to adequately conceptualize the
problem. It is the thesis of this research that spouses
tend to be homogamous in terms of overall level of adjust—
ment, while they might be homogamous or heterogamous in
terms of specific characteristics or traits. The instru—
ment used to investigate this hypothesis is the Tennessee
Self Concept Scale, which is described in the method

chapter.

Méfital Happiness

Marital "success" was identified by Tharp (1963) as
theISecond—most basic variable of interest in marriage
research (after the homogamy-heterogamy issue). One of
the eaifliest attempts to measure this construct was made
byllivis (1929) who asked subjects: "Is your marriage
happy or unhappy." Kirkpatrick (1937) identified marital

suCCeEHs with the social stimulus value of the marriage

when he evaluated marriages by asking acquaintances of
marital couples to rate that marriage. Terman (1938)
contended that marital success is a subjective phenomenon
which therefore can only be measured by self-report of
the experience of the marital partners.

Most current approaches to this variable of success
consist of self—report questionnaires in which the basic
question of Davis (1929) is still asked except that a
variety of aspects of the marriage are evaluated. An
example of this is the instrument developed by Locke &
Wallace (1959). This questionnaire (the LW), reproduced
in Appendix A (p. 12A) has been used extensively, also as
an independent variable. Basically, the approach of the
LW is to question subjects directly concerning their
expressed satisfaction with the marriage, the degree of
cooperation, conflict, and similarlity of attitudes
between them.

A somewhat more indirect procedure was developed by
van der Veen, Huebner, Jorgens, & Neja (196A). This
instrument, the Family Concept g sort, uses a g sort of
statements descriptive of family relationships, percep-
tions and attitudes. Degree of family success or adjust-
ment is assessed by the degree of similarity of the sub-
ject's family rating with that of the ideal family as
defined by its description, in terms of the same Q sort,

by a group of judges.

Recently a number of authors have taken issue with
this general approach of marriage assessment. Ryder,
(1966), e.g., stated:

It may be seen that a fair amount of work has gone
into the irrational goal of discovering the
correlates of successful marriage. What is a
successful or good marriage? Is it one that ends
in death rather than in divorce? Not exactly. Is
it one where husband and wife tell you that they
are satisfied? Perhaps. Success depends, does it
not, on what one believes marriage ought to be.
And what marriage ought to be is not a matter of
descriptive fact. One's findings must depend in
principle on whose ideals are incorporated into the
research. The difficulty cannot be resolved by
judging each couple under study in terms of their
own values, since this is simply smuggling in the
ideal that married couples should be whatever they
want to be (p. 565).

Bowman (1956) pleaded with researchers to abandon the
questionnaire method in the study of marriage dynamics and
marital adjustment. He challenged the assumption that the
dynamics of family roles and relationships are simple,
uncomplicated phenomena or that informants possess an
articulate understanding of these phenomena. Similarly,
Fontana (1966) states that retrospective studies assume
that:

(1) People conceptualize their lives in terms of
the language used by the investigator so their
understanding of the question is similar to that
of the investigator; (2) People can accurately
recall events and feelings of many years past
with minimal forgetting; (3) People will report
unpleasant events without selective forgetting,
defensive distortion and justification of actions
by inaccurate elaboration; and (A) People will
report past events unaffected by social desirabil-
ity or other response sets (p. 215).

10

Lively (1969) stated: "The conclusion is that the
continued use of 'marital happiness,' 'marital success' or
'marital adjustment' is detrimental to the development of
precise analyses and theoretical formulations of marriage
and interactional behavior (p. lAA)."

Three studies support this general disillusionment
with self-report measures. Kenkel & Hoffman (1956)
instructed couples to assume a gift of $300 and to discuss
how they might spend it. Before and after the actual dis-
cussion spouses were instructed to characterize their own
role in the discussion in terms of: (a) total number or
ratio of initiated actions; (b) number or percentage of
actions initiated consisting of giving ideas and sugges-
tions and (c) those actions which contributed to the smooth
functioning of the session. The actual discussion was
recorded and analyzed using Bales' interaction categories
(Bales, 1950). The authors reported low accuracy of
spouses in predicting or postdicting their own role.

Similarly, Olson (1968) compared self-report and
behavioral measures of power in marriages, by instructing
35 couples, all expecting a first child in the near
future, to answer a questionnaire consisting of 27 items
dealing with issues which needed to be decided soon after
the child was born. For each issue, §S were instructed
to: (a) predict who would make the decision; (b) state

who had the legitimate right to make the decision;

ll

(c) rate how well they understood spouse's preference on
that item and (d) state how relevant the item was to them
as a couple. The questionnaire was followed by a Revealed
Differences Task for the same items. Olson (1968)
reported no significant relationships between predicted
and actual power. Spouses did not differ along sex lines
in ability to predict decisions.

Weller & Luchterhand (1969) obtained ratings of
family functioning for 39 families from two separate
sources. The same categories were used in both ratings.
One rating was made by a family case-worker who had worked
with the family a median of 31 times. The case-workers
had not been informed in advance that these ratings would
be expected. A second rating was made by a social worker
who interviewed the same families specifically to gather
the necessary information for the ratings. A comparison
of the two ratings showed significant correlations in
only two of the eleven aspects of family functioning rated.
These were individual behavior and adjustment of the father
and the source and amount of income.

A number of experimenters have also reported the
striking tendency of marital partners to present a united
front, to the point of feigning agreement and deliberate
distortion (see below, Vidich, 1956; Ryder & Goodrich,

1966).

l2

Interaction Studies

The interaction studies escape some of the shortcom—
ings of retrospective methods. The interaction method is
characterized by direct observation and/or recording of
the interaction between marital partners. Usually this
occurs in a laboratory setting and a variety of tasks,
designed to stimulate interaction (most frequently verbal
discussion) have been used. Jackson (1967) attributed
the emergence of family interaction studies to "the
accident of conjoint family therapy (p. 36)"; i.e. because
therapists were seeing family units as a whole rather than
piecemeal in individual therapy. (See also Haley, 1962;
Riskin, 1968; Levinger, 1963; Rabkin, 1965.)

Another significant factor contributing to the rise
of the interaction study approach to family and marriage
dynamics was most likely the methodology developed by
small group research (see Cartwright and Zanders, 1962).
In reviewing this approach, Fontana (1966) identified the
following assumptions:

(a) . . . reaction patterns in the experimental
setting are the usual family patterns and the
subjects' usual behavior is not altered by the
knowledge that they are being studied by pro-
fessional experts; (b) reaction patterns remain
stable over time; (c) families react to the
experimental task as they characteristically
react to most tasks and (d) patterns remain the
same when some members are absent (p. 217).

There has been relatively little research on these assump-

tions. An exception is a study by Moore (1966) in which

13

the interaction of 15 families was analyzed on two occa-
sions, six to eight weeks apart. Moore (1966) interpreted
the results to indicate relative consistency in interaction
patterns over time.

The earliest marital interaction study in a labora-
tory setting appears to be the study by Strodtbeck (1951).
Spouses were instructed to nominate three families with
whom they were familiar. After agreement had been reached
on three families whom they knew best, spouses were
instructed to rank these families on a number of dimensions
such as ambitiousness, happiness of children, etc. This
initial ranking was done separately and independently by
each of the two spouses. When these rankings were com-
plete, spouses were brought together and informed of each
others rankings and instructed to resolve any differences.
The discussion which ensued was analyzed using Bales'
interaction categories (Bales, 1950). Strodtbeck (1951)
reported that spouses who talked most also "won" most
decisions or had the most influence in final answers. This
method, named the Revealed Differences Task (RDT), was
used in a number of later experiments.

March (195A) investigated husband—wife interaction
around political issues using an essentially Revealed Dif-
ferences Task. He reported a tendency of marital couples
to escape conflict by allocating decision power to each

other for different areas. Vidich (1956) modified the

1A

RDT by remaining with the couple during the task and point-
ing out their differences as they occurred. During this
experiment two observers were in the room, one to point out
the differences and one to observe gestures and to operate
a tape-recorder. Vidich (1956) reported considerable
difficulty with this technique since most couples found it
difficult to interact together without directing comments
at the observers. He questioned the value of direct obser-
vations because of the artificiality of the situation. He
also noted the tendency of couples to present a united, not
necessarily accurate, and socially desirable front to the
experimenter.

The small group research, especially that of Bales
(1950), led to the identification of two leadership roles,
the instrumental (i.e. task oriented) and the expressive
(i.e. socio—emotional oriented) roles. Parsons (1955)
suggested that the instrumental and expressive roles in
the family are filled by the husband and wife respectively.
This suggestion was investigated by Bachove and Zubaly
(1959) by analyzing the discussion of normal family triads
(father, mother and son) in response to hypothetical
family situations and TAT cards. They reported that the
father and mother did indeed emerge as the instrumental
and expressive leaders respectively. Levinger (1959)
compared normal and abnormal families in a similar experi-

ment and reported a higher degree of maternal negative

15

emotional behavior and a higher maternal participation rate
in clinic families. These results were interpreted as
"role reversal" in clinic families with a passive father
and a dominant mother, similar to the "schizophrenogenic
mother" notion of Fromm—Reichman (19A8). This was followed
by a number of studies which investigated this construct,
conceptualized as a precusor or determinant of schizo-
phrenia in at least one child of the family (see Caputo,
1963; Cheek, 196A; A. B. 1956; Farina, 1960; Farina, Storrs,
& Dunham, 1963). Fontana (1966) after critically evalu-
ating these experiments concluded "there is no evidence

for the proposed 'schizophrenogenic' pattern of dominant
mother—~passive father (p. 225)."

The interest in this notion was valuable since it
resulted in considerable data on family interaction.
Variables of interest were also no longer restricted just
to power indices or Bales' Interaction Categories. Acti-
vity variables, such as rate of interaction, number of
times spoken, amount of silence, etc. were used by Haley
(1962), Lennard, Beaulieu & Embrey (1965), and Cheek
(196A). Indices of conflict were investigated by Farina
(1960), Caputo (1963), and Lennard et a1. (1965). A
variety of indices of clarity of communication were devel-
oped by Fisher, Boyd, Walker, & Shear (1959), Stabeneau,
Tupin, Werner, & Pollin (1965), Beavers, Blumberg, Timken,

& Weinen (1965), and Caputo (1963). Rigidity of

16

communication was investigated by MacKenzie 1968; Haley
(196A, 1967), Stabeneau, et a1. 1965. These studies will
not be reviewed in detail here since they are only tangen-
tially relevant to the present study. One exception is
the research program of Ferreira and his associates since
some of the interaction variables used in the present
study were developed in that project. The research method
of Ferreira is also in most ways representative of family
interaction studies.

Ferreira (1963) instructed family triads to indepen-
dently answer 16 emotionally neutral items by ranking 3
solutions in decreasing preference. (Example: "If you
had to order something to drink which would you choose:
coffee, milk or tea?") Upon completion the family was
brought together and instructed to arrive at a consensual
decision for the same items.’ The analysis consisted pri—
marily of a comparison of individual and joint answers.
Ferreira identified: (a) unanimous decision or spontane—
ous agreement; i.e. members agreed without previous discus-
sion; (b) majority decision; i.e. joint answer was the same
as the initial answer for two of the three members; (c)
dictatorial decision; i.e. joint answer was the same as
initial answer of one member; (d) chaotic decision; i.e.
joint answer different from all initial answers. Results
were interpreted to suggest among other conclusions: "that
normal families agreed spontaneously, significantly more

often than abnormal families (p. 72)."

17

A later study (Ferreira & Winter, 1965) followed the
same basic design. Subjects were 50 normal; 15 schizo-
phrenic; l6 delinquent and AA mal-adjusted family-triads.
Subjects were instructed to (a) independently and then
(b) jointly choose three preferred and three non-preferred
from among ten alternatives to seven questions. The ques-
tions were essentially the same as those of the earlier
(Ferreira, 1963) study. Analysis was in terms of a Spontan-
eous Agreement score (number of unanimous decisions);
Decision Time (time needed to complete the joint question-
naire); and Choice Fulfillment (computed for each member,
consisting of number of initial choices which became family
choices). The authors concluded that "normal families,
when contrasted with abnormal ones, were shown: (a) to
have a much greater agreement in what their members liked
or disliked, prior to any exchange of information, (b) to
spend less time in the reaching of family decisions and
(c) to arrive at more appropriate decisions in terms of a
better fulfillment of the family members' individual
choices (Ferreira & Winter, 1965; p. 220)."

In yet another study (Ferreira, Winter & Poindexter,
1966) the same family-triads were instructed to produce
stories to three TAT card sequences. Among other results
the authors concluded that the normal families needed
less time for this task while the abnormal families' dis-
cussion contained a significantly greater number of

silences.

18

Haley (1962) and Cheek (196A) also reported that
clinic families were less active than non-clinic families,
i.e. more time was needed to complete a given task.

Bodin (1966) also demonstrated higher overall parental
agreement and more efficient joint decision making in real
than in artificial families.

A novel approach, called Interaction Testing, was
developed by Roman & Bauman (1960). This method consisted
of two phases: (a) standard administration of individual
psychological tests (both projective and intelligence),
followed by (b) administration of the same test to both
spouses where only the mutually agreed upon response was
recorded. Most of their research used the Wechsler-
Bellevue Comprehension and Similarities Subtests. This
procedure yielded four sets of data: (a) individual proto-
cols, (b) joint protocol, (c) comparison of individual and
joint protocols, and (d) the interaction process (Bauman &
Roman, 1966).

A comparison of individual and joint protocols led
to four scoring categories: (a) dominance (joint response
contains one individuals response in the absence of other's
response); (b) combination (joint response contains ele-
ments of both individual responses); (c) emergence (joint
response contains a new idea or concept not present in
either individual answer); (d) reinforcement (joint

response is same as both individuals' response). In

19

addition, the quality of the response from individual to
joint protocol was evaluated. A plus was scored if the
joint response was better than either individual's response.
A minus was given if the joint response was poorer than at
least one of the individual responses and a zero reflected
no change in quality of response. Bauman & Roman (1966)
administered this procedure to a sample of 50 couples of
which one spouse was an inpatient in a metnal hospital.

The primary variable of interest was pattern of dominance
and its determinants which were hypothesized to include:
rational, (i.e. dominance based on competence) and irra-
tional considerations (i.e. dominance based on husband
status, non—patient status or recorder status). The authors_

reported that all these factors were significantly corre-

lated with dominance (husband status 3 = .AA; competency
(based on individual IQ score) 3 = .39; nonpatient status
5 = .30; recorder status 3 = .26). These determinants

acted independently.

This research led to the development of a decision
efficiency concept. The "intellectual efficiency" of a
marriage relationship was defined as the difference between
a couple's potential joint IQ (i.e. the "best score" which
a couple would obtain if they consistently selected, in
interaction, the better of their two individual responses)
and their actual IQ. Bauman & Roman (1968) reported the

results of an unpublished study which suggested that degree

20

of efficiency is directly related to degree of individual
pathology.

The most comprehensive study of marriage is currently
in progress. It is sponsored by the Family Development
Section of the Child Research Branch, National Institute
of Mental Health, presently under the direction of R. G.
Ryder. The stated purpose of the project is to develop a
"psychosocial taxonomy of marriage." At this time a number
of reports have appeared with results of a pilot study of
50 couples, who were all white, middle-class and in the
fourth month of a first marriage. A larger sample of two
thousand couples is presently being studied. In the pilot
study, couples were studied by extensive individual and
joint interviews which dealt with their personal back—
ground as well as their current functioning and plans for
the future, a modified Revealed Differences Task (the Color
Matching Technique), questionnaires (including the Locke-
Wallace) and role playing tasks.

Goodrich & Boomer (1963) reported the method and
initial results of the Color Matching Technique (CMT).
Spouses were seated at opposite sides of an easel which
displayed colored sheets of paper. Each sheet was of a
different color, identified with a number. After the
experimenter had displayed and removed a stimulus color,
spouses were asked for the number of the color on their

easel which best matched the stimulus color. Spouses were

21

told that only the responses on which they agreed would
"count" toward their "score." Of twenty such matches, ten
were impossible since the same colored sheets were numbered
differently on the two sides of the easel. The discussion
which ensued was recorded. Goodrich & Boomer (1963) con-
ceptualized observed differences primarily in terms of
"coping style." It was noted that some couples achieved
perspective on the situation while others did not. Simi-
larly, some couples maintained esteem for self and other,
while others did not.

When a couple takes this position (looks at the

test as a whole) it is based upon trust in each

others' perceptions and judgment and in a greater

willingness to consider that the difficulty lies

in the situation rather than themselves; there-

fore, we consider it a sign of mutual confidence,

and usually a sign of coping adequacy (p. 22).

Ryder & Goodrich (1966) reported a more detailed
analysis of the findings. Variables used in the analysis
of the discussion of the couples in response to the CMT
included: (a) number of statements; (b) husband initiation;
(c) husband dominance; (d) errors (errors were differen-
tiated as to whether they were made before or after the
other spouse had spoken); (d) task discussion (several
dimensions); (f) disapproval of spouse; (g) laughter; and
(h) avoidance of structure. Ryder & Goodrich (1966) noted
a striking tendency of spouses to distort or alter choices

to avoid disagreement; i.e. there were many more errors

after the other spouse had indicated his/her choice than

22

there were errors before the other spouse had spoken. A
factor analysis of the variables yielded two main factors:
(a) rationality versus affectivity and (b) verbal fluency.

In yet another report, Ryder (1968) stated that
"married strangers" (i.e. subjects were married but not to
their partners for the CMT) were much more cheerful,
cooperative and polite to one another than were married
couples.

Ryder (1969) summarized the findings of the pilot
study. The data of each method of assessment, i.e. inter-
views, questionnaires, etc., was factor-analyzed and the
four most prominent factors for each set of data were
retained and subjected to another "cross-method" factor-
analysis. This method yielded four main factors: (a)
closeness to husband's family; (b) role orientation; (c)
complaints about the marriage; and (d) closeness to wife's

family.

Summary of Literature Review

The clinical literature abounds with generally unsub-
stantiated claims that spouses tend to be of equal levels
of immaturity; however, the research in this area is very
limited and has used primarily instruments designed to
measure "traits" or characteristics rather than overall
functioning in a search for "proof" of the homogamy or

heterogamy hypothesis.

23

Recent literature also reveals an increasing disillusion-
ment with traditional methodology in marriage research.
There is a discernable movement away from self-report mea-
sures which are being replaced by laboratory observation

of actual family functioning. The interaction approach

is still relatively recent in origin and is still search-
ing for methodology and meaningful variables. The inter-
action approach is typified by an emphasis on interpersonal
variables; i.e. those which are manifested only in inter—
personal behavior and go beyond "intra-psychic constructs"
(see Framo, 1965, p. A51). However, a number of researchers
have analyzed interactional processes in terms of intra—
psychic constructs (e.g. Goodrich & Boomer, 1963; who
translate "coping adequacy" of couples into "ability to

maintain esteem for self and other" of individuals).

Purpose
The present study was designed to investigate: (a)
the notion that spouses tend to be equally mature or
immature; (b) self-report and interactional methods of

marital assessment and the relations among these methods.

CHAPTER II

METHOD

Subjects (SS, i.e. husband-wife teams) were recruited
from a clinic and non-clinic population in order to maxi—
mize range of scores. An identical procedure was adminis—
tered to all Ss. This consisted of: (a) the Tennessee
Self Concept Scale as a measure of overall individual
adjustment; (b) the Locke-Wallace scale of marital adjust-
ment and the Family Concept Inventory as self-report mea-
sures of marital functioning; (c) three modified Revealed

Differences Tasks which were tape-recorded.

Instruments

Tennessee Self Concept Scale

The TSCS was used to measure level of individual
emotional maturity or adjustment. The TSCS, developed by
Fitts (1965), is a short (100 items, average administra—
tion time is approximately 13 minutes) questionnaire
designed to describe an individual's self—concept.
Several scores are derived from the standardized scor-
ing procedure. The single most important score is Total
P which reflects the overall level of self-esteem. This

score is broken down by identifying different sources of

2A

25

esteem, both from an internal and external frame of refer—
ence. The scores Identity, Self Satisfaction and Behavior
make up the internal frame of reference. Respectively, S5
describes or rates his basic identity (what he is as he
sees himself), self satisfaction or self acceptance, and
how he acts. The external frame of reference consists of
Physical—, Moral Ethica1-, Persona1-, Family—, and Social-
Self. The TSCS also yields seven empirical scores which
are based on item analyses of the performance of several
norm groups. These scales are: Defensive Positive;
General Maladjustment; Psychosis; Personality Disorder;
Neurosis; Personality Integration and Number of Deviant
Signs. The last is simply a count of the number of
deviant features on all the other scores of the instrument.
According to Fitts (1965) the Number of Deviant Signs
Score "is the scale's best index of psychological disturb-
ance. This score alone identifies deviant individuals
with about 80% accuracy (p. 5)." Other scores include the
Self-Criticism Score, which is made up of items from the

L scale of the Minnesota Multiphasic Personality Inven-
tory. The Variability Score reflects the consistency or
integration of a person's self—concept. The Distribution
Score is a response—set measure and is interpreted as a
measure of the individual's definiteness or certainty

when describing himself. The True-False ratio is also a
response— set measure. The Net Conflict Score differenti-

ates acquiescense conflict from denial conflict (i.e.

26

over-affirmation of positive attributes versus over-denial
of negative attributes). The Total Conflict Score reflects
the confusion, contradiction and general conflict in self-
perception.

The mean scores of a standardization group, a psychia-
tric patient group and a "personality integration group,"
the standard deviations and reliability coefficients (based
on test-retest over a two week period with H.” 60) are
presented in Table 1. An inspection of this table suggests
significant differences among the three groups for most
variables. Fitts (1965) has also published much other
information, including correlations with various other mea-

sures which supports the validity of the TSCS.

Locke—Wallace Scale of Marital Adjustment

The LW consists of a number of questions borrowed
from a variety of earlier, similar instruments. It is
reproduced in Appendix A, p. 12A. For a sample of 236
individuals, Locke & Wallace (1959) reported a split—half
reliability of .90. The validity of the test was investi-
gated as follows:

"Forty-eight of the 236 subjects were known to

be maladjusted in marriage. Extensive case data
[not further described] corroborated this for
thirty-one of the persons, twenty-nine of whom
were clients of a clinic. Eleven more cases were
recently divorced, and six were separated, making
a total of twenty-two males and twenty-six females
in the maladjusted group. This group of forty-
eight was matched for age and sex with forty-
eight persons in the sample judged to be excep-
tionally well-adjusted in marriage by friends who
knew them well (p. 25A)."

27

.Ammmfiv mupam some stratum tapas

nouoz

.omoumcH poppoamm who: mcmfiomz
.Hsmm:acmos we on cofipmfi>oo onmocmpm new some you oo3mxm oop sumo mo cowpsnappmamm

 

 

om. m.m sm.z m.mm mmcmfim acaa>ta so tanssz
om. mm.m o.ma mw.m ma.oa AH.: ae.m cofipmawmch mpaamcomtta
Hm. 3H.» me.am OH.HH Hm.=m H.ma m.m~ mfimotsmz
mm. mw.m ma.mm NA.HH mm.ofi m.ma o.mo twosomao ASAHmcomttm
mm. mo.m mm.m= ma.o oa.ms =.m s.m: mfimonoamd
Am. mo.» 30.:0H mH.m om.mm s.mH m.mm pctspmsnomamz Hmtmcmo
om. Hm.m op.mm mm.ma o:.:m m.aa m.Hm t>apamoa t>amctemo
mm. Ha.om OH.OMH mfi.am :a.oma H.Hm =.HNH soapsnatpmao
so. om.e no.5m m=.mH mm.ma m.:a m.Hm suaaanmapm>
om. Hm.m 53.3A mm.~ aa.mm w.oa o.mm Hmfioom
am. am.» ma.»~ m:.m mm.o> m.oa m.=m zaaamm
mm. mm.m mp.ae as.» mm.ao m.HH m.om Hmcomtma
om. om.» me.ms oe.m mm.oe o.HH «.mm Havanpmuamtoz
Am. ma.m mo.ms No.5 me.a~ H.HH m.sm Hmoamsna
mm. mw.m oo.mma NN.HH Ho.mHH a.ma o.moa AOH>mstm
mm. :H.NH mm.oma m~.ma Am.moa ~.~H H.ma coapomemapmm
Hm. mm.m m:.mma om.m oa.~mfl ~.mH m.mHH suapcteH
mm. m:.mm Ho.mem os.om Am.m:m m.:: o.mmm m>apamoa Hmpoe
as. mm.m oo.mm Hm.w oa.om m.HH H.mm poaaecoo Hence
:5. ma.m ma.mat Ho.mH Hm.: I m.ma o.m pOHHecoo ptz
mm. NH. mm. mm. mo.H oz. AH.H oapam tmaamttate
mp. mm.m Am.mm op.m am.mm m.m. o.mm amaoapaao eatm
ASAHapmfiHtm .o.m cat: .a.m cats .a.m cat: ttoom

 

Amuv axons Hm

Awmmv Q2090 Enoz

Ammmv daoao scraped

 

masopo oopce mom mowoom meme mo zuHHHanHom new mcoaumfi>oo

H mHDwB

opmocwum .mcmoz

28

Mean scores were reported to be 135.9 and 71.7 for the
well- and mal-adjusted groups respectively which was inter-
preted to show that: ". . . this short marital adjustment
test clearly differentiates between persons who are well-
adjusted and those who are maladjusted in marriage (p. 255)."

Hawkins (1966) investigated the possible confound-
ing influence of the social desirability of the LW items.
The LW and the Marlowe—Crown social desirability scales
were administered to A8 couples. Correlations between the
scores were reported to be .31 for males and .37 for
females. These correlations, although statistically
significant, led to the conclusion that social desirability
is not a major factor in the LW score.

A number of experiments used the LW as an indepen-
dent variable. Murstein (1961), in an investigation of
the pattern of psychological variables between spouses,
used the LW as a criterion measure of marital adjustment.
Similarly, Katz (1965) differentiated happily versus
unhappily married couples using the LW. Navran (1967)
also used the LW as a measure of marital adjustment and
reported positive correlations between the LW and a self-
.report assessment of communication effectiveness within

the marriage.

Family Concept Inventory
The FCI, reproduced in Appendix A (p. 126), was

developed from the original Q sort constructed by

29

van der Veen et a1. (196A). The original Q sort contained
80 statements, however, only A8 of these entered into the
scoring. Van der Veen et a1. (196A) reported significantly
different mean adjustment scores for well-adjusted and
mal—adjusted families (means were 35.2 and 27.9 respec-
tively; N = 20). Van der Veen et a1. (196A) also reported
a significant positive rank-order correlation between the
LW and the Q sort (3 = .67; N = A0). Van der Veen &
Ostrander (1961) reported a median test—retest correlation
of .7 over a four week period. Hofman (1966) administered
both the Q sort and the critical A8 items in a true—false
form to a sample of 25 couples and reported a correlation
of .72 between the two forms and an internal consistency
index of .8A for the true—false form. Palonen (1966)
developed a five-choice form for the same A8 items (the
FCI) and reported a split-half reliability of .85 (N = 80).
Several studies have used both the LW and some form of the
FCI on the same group of subjects. Correlations between
the two measures as well as between husband and wife are

summarized in Table 2.

Revealed Differences Tasks

Three RDTs were used. These are reproduced in Appen-
dix A (pp. 120-123. All three present a hypothetical
problem-situation which could arise in a marriage.
The first deals with allocating money, the second with

childrearing policies and the third with methods of birth

30

Table 2

Correlations between Husband's and Wife's Marital
Satisfaction Scores and between the Locke-Wallace
and Family Concept Instruments

 

 

 
      

LW FC-Q sort FC-TF FCI
Locke-Wallace Scale .281

.582 .681

.58 67 .761
Family Concept
Q-Sort
Family Concept
True-False

2

Family Concept .65
Inventory 726

 

Note - Husband-wife correlations are below the

diagonal.

1. From Hofman (1966) N = 25

2. From Palonen (1966) N = A0

3. From Van der Veen (196A) N = 20
A. From Hoeg (1965) N = 22

5. From Powell (19657 N = 23

6. From Updyke (1968) N . 23

a. Low adjustment group

b. High adjustment group

31

control. Each task has a list of eight solutions. §s were
instructed to rank these solutions in order of preference.
After this had been done, §s were instructed to answer the
same tasks jointly (see procedure below). This procedure
yielded three rankings for each of the RDTs, one by each
of the spouses and one reflecting their joint decision.
The discussion which ensued was tape-recorded and then
typed. After the first typing it was corrected twice by
two different judges and then again typed. Two sets of
interaction measures were derived from this procedure:
(a) decision efficiency measures and (b) discussion ratings.
These will be discussed in turn.

Decision Efficiengy Measures were largely borrowed
from Ferreira (1963) and Ferreira & Winter (1965).
Decision Time (DT) was defined as the time, in seconds,
needed by §s to complete the three RDTs. DT was measured
by timing the appropriate parts of tape—recordings. Spon-
taneous Agreement (SA) was defined as the degree of
agreement between spouses prior to any consultation. SA
was obtained by a rank-order correlation between spouses'
individual rankings. Choice Fulfillment (CF) was defined
as the degree to which individual choices also became
joint choices. CF was obtained by a rank-order correla-
tion between §s' initial choices and joint choices. Pre-
vious research has shown that clinic families on the
average need more time to complete a given task than normal

families (see Ferreira, 1963; Ferreira & Winter, 1965;

32

Ferreira, Winter & Poindexter, 1966; Haley, 1962; and Cheek,
196A). Normal families also have shown greater SA than
abnormal families. (See Ferreira, 1963; Ferreira & Win—
ter, 1965 and Bodin, 1966). Non-clinic families, compared
to clinic families, also Show a greater degree of CF.
(See Ferreira, 1963 and Ferreira & Winter, 1965).
Discussion Ratipgs were based on a system developed
by Argyris (1965, A, B, and C). Central to this system
is the concept of competency in interpersonal relation-
ships. Argyris defined interpersonal competence as:
"1. One's awareness of relevant factors (relevant factors
are those that have an effect), 2. problems are solved in
such a way that they remain solved, 3. with a minimal
deterioration of the problem-solving process (1965 B; p.
59)." This definition led to a system of categories of
interpersonal verbal behavior. Such behaviors as "owning,
being open, and experimenting" are characteristic of and
increase interpersonal competency, while "not owning, not
being open and rejecting eXperimenting" are detrimental
to the problem solving process. On the interpersonal
level, the same categories are defined as "helping others"
or "not helping others to own, be open and experiment."
In addition, each behavior is also rated as involving
either feelings or ideas. The complete system of cate-

gories as used in this study is presented in Table 3.

33

 

 

 

 

 

.Ammmav maszws< Eon: oouamom manna "ouoz
NI m u:oEfi:ooxo on mat a w:au:oefipoqxo
Ox? at H mso:po dam: uo: :HI H w:fiuoonon
e
a. on a code on or man a
w mt H mumsuo dam: uo: mt fl :oao no:
Tr
who:o>auoommm M mscwz
oommosooa w
3 mt m :30 on :HI 9
8 ml a msonuo dam: no: mt a w:H:zo no:
osoN
m m :30 m m
Mus m H on wsocuo dam: H a w:a:3o
Tea
NM. 3 : coao to S :
mww m a on msonpo dam: m H who::oqo
B
wmoco>fiuoommm m u moan
oommoso:H %
4x ma m u:oEH:ono ma m
u a on wsocuo mac: n H w:Hp:oeH:ooxo
monouso unmfioz Hm:om:oo:ou:H unwfios Hmsofi>ao:H

 

ho:opoQEoo Hm:om:oo:op:H ou nonmaom :oa>m:om mo moasowopmo

m manna

3A

The system provides numerical values for each rating
(see Table 3). An individual's index of interpersonal com-
petency (IC) is a ratio of weighted sum of ratings over
highest possible weighted sum of ratings; i.e. number of
units scored multiplied by 16. This index behaves as the
correlation index with minus scores indicating low inter-
personal competency and plus scores indicating high inter-
personal competency with minus 1 and plus 1 as limits.
Argyris (1965A) reported a minimum percentage agreement
of 86 between experienced raters before any discussion of
discrepancies. Argyris (19658) also found that 1 group
members' 10 scores correlated significantly with staff
interpersonal competency ratings for the same members. The
staff members were not familiar with the system of cate-
gories. Other data (Argyris, 1965A) suggested that members
with higher IC scores showed more growth and greater satis-
faction with the 2 group sessions than those with low IC
scores. In summary, the preliminary research reported by
Argyris suggests that the system of categories can be
reliably used to measure an important aspect of individual

and group communication abilities.

Subjects
A Clinic (N = 15) and Non-Clinic (N - 15) sample of

couples were recruited. The Clinic Group was obtained
through the cooperation of therapists at a Community

Mental Health Clinic. Criteria for the clinic sample were

35

that one of their children had been referred to the clinic
and that they had been married for at least four years.
After therapists nominated couples, N contacted them by
phone and asked them to participate in a research project
which would require approximately an hour and a half. S3
were told that they would be asked to complete a number of
questionnaires and to discuss some topics. Participation
rate of those contacted was approximately 80%. Only one
couple flatly refused to participate and two other couples
did not participate because of time-scheduling problems.
§s were paid five dollars for their participation. The
Clinic Group also included two couples from an ongoing
marital therapy group who were not paid directly but were
remunerated by a reduction in fees.

The Non-Clinic Group was recruited with the coopera-
tion of two local ministers of a moderately conservative
and a liberal church, both Protestant. In one instance,
the minister mailed a letter explaining the project to
about twenty couples whom he thought to be well-adjusted
and happily married. Ten of these responded. Letters were
also sent to nine couples nominated by the other minister
and six of these responded. It was discovered that three
or these sixteen couples had one child who was, or had
beer“ referred to a mental health clinic for what appeared
t“) be emotional difficulties. These three couples were

1rHzluded in the Clinic Group. Three more non-clinic

36

couples were recruited through the cooperation of a teacher
at a nursery school. Non-clinic couples were paid fifteen

dollars for their participation.

Procedure

§s were seated in the same room either at oppositely
facing desks or across a rather large table. §s were pre-
sented with a booklet which included a general introduction
and instructions followed by seven tasks. Tasks l, 2, and
3 consisted of the three Revealed Differences Tasks. Task
A consisted of the LW and Task 5 of the FCI. The TSCS
constituted Task 6. ‘Qs were instructed to complete the
tasks separately without consulting each other, suggesting
or discussing the answers. They were also informed that a
tape-recorder (clearly visible) was present in the room
and would be used later. At that point N left the room
informing §S that he was available in the next room if
problems arose. After §S had finished the six tasks they
notified N who quickly checked their booklets to ensure
that these had been completed properly. N then engaged the
tape-recorder and instructed §s to again complete Tasks 1,
2, and 3 but to produce a ranking of solutions for each
of these tasks which would be satisfactory or acceptable
to both. After answering any questions Q again left the
room. Upon this joint completion of Tasks l, 2, and 3
§s again notified N who then instructed §s to complete a
short demographic questionnaire. Payment of §s marked the

end of the procedure.

Hypothesis 1.

CHAPTER III

HYPOTHESES

Mean MA scores, both LW and FCI, are signi-
ficantly higher for the Non-Clinic Group
than for the Clinic Group.

Clinic Group mean scores are significantly
higher on the pathognomonic scales of the
TSCS, especially NDS.

Non-Clinic Group mean scores are signifi-
cantly higher on the self-esteem scales of
the TSCS, especially TP.~

Mean SA scores are significantly higher for
the Non-Clinic Group than for the Clinic
Group.

Mean CF scores are significantly higher for
the Non—Clinic Group than for the Clinic
Group.

Mean DT scores are significantly higher for
the Clinic Group than for the Non-Clinic

Group.

37

10.

11.

12.

13.

1A.

15.

16.

38

Mean 10 scores are significantly higher for
the Non-Clinic Group than for the Clinic
Group.

There is a positive correlation between
spouses' TSCS scores, specifically TP and
NDS.

There is a positive correlation between TSCS
self-esteem scores and MA scores.

There is a negative correlation between TSCS
pathognomonic scores and MA scales.

There is a positive correlation between MA
scores and positive interaction scores; i.e.,
SA, CF, and IC.

There is a negative correlation between MA
and DT scores.

There is a positive correlation between TSCS
self-esteem and positive interaction scores.
There is a negative correlation between TSCS
pathognomonic scores and positive interaction
scores.

There is a positive correlation between DT
and TSCS pathognomonic scores, and a negative
correlation between UT and TSCS self-esteem
scores.

There is a positive correlation between IC

and CF.

CHAPTER IV
RESULTS

Demographic Variables

Comparative means of demographic variables for Clinic
and Non-Clinic Groups are presented in Table A. Clinic §
were significantly older (p < .001) and had been married
longer (p < .05). There were no statistically significant
mean differences for annual income, educational level,
number of children or social class (social class was deter-
mined after Hollingshead, 1959). When males and females
were considered separately, non-clinic females had a

significantly higher educational level (p < .05).

Marital Adjustment Scales

Hypothesis 1 predicted significantly higher mean
Marital Adjustment (MA) scores for the Non-Clinic than the
Clinic Group. The results, presented in Table 5, con-
firmed this hypothesis. The Non-Clinic Group obtained
significantly higher mean LW and FCI scores than the Clinic
Group (E < .01 and p < .001 respectively; one tailed 3
tests). These differences still held when males and
females were considered separately; however, the signifi-

cance levels were reduced (see Table 5).

39

A0

oaaoouozu .Hoo. v.m...

 

 

oat» oIo
ammo w ooaaeouoze .Ho. v we.
use» p ooaaeotozo .mo. v m.
w.m mm.ma m.=m N>.mH H.0m mmeo Hwfioom
m. mN.H >.m m>.H m.m concafico Mo #09532
ceases:
mm.m mm.m ~.oa Hm.m m.mH memo» no sonesz
Azaams::m
u.m NH.: 0.0H mm.m ~.ma mo:mm:o:uv osoo:H
*m.a ma.m m.mH om.m m.mH moamsom
H.m mo.m 5.5H mo.: m.mH mmamz
m.H nm.m m.ma NH.m e.ga :oautosom
a:m.m mu.m >.Hm mm.m m.oz monEom
aaw.m em.> H.mm mm.m m.a= moan:
***N.w wn.w m.mm Hm.m N.H= mw<
oo:o:mmmaa
:moz mm :moz mm :mm: manmfism>

 

ozosu 0H:HH0I:oz

 

ozone ofi:HHo

 

: oases

m:oHumfi>om o:mo:m»m o:m m:moz .moanmfism> ofisowswosoo

A1

So. v a...

 

 

women m ooafieutoco Haa moo. v as.
mo. v.ma
.e.m.mm me.oH H.omH e=.oH o.mma modest:
.a.o.a~ oa.oa H.mmH om.oa m.oma meat:
.e.o.om me.oa o.oma mm.oH o.oma stoocoscH oooocoo eases:
.o.oH AH.HN A.HNH em.mm s.moa modest:
.m.mH oo.ma =.mma mo.oa H.oHH mode:
*.A.ma om.sa H.NNH Hm.mm s.ooH oaeom ooeHHeztoxoog
oo:o:ommam
:moz mm :moz mm :mm: mapwfism>

 

 

moons oH:fiHoI:oz ososc 0H:HHO

 

m:oapmfi>oa osmp:mum o:m m:moz .mosoom p:oEpmzno¢ Havana:

m magma

A2

Inspection of the correlations between MA and demo-
graphic variables (see Table 28) revealed statistically
significant correlations between FCI and number of years
married and age (g =-.39 and -.Al respectively). Since
clinic couples had been married longer than non-clinic
couples, and since these variables were negatively corre-
lated, the difference in MA scores between groups could
be explained as due to the difference in number of years
married. To control for the number of years married, an
analysis of covariance, with number of years married as
covariate, was used. The analysis is summarized in Table
6. The highly significant Group effect suggests that
mean MA differences were not simply attributable to age

differences.

Table 6

Analysis of Covariance, Family Concept Scores with
Number of Years Married as Covariate

 

 

Source g; N§ N p
Sex (A) 1 8.1 NS
Groups (B) l 9l6A.6 30.32 v .001
A X B l 209 NS
Error ' 56 302.3

 

Product-moment correlation coefficients, multiplied
by 100, between LW and FCI scores are listed in Table 7 for

all Ss and also for sexes and groups separately.

“3

Table 7

Correlations between Locke Wallace Scale and
Family Concept Inventory

 

 

Subjects , N 3* p
All Subjects 60 55 .0005
Males 30 61 .0005
Females 30 52 .005
Clinic Group 30 A5 .05
Males 15 60 .05
Females 15 33 N§.
Non-Clinic Group 30 Al .05
Males 15 A3 NS
Females 15 50 NS

 

*3 multiplied by 100

Tennessee Self Concept Scale

Hypotheses 2 and 3 predicted significantly higher
pathological scores for the Clinic Group and significantly
higher Self—Esteem scores for the Non-Clinic Group. TSCS
means are presented in Table 8. Both predictions were
confirmed. The Non-Clinic Group reported significantly
higher self-esteem in all areas than did the Clinic Group,
with the exception of the Moral Ethical-Self Score which
did not differentiate the two groups. The differences
between groups held when males and females were considered
separately (see Tables 9 and 10).

The clinical scales (Defensive Positive, General
Maladjustment, Psychosis, Personality Disorder, Neurosis,

Personality Integration, and Number of Deviant Signs)

AA

~00. v a...

 

 

memo» o ooaaeo oco Ho. v 0..
mo. v an.

em.» mH.0H m.0m mm.mH 0.0: moCMHhm> H0909
10.: 0m.0 m.hm m~.0 m.Hm uofiau:oo H0009
0.0 a0.0 m.mHI 00.HH 5.0 I uoaamcoo 002

N. ma.m 0.0m :N.m 0.0m Emaoaufiho maom
.e.m.o mp.m m.o o0.aa o.ma woman oceasoo so toosoz
.:.~ om.m e.ma ma.o m.HH cofioeawoooH soaaeoomtoa
gmao.aa no.0 0.:0 0H.oa 0.me memossoz
.a0.m oo.o 0.5» No.0 m.m~ tootomao soaaocomtom
*300.: 0m.m 0.0: 50.0 0.»: mamonohmm
.e.o.0 os.0 m.oo ao.0 m.ao outsomsnoofioz Heaocoo
a=.m :0.» 0.00 mo.oa 3.»: o>apamom o>«m:mmoa
0.0 0:.zm 0.00H H0.0H 0.00H :OHusnfinpmHQ
:0. 0H. 0. mm. 0. oapmm omawmtoshe
***~.m 00.0 0.0» 00.0 0.30 maom Hmfioom
memm.0 no.0 H.me 0m.» 0.00 uaom zHaEmm
aao0.0 sm.0 0.~0 0~.0 o.H0 :aom Hmcomsom
m.m om.0 o.ms NH.o m.os :Hom Hooecom Homo:
3*H.m 0n.m 0.00 03.0 0.20 “H00 Hecammnm
aeam.w >H.0 5.0HH H0.0 0.00H soa>o:om
***0.0H m0.HH :.0HH 00.HH 2.00H :OHpommmemm «How
esau.0 NH.~ 0.0mH 0:.0 0.0HH auHu:oUH
exam.>m >>.mm m.mmm 0m.mm 5.0mm o>HuHmom H0009

00:0:000H0

:00: mm :00: _mm :00: oamom

 

 

030:0 0H:HH0I:02 030:0 0H:fiH0

 

mpoonnsm HH< .w:oapwa>00 0:00:0p0 0:0 m:moz .oHMow

0 tapes

0000:00 maom oommo::oa

H00. v a...

 

 

 

 

omoo o ooflaeo oco Ho. v we.

00. v m.
0.m 0H.e m.om oo.o: o.:s ooceaao> Hosea
0.N 00.0 0.0m 00.0 0.0m uofiahcoo H0poa
0.0 5.0a H.:HI H5.0 0.0 I poHHm:00 002
:.H 00.: H.5m 0N.0 5.0m Emacapahuthamm
.em.o Ho.m o.~ oH.oH N.HH woman ocea>o0 so toosaz
00.0 05.0 0.:H 00.: NH :ofiu0swou:H >00H0:om:om
0.0.0 ma.0 0.00 05.5 5.05 mamossoz
.0.: mm.0 m.m5 5m.0 o.o5 twosome: soaaocomtod
0.0.: 05.: H.m: 00.0 0.5: mfimo:ozmm
000.0 50.0 0.00 5m.0 5.H0 u:08003000H02 Hmhmcmc
:.m 0m.0 0.m0 00.5 0.0: 0>Hpamom o>amcwmoa
00.0H 50.:H 0.50H 50.0H 0.:0 :oauanfihumaa
mo. 0H. 00. 0H. 00. oHp0m oma0mtosne
000.0 :H.0 0.05 00.0 5.:0 uaom H0aoom
0300.5 05.: H.05 Hm.5 0.50 mamm hHHE0m
0.0.0 0:.0 5.50 Hm.0 5.H0 «H00 H0:ompmm
.m.m 00.: :.05 00.5 m.00 hamm Hmofinpm H0902
30.0 0H.0 m.H5 00.0 m.00 mamm Hmofimhnm
00.0 00.5 0.0HH 00.: 5.0HH soa>0com
$00.0 0:.HH 0.0HH H0.5 5.HOH :OHuomhmemm hamm
3:00.HH H5.0 m.0ma 00.0 m.5HH zpfiucmoH
auma.0m ma.am 5.:0m :0.0H 0.0mm o>fiufimom H0908

oocmsmumao

:00: mm :00: mm :00: 0H0om

dooto oficfiaoncoz masoto oacfiao

 

mma0z .m:oHu0H>00 0:00:000 0:0 0:00: .0H0om :000:00 «H00 00000::09

0 0H909

A6

Ho. v 0..

 

 

one: o 00:00» 0:0 0o. v m.
.m.o: 00.m: H.o: o0.0: o.H0 oocaaae> Hence
.m.0 50.5 0.50 0:.0 0.00 ooHHOcoo Hence
0.: 0:.0 m.o:I 0o.:: o.o I poa:::oo :02
:.H Hm.: 0.00 00.0 3.50 smaoao«:oI:Hom
..:.oH :m.: 0.0 :o.m: 0.0: 020:0 ocofi>oo :o toneaz
m.m 00.0 0.0: Hm.: 0.o: co:»e:woocH sofiaecomaoa
..H.m: 00.0 H.:0 :H.NH o.m5 mamo:oez
.0.0 5H.o H.o0 :0.o 0.05 twosomfia soaaecom:00
.m.= No.0 0.0: 00.: 0.5: mamonozma
.eo.5 50.5 0.50 05.5 o.oo ecospmaoooaez H0:oceo
.0.5 00.5 0.00 mo.:: 0.0: o>:p:000 e>fimco:oa
0.: oo.mm 5.:H: :5.o: 0.0o: coaooofiaomfln
0o. 2:. mo. 00. 00. 00000 oedemIoste
.0.0 00.0 5.00 :H.0 0.00 0:00 Hefioom
0.0.0 0:.0 0.05 03.5 0.00 :Hom adage:
ega.5 50.5 0.50 5H.0 :.o0 :Hmm H0:00:00
H.m 00.0 0.:5 0:.0 «.05 0:00 Heofinom He:o:
:.0 0:.0 0.00 00.:: 3.00 0:00 Heoammna
.e0.o m5.o: H.5HH :N.HH 0.5o: :o:>osem
e.:.:: mo.HH m.o:: 50.3: H.oo cofiooo:m:oem 0:00
.em.0 05.5 0.50: 0o.o: 0.0:: soapstoH
.a0.mm 00.00 5.000 0:.Hm 0.0mm e>Hofimod Hope:
mocmhthHQ

:00: mm :00: .mm :00: 0H000

 

020:0 oficdaot:oz

 

050:0 00:0H0

 

0H magma

00:0500 .m:oap0fi>00 0:00:000 0:0 0:00: .0H000 0000:00 :H0m 00000::09

A7

differentiated the two groups at statistically significant
levels in the prediCted direction; i.e. the Clinic Group
scored in the more pathological direction. The only
exception was the Defensive Positive Score which was signi-
ficantly higher for the Non-Clinic Group. It should be
noted that General Maladjustment, Personality Disorder,

and Neurosis are inverse scales; i.e. a high score is
indicative of less pathology.

The response-set measures and empirically defined
scores (Total Conflict, Net Conflict, True-False ratio,
Distribution, and Total Variance) did not differentiate
the two groups as clearly. The Clinic Group scored signi-
ficantly higher on Total Conflict and Total Variance.

Product-moment correlation coefficients, multiplied
by 100, among TSCS scores for all §s are presented in

Table 11.
Interaction Scores

Spontaneous Agreement

SA scores were obtained by multiplying by 100 the
rank-order correlation coefficients between spouses'
initial rankings for each of the three tasks. Hypothesis
A predicted significantly higher mean SA scores for Non-
Clinic than Clinic couples. This prediction was only
partially substantiated. SA means are presented in

Table 12. The Non-Clinic Group obtained a significantly

A8

0500a 0:a000a:0>0 :0 0030000 2

ma:

\4

00

m ”a0. w 0 “mm. M M ”00. w m "mm. M M

>ansow:30n a0a0om 000:: 00 meow 0:050 0:0au0a0::00*

 

00:0a:0> a0009

 

o0 ooaH0coo aeooe
0a omt poaaucoo :02
0: am :01 Emaoapapo Mamm
:0 0: 0a 0a 0:0:0 p:0a>00 00 :00532
50t 00I mat 0mt m0t :0au0:00p:a :paa0:00:00
a0t 5mt mat 5mt 00I 00 . ma00::0z
mmt 0at mat omt mmt 0m 00 :00:00ad :0aa0:00:00
00 0at 0m 0mt m0 mot 0:I :at manocommm
a:t mat amt mo a0t mm 55 50 00I 0:08:03000a0: a0:0:00
a0t 5mt :a 0:I 00I a: 00 00 amt mu 0>apa000 0>awc0000
:m 0m mo 50 mot mmt 0m 00 m:t 00 00 :Oausna:pwa0
0m m0 a0 a0 0a mmt 00 a0 mm 00I 00 00 oap0m 00a0m 03:9
amt 5at 0at 00 m:t mm m5 0: m0t a0 00 5: met *0a00 amaoom
00I mm: aat 0at 00I mm :5 05 0mt 00 m0 5n :ct :0 *0a00 :aaE00
00I 00I 0at 0at a0t 0: 00 m0 0:I m0 55 0: 00 50 :5 *0a0m a0:00:00
aat 00I :ot oat mmt 0a 00 :0 0at m0 0: 0: aa 0m 00 m0 *0a00 a00a£pmta0:0:
00I 0mt mat oat 00I :: a5 0m 00I 00 00 00 00I a0 :m m0 mm *0a0m a00a0000
0:t 0mt awt 5at 00I 5m 00 00 5:I :0 :0 5: aat 05 50 05 00 m5 *:oa>0:0m
:0I m:l :OI mml 00I 0: mm 05 mml m5 m5 mm mo :0 ca 50 o0 :0 05 *COHpommmemm mamm
amt 00I oat 0a 00! mm 55 a0 00I 00 :0 mm 00 05 a5 a0 mm 0m 00 00 thuwucmch
mml om! mat mat 00! m: 00 m5 am! 00 05 o: oo 05 a0 N0 00 m5 om am 00 00>Huamom H0009
>9 QB 02 um mQZ H0 02 D0 mm EU 0Q HQ 0% mm mu W0 02 mm mm mm DH

 

00000030 aa: ”00:000 0a0om 0000:00 0a00 00000::09 0:080 0:0a00a0::00

aa 0a009

“9

higher mean SA score on Task 3 (selection of birth control
methods) and also a significantly higher total mean SA.
There were no significant mean SA differences between the

two groups on Tasks l and 2.

Table 12

Spontaneous Agreement, Means and Standard Deviations

 

 

 

 

Clinic Group Non-Clinic Group
Task Mean §Q Mean §2’ Mean
Difference
1 53 31 61 21 8
2 77 16 73 20 H
3 35 35 63 21 28*
total 55 17 66 12 . 11*

 

*2 < .05; one-tailed t test

SA scores were also analyzed using a Group X Task, repeated
measures, analysis of variance (Winer, 1962; p. 306). The
Nflnmary of this analysis is presented in Table 13. The
main Task and the Task X Group effect were statistically
Sigrufdcant, while there was no significant Group effect.

3A scores are also presented in Figure l.

50

 

 

 

Table 13
Spontaneous Agreement, Analysis of Variance
Source g; MS 3 3
Between Subjects 29
Groups (A) l 1285 1.99 NS
Subjects within
Groups (B) 28 6“?
Within Subjects 59
Tasks (C) 2 5236 8.5“ <.Ol
A X C 2 1985 3.2N <.05
B X C 56 613
Figure l

Spontaneous Agreement Scores for Groups and Tasks

 

Spontaneous Agreement

80

7o , ...____

C)’.",1>;V”’ \\\
60 x,/’/’ \\\

-~~--~O

50 \
\
MO . ‘x
30
Task 1 Task 2 Task 3

x - clinic group
0 = non-clinic group

51

Choice Fulfillment

CF was obtained for males and females separately by
multiplying the rank-order correlation coefficient between
Ss' initial ranking and the joint ranking for each of
these tasks. Hypothesis 5 predicted significantly higher
mean CF scores for Non—Clinic than Clinic gs. This pre-
diction was not substantiated. CF means are presented in
Table 19.1 There was no statistically significant mean

difference for any of the tasks nor for the total CF score.

Table 1n

Choice Fulfillment, Means and Standard Deviations

 

Clinic Group Non-Clinic Group

 

 

 

Subjects Task Mean SD Mean SD Mean
Difference
Males
l 80 22 87 ll 7
2 87 ll 82 38 5
3 7A 27 67 33 7
Total 80 12 79 15 1
Females
l 76 25 72 20 A
2 91 O6 79 38 12
3 62 3H 7H 39 12
Total 76 17 75 12 1

CF scores were also analyzed using a Group X Sex X
Task, repeated measures, analysis of variance (Winer, 1962;
D. 337) which is summarized in Table 15. The main Task
Effect contributed significantly to the variance. None of

the other main or interaction effects was significant.

52

Table 15

Choice Fulfillment Scores, Analysis of Variance

 

 

Source 22 Mé, E E

Between Subjects
Groups (A) l 69 NS
Sex (B) l 631 NS
A X B l 2 NS
Subjects within

Group (C) 56 771

Within Subjects
Tasks (D) 2 3733 4.95 <.05
A X D 2 597 NS
B X D 2 “59 NS
A X B X D 2 927 1.23 NS
D X C 112 75“

 

Decision Time

DT scores were obtained separately for the three
tasks by timing the relevant sections of the tape—record-
ing. Only that discussion related directly to the solu-
tion of the problem was included. Hypothesis 6 predicted
significantly higher mean DT scores for Clinic couples
than for Non-Clinic couples. This prediction was not
substantiated for any of the tasks nor for the total time.
Mean DT scores are presented in Table 16. There were no
significant differences between the two groups.

DT scores were also analyzed by means of a Group X
Task, repeated measures, analysis of variance (Winer, 1962;
D- 306) which is summarized in Table 17. None of the

effects was statistically significant.

53

Table 16

Decision Time, Means and Standard Deviations

 

 

 

 

 

 

 

 

Clinic Group Non-Clinic Group
Task Mean SD Mean SD Mean
Difference
1 173.5 79.59 203.2 9A.62 39.7
2 2A5.3 138.62 200.7 108.29 AA.6
3 287.0 205.29 218.3 101.20 68.7
Total ' 705.9 317.3“ 627.1 236.21 78.8
Note--Time scored in seconds
Table 17
Decision Time, Analysis of Variance
Source g; M§ E 2
Between Subjects
Groups (A) 1 15A97.3 NS
Subjects within
Groups (B) 28 26082.9
Within Subjects
Tasks (C) 2 28629.0 2.A6 <.l
A X C 2 21955.7 1.89 <.25
B X C 56 11611.9
Dominance

 

An additional interaction measure was developed to
measure the degree of influence each spouse exerted on the
final or joint ranking, compared to the influence of his/
her spouse, taking into account the degree of initial

agreement. This index of dominance (DOM) was defined as

5A

Chm _ CFf

100 — SA '
dominance of husband; i.e. the final ranking is deter-

A DOM score approaching +1 indicates

 

mined primarily by husband's initial ranking. A DOM

score near zero indicates compromise or a nearly'equal
degree of influence on the joint decision. Wife dominance
is indicated by a DOM score at or near -1. The actual
numerical values obtained as DOM scores were categorized
as +1, 0, and -1. Limits for these categories were arbi-
trarily set at +.33 and -.33. DOM frequencies are pre-

sented in Table 18. These frequencies were analyzed using

 

a Chi-square technique. For Task 1 there was no relation-
ship between group membership and dominance pattern.
However, there was a clear tendency for all couples toward
husband dominance on Task 1 (pg< .005). There was no rela-
tionship between group membership and pattern of dominance
0n Tasks 2 and 3 nor was there a tendency of all couples
to favor any of the dominance pattern when summed across
tasks.

DOM scores were further analyzed using a Group X
Task, repeated measures, analysis of variance, which is
Summarized in Table 19. None of the effects was statis-

tically significant.

Frequencies of Dominance Patterns

55

Table 18

 

 

 

 

 

 

 

Husband Wife
Task Group Dominance Compromise Dominance
1 Clinic 8 2 5
Non-Clinic 10 3 2
2 Clinic A 7 A
Non-Clinic A 6 5
3 Clinic 5 6 A
Non—Clinic 5 2 8
Table 19
Dominance Scores, Analysis of Variance
Source g; MS 5 9
Between Subjects
Groups (A) l 0
Subjects within
Groups (B) 28 .88
Within Subjects
Tasks (C) 2 1.75 3.07 <.1
A X C 2 1.7 3.00 <.1
A X B 56

56

.——-9

Interpersonal Competency

Interpersonal Competency ratings were obtained from
the typescripts by an advanced graduate student in Clini-
cal Psychology and the investigator. Both raters rated
all type-scripts independently. Mean agreement between
raters was 81.2%. Disagreements in ratings were discussed
until resolved. IC scores were computed from mutually
agreed-upon ratings, and were computed for male and females

Separately for each task. The IC scores of spouses (ICm
and ICf) were combined for the couple scores (ICC).

Hypothesis 7 predicted higher mean IC scores for the
Non-Clinic Group than the Clinic Group. This prediction
was not supported. Means are presented in Table 20.

There were no statistically significant mean differences
between the groups for any of the tasks, nor was there a
difference between total IC scores (see Table 20).

The IC scores were further analyzed using a Sex X
Group X Task, repeated measures, analysis of variance
(Winer, 1962; p. 337). The summary of this analysis is
presented in Table 21.

Total number of units rated and frequency of ratings
in each category are presented in Table 22, For both
Eroups, 65% of the ratings were in the "own ideas" cate—

gory, Significantly more feeling scores were assigned to

Interpersonal Competency, Means and Standard Deviations

57

Table 20

 

Clinic Group Non-Clinic Group

 

 

 

 

 

 

Subjects Task Mean SD Mean SD Mean
Difference
Males 1 35 35 26 31 9
2 1A 38 19 37 5
3 3A “3 33 31 1
Females 1 AA AA A2 26 2
2 39 A1 3A 32 5
3 32 A3 33 35 1
Couples 1 39 25 3A 16 5
2 23 38 27 23 A
3 3A 31 33 23 l
Males Total 29 26 25 2A A
Females Total 36 28 35 21 1
Couples Total 32 22 3O 12 2

 

Note--Actual scores were multiplied by 1000.

Table 21

Interpersonal Competency, Analysis of Variance

‘1

 

Source <_i_f; MS 3 2

Between Subjects
Groups (A) 1 862.A NS
Sex (B) 1 2993.1 NS
A X B 1 2319.5 1.Al NS
Subjects within

Groups (C) 56 2119.7

Within Subjects
Tasks (D) 2 17AA.5 1.61 NS
A X D 2 293.2 NS
B X D 2 1115.6 NS
A X B X D 2 628.1 NS
C X D 112 1086.8

.1,

w—
,u

 

58

Table 22

Frequencies of Interpersonal Competency Ratings

 

 

 

 

 

 

Ratings Clinic Non-Clinic
Group Group
Ideas Positive 1852 1529
Own 1500 1236
Open 3AA 288
H. own 8 5
Negative A03 337
N. own A2 36
N. open 0 3
N. H. own/open 361 298
Feelings Positive 65 29
Own 62 27
Open 2 - 2
H. own 1 0
Negative 18 2
N. own 10 2
N. H. own 8 0
Total Ratings 2338 1897

 

the Clinic Group than to the Non-Clinic Group (X2 - 1A.67
With l g;; p < .005). The distribution of positive versus
negative feeling scores was independent of group status
(52 - 3.55 with 1 93;; NS).

Product-moment correlation coefficients, multiplied
by 100, between interaction scores are presented in
'Pables 23 and 2A, for total interaction scores and for

each task separately.

59

.ma00050000

 

 

x000 £000 000 000000 :0a000n000a 000:» 00050 0:0a00a00000 00200000 :m 0an09
cmaamptozp .ao. v_m..
omaamutozp .00. v 0.
om t m
05050000500 a0comp0qn000a
5at 0050:0500
mm mot 0205000m< 03005005000
mot mot ammt 05a9 coa0a00a
.mm .o:t ..a0 .mmt augmeaaaaasm moaoco
mo .0: mm mat 5m gunmeaaaaasm 000050
came mo am: omt mxom oa 05050000500 a020050000ucH
00:5 wmt mo oa mo mot mo 55050000500 a0000a0000uCH
0 m 5 0
0a :00 40 90 mo mo 0H x0oca

 

005000 amuoe coapo0p005a wco50 0:0ap0a0ppoo

mm 0a909

 

 

 

om u m
DmaaquOZa .Ho. v dz:
cmaamatozp .00. v 0*
m
wot m
mm ma a zoo
amt mo :at m
50. 5o mat ma 0
act :0 ca mo 50. a <0
mot mo. 00 mm- act 02 m
mo :0. Nat mm- .:o0t co mm m
50 mo mm 50 mo *o:t *mm 5a a 50
«*owt mat omt *xmm mot mo mmt mot oat m
n0 5a- mm- 00 Na 0.00 no mot cat «at :a m 0
r0 mo mo .m:t mo mo *xm0 mat 0o- *xumt so act a 00
*5m mot mo- 50 amt mat aat 0 0a so .a. a1- m
wmt 5m 00 om *0: ca 00. :at ac :a **.a :2- not 0 5
mm 00 .*O5 mat mot **00 50 :at mo oat ca ac aat :c a 00
mmt om 00 *0m :a :0 amt mat 0o- *:: :at .2 00 cm mat m
50 mat mo am :0 00 0a 0a. cot mat 0a ,2 aa 00 :at aa 0 0
mmt 5a 5a- ma am mot mot :0t 50 *aa ma co (mt mm amt 5m oat a 0a
mo. **0: 5ot 0a mo 0a 00 a0 am 5ot oat ea 0a 0a :Ot ac mm amt m
00 oo mmt mo 50. mo- ca :0 00. aa 0:. :0 co mmt *::t mot 0a 0st :0: m 5
act mmt **mmt aot met mat mo 5a mat 00 ac an ma a0- *amt «at me 0mt ma .m: a 0a
m m a m m a m m a m a a m m a m m a m a :009 xmuca
200 «0 a; 000 500 . 00a E0a

 

x005 £000 000 000000 :0a00000pca 0:050 0:0a00a05500

61

Hypothesis 16 predicted IC to be positively corre-
lated with CF. This Hypothesis was supported for females

but not for males. ICf 3 and ICf total correlated signi-

ficantly with CF 3 and CF total respectively but there was

a significant negative correlation between 10m 1 and CFm 1

(see Tables 23 and 2A).

Interspouse Correlations

Hypothesis 8 predicted spouses' total self-esteem
(TP) and overall pathology (NDS) to be positively corre-
lated. Interspouse correlations for these and most other
individual scores are presented in Table 26. The predic-
tion was generally substantiated. The TP interspouse
correlation was significantly different from zero, as were
the Identity, Self Satisfaction, Behavior, Personal-,
Family-, and Social-Self scores (see Table 26).

However, the NDS interspouse correlation was -.O3;
riot significantly different from zero. This correlation
Inay be somewhat misleading since the distribution of NDS
Scores is highly skewed, approaching a J curve. To further
investigate the relationship between spouses' NDS scores,
each S's NDS score was classified as + or -; i.e. S's NDS
Score was less than or exceeded the critical score of 10.
“he results of this categorization are presented in Table

25.

62

Table 25

Pattern of Couples' Number of Deviant Signs Scores

 

 

 

Clinic Group Non-Clinic Group
Husband + + - — + + - -
Wife + - + - + - + -
Frequency A 5 A 2 l2 3 0 0
Table 26

Interspouse Correlations

 

 

All Clinic Non-Clinic
Subjects Group Group
(30) (15) (15)
Total Positive .A8** .2A .38
Identity .37* .11 .27
Self-Satisfaction .A0* .lA .A3
Behavior .37“ .33 .19
Physical Self .25 - .10 .30
Moral-Ethical Self .23 .18 .28
Personal Self .A5* .30 .31
Family Self .53" .AO .lA
Social Self .39” .11 .A6
True-False Ratio .22 .16 -.05
Distribution .17 .A2 .32
Defensive-Positive .A9** .37 .63*
General Maladjustment .26 -.01 .1A
Psychosis .05 -.22 -.08
Personality Disorder .A1* .16 .58“
Neurosis . .52** .32 .36
Personality Integration .06 -.21 .32
Number of Deviant Signs -.03 -.38 .00
Self Criticism .AA* .80" .11
Net Conflict .09 .01 .23
Total Conflict .05 .01 -.08
Total Variance .17 .0A .27
Locke—Wallace Scale .AA* .A2 .25
Family Concept Inventory .62** .36 .A3
Interpersonal Competency l -.26 -.19 -.A3
Interpersonal Competency 2 .16 .3A -.05
Interpersonal Competency 3 .01 .02 .00
Interpersonal Competency Total .02 .3A -.A3

 

*p < .05

“.2 < .01 two-tailed test

63

When the couples were categorized as "spouses similar"
versus "spouses dissimilar," in terms of NDS, it yielded a

pattern as presented in Table 27.

Table 27

Similar versus Dissimilar Couples in Both Groups

 

 

Clinic Group Non-Clinic Group
Spouses Similar 6 12
Spouses Dissimilar 9 3

 

0n the basis of Chi-Square test, the null—hypothesis
of no association between group status and pattern of
similarity can be rejected (32 a 5.00; g; = l p < .05).
When the two groups were combined there were 18 "homogen-
eous" and 12 "heterogeneous" couples. This ratio of .6
was not significantly different from .5 (g a 1.22;

p < .39).

The interspouse correlations for Defensive Positive,
Personality Disorder and Neurosis were significantly dif-
ferent from zero in the predicted direction. None of the
interspouse correlations of response-set measures were
significant. .The very high (3 I .80) interspouse correla-
tion on Self-Criticism for the Clinic Group was not

repeated in the Non-Clinic Group.

’6“

Both the LW and FCI interspouse correlations were
significantly different from zero. There were no signifi-

cant interspouse correlations for the IC measure.
Relationships among Variables

Demographic Variables and Marital Adjustment Scores

Product-moment correlation coefficients, multiplied
by 100, between demographic and MA measures are presented
in Table 28.

Age and number of years married were negatively
correlated with FCI and to a lesser degree positively
correlated with LW. Income tended to be negatively corre-
lated with both LW and FCI, however, there was considerable
variation in this relationship between groups. Number of
daughters tended to be positively correlated with LW.
Demographic Variables and Tennessee
Self Concept Scale

Product—moment correlation coefficients, multiplied
by 100, between demographic and TSCS scores are presented
in Table 29. Most of the correlations were low, with few
statistically significantly different from zero. NDS was
positively related to age, and negatively to level of
education, which in turn was positively related to Per-
sonality Integration and negatively related to Total Con-
flict. Also, annual income was positively related to

Psychosis.

65

 

 

 

 

 

 

ammo omfifimpuozp Ho. v my» “mo. v m.
cm u 29 .om a 2m
ma .amm mm HH ma- ma mH 3H Ho ma cmaoafino
so .mm *sm mm Ho mm am am ma .mm mnmpnmsma
NH mm aa mo- om- mo- OH- NH- man OH- meow
no man: mm! mm *tmml 0H mml mo xtmml ma Umfihhmz mhwmw
so- Nm mm- .03. was- man mm- om- .mm- Hm- meoocH
mm- mo mo- 0H- ma Ho so mo: OH mo coaumosom
Ho .H: ma: 3mm **smu mo mm: mo ..H:u mo mw<
Hum 3m Hum :4 Hum :4 Hum 34 Hum 34
OHcHHoncoz poacfiao pmmHmEmm nmmfim: «muomfinsm HH<

 

monommmz pcosum3n6< Haufipwz new oasampwosmo cmmzpmn mCOdumHmhnoo

mm wands

Correlations between Demographic Variables and

66

Table 29

Tennessee Self Concept Scale; All Subjects

 

 

Years
Age Education‘ Income -Married~

Total Positive -19 -02 ~12 L014
Identity ~17 ~08 ~l3 ~0U
Self Satisfaction ~16 -02 ~09 -03
Behavior ~15 ~02 ~14 —ou
Physical Self ~21 01 -11 ~03
Moral Ethical Self O3 -O2 O6 16
Personal Self ~22 ~13 ~09 ~08
Family Self ~18 O7 -21 —15
Social Self -08 -Ol -09 -Ol
True—False Ratio 11 ~22 09 12
Distribution 09 ~29 ~06 12
Defensive Positive ~07 ~09 ~21 03
General Maladjustment ~12 00 ~13 05
Psychosis 2M 08 29* 17
Personality Disorder ~06 ~05 00 03
Neurosis ~3l* ~03 ~l7 ~20
Personality Integration ~10 3H** 12 03
Number of Deviant Signs 37** ~28* 19 2M
Self-Criticism ~07 10 03 ~07
Net Conflict 09 00 16 08
Total Conflict 18 ~37** ~20 08
Total Variance 17 ~11 ~02 05

 

y_=6o

*p < .05
**E < .01

two—tailed test

67

Demographic Variables and Interaction Scores
Product-moment correlation coefficients, multiplied

by 100, between demographic variables and interaction

scores are presented in Table 30. Number of children was

correlated positively with CFm 3, CFm total, DOM 3, DOM

total, and with DT 1. Number of children was negatively

correlated with CFf 3. Number of years married was posi—

tively related to DT 2 and DT total. Annual income was
related positively to DT 3 and DT total. Social Class

was correlated negatively with SA 2.

Marital Adjustment and Tennessee Self Concept Scale

Product-moment correlation coefficients, multiplied
by 100, between MA and TSCS scores are presented in Tables
31 and 32. Hypotheses 9 and 10 predicted positive corre-
lations between MA and TSCS self-esteem scores, and nega-
tive correlations between MA and TSCS pathology scores.
These hypotheses were on the whole confirmed. TP was
positively correlated with LW and FCI, as were a number of
the sub-scores; viz. Identity, Behavior, Family Self, and
Social Self. In addition, the FCI score was also posi-
tively correlated with Self-Satisfaction, Physical Self,
and Personal Self.

Conversely, LW and FCI scores were on the whole
negatively correlated with positive indices of pathology

on the TSCS. The only exception was again the Defensive

68

Table 30

Correlations between Demographic Variables
and Interaction Scores

 

 

 

 

 

 

 

 

Annual Years Number of Social
Index Task Income Married Children" Class
Spontaneous l 1U ll ~05 ~08
Agreement 2 10 ~11 04 ~38*
3 ~11 ~02 ~09 ~08
Total 0A 00 ~08 ~25
Choice 1 ~01 ~13 23 ~06
Fulfillmentm 2 25 10 22 05
3 ~20 11 36* 29
Total 01 10 54** 23
Choice
1 ~07 ~05 ~21 ~01
Fulfillmentf 2 22 19 19 ~08
3 -05 03 ~HO* -05
Total 05 10 ~26 ~08
Dominance l 20 02 3A ~27
2 06 ~06 09 ~04
3 —1o 09 56** 23
Total l8 12 58** 00
Decision 1 13 22 36* ~11
Time 2 O7 “0* ~01 22
3 A9** 29 ~08 ~12
Total 36* “2* 07 ~01
Interpersonal l 06 ~01 ~25 15
Competencec 2 lb ~05 ~03 ~18
3 08 ~20 ~01 ~31
Total 22 ~11 ~12 -3u
N. = 30
*p < .05
**E < 01

Correlations between Marital Adjustment

69

Table 31

and Tennessee Self Concept Scale

 

 

 

 

 

All Subjectsa Malesb Femalesb
LW FCI LW FCI LW FCI.

CPotal Positive 35** 55** 31 H9** 37* 61**
IEdentity 3N** U9** UU* us* 28 53**
ESelf Satisfaction 36* U7** 1“ U2* 31 53**
IBehavior 33** 51** 19 A0* 39* 61**
Eflnysical Self OO 26* ~09 O7 O2 41*
Ddoral Ethical

Self 25* 19 32 18 28 ”3*
I?ersonal Self 28* “8** l9 “0* 32 56**
IFamily Self 51** 71** 59** 75** 47* 67**
ESQCial Self 32* 44** 25 “2* 35* N6**
TPrue-False 01 01 -Ol 10 O6 05
IDistribution 28* 25* 22 40* 35 18
[Defensive

Positive 32* 37** 18 2M 38* 48**
General

Maladjustment 38** 51** 37* U3* 39* 58**
I?sychosis ~13 ~32* ~08 ~20 ~17 ~UA*
P ers onali ty

Disorder 32* 38* 30 30 37* u7**
IVeurosis 26* 57** 18 51** 30 6M**
IPersonality

Integration 10 19 09 05 08 3A
Nunmer of

Deviant Signs ~18 —uo** ~2H ~20 ~13 ~56**
Seli'Criticism ~20 -O9 -2l ~06 -19 ~12
Net Conflict 03 Ol 00 ~01 O7 02
'Total Conflict ~17 ~22 00 —ou ~26 ~39*
'Potal Variance -07 -32* 01 -20 ~08 —u2*

‘

ag 6o

L1=30

70

Table 32

Correlations between Marital Adjustment and
Tennessee Self Concept Scores (Continued)

 

Clinic Group Non-Clinic Group

 

 

 

 

LW FCI LW FCI
U?otal Positive 12 23 31 51**
IEdentity 12 15 30 ”1*
ESelf Satisfaction 07 26 20 3M
I3ehavior 12 16 34 62**
IPhysical Self ~29 O2 10 21
Pdoral Ethical Self 05 ~06 H8** UA*
IDersonal Self 1U 29 12 27
Ifamily Self A6** 51** 30 62**
ESOCial Self 12 O2 31 58**
fPrue-False 05 03 08 27
IDistribution 02 -1u H8** H9**
IDefensive Positive 38* 25 01 26
C§eneral Maladjustment 18 16 36* 52**
I?sychosis 07 08 ~02 ~27
I?ersonality Disorder 10 12 “1* 41*
lJeurosis 1A 37* 02 36*
I?ersonality Integration ~09 05 ‘12 ~10
lJumber of Deviant Signs ~02 ~16 03 ~06
fSelf Criticism ~u8** —27 11 05
IVet Conflict 05 16 18 1a
rTotal Conflict ~1U ~13 ~19 02
'Total Variance 02 ~30 06 ~01

i=30
*p < .05

**p < .01

71

Positive score. High LW and FCI scores were associated
with less pathognomonic scores on the General Maladjust—
rnent, Personality Disorder, and Neurosis Scales. High
IPCI scores, in addition, were significantly related to
Zless pathognomonic Psychosis and Number of Deviant Signs
Scores.

Of the response—set measures, the Distribution
£3core correlated positively with LW and FCI and Total
\fariance correlated negatively with FCI.

It is clear that on the average, the FCI correlated

rnore highly with the TSCS than did the LW.

Ddarital Adjustment and Interaction Scores

Hypothesis 11 predicted positive correlations between
PTA and positive interaction scores (SA, CF, and IC).
Iiypothesis 12 predicted negative correlation between MA
zand DT scores. Neither hypothesis was supported. Product-
Inoment correlations, multiplied by 100, between MA and
iJiteraction scores are presented in Tables 33 and 3A.
rI‘able 33 presents correlations between couples' interaction
EScores and couples' MA scores (the latter were obtained
by summing spouses' individual scores).

There were no significant correlations between indi—

Vidual interaction and MA scores (see Table 3“).

72

Table 33

Correlations between Marital Adjustment
and Interaction Scores

 

 

 

 

 

 

 

 

Males Females Couples
‘VEtPiable Task LW FCI LW FCI LW FCI
Sg>ontaneous l 00 13 0“ ~02 02 06
.Agreement 2 06 ~18 ~11 ~06 ~04 ~13
3 37* 19 27 35 36* 30
total 27 13 17 20 25 19
Dcnninance l 03 31 15 15 12 26
2 -O8 O6 18 14 O9 11
3 -01 23 O7 O5 0“ 15
total ~01 18 2“ 1H 16 18
De<1ision 1 26 22 26 O2 31 13
'Time 2 ~21 ~13 OH -25 -O8 -21
3 -30 ~18 ~18 -U5* ~27 -35
total ~19 -09 ~01 ~37* ~09 ~26

i = 30

73

Table 3%

Correlations between Individuals' Marital
Adjustment and Interaction Scores

 

 

 

Variable Task LW FCI
Chuoice Fulfillmentma l 03 32
2 ~17 ~2H
3 15 17
total 01 ll
Ckuoice Fulfillmentfa 1 ~12 ~09
2 ~OA ~29
3 OH 01
total —ou ~17
Irlterpersonal Competencyb 1 0“ ~16
2 ~22 ~03
3 00 18
total ~11 ~02

afl = 30

by; = 60

Terniessee Self Concept Scale and Interaction Scores
Hypothesis 13 predicted positive correlations between
TSOES self-esteem and positive interaction scores (SA, CF,
and.IIC). This hypothesis was only minimally supported.
PrCKhict—moment correlation coefficients, multiplied by 100,

are presented in Tables 35, 36, 37, and 38- TPf was

Cor'I‘elated positively With SA 3, but there were no signi—

ficant positive correlations between TPm or f and SA tOtal:

CF total, or 1C total. Personality Integrationm was

74

 

 

Ho. v m**
mo. v m*
om u 2
No- mHu mN- *N:. :H: mocchm> Hmpoe
NH- NH 00 *ms. NH QOHHocoo Hmpoe
mH mH oo **Nsn mH- HOHHNcoo poz
mo mo *0: mm: so: EmHOHpHHo maom
NH: mN HH- **om| mo- mcme pcmHsmm mo Hmnszz
wHI Hon mm *2: mo COHomemocH zpflamcompmm
NN OH: mo: Nm No- mHmoszmz
NH: mm so: mm :H twosomflm thHmCOmpom
3N: Ho: HH mo: 3 mHmocosmm
as :H MH mN mH: pamspmSWUNHmz Hmsmcmm
AH ma mm: :m NH: o>HpHmom m>fimcmmma
mo mm Ha ma: non COHuSQHmeHQ
om mm won **om| sml OHpmm mmfimmumspe
mo NH mo mm so: m>fip~mom Hmooe
mocwpoasoo mEfiB mocmsflsoo pcosmmpw< pcmEHHHmasm
Hmcompmopoch coamflooa msomcmpcoom mowoco

 

moaoom Hapos QOHpomHmch ocm

momE mam: ompomamm cmmzpmo mCOHomHohhoo

mm mamas

75

Ho. v m**

 

 

 

 

 

 

 

 

mo. v Q:
om n m
mHI mml mm wml mH OHI xHI mol Hml NM! ncl *Hzl mHI HH SNI mocmHLm> HMpCE
MHI le Nm :0! mm mo NH :0! cc *wwl cm! cal we cc mH poHHLCCQ pr09
HH OH NOI 0H we we NH wHI Ho mml dml :ml 0 *Gzl mHI poHHucoo um:
mOI :0 oo :0! co Hm mel mm *H: mm! Hm! Gal oml CH mH - EwHoHpHLO Mme
NMI 001 OH om mo HH OH QM! do *Oal QCI **mul me me mH- wcmHm mcmH>wfi mo LOQEDZ
no HOI **0:I me! we! :H MOI me am *Nw no H, HH mcl mo COHmemmch >HHHmCCmme
Hm Hm :0! mHI mo! 00 HO mH CHI *(m wal cm HH mml mo - . mHmOLsmz
mo oml NHI, :H mH *am 00 mo OH *mm mHI mc 00 mo HH LocLCwHQ >HHHmcow$¢m
mH- mN- mN- NO ON- HH 0H HH- CH No- *cm *Hs- AH as mo- ‘ 38:03.1
wH Hm QHI mo so m HH cm non cm [H1 am mc- mm: mH ocoaomshomHmz Hmtocco
mm :H mOI HH OH HO CHI mm *wml Hw mm! mm mH mml OCI ®>Hproa ¢>Hmcmwma
om 00 :H mo *xo: :m mo 0H mo moi *xsul CH mH oml NH :chseHLomHQ
mH mH HH 2H Hm OH mH men wen oH- **cm- da- mm **mm- mH- oHpmm mmHmm-ozcs
OH PH OH! Ho mH wm HOI om mcl mm Us! um QC mml OH ®>HpHmOm kuCP
m m H m m H m m H m n H m m H
mocopmasoo mEHB mocmcHEoo pccficmsm< ucmEHHHLHzm
Hmcomgmquch coHMHowQ msomcmpccon ooHoco
mmcoom :oHpomLoozH ocm moms mHmz wouomHom coozpmn wcchmHmLLou

mm @2st

76

Ho. v NT...

 

 

mo. v m...
om u 2
0H NH so *ozl so mocmem>
mH- oH HH mH- oo HOHHmcoo Hmpos
:o HH oo Hm- mo- uoHHmcoo poz
mN mo mN oNI HHI EmHOHpHHo MHmm
No oo NH oH- NN mcme ocmH>oQ mo Hmossz
NN NN- mH- **mz sH cOHHNHmmch spHHNcomHmN
No om- mo *om oHI mHmoHomz
Hm mo- mo 0H HH- HmsHomHo spHHNcomHmN
NH mo HN- oH- mN mHmocosmN
mo so- so sN sH- pcmspmsnsmHmz Hmpmcmo
:o- so- so- sN HH- COHHHmom m>Hmcmmmo
moi Ho mN moi mH- :oHpsoprmHo
so NH oo om- oN- OHHNm mmHmm-osms
mo sH- oo Nm mH- m>HpHmom Hmpoe
socmpmosoo mEHB mocmcHEom pomsmmpm< pCmEHHHmHzm
HmcowpoopousH COHmHomo msoocmpcoom moHogo

 

mmpoom COHuomHmch ocm moms mHmEmm omuoonm :mmzpmo

sm mHomB

mCOHpmHmppoo

77

HO. v W**

 

 

 

 

 

 

 

 

mo vm*
om um
oH mo oNi Noi mm moi oo- HHi. Hoi *om- omi oo- mH moi :oi mocwHLm>
NH- mH oNi :o :N oHi HH moi NH NH- sHi oH Ho mo soi HOHHocoo Hmpos
mo ooi Hoi moi *om moi HH sHi No HNi moi ooi soi mHi Ho poHHocoo p62
sN mo :H mo HH mo OH NOi sH :Hi sNi ooi moi sHi :oi : EmHCHoHLo onm
zo- moi .uN Hoi sN sHi NH HHi Hoi Nmi soi mH NH mH NH acaHm.m:mH>o@ so Lopez:
NN HH oo NHi *zzi mH oHi mH Noi xxmz *om Ho oo NN Ho :oHpmsmmch spHHNCOmea
mOi mH sHi :Ni mmzi mo oo oH mo **s: CH oo moi Ho mHi . mHmossmz
mH so oN oo sN-. oN HH- *Ho so so Ho mo- mo oH- oH- twosomHo soHHoccosmN
HH moi om Hoi oN Hoi sNi mo Ni MH- mo mHi a NH No mHmocosmm
No mo mHi moi oHi :H mo :H mo *sm coi No HHi mHi ooi ocoEumsnomHmz Hmsmcmo
Noi moi oHi oHi mHi mH moi oH mo *N: Hoi Noi moi oHi NHi m>HuHmca w>Hmcwooo
oo NHi oo soi oH HHi mH oH No oo sNi mH OHi moi moi coHuanLomso
Hoi oo Noi sHi *sm mH oH moi oo- ooi :Ni 0N- moi HNi NHi oHomm mmHmmimzss
no so ooi HHi Nmi HH Ho mN oo *o: no mo HHi moi soi m>HuHm0d Hopes
m N H m N H m N H N N H m N H
mocmquEoo mEHs oocmcHEoc oowsmosmq pcoEHHHstm
chompoogmucH conHomo mscmcmpcccm ooHoco
mmooom :chowLmocH now moms chscs ooHocHoo ocmsumn mcoHpmHoLpoo

mm

oHoos

78

correlated positively with SA 3 and SA total, as predicted,
but negatively with IC 1. Similarly, Personality Integra—

tionf was correlated positively with SA 2, SA 3, and SA

total.

Hypothesis 1A predicted negative correlations between
TSCS indices of pathology and positive interaction scores
(SA, CF, and IC). Psychosism correlated negatively with
SA 1, as predicted, but correlated positively with SA 2.

Personality Disorderm and Neurosism (both inverse scales)

were correlated positively with SA 3 as predicted. The

correlation between Number of Deviant Signsm and SA 1, SA

3, and SA total were consistently negative. There were no

significant correlations with IC and CF.

TSCS pathology scores of females did not correlate
as consistently with interaction scores as did those of

the males. General Maladjustmentf and Neurosisf (both

inverse scales) were correlated positively with SA 3 and
SA 1 total, respectively, as predicted. However, there

was a positive correlation between Defensive Positivef
ENNi SA 3. Number of Deviant Signsf did not correlate

Witki any of the interaction indices.

Hypothesis 15 predicted DT to be positively related
to TTSCS pathology scores and negatively related to TSCS
selfl-esteem scores. DT l was positively correlated with

Perscnqality Disorderm (inverse scale) as predicted. Also

79

in support of the hypothesis were the negative correla-

Neurosis (inverse

tions between Personality Integrationf, f

scale) and DT 2.

CHAPTER V

DISCUSSION

Selection of Subjects

It is clear that the Clinic and Non-Clinic Groups
differed significantly on a number of demographic varia-
bles, most notably age, number of years married and to a
lesser extent, level of education. Clinic gs were gener-
ally older and had been married longer. While participa-
tion was high among clinic couples, the participation
rate of the non-clinic couples could not be accurately
determined since it was unknown how many initial letters
were sent by one of the ministers. However, the fact
that six couples responded, of nine nominated by the
other minister and contacted by E, suggests a lower
participation rate. It is unknown how this difference
in response affected the sample characteristics. In
addition, clinic §s were drawn from an industrial city
While non-clinic SS were obtained from a university
tOWTl. Although these group differences constituted an
unfOlf'tunate confounding factor, observed differences
in I’eported individual and marital adjustment cannot be
eXplflined as due only to demographic background. This

eXplErnation was excluded by the results of the analysis

80

81

of covariance of MA scores, controlling for years married
(Table 6). Moreover, demographic variables did not
correlate significantly with most of the dependent meas-
ures (see Tables 28, 29, and 30). In addition, the
purpose of this study was not just to compare groups
matched on demographic variables; rather, the selection
procedure was in part designed to maximize the statis-
tical power of the correlational analyses by increasing
the range of scores. For this purpose the groups were

frequently pooled.

Marital Adjustment

Both the LW and FCI differentiated between the
two groups at statistically significant levels (see
Table 5). Assuming that relatively low scores on the
LW or FCI indicate a disturbed marriage, these results
lend credibility to the statement by Framo (1967) that:
"Whenever there are disturbed children there is a dis-
turbed marriage...(p. 154)." It must be remembered
that in the vast majority (13 of 15) of clinic couples
the only selection criterion was that at least one of
their children had been referred to a clinic for
emotional difficulties. These Ss had not necessarily
admitted that their marriage was unsatisfactory.

Given the fact that mean LW and FCI scores differed
Significantly in the two samples, the question still

remained as to how well these instruments discriminated

82

clinic from non-clinic subjects. Of 30 clinic §53 9
scored below a raw score of 95 on the LW while 2 of 30
non-clinic SS scored that low. The FCI discriminated
even better. Whereas 22 clinic SS scored below a raw
score of 135, only 2 of the non-clinic gs did.

The correlations obtained between the LW and FCI
(see Table 7) were somewhat lower than was to be expected
from previously reported investigations (see Table 2).
There were no significant differences in these correla-

tions between groups or sexes.

Tennessee Self Concept Scale

Of 15 statistically significant mean differences
on the self-esteem and clinical scales of the TSCS, 1A
were in the predicted direction. On the Moral Ethical
Self Score, the only index which differentiated between
sexes, the mean difference was not significant. There
was a statistically significant mean difference between
the two groups on the Defensive Positive Scale; however,
the Non-Clinic Group obtained a more pathological score
than did the Clinic Group. Fitts (1965) described this
scale as "...a more subtle measure of defensiveness than
the Self-Criticism Score" (p.5). The Defensive Positive
Score is based on 29 items which differentiated a group
of "...100 psychiatric patients whose Total P Scores
were above the mean for the Norm Group (p. 5)." However,

it must be noted that Fitts (1965) reported that a

83

"Personality Integration Group" (N = 75) scored signifi—
cantly higher on the Defensive Positive Score than did
a "Patient Group" (N = 363). The respective means are
listed in Table 2. The Defensive Positive Score was
derived in a manner similar to the MMPI K scale with which
it is correlated (3 = .29; Eta = .A2; N = 102; Fitts,
p. 25). The latter is sometimes interpreted (e.g.
Heilbrun; 1961) as an "Ego-strength" measure, so the
higher Non-Clinic Group score may be appropriate.

In general, the Clinic Group most closely
approximated the pattern of TSCS scores as reported
by Fitts for a psychiatric patient group (see Table 2).
The Non-Clinic Group's profile of scores was most
similar to the Personality Integration Group, although
the level of scores most closely approximated the Norm
Group (see Table 2).

Again it should be noted that the vast majority
of clinic SS did not necessarily consider themselves
to be individually maladjusted. They were designated
as clinic §s because they, or someone else, had referred
their child to a clinic. However, these parents, as
individuals were significantly less well adjusted, as
measured by the TSCS, than were parents whose children
were not referred. Framo's statement (1965) that there
is a disturbed marriage whenever there are disturbed
children might well be expanded to "Whenever there are
disturbed children there are disturbed parents engaged

in a disturbed marriage".

8A

The intercorrelations among TSCS scores (see
Table 11) were essentially similar to those reported

by Fitts (1965).
Interaction Scores

Spontaneous Agreement

Of the interaction scores, SA differentiated the
two groups most successfully. As predicted (hypothesis
A) the Non-Clinic Group's mean SA was higher than that
of the Clinic Group on Task 3, also on the total SA
score. The Non-Clinic Group also achieved a (not signi-
ficantly) higher SA score on Task 1, but this trend was
reversed on Task 2 where the Clinic Group showed higher
initial agreement between spouses. In an analysis of
variance of the SA scores (see Table 13) the Task and
the Task X Group effects were statistically significant.

However, interpretation of these results must be
guarded because of two confounding factors. First, the
present design did not control for a possible order
effect. All SS completed the tasks in an identical
sequence. Secondly, a review of the distribution of
rankings assigned to each item, as presented in Table
39, suggested that the three tasks were not comparable.
Tasks 2 and 3 each contained one item which the vast
majority of gs ranked identically, thus spuriously
increasing the SA score. Similarly, Task 2 contained

20 non—used item-rank combinations, compared to 10 and

85

Table 39

Item Rank Frequency

 

Item

Rank

 

96“.- 73786
1

lhfilOSBS/U
111

19824691
11

l23hfi5678

Task 2

36859720
1 1

20395100
22

00000055
5

OOO2-IH275
l3

l23i45678

Task 3

09723270
111
3602u230
llll
noun/.1854.“-
l 2

00001162
5

18789272
1 l

i-
1A
A
i-

22

2500

8
11
1A
20

12385678

 

86

3 for Tasks 3 and 1 respectively. Any conclusions
regarding initial agreement between spouses, as well
as CF, and DT, across tasks must, therefore, be very
tentative.

The significant Task X Group effect leads to the
conclusion that it is important to consider in which
content area SA is measured. It seems ironic that
with the pedagogue's emphasis on inter-parental consist-
ency, children with emotional difficulties should have
parents who agreed more on desirable parent-child
interactions than did parents of children not referred
to a clinic. However, it is also possible that those
parents who had been sufficiently motivated to avail
themselves of clinical consultation had become more
sensitized to their parental role. Contact with a
clinic, or the decision to cooperate with a clinic, may
well have led to increased discussion of disciplinary
methods between parents, if only in an effort to portray
a "united front" or to do the "correct" thing, and thus
a higher SA score.

Task 3, which discriminated most successfully
between groups, may also be characterized as the one

which engaged §s most directly and personally.

thice Fulfillment
The two groups did not differ significantly in
terms of mean CF scores for any of the tasks (see Table

14). An analysis of variance of CF scores (see Table

87

15) showed a significant main Task effect. This may
most parsimoniously be explained by the fact that CF
and SA were positively related (see Table 23-and 2A).
It seems logical that as SA increases, CF increases.
The significant Task effect was, therefore, most
likely due to the differences in SA scores between
Tasks, which in turn were at least in part due to

the non-comparability of the relative attractiveness
of the items on the tasks.

The CF measure appeared to be less powerful than,
and essentially redundant with, the SA index with which
it was significantly correlated. Intuitively it appears
logical to correct the CF score by subtracting SA. A
difficulty with this procedure is that then couples with
high SA scores would be much more likely to obtain low
CF scores. High CF scores would similarly be obtained
by cOuples with low SA scores. In addition to the
relationship between SA and CF, a S's CF score is also
dependent on CF of spouse. The DOM measure is a more
meaningful index which takes into account CF of both
patterns as well as the initial agreement. CF was
retained in the analysis because it is an individual

score, while SA is a-couple score.

Qggision Time
Ferreira & Winter (1965) reported that normal
families required significantly less time than did

clinic families to complete a task similar to those

88

used in the present investigation. This difference
was replicated in the same sample of families using a
TAT task (Ferreira et a1., 1966). Similar conclusions
were reached by Haley (1962) and Cheek (196A). This
difference in time needed for task completion was not
verified. There were no mean DT differences between
groups (see Tables 16 and 17). It should be noted
that the very high variance almost precluded any
significant mean differences. A possible explanation
of this discrepancy in reported findings might be due
to differences in instructions. No attempt was made
to convey to Ss that speed was desirable, neither
did S make a conscious effort to convey the opposite.
It is unlikely that Ss felt that time was of importance.
Only one reference to time was made in the instructions
(see Appendix A, p. 128), Ss were told that they need
not time themselves on the TSCS.

An added consideration is that non-clinic Ss
were paid $15, while clinic Ss were paid only $5.
Cognitive dissonance theory would predict that those
Ss who felt that the remuneration was excessive
would spend relatively more time on the requested
task (see Adams & Rosenbaum, 1962). If Ss from the
an-Clinic GrmnDindeed felt that they were being paid.
too much, it would tend to reduce any differences
between the two groups, given that there is a real

difference. Non-clinic Ss were not aware that other

89

Ss had been paid $5, rather than $15. Although a number
of both clinic and non—clinic Ss commented that payment
was not necessary, in most cases it appeared to be a

polite, expected comment. No Ss refused payment.

Dominance

 

The DOM measure was formulated post hoc to assess

 

the relative degree of influence each spouse asserted
on the joint decision, taking into account the degree
of initial agreement. Since DOM was defined as a ratio
of rank-order correlation-coefficients it was considered
a nominal measure. Chi-square analyses of frequencies
of DOM scores (see p. 55) suggested no relationships
between relative influence and sex status or group
membership. There was, however, a clear tendency for
husbands to be more influential on Task 1 regardless

of group membership. This tendency was not evident for
the other tasks.

The results suggest that the allocation of decision
power or dominance was related to content of the task.
This conclusion is similar to that of March (1957) that
spouses tend to escape conflict by allocating power to

each other in different areas.

Interpersonal Competency
The IC scores of individuals or couples did not
differentiate between groups for any of the tasks (see

Tables 20 and 21). The prediction that non-clinic

90

couples would show a greater degree of interpersonal
competency during the task discussion was not confirmed.
However, an analysis of frequencies of ratings in each
category (see Table 22) resulted in a significant'
difference between groups in terms of percentage of
ratings assigned in feeling categories rather than idea
categories. The latter finding supported the conclusion
of Goodrich & Boomer (1963) and Shuham (1968) that
clinic families, as compared to normal families express
more, primarily negative, affect. The Non-Clinic Group
remained task oriented to a greater degree than did the
Clinic Group, which resorted more frequently to the
expression of affect.

The non-differentiation between groups by the IC
summary score may have been due to the procedures
utilized. The rating procedure was such that all initial
disagreements between the two raters were resolved.
Although there were a number of simple, obvious mistakes
where there was no question as to what the rating should
be, there were even more instances where the disagree—
ment was based on differences in interpretatiOn of what
was being said or what the effect of what was said
would be on the other spouse. In many of these instances,
agreement was obtained by deciding on the more conserv-
ative (i.e. less heavily weighted) rating. Most disagree—
ments occurred in "feeling" and other heavily weighted

categories. Since these were often resolved by

91

compromise on a less heavily weighted score, the dis-
criminating power of the IC summary score may have been
minimized inadvertently.

The non-differentiation of the IC summary score
may also be partly due to the purpose for which it was
originally developed. Argyris (1965) developed and
validated the rating procedure on the communication
in T groups; i.e. SS 222 groups. Ryder (1968) showed
that the verbal communication between "married strangers"
was considerably more polite, with more effort at
clarification than that between spouses. The tendency
of families to use more "short-hand communication" than
29 299 groups most likely also resulted in a less valid

rating.

lntercorrelations of Interaction Scores

 

Some of the intercorrelations of interaction total
scores (see Table 23) were spuriously high. The signifi—
cant correlations of ICC with 10m and ICf resulted
because ICC was a combination of 10m and ICf. Similarly,
the significant correlations between DOM and CFm and CFf
were due to the definition of DOM (see page 55). The
significant negative correlations between DT and SA were
to be expected since a higher degree of SA resulted in
fewer disagreements to be resolved. Also, SA was expected
to be positively correlated with CFm and CFf since the

higher the SA, the less the necessity of extensive changes

from individual to joint rankings.

\O
[\J

The remaining significant correlations of Table
23 were associated with indices of female functioning.
le was positively related to CFf and SA. CFf was
negatively related to DT. An inspection of Table 2A,
which presented intercorrelations for each task, revealed
that the positive correlations of ICf with CFf and SA
were most pronounced in Task 3. The negative correlation
of CFf with Dt was more consistent across the three tasks.
Table 2A also contained many spuriously high

correlations. These included CFm and CF with SA and

f
DOM, and SA with DT. The intercorrelations across tasks
for the same interaction index, suggested that the 10m
measure was the most stable, followed by DT. The other
measures showed little consistency in ranking Ss across
tasks.

After setting aside those correlations between
logically related measures (SA with CFm, CFf and DT;
DOM with CFm and CFf), and those between the same
measures across tasks, 1AA intercorrelations remain.
Of these, nine were significantly different from zero
with 23:.05. With the .05 significance level, approxi-
mately 7 of 1AA correlations are expeCted to be signifi-
cant due to random variation. Of the nine significant
correlations, three were between measures but across
tasks; e.g. ICm 2 was positively related to CFm 1. These
relationshipsare difficult to explain and are probably

mOSt safely attributed to chance. Although the same

reasoning might be applied to the remaining significant

3" -"l m Y'RJI'? -

 

1i.

93

relationships, it must be noted that the 10m measures
tended to be negatively related to CFm and DOM, while

the 1C index, on the other hand, tended to be positively

f
related to CFf. Hypothesis 16 predicted a positive
correlations between 10 and CF scores. This hypothesis
was based on the assumption that high interpersonal
competency includes the ability to obtain personal
rewards in an interpersonal situation. Given the present
measures of the constructs interpersonal competency and
personal satisfaction, this hypothesis is not tenable.
Although the predicted relationship was obtained between
le and CFf,
ICm and CFm tended to be negatively correlated. The

it was obtained for Task 3 only. Moreover,

added consideration that ICm tended to be negatively
related to DOM, in which a high score indicates husband
dominance suggests that interpersonal competency does
not necessarily result in personal reward at the expense
of the other person.

The negative correlation between CF and DT may
well be an artifact since CF is logically related to
SA which in turn is negatively related to DT. The
significant correlation between CFm 2 and CFf 2 was most

likely due to the high average SA obtained on Task 2.

Interspouse Correlations

The hypothesis that spouses tend to occupy

similar positions on a global measure of individual adjust—

ment or maturity was largely confirmed (see Table 25).

 

 

 

9A

Although the correlations between spouses' NDS scores
were not significant, the Chi-square analysis of the
NDS score (see Table 27) provided support for the
hypothesis. The results suggest that the homogeneity
hypothesis is tenable when overall emotional functioning
of the spouses is assessed, rather than personality
traits or psychic needs.

The interspouse correlations obtained for LW and
FCI were generally comparable to those reported elsewhere
(see Table 2), although the interspouse LW correlation
was considerably greater than that obtained by Hofman
(1966). This is somewhat puzzling since the administra-
tion was essentially identical; i.e. there was little
opportunity for spouses to collaborate. The observed
interspouse FCI correlation was almost identical to that

obtained by Palonen (1966).

Relationships-Among Variables

Demographic Variables and Marital Adjustment

 

A consistent relationship found between demographic
variables and marital adjustment was that age and number
of years married (which are of course very highly corre—
lated) were negatively correlated with FCI. These negative
correlations were significantly different from zero at
the .01 level for all Ss. The relationship tended to
be more pronounced for females than males. However,
when the samples were considered separately the relation-

ship was essentially random for the Non-Clinic Group

95

and only minimally present in the Clinc Group. This
negative correlation between length of marriage and
FCI scores was also reported by Updyke. (1968).

This evidence could lead to the conclusion that
the longer people are married, the less well-adjusted
they become in marriage. However, a safer, and more
accurate, generalization is that the longer people
are married the less they perceive their marriage in
a manner congruent with professionals' description of
an ideal marriage. The negative correlation between
length of marriage and MA scores was not replicated
with the LW, which instead tended to correlate positively,
albeit non-significantly, for all S5 with length of
marriage. This might be interpreted to mean that as a
marriage endures, the expectations and perceptions of,
and attitudes toward, the marriage change although this
does not necessarily signal decreased marital satisfac—

tion or happiness.

Dempgraphic Variables and Interaction Scores

 

An inspection of the correlations between demo~
graphic variables and interaction scores (see Table 30)
revealed that number of children accounted for more
variance of interaction scores than did the other
demographic variables. Number of children correlated
positively with husband dominance on Task 3, and to a
lesser extent on the other tasks, as evidenced by the

positive correlations between number obeildPen and CFm 3:

96

CFm total, DOM 3 and DOM total and the negative correla-
tion between number of children and CFf 3. Number of
children was also related to a longer discussion on Task 1.

Decision time was related to annual income on Task
3, years married on Task 2 and number of children on
Task 1. SA was related negatively to social class on
Task 2. It is difficult to propose logical explanations
for these relationships. However, it is quite clear
that on the average the number of children in a family
is associated with the dominance pattern between the
spouses. The husband tends to be more dominant in
families with more children.

This relationship was unexpected. A possible
explanation is that in families with more children,
as in a group with relatively more members, the task
leader's (husband's) role behavior diverges increasingly

from the behavior of the other members (see Bales, 1950).

Marital Adjustment and Tennessee Self Concept Scale

 

Most of the correlations between MA and TSCS scores
were as predicted (see Tables 31 and 32); i.e. MA scores
were on the whole positively related to measures of
individual adjustment and negatively related to individual
maladjustment. A note of caution in interpretation of
these results is suggested by the positive correlation
between the Distribution Score and the MA scores. The
Distribution Score measured the tendency of SS to use

extreme answers (which are more heavily weighted).

Therefore, the positive correlations between TSCS and
MA scores were at least in part due to response style,
since the instruments are similar in format. However,
it was concluded that marital adjustment, as measured
by the LW and FCI, is at least in part determined by
individual adjustment, or vice versa. An individual
who reported himself to be individually well adjusted
also tended to perceive his marriage more congruently
with professionals' definition of an ideal marriage
than did the less well adjusted individual. He also
tended to express more satisfaction with his marriage
and less open conflict with his spouse.

These results support the contention that individ-
ual adjustment determines, at least in part, marital
adjustment. To more precisely assess the association
between individual adjustment and reported marital
satisfaction, a multiple regression method was used.

Both the LW and FCI individual scores were used
separately as dependent variables, with Sex, Total
Positive, Distribution, Defensive Positive, General
Maladjustment, Psychosis, Personality Disorder, Neurosis,
Personality Integration, Number of Deviant Signs, and
Self Criticism as independent variables. These variables
together accounted for 28% of the variance of the LW
score and for 39% of the FCI score. This compared rather
unfavorably with the amount of variance of LW and FCI

accounted for by the Family Self Score, which was the

98

single TSCS score which correlated most highly with LW

and FCI (see Table 31). The Family Self score accounted
for approximately 25% of the LW variance and approximately
A9% of the FCI variance.

The correlations between the TSCS Family Self Score
and the FCI were of approximately the same order as the
reported reliabilities of the FCI (see van der Veen &
Ostrander, 1961; Palonen, 1966). Although the LW did
not correlate as highly with the TSCS as did the FCI,
the over all pattern of correlations was approximately
the same. These results raise serious doubts about the
usefulness of the MA questionnaires. The FCI, while
it discriminated more clearly between adaptive and non-
adaptive marriages than did the LW, also appeared to be
primarily redundant with the Family Self Score of the
TSCS. .

In this context it is of note that Vacchiano &
Strauss (1968) identified 22 factors in TSCS responses
of 260 college—age S5. The most prominent of these,
accounting for 30% of the common variance, reflected a
negative family concept. The second factor, accounting
for 6% of the variance, consisted of a positive attitude
towards the family of orientation.

The correlations showed that for females, the
Physical Self and Moral Ethical Self Scores were related
significantly to MA scores, which was not true for males.

It also appears that the relationship between individual

99

adjustment and marital adjustment was of a higher order
for females than males and was also more pronounced for
the Non—Clinic than the Clinic Group. Similarly, the
FCI scores were more significantly related to TSCS
scores than were LW scores. This may have been due
primarily to response-style since the FCI and TSCS are

more similar in format than are the LW and TSCS.

Marital Adjustment and Interaction Scores

 

The correlations obtained between MA and inter-
action scores (see Tables 33 and 3A) did not support
the hypothesis that marital satisfaction, as measured
by the LW and FCI, is positively related to interaction
indices. Hypothesis 11 predicted positive correlations
between MA and SA, CF, and IC and a negative correla-
tion between MA and DT. These relationships were
minimally present in Task 3 where LWm and LWC were

positively related to SA, and FCI was negatively related

f
to DT. There were no significant relationships between
MA and CF and 10. There were also no significant

relationships between MA and DOM.

Tennessee Self Concept Scale and Interaction Scores

 

It is clear from an inspection of Tables 35, 36,
37 and 38 that the SA index accounted for more of the
variance in TSCS scores than did the other interaction
indices. This was true both of males and females. The

relationships between SA and TSCS scores was most

100

pronounced in Task 3. Considering just Task 3, SA and
TSCS scores of females were on the whole associated as
predicted. SA was positively related to Total Positivef,
Defensive Positivef, General Maladjustmentf, Neurosisf

and Personality Integrationf. Correlations between SA

and Psychosis Personality Desorderf and Number of

f-
Deviant Signsf, although not significant, were all in
the predicted direction.

The correlations between SA and TSCS scores varied
considerably among the three tasks. The present data
leaves it unclear whether this was primarily artifactual
because of the non-comparability (i.e. in terms of
response alternatives) of the tasks (see p. 89) or
whether it was due to the differences in content of the
tasks. However, the significant Group X Task effect
in the variance of SA scores (see Table 13) suggests
that it may well have been due to the content area. The
variability of correlations of TSCS scores across tasks
was also evident for the other interaction indices.

Of the TSCS scales, only Personality Integration
was correlated significantly with SA total for both
males and females. In addition, Number of Deviant Signsm
and Neurosisf correlated with SA total. Somewhat surpris-
ing were the significant correlations between SA total
and True-False Ratiom, Net Conflictm, Total Conflictm
and Total Variancem. This suggested that males who
tended to over—affirm their positive attributes relative

to denial of negative attributes, and who reported a

101

l

greater degree of confusion, contradiction, conflict and
lack of integration, tended to be less in spontaneous
agreement with their spouse.

In addition to these relationships, another
response-style measure, Distributionm, was negatively
correlated with SA 2 and positively correlated with DT 2.
These relationships were not apparent in the other tasks.
Males who tended to be more definite and certain of their
self-perception; i.e. they tended to use more extreme
ratings, were less in spontaneous agreement with their
spouses on Task 2 than were males who were less definite
and tended to play it safe. This lesser degree of
spontaneous agreement in turn necessitated a greater
amount of time for the solution of Task 2.

The correlations between response-style measures
of females and interaction scores were generally in the
same direction as those for males, but the relationships
were not as pronounced for the SA score as they were for
the DT scores. True-False Ratiof and Net Conflictf were
both positively correlated with DT 2; i.e. the more
acquiescent the wife the more time required for solution
of Task 2. DT 2 was also negatively correlated with
Neurosisf and Personality Integrationf; i.e. extended
discussion on Task 2 was associated with couples in
which the wife tended to be dissimilar to a norm group
of neurotics and similar to a personality integration

norm group.

102

Husband dominance was correlated positively with
Self Criticismm on Task 1 and with Personality Disorderf
on Task 2. The positive relationship between males'
willingness to be self—critical and husband dominance
was supported by negative correlation between Defensive
Positivem and DOM 1.

Interpersonal Competencym l correlated negatively
with Personality lntegrationm. It seems somewhat ironic
that of 10A correlations between 10 and TSCS the only

one which correlated significantly did so in the

opposite direction than was to be expected.

CHAPTER VI

CONCLUSIONS

The assessment of marriage remains a difficult task.
In the present investigation, questionnaire methods; i.e.
the FCI and LW, were found to be more efficient and
accurate in differentiating clinic and non-clinic marriages
than were the interaction measures of decision and
communication variables. The results suggested that the
FCI discriminated more accurately and with less overlap
than did the LW between well-and mal-adjusted marriages.
When comparing correlations between individual and
marital adjustment scores versus interaction and marital
adjustment scores, it was clear that marital adjustment,
as measured by the LW and FCI, was significantly more
strongly related to individual adjustment, as measured
by the TSCS, than to the interaction indices. It is
tempting to conclude from this that therefore the "best"
indirect method of predicting marital adjustment is to
assess the individual adjustment of spouses. However,
a serious confounding factor in such a conclusion was
the methodological similarity between assessment of

marital and individual adjustment.

103

10A

The results supported the homogamy theory; i.e.
on the average, spouses who were individually well-
adjusted tended to be married to a person who was also
well-adjusted. Similarly, spouses who were individually
well-adjusted reported a significantly higher degree
of satisfaction with their marriage than did less well-
adjusted individuals. Well-adjusted individuals also
tended to perceive their marital relationship as closer
to a description of the ideal family by professional
family counselors than did those less well—adjusted.

While self-esteem and clinical scores of the
TSCS were significantly related to marital adjustment
scores, the response—style measures of the TSCS tended
to be significantly related to decision process
variables. FCI scores correlated negatively and signifi-
cantly with the length of marriage. Interaction indices
were most significantly related to number of children
in the marriage.

The frequently stated hypothesis that individual
mal—adjustment tends to be transmitted from parents to
at least one of their children was supported. The
obtained results yielded few clues as to the nature of
the process through which this occurs; however, the
hypothesis that emotional difficulties of children
are related to the emotional atmosphere of the family

of orientation received some indirect empirical support.

105

A system of categories developed to assess
individual and group interpersonal competency from
a rating of verbal interaction was found to be not
very useful in discriminating clinic from non-clinic
marriages. However, clinic couples were proportion-
ally more frequently assigned feeling rather than idea
scores, relative to non-clinic couples.

Of the interaction indices used, spontaneous
agreement between spouses in selecting action alter-
natives, and frequency and quality of feeling-expression
in verbal communication, appeared to be most relevant
to the differentiation of clinic and non-clinic
marriages. The results suggested that when assessing
and reporting interaction between spouses it is highly
desirable to consider and describe the task content.
Interaction indices of discussion and selection of
birth control methods, when prevention of pregnancy
was mandatory in a hypothetical situation, were more
divergent for clinic and non—clinic groups of marriages
than were indices of discussions regarding financial

management and parent-child interaction.

Suggestions for Further Research
Although the present investigation showed that
self-report MA measures differentiated more successfully
than interaction variables between what were considered
to be well—and mal-adjusted marriages by the criterion
of child adjustment it could not be determined to what

extent reported marital adjustment or satisfaction was

106

confounded with individual adjustment. In order for a
construct of marital adjustment to be useful, it appears
to be highly desirable to distinguish and separate it
from individual adjustment, even though both variables
may be interrelated. One way in which this might be
accomplished is to compare groups of marriages, matched
on individual adjustment of spouses, but distinguished
by emotional adjustment of children. An item or factor
analysis of responses to a pool of items, drawn from
existing MA scales and related sources may also identify
or suggest variables basic to marital adjustment. Even
more preferable would be a longitudinal investigation
such as that carried on by the Family Development Section
of the Child Research Branch, National Institute of
Mental Health (see above, p. 20) which circumvents

assumptions regarding causality necessary for SS hoc

 

investigations.

A difficulty which remains in cross-sectional
studies is the identification of well-and mal—adjusted
marriages, independent of the instruments of interest.
The usefulness of adjustment of children as criterion
of the quality of the marriage is limited, in part
because it is dichotomous. However, it is deemed
preferable to imminent marital dissolvement as a
criterion since that decision will almost certainly
significantly alter the relationships and attitudes

of the spouses. In addition, the population of

107

interest is not primarily those spouses who have
decided to obtain a divorce. It seems more useful for
therapeutic purposes to be able to identify marriages
with destructive elements before these dissolve the
relationship.

Investigations of self-report measures should
not preclude the analysis of interaction between spouses.
It is admittedly easier to obtain test scores than to
evaluate interaction. However, since most marital
therapy is focused on behavior between spouses, it
seems more important to investigate spousal interaction,
identify destructive elements, and establish baselines,
rather than compare test scores.

It is suggested that further investigation of the
interaction between spouses direct itself primarily to
the analysis of expression of feelings. The statement
of Goodrich & Boomer (1963) that couples can be differ—
entiated by their ability to maintain esteem for self
and other, even in the face of overt disagreement, also
appears to be an important dimension for research.

Along similar lines, another useful framework might be
that of Berne's dichotomies of "I'm OK ~~ I'm not OK"
and "You are OK ~~ You are not OK" (Berne, 1966).

The present research indicated that marital
behavior is determined in part by content of the task.
However, when a variety of tasks are used it is advisable

to ascertain the comparability of tasks in different

108

areas in terms of item attractiveness, distribution
of responses, etc. It is suggested that interaction
between spouses concerning subject areas most
directly and personally affecting them may be most
fruitfully investigated.

Relationships between MA indices and demographic
variables indicate that whatever marital adjustment
consists of, it is not impervious to situational factors
such as number of children, length of marriage, etc.
Further exploration of these relationships seems desirable.

A most important consideration for further research
appears to be a re—evaluation of the marital adjustment
construct. It remains an overly simplified, global
construct. The correlations found between the TSCS
and MA questionnaires approach the reliabilities of the
latter, suggesting that the MA questionnaires are
primarily redundant with a more adequately standardized
and validated instrument. The same consideration also
raises doubts about the validity of the marital adjustment
construct. It is felt quite strongly that what is
needed instead is further investigation of relation—
ship and system variables and the relations of these
to individual (whether spouses or children) and marriage

variables.

REFERENCES

109

REFERENCES

Adams, J. 8., & Rosenbaum, W. B. The relationship of
worker productivity to cognitive dissonance.
Journal of Applied Psychology, 1962, NS, 161 — 16A.

 

Argyris, C. Organization and innovation. Homewood, Ill.:
Richard Irwin, Inc., 1965. (a)

 

Argyris, C. Explorations in interpersonal competency:
I. Journal of Applied Behavioral Science, 1965, l,

 

Argyris, C. Explorations in interpersonal competency: 11.
Journal of Applied Behavioral Science, 1965, l,
255 ~ 269. (c)

 

Bachove, G., & Zubaly, B. M. Exploratory study on family
interrelationships in a small sample of non-clinic
families. Unpublished master's thesis, Bryn Mawr.
Cited in I. Boszormanye-Nagy, & J. L. Framo (Eds.),
Intensive family therapy. New York: Harper & Row,
1965.

 

Bales, R. F. Interaction process analysis. Cambridge,
Mass.: Addison—Wesley, 1950.

 

Bauman, G., & Roman, M. Interaction testing in the study
of marital dominance. Family Process, 1966, S,
230 ~ 2A2.

 

Bauman, G., & Roman, M. Interaction product analysis in
group and family diagnosis. Journal of Projective
Techniques and Personality Assessment, 1968, 3g,

331 - 337.

 

 

Beavers, W. T., Blumberg, S., Timken, K. R., & Weiner, M. F.
Communication patterns of mothers of schizophrenics.
Family Process, 1965, N, 95 ~ 10A.

 

Berne, E. Principles ofggroup treatment. New York: Oxford
University Press, 1966. p. 260.

 

110

111

Bodin, A. M. Family interaction,gcoalition, disagreement
and compromise in problem, normal and synthetic
family triats. (Doctoral dissertation, State Uni—
versity of New York at Buffalo) Ann Arbor, Mich.:
University Microfilms, 1967. No. 66 — 7960.

 

 

 

Bowen, M. A family concept of schizophrenia. In D. D.
Jackson (Ed.), The etiology of schizophrenia. New
York: Basic Books, 1960.

 

Bowman, L. Research in family dynamics: A criticism and
a proposal. Social Forces, 1956, 33, 201 ~ 207.

 

Burgess, R. W., & Wallin, P. Courtship, engagement and
marriage. New York: Lippincott, 195A.

 

Caputo, D. The parents of the schizophrenic. Family
Process, 1963, S, 339 ~ 356.

Cartwright, D., & Zanders, A. Group dynamics: Research
and theogy. Evanston, 111.: Row—Peterson, 1962.

 

 

Cheek, F. E. The schizophrenogenic mother in word and
deed. Family Process, 196A, 3, 155 ~ 177. (a)

 

Cheek, F. E. A serendipitous finding: Sex roles and
schizophrenia. Journal of Abnormal and Social
Psychology, 196A, 69, 392 ~ A00. (b)

 

 

Davis, K. B. Factors in the sex life of twenty-two hundred
women. New York: Harper & Brothers, 1929.

 

Dean, D. G. Emotional maturity and marital adjustment.
Journal of Marriage and the Family, 1966, SS, A5A —
357.

 

Eshleman, J. R. Mental health and marital integration in
young marriages. Journal of Marriage and the Family,

1965, g1, 255 - 262.

 

Farina, A. Patterns of role dominance in a conflict in
parents of schizophrenics. Journal of Abnormal and
Social Psychology, 1960, SN, 31 ~ 38.

 

 

Farina, A., Storrs, C., & Dunham, R. Measurement of family
relationships and their effects. Archives of General
Psychiatry, 1963, 9, 6A ~ 75.

 

 

Ferreira, A. J. Decision making in normal and abnormal
families. Archives of General Psychiatry, 1963, S,
68 ~ 73.

 

112

Ferreira, A. J., & Winter, W. D. Family interaction and
decision making. Archives of General Psychiatry,
1965, 1;, 21A ~ 223.

 

Ferreira, A. J., Winter, W. D., & Poindexter, E. J. Some
interactional variables in normal and abnormal
families. Family Process, 1966, S, 60 - 75.

 

Fisher, 8., Boyd, 1., Walker, D., & Sheer, D. Parents of
schizophrenics, neurotics and normals. Archives of
General ngchiatry, 1959, l, 1A9 ~ 166.~

 

 

Fisher, S. F., & Fisher, R. L. The complexity of spouse
similarity and differences. In G. H. Zuk & I.
Boszormanye-Nagy (Eds.), Family therapy and disturbed
families. Palo Alto: Science & Behavior Books, 1967.

Fitts, W. H. Tennessee self—concgpt scale: A manual.
Nashville, Tenn.: Counselor Recordings and Tests,

1965.

 

Fontana, A. Familial etiology of schizophrenia: Is
scientific methodology possible. Psychological
Bulletin, 1966, SS, 21A - 227.

Framo, J. L. Rationale and techniques of intensive family
therapy. In I. Boszormanye-Nagy, & J. L. Framo
(Eds.), Intensive family therapy. New York: Harper
& Row, 1965.

 

Fromm-Reichman, F. Notes on the development of treatment
of schizophrenics by psychoanalytic psychotherapy.
Psychiatry, 19A8, 11, 263 ~ 273.

 

Goodrich, D. W., & Boomer, D. S. Experimental assessment
of modes of conflict resolution. Family_Process,
1953, a. 15 - 2A.

Haley, J. Whither family therapy? Family Process, 1962,
g, 69 - 100.

 

Haley, J. Family exoeriments: A new type of experimenta-
tion. Family process, 1962, l, 265 ~ 293.

 

Haley, J. Research on family patterns: An instrument
measurement. Family Process, 196A, 3, Al ~ 63.

 

Haley, J. Speech sequences of normal and abnormal families
with two children present. Family Process, 1967, S,
81 ~ 97.

113

Hawkins, J. L. The Locke marital adjustment test and social
desirability. Journal of Marriage and the Family.
1966. 2.61. 193 - 195.

Heilbrun, A. B.. Jr. The psychological significance of the
MMPII<scale in a normal population. Journal of Con-
sulting Psygholggy, 1961, SS, A86 - A91.

 

Hoeg, B. L. Marital adjustment and the comparison of
personality descriptions of mates. Unpublished mas-
ter's thesis, Michigan State University, 1966.

Hofman, K. C. An investigation of the construct validity
of marital adjustment and the similarity between
marital adjustment of spouses. Unpublished master's
thesis, Michigan State University, 1966.

Hollingshead, A. B. Two factor index of social position.
1965 Yale Station, New Haven, Conn.: August B.
Hollingshead, 1959.

Jackson, D. D. Aspects of conjoint family therapy. In
G. H. Zuk and I. Boszormanye—Nagy (Eds.), Family
therapy and disturbed families. Palo Alto:
Science & Behavior Books, 1967.

 

Katz, M. Agreement on connotative meaning in marriage.
Familnyrocess, 1965, 5, 6A ~ 7A.

 

Kenkel, W. F., & Hoffman, D. K. Real and conceived roles
in family decision making. Marriage and Family

Living, 1956, SS, 311 - 316.

Kirkpatrick, C. Community of interest and the measurement
of marriage adjustment. The Family, 1937, SS,

La Forge, R., & Suczek, R. F. The interpersonal dimension
of personality, Ill. An interpersonal check list.
Journal of Personality, 1955, SN, 9A.~ 112.

 

Lennard, H. L., Beaulieu, M. R., & Embrey, N. G. Inter-
action in families with a schizophrenic child.
Archives of General Psychiatry, 1965, SS, 166 ~
183.

Levinger, G. Supplementary methods in family research.
Family Process, 1963, S, 357 ~ 366.

 

11A

Lidz, T., Cornelison, A. R., Fleck, S., & Terry, D. The
intrafamilial environment of schizophrenic patients:
II Marital schism and marital skew. American Journal
of Psychiatry, 1957, 11A, 2A1 ~ 2A8.

 

Lively, E. L. Toward concept clarification: The case of
marital interaction. Journal of Marriage and the
Family, 1969, SS, 108 ~ llA.

Locke, H. J., & Wallace, K. M. Short marital-adjustment
and prediction tests: Their reliability and validity.
Journal of Marriage and the Family, 1959, SS, 251 ~
255.

MacKenzie, M. H. The interpersonal behavior of normal and
clinic family members. (Doctoral dissertation,
Michigan State University) Ann Arbor, Mich.: Univer-
sity Microfilms, 1968. No. 68 ~ 11072.

 

March, J. G. Husband-wife interaction over political
issues. Public Opinion Quarterly, 195A, S1, A61 ~
70.

Moore, M. R. Consistency of interaction in normal and
clinic families. (Doctoral dissertation, Michigan
State University) Ann Arbor, Mich.: University
Micrefilms, 1967. No. 67 ~ 758A.

 

Murstein, B. I. The relation of mental health to marital
choice and courtship progress. Journal of Marriage
and the Family, 1967, 29, AA7 ~ A51.

Murstein, B. I. The complementary need hypothesis in newly
weds and middle-aged married couples. Journal of
Abnormal and Social Psychology, 1961, SS, 19A ~ 197.

Navran, L. Communication and adjustment in marriage.
Family Process, 1967, S, 173 ~ 18A.

 

Olson, D. H. Decision makipg in couples: A comparison
of self-report and behavioral measures ofppower.
(Doctoral dissertation, Pennsylvania State Univer—
sity) Ann Arbor, Mich.: University Microfilms, 1968.
No. 68 ~ 8729.

Palonen, D. Interpersonal perceptions and marital adjust-
ments. Unpublished master's thesis, Michigan State
University, 1966.

‘115

Parsons, T. Family structure and the socialization of the
child. In T. Parsons & R. F. Bales (Eds.), Family,
socialization and interaction process. Glencoe, Ill.:
Free Press, 1955.

Powell, D. M. A personality inventory approach to the
study of marital adjustment. Unpublished master's
thesis, Michigan State University, 1965.

Rabkin, L. Y. The patients' family: Research methods.
Family Process, 1965, S, 105 ~ 132.

 

Riskin, J. Methodology for studying family interaction.
Archives of General Psychiatry, 1963, S, 3A3 ~ 3A8.

Roman, M. & Bauman, G. Interaction testing: A technique
for the psychological evaluation of small groups.
In M. Harrower, P. Vorhaus, M. Roman, & G. Bauman
(Eds.), Creative variations in the projectiyg
teghniques. Springfield, Ill.: Charles C. Thomas,
19 0.

 

Ryder, R. G. The factualizing game: A sickness of psycho-
logical research. Psychological Raports, 1966, lg,
563 - 570. '

Ryder, R. G. & Goodrich, D. W. Married couples' responses
to disagreement. Family Process, 1966, S, 30 ~ A2.

Ryder, R. G. Husband-wife dyads verses married strangers.
Family Process, 1968, 1, 233 ~ 238.

 

Satir, V. M. Conjoint family therapy. Palo Alto: Science
& Behavior Books, 196A.

Stabeneau, J. R., Tupin, J., Werner, M., & Pollin, W. A
comparative study of families of schizophrenics,
delinquents and normals. Paychiatpy, 1965, SS,

A8 ~ 59.

Strodtbeck, F. L. Husband-wife interaction over revealed
differences. American Sociological Review, 1951,
lS, A68 ~ A73.

Terman, L. M. Paychological factors in marital happiness.
New York: McGraw-Hill, 1938.

Tharp, R. G. Psychological patterning in marriage.
Psychological Bulletin, 1963, fig, 97 - 117.

 

116

Updyke, P. R. Family and role satisfaction among young
married women. Unpublished master's thesis, Michigan
State University, 1968.

Vacchiano, R. B., & Strauss, P. S. The construct validity
of the Tennessee Self Concept Scale. Journal of
Clinical Paychology, 1968, SS, 323 ~ 326.

 

van der Veen, F., Huebner, B., Jorgens, B., & Neja, P., Jr.
Relationships between the parents' concept of the
family and family adjustment. American Journal of
Orthopsychiatry, 196A, SN, A5 ~ 55.

 

van der Veen, M., & Ostrander, K. Development and initial
use of a family concept Q sort with clinicians and
clients of the Dane County Guidance Center. AUnpub~
lished master's thesis, University of Wisconsin,
1961. Cited in F. van der Veen, The parents' con~
cept of the family unit and child adjustment.
Journal of Counseling Psychology, 1965, lS, 196 ~ 200.

Vidich, A. J. Methodological problems in the observation
of husband-wife interaction. Marriage and Family
Living, 1956, 18, 23A ~ 239. '

Weller, L., & Luchterhand, E. Comparing interviews and
observations on family functioning. Journal of
Marriage and the Family, 1969, Sl, 115 ~ 122.

 

 

Winer, B. J. Statistical principles in experimental design.
New York: Harper, 1958.

Winch, R. F. Mate ~ selection: A study of complementapy
needs. New York: Harper, 1958.

 

“‘th

“. ‘u '..‘.

APPENDICES

117

i.e.-A

w," ,4

APPENDIX A

INSTRUCTIONS AND TASKS ADMINISTERED TO SUBJECTS

118

DIRECTIONS

Psychologists have studied and tried to understand
the individual person for a long time. Some important
discoveries and gains in helping people who somehow find
life difficult have been made. It is now realized that a
person's marriage or the marriage of his parents may have
a great deal to do with how he feels. This prOcedure in
which you have been asked to participate is designed to help
us understand what happens between two married people- The
fact that you were asked to participate in this procedure
does not necessarily mean that we feel that there is some~
thing wrong with your marriage. At this point we are only
concerned with ySaS happens, rather than attempting to
judge it as good or bad.

There are several tasks which you will be asked to
complete in this procedure. Most are contained in this
booklet. The blue booklet, also on your desk, you will
not need till later. Read the directions for each task as
you are ready to complete that task., Do not read ahead in
the directions for the other tasks; instead, do them one
at a time in the order presented in this booklet.

It is important that you do not help each other
except in case one of you does not understand the direc-
tions of a task. It is also important that you follow

the directions carefully and answer all of the questions

119

 

120

without leaving apy blank spaces. It is not necessary to
work as fast as possible, so take your time and consider
each question carefully.

All information will be held strictly condifential.

Revealed Differences Tasks

This task consists of three different situations.
They most likely will never occur in your life, but they
do happen to some people. Please imagine that the situa—
tions described actually SS happen to you. Your task is
to try to imagine how you would feel in each situation and
how you would most likely react or try to solve the pro~
blem. To make it somewhat easier, each situation includes
a list of possible solutions. Your task is to rank these
solutions; i.e. put a number 1 by that solution which you
feel you most likely would do, a number 2 by the next best
solution, etc. until all possible solutions have a number
indicating how likely you would be to use it.

Do these one at a time, do not read ahead in the
directions before you have finished each task.

Also, please remember that the more seriously you

do each of these tasks, the more value it will have.

ue—HJ

121

Imagine that one day you come home and find a letter
in your mailbox which informs you that you have won ten
thousand dollars cash in a sweepstakes. You are of course
elated and very excited, especially since you had already
forgotten that you had entered the contest. Below you
will find a list of what people might do with ten thousand
dollars. Your task is to put a number 1 by the item most
attractive to you, a number 2 by the next most attractive,
etc., until all items have a number indicating your order
of preference. Assume that you received the money today;
i.e. with your present living quarters, present bank
account and general financial situation. Be sure that
each item has a number from 1 to 8, where l is the most

and 8 the least attractive item.

Take a vacation

Invest the money

Pay overdue bills

Buy sporting equipment

Buy a boat

Redecorate the house

Build a den in your basement

Use it for a downpayment on a new house

YE..-

0‘
._A'.. .

122

Imagine that you have been married for 15 years and
have 2 children, a boy and a girl aged 12 and 10 respec-
tively. Both of them have been receiving some informa-
tion about human reproduction in the classroom as well as
from other children. You accidentally overhear them dis~

cussing it and you realize that their information is quite

incorrect and misleading. They are not aware that you over—7

heard them. How would you handle this situation? Please
answer this question by ranking the solutions listed
below. Place a number 1 by the solution you feel would be
the best way to handle it, a number 2 by the next best way,
etc., until all 8 items have a number from one to eight.

Do nothing, ignore it.

Reprimand them, and forbid them to talk about such
subjects.

Walk away, but tell you spouse and ask him/her to
talk to them later.

Walk in and tell them how they are incorrect and
explain it to them.

Attempt to find out which teacher gave them the
incorrect information and report it to the principal.

Walk away but later talk to your son/daughter and
ask your husband/wife to talk to the other child.

Ignore it but tell your spouse what you heard.
Buy some books on the subject and leave them where

the children could easily find them, so that they
could get better information on their own.

123

You have been married for 10 years and have 3 child-
ren. There were some complications with the last birth
and your doctor, after having taken a number of tests,
advises you that if you have another child the chance is
very high that it will be physically deformed or mentally
retarded. He advises strongly that you not have anymore

children. A list of methods of birth control methods and

7‘.
._ _ i. ,1
i

other possibilities is below. Please place a number 1 by
the method you would most likely pick, a 2 by the next most
acceptable method, etc., until you have rated all eight 7 3'1
choices in terms of how likely you would be to use it.

An intra-uterine device, or "loop", or "coil" (an

artificial device installed by your doctor in the

female, and must be removed by him.

Birth control pills (to be taken almost every day

by the female for the rest of her years—or until

past menopause).

Relatively minor surgery performed on the male
(sterilization).

Surgery on the female (sterilization).

Refuse his advise and continue in a normal sexual
relationship and take the chance of another
pregnancy.

Go to another doctor.

Use contraceptive jelly or foam.

Use prophylactics (also known as rubbers or condoms).

12A

Locke-Wallace Scale

Encircle the dot on the scale below which best describes
the degree of happiness, everything considered, of your
present marriage. The middle point, "Happy," represents
the degree of happiness which most people get from marri—
age, and the scale gradually ranges on one side to those
few who are very unhappy in marriage, and on the other,
to those few who experience extreme joy or felicity in
marriage.

Very Happy Perfectly
happy Happy

State the approximate extent of agreement between you and
your mate on the following items. Please encircle the
appropriate dots.

Almost
Almost Occa- Fre- Always Always
Always Always sionally quently Dis- Dis-
Agree Agree Disagree Disagree agree agree

Handling family
finances:

Matters of
recreation:

Demonstrations
of affection:

Friends:
Sex Relations: - - - . - .

Conventionality
(right, good
or proper
conduct)

Philosophy of
life

Ways of dealing
with in-laws ' ' ' ' '

When disagreements arise, they usually result in:
husband's giving in , wife giving in , agreement by
mutual give and take .

(Continued)

 

 

125

Do you and your mate engage in outside interests
together? All of them , some of them , very few of
them , none of them ?

In leisure time do you generally prefer to be on

the go" , to stay at home ?

Does your mate generally prefer: to be "on the go"
, to stay at home ?

Do you ever wish you had not married? Frequently
, occasionally , rarely , never .

If you had your life to live over, do you think you
would: marry the same person , marry a different per—
son , not marry at all

Do you confide in your mate: almost never ,
rarely , in most things , in everything ?

3“ — i I

avew.m_g

 

126

FAMILY CONCEPT INVENTORY

Instructions: Indicate the degree of
your agreement or disagreement with
each of the following items as it
applies to your immediate family
(husband or wife and children) and
encircle the 1etter(s) representing

 

the appropriate response. First

impressions are satisfactory, and

mmost people are able to complete

this inventory in ten minutes. It is

quite important that you give a

response to each item, even though it

may sometimes be difficult to make a

decision.

1. We usually can depend on each other.

2. We have a number of close friends.

3. We feel secure when we are with
each other.

A. We do many things together.

5. Each of us wants to tell the
others what to do.

6. There are serious differences in
our standards and values.

7. We feel free to express any thoughts
or feelings to each other.

8. Our home is the center of our
activities.

9. We are an affectionate familv.

10. It is not our fault that we are

7 having difficulties.
11. Little problems often become big

ones for us.

12. We do not understand each other.

13. We get along very well in the
community. _

1A. We often praise or compliment
each other.

15. We do not talk about sex.

16. We get along much better with
persons outside the family than
with each other.

17. We are proud of our family.

18. We do not like each other's friends.

19. There are many conflicts in our
family.

20. We are usually calm and relaxed

» when we are together.

21. We respect each other's privacy.

(Continued)

all

303?; [Strongly Agree

SA

SA
SA

SA
SA
SA

SA
SA
SA

SA

SA
SA

I09

”‘3 [Tend to Agree

mm

Neither Agree
Nor Disagree

ZZ Z ZZ Z ZZ Z Z Z ZZ ZZ l

ZZ Z ZZZ

‘0.-

QaQa Q. QC.- Q-a CLO-a Q» Q.- Q- 910-: QaCL 'Tend to Disagree

CLO- C1- CLO-Q-

Cé’g IStrongly Disagree

SD

SD
SD

SD

SD
SD

SD
SD
SD

SD
SD
SD

SD

SD
SD

 

22.

23.
2A.

25.
26.
27.
28.
29.
30.

31.
32.

33.
3A.

35.
36.
37.

38.
39.

A0.
A1.

A2.
A3.

AA.

A5.
A6.

A7.
A8.

127

Accomplishing what we want to do
seems to be difficult for us.

We tend to worry about many things.

We are continually getting to
know each other better.

We encourage each other to develop
in his or her own individual way.

We have warm, close relationships
with each other.

Together we can overcome almost
any difficulty.

We really do trust and confide
in each other.

The family has always been very
important to us.

We get more than our share of
illness.

We are considerate of each other.

We can stand up for our rights
if necessary.

We have very good times together.

We live largely by other people's
standards and values.

Usually each of us goes his own
separate way.

We resent each other's outside
activities.

We have respect for each other's
feelings and opinions even when
we differ strongly.

We sometimes wish we could be an
entirely different family.

We are sociable and really enjoy
being with people.

We are a disorganized family.

We are not really fond of one
another.

We are a strong, competent family.

We just cannot tell each other
our real feelings.

We are not satisfied with anything
short of perfection.

We forgive each other easily.

We usually reach decisions by
discussion and compromise.

We can adjust well to new
situations.

Our decisions are not our own, but

are forced on us by circumstances.

SA
SA

SA
SA
SA
SA
SA
SA

SA
SA

SA
SA

SA
SA

SA

SA
SA

SA
SA

SA
SA

SA

SA
SA

SA
SA

SA

Z Z Z ZZ ZZ Z Z Z Z Z Z ZZ

Z Z Z ZZ Z ZZ ZZ Z Z

O.- Q- Q: C1-C3. QC)» 0.- Q- Q Q.- Q- Q.- CLO;

Q.- Q- Q- 90: Q- Q-Q {10.- Q 0»

SD
SD

SD
SD
SD
SD
SD
SD

SD
SD

SD
SD

SD
SD

SD

SD
SD

SD
SD

SD
SD

SD

SD
SD

SD
SD

SD

128

This task is somewhat different. This is the blue
booklet with Tennessee Self-Concept Scale on the front
page. Please do not write anything in this booklet.
Other people will have to use it also. Inside the book-
let you will find a form which is to be used for your
answers. Along the right hand side of the form you will
find a space for your name, age and education. Please
fill in these spaces. There is also a space for timing
wbut you need SSS time yourself on this task. The direc-
tions are the inside of the front cover of the test
booklet, please read these carefully.

When you have completed this task please wait until

your husband (wife) is finished and signal the assistant.

APPENDIX B

SCORES OF ALL SUBJECTS

129

1Q30

HHm .mmHmE mam m

r n .5- "iI.H.O.J...-’dn.}

 

mHmopsoz

HmoLONHo mpHchomLcd

. mHmocosnm
ucmspmonomHmz Hmtmzoo
m>HpHmom i m>Hmcoomo
mocmHHm> Hmocs

poHHocoo Hmocs
:oHumcwoocH NNHHmcowLm;
mcme pcmH>mo so Hoots:

wH
sH
wH
NH
:H
mH
NH
HH
CH

onN HNoHeoN-Hmeoz -

coHooNomHoeN-oHam -

omLmoE::iooo HH< ”NHH: mH: so NHNHNHCNEEH soroHHos

onN HaHoom -
oHem sHHeem -
sHmm Hmc0mpmm i

oHoN HNOHNHNN -
LoH>mcom i

(\JMJLDKONCDO

soHocooH -
meHonoN Hoods - H oHoNHcm>

.memEmu mam m omLoQE::ico>o
.oneHo ooonHH NH cameos: - .

 

 

NN HN NN so NN NN NN NH N oN ss NN Ns sN sHH soH NNH ssN om
NN ss NN HN Hs sN Hm N sm ea NN NN ea sN NHH oo NoH som oN
Hm as Ns HoH NN NN sm NH NH Ns Ns Ns Ns Ns HNH sHH eNH sNN NN
om HN Ns so sN so NN HH. H st NN Hs on NN NHH NHH NNH sNN sN
Ns ss Ns so NN sm NN o s oN os os ms sN oHH soH NNH omm NN
ms NN Ns so NN oN as N sN oo es NN Hs eN NoH NoH NNH NNN NN
Ns ss Ns No NN HN oN sH N Ns Hs sN es sN sHH NoH sNH ssm sN
sN. NN ss oo sN Nm NN NH m HH es sN Ns ss oHH NoH sNH st NN
Ns ss oN oo NN om NN NH s or sN NN Ns ss sHH soH oHH osN NN
ss oN Ns Ho HN HN NN HH s so NN NN NN as NHH so sHH sNN HN
Hs NN sm Ho sm ss sm oH N No eN aN sN sN oHH No sHH NHN oN
NN NN NN Ns .sm HN NN s sN oN eN mN HN NN ooH oN so NNN oH
Hs NN Ns sN NN sN as N sm on NN NN N NN ooH sN oN mHm NH
Ns ss ms Ne NN ss NN NH s No Hs mu Ns os sHH ooH NHH Hsm sH
om sN Ns NoH NN NN sm HH N s ss Hs Ha Ns HNH NHH oNH Nsm NH
sN ss ss Ho sN sm HN NH N aN Ns sN NN Hs NHH HHH NHH msm NH
NN Ns oN NN ss NN sN o NH NN HN NN ss oN No me HNH oom sH
oN NN Ns NN Ns Hs NN NH N so No HN NN os NoH mo NHH sHN NH
NN as Ns dN mm NN oN N Nm Hs oN NN HN NN NN N oNH NsN NH
Ns NN Ns so Ns NN oN sH N os sN HN NN NN NHH NoH NHH NNN HH
NN NN NN oN Nm NN om m sm sN sN Ns NN oN NoH Ns No ssN oH
ss HN Ns ooH NN sN Ns o NH NN so NN as oN oHH HoH NNH Nsm o
oN Ns sN Ns NN NN NN N Nm HN NN sN ss mm NN NN soH oNN N
ms HN Ns No ms ss NN NH N Ns os NN sN NN NoH NoH NHH NNN s
ss Ns ss oo HN oN sN sH a sN Hs HN NN sN NoH NoH sHH NNN N
Ns HN NN Ho ms NN NN NH sH NN sN oN NN oN NoH No soH oHN N
Ns as ms so as os Hm NH m sN NN sN HN as NHH NHH sNH st s
ms Ns Hs No ss ss mm s N N. oN HN Ns Hs NHH ooH NNH omm N
NN ss ss oo Hs Nm om NH NH NN NN sN Ns NN soH NoH HHH NHN N
Hs ss NN No sN NN NN sH NN sN sN NN Hs NN NoH ooH NHH on H
NH sH NH NH sH NH NH HH oH o N s N N s N N H :
moHomHLm>

oHoEmm oHcHHo mmHmom pooocooioHom mmmwoccms .H

mmmcow H<DQH>HQZH

131

mHmoczoz

HmoLomHQ NHHchowcwm
meozona

pcprmSHUNHmz Hmtocoo
m>HuHmom i m>choomo
mocmem> Hmooe

uoHHocoo Hmuos
coHuwawmocH NNHHmccoLoo
mcme ucmH>wo so Lasso:

mH
sH
NH
mH
:H
mH
NH
HH
OH

“How HmHoom
onm sHHEwnH
sHmm HNCOmHmm

oHoN HmoHeom-Hmeoz -
onN HeeHmseN -
LoH>m£mm i
eoHooNoaHoNN-NHNN -

HHw .onmE one m ompmoESCiooo HH< mmmHz wH: so sHopmHooEEH omonHoo

seHoeooH
m>Honoo Hmoos

mmzmxosoom

H mHomHee>

.monEos mom m omtmoESCIco>m
NH oceemez - *

.omeHo ooomHH

 

 

NN Ns sm No as Nm sm NH s sN as as o: NN sHH NoH omH NNN oN
mN Ns ss mm as Nm HN .sH N so Ns NN NN NN NHH NoH NHH smm oN
Ns HN _Ns soH NN NN mm o s ss Ns NN HN so sNH oHH NNH st NN
NN Ns Ns NN sN mm NN sH N ms ss ms Ns Ns HHH NNH HNH st sN
ooH NN sN sHH oN sN oN NH N NN NN HN NN es NsH sNH HsH .oHs NN
NN . sN ss ooH HN oN .mN oH N as ss NN as os oHH NHH oNH HNN NN
so No Hs NoH HN mm NN HH N NN Na as sN as HNH NNH NsH Nos sN
No NN Nm NHH Ns Nm NN N oH Ns sN Ns ms oN oNH NNH NNH NNN NN
ms NN om oo NN NN oN HH N No do NN sN NN NoH No NNH Hmm NN
ss Ns ss no Hs Ns sN NH s NN s eN Hs NN NoH soH NHH Hmm HN
Ns Hs Nm NN ss os NN sH N eN as NN N NN NHH soH NNH ssm oN
Ns NN os so os Ns Nm oH H a; so co NN so soH No NNH st. as
so ss ss No NN NN NN NH H as ms o. os ms eHH NHH NNH Nsm Ns
No Hs Nm Na NN sm NN oH N as as Ns ms ss NNH NNH HNH Nsm ss
Na NN ss NoH NN om N sH s as NN ss NN Ns ooH sNH sNH NNN Ns
HN Ns Ns No HN os NN NH o sN ms so NN os HHH NoH NNH Nsm Ns
HN HN Ns so Ns as NN NH H No es do oN Hs NHH oHH oNH NNN ss
NN NN Ns so NN NN sH HN s Ns ss Ns ss ss oHH HNH NNH Nsm ms
HN Ns ms NN sN om sN oH N do s_ so Ns sN HHH NoH NNH Nsm Ns
No ms ms NoH sN NN sN NH H as s as ss Ns oNH NHH NNH NNN Hs
HN ss sN sN sN Hm NN sH sH NN os NN Hs NN NoH NoH sHH NNN os
NN HN oN HoH sN mm NN NH o Hs Ha NN N NN eHH sHH oNH NNN am
os Ns N No Hs NN sN s NH sN as so os NN HHH so oHH st Nm
Ns sN Hs No Ns Ns NN HH N so NN Ne HN so soH No NNH NNN sm
sN NN as so NN Ns sN NH s No ss Hs sN NN oNH NHH NNH NNm Nm
oN os Ns No Ns NN NN sH H sN N. HN ms os NHH No NNH Nsm Nm
NN NN Ns No NN om ms HH s sN ms NN NN NN HHH HoH NNH NNN sm
sN ss Ns so mN om om NH H NN Ns oN NN N HHH NHH oHH Nsm mm
ss Ns su Ho Ns NN NH NH NH NN NN NN NN NN NoH soH NHH NNN NN
sN Ns sN NoH Ns ms Nm NH N as ss aN NN ss NNH soH HNH NNN Hm
NH sH NH NH sH NH NH HH oH o N s N N s m N H z

mmH pm Hch>

oHoemm oHcHHoicoz

”onom HomecooioHom mcwmmccms .N

1132

 

 

wHQEmm OHCHHQ

N xmms pcmEHHHNHsm cmmz I NN
N xmms acmEHHHNHsm cwmz I NN
m xmmE NocmquEoo HNCONLmQmecH I H xmmE acmEHHHNHzm ummz I Hm
m xwme NocwquEoo HNCONLmamecH I pamocco NHHENN I cm
H xwme Nocmumano HmcomLmQmecH I mHmom momHHm3Imxooq I OH mHanLm>
NH I ss sN os oOH Hs NNH oNH oN HN NN Ns No ooH NH NsH sHH oN
ss NN NN ss NN NN NNH sNH NN Ns I so I No NN Ns NN NNH NNH 0N
HoH HoI os ooH NN Ns NsH NNH NN NN NN NN ooH No ooH ssH sHH NN
NN sH NH NN ooH Ns ssH No sN NoH NN sN No ooH NN HNH sNH sN
HN NN NN ms NN NN ssH NsH NN NN I oH NN I sH I HN NN NNH NHH NN
sN I NNI NNI HN NN NN NsH NsH NN sN NN I NN ooH NN No NNH oNH NN
so NoI sN HN NN NN NNH NNH sN NH I so oNH NN Ho No NHH sNH sN
NN NN HN sN NN NN osH oHH NN NH os I HN NN NN ms soH NHH NN
NN os oN NN ooH Ns NNH NNH NN NN sN sN Hs Ho Hp HNH NN NN
NN NN NN .ss sN NN NNH Ha HN NN so NN oN No NN NN as HN
NN NN sHI NN NN NN NNH HNH CN oN NN so NH No on NNH NoH oN
Ns No NN HN HN ooH NNH sNH cs N I NN I NN Na No No sNH NN NH
so NN sN HN NN ooH NsH NN as NH Ho ss NI NN NN sHH sN NH
NN Ns oN oN HN HN NNH NHH ss NN sN I NN No HN ooH sHH NHH sH
so Ns Ns HN I NN HN NNH HsH Ns HN NoH sH NN Hg No oNH NoH NH
NN ssI NNI NN I ooH NN NsH sNH Ns Hs NH :c I ss Ns No NsH NN NH
NN NN oN NN ss Ns NsH NNH ss HN Ns om NN HN No sHH so sH
NN ss sN NN IN NN sNH sNH ms 0N MN NN Nd Nu Hm CNH OCH NH
soHI NH NH HN NN I H. NNH 0N as N ss NN N NN NN NoH N NH
NN Ns sN NN Ns I N ssH NNH Hs No HN NN NN ON No CNH NHH HH
Hs oN Hs NN NN N NsH sNH as NH I sN I No NN Ho HN NHH HNH oH
ss NN NH NN mm mm NNH sNH cm NH I No Hs NN Hs NN oHH sHH c
oN NN Ns ooH ooH NN NsH oHH 0N NN sN I ssH NN NN Na sHH NHH N
No I NH NN NN No Hs NHH NHH sN OH I so I No I sH ocH No NNH sHH s
NN NN NN NN NN oN HNH NNH NN Ns sN sN NN No sq HNH HNH N
NN NNI NHI Ns NN NN NsH sNH NN NN NN HN I No ooH No NsH scH N
sH NHI sN ooH NN Ns NsH Nu sN ss No N NN NN Ho soH 0N s
HN sH sN NN NN Ho NHH ooH NN oN 0H NN Ns ooH No NN scH N
_HN oo oN oOH NN HN ssH NoH NN NN NN os sN NN sN Ho NN N
sN so HN Ns NN NN NNH NHH HN HN Ns NN NN NN Ho NHH No H
NN NN sN NN NN HN oN NH : NN NN sN NN NN HN oN oH z

mmHanLm>

.m

133

COUPLE SCORES

1. Clinic and Non+C1inic Sample Characteristics

 

 

 

Variables
Age Education
Number of Years Social

N H w H w Income Children Married Class
Clinic Sample

1 “3 H3 20 15 2O 2 21 1
2 3H 35 2O 16 10 3 1h 1

3 33 33 20 1h 12 u 11 1

u 51 MM 20 16 15 2 15 l
5 50 50 1h 13 9 1 25 u
6 M3 U1 16 12 15 7 21 3
7 H1 38 18 1h 16 3 16 2

8 US M9 20 16 17 3 20 1
9 U3 “8 10 12 11 3 13 U
10 “l 33 13 12 1a 5 6 3
11 31 31 13 1M 12 2 12 3
l2 “5 U8 13 13 10 u 24 3
13 M7 u2 8 12 11 7 18 5
1H U2 38 12 8 8 H 15 U
15 38 35 18 16 11 3 13 2
Non-Clinic sample

16 29 29 18 1a 10 2 8 3
17 32 28 20 1D 5 2 9 2
18 3O 3O 2O 16 16 2 9 1
19 30 30 16 18 9 2 9 3
20 36 37 2O 13 12 6 16 1
21 33 31 2O 18 6 1 6 2
22 52 H6 20 18 17 3 25 1
23 31 30 2O 16 13 3 10 1
2H 27 26 1a 12 l“ 1 5 3
25 35 36 18 18 9 3 11 2
26 3O 28 2O 16 3 2 8 2
27 32 30 12 13 10 u 11 N
28 “7 H6 l2 12 13 u 2“ 3
29 27 25 15 15 8 2 6 h
30 26 24 2O 1“ 5 3 5 2

13M

2. Revealed Differences Task Scores

Variables
Spontaneous
Agreement Dominance Time
N 1 ‘2 3 1 2 3 Total 1 2 3 Total

 

Clinic Sample

1 57 86 u5 + — — - 292 205 9uu_ luul
2 79 95 21 0 + 0 0 186, 285 392 863
3 81 98 33 + + + + 150 152 225 527
u 81 88 -12 + + — + 118 178 169 A65
5 52 57 9 - - 0 - 120 665 363 1118
6 62 62 17 + - + + 265 330 287 882
7 29 81 96 - 0 0 0 125 18a 157 u66
8 91 55 11 + 0 0 0 90 307 306 703
9 68 55 50 + 0 + + 118 178 169 H65
10 55 86 10 + 0 + + 100 186 282 568
11 36 81 . 29 - 0 0 0 105 221 18a 510
12 62 83 7M - 0 0 0 213 195 27h 682
13 1h 57 -1M + 0 + + 360 365 367 1092
in 62 98 98 - + — — 162 83 101 3H6
15 -31 76 60 0 — - - 199 1H6 85 u30
Non-Clinic Sample
16 7h 62 as + 0 - 0 170 230 160 560
17 ”5 95 95 + 0 - O 165 95 H2 302
18 21 95 71 + 0 - 0 u10 200 320 930
19 60 95 38 - - - - 175 137 252 564
20 71 81 91 + 0 + + 310 265 120 695
21 38 55 zu 0 0 + 0 217 250 305 772
22 67 62 76 + + - + 355 365 113 1133
23 71 95 ”8 + + 0 + 130 67 281 R78
2h 91 71 67 - - - - H7 155 216 uua
25 86 62 81 + 0 - 0 121 185 121 A27
26 62 57 52 + - 0 0 165 135 123 123
27 50 67 52 + + + + 253 131 211 628
28 86 76 88 + - + + 158 1“? 105 410
29 6H 29 69 0 + - 0 195 185 275 955
30 31 98 #3 0 - + 0 252 163 267 682