University of Southern California Dissertations and Theses

Complementarity, Homogeneity, Heterogeneity And Marital Stability

(USC Thesis Other)

Complementarity, Homogeneity, Heterogeneity And Marital Stability

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been
microfilmed exactly as received
69-625
LOSSNER, Walter Martin, 1912-
COMPLEMENTARITY, HOMOGENEITY, HETERO
GENEITY AND MARITAL STABILITY.
University of Southern California, Ph.D., 1968
Sociology, family
University Microfilms, Inc., Ann Arbor, Michigan
© Walter Martin Lossner 1969
ALL RIGHTS RESERVED
COMPLEMENTARITY, HOMOGENEITY, HETEROGENEITY
AND MARITAL STABILITY
by
Walter Martin Lossner
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Sociology)
June 1958
UNIVERSITY O F S O U T H E R N CALIFORNIA
THE GRADUATE SC H O O L
UNIVERSITY PARK
LOS ANGELES. CALIFORNIA 9 0 0 0 7
This dissertation, written by
M 3 ^ R ..M 2 l BTJLN..Jj QSSNEH..............................
under the direction of A..JLSL Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by the Graduate
School, in partial fulfillment of requirements
for the degree of
D O C T O R O F P H I L O S O P H Y
DateAme.'...J333.
DI^ERTATIO^-CQM M ITTEE
//Py 1 c n ^ a i r m a n /
U i-;~ x. J t J l
TABLE OF CONTENTS
|Chapter
I. THE PROBLEM AND DEFINITION OF TERMS USED ....
The Problem
The Hypotheses
Definitions of the Terms Used
The Conceptual Framework and the Variables
| II. A REVIEW OF THE LITERATURE ...................
Some Basic Studies on Marital Adjustment Tests
Some Studies of Role, Personality, Empathy, and
Marital Adjustment
Other Studies Related to Marital Adjustment
Some Studies on Homogamy and Heterogamy
Summary
III. THE SAMPLE AND THE METHODOLOGY USED .........
The Sample
The Methodology Used
i The Interpersonal Check List
i Diagram of ICL by Octants and Quadrants
i The Marital Stability Test
Summary
IV. THE FINDINGS OF THE STUDY.........................92
Homogajny in Both Dominance and
Affiliativeness
Homogamy in Affiliativeness and
Heterogamy in Dominance
Heterogamy in Both Affiliativeness
and Dominance
Heterogamy in Affiliativeness and
Homogamy in Dominance
Summary
V. EMERGENT FINDINGS ............................ 114
VI. DISCUSSION OF THE FINDINGS ........... 118
VII. IMPLICATIONS FOR FUTURE RESEARCH .............. 122
VIII. LIMITATIONS OF THE STUDY . . ......... . 124
IX. CONCLUSIONS................................ 126
APPENDIX......................... 129
BIBLIOGRAPHY............................................ 132
iii
LIST OF TABLES
Table Page
1. Percentage of Men and Women in Each Group
(Stable and Unstable) in Various Age
Groups.........., ......................... 67
2. Number of Children of Stable and Unstable
Marriages, by Percentages .................. 68
3. Length of Marriage of Stable and Unstable
Couples, by Percentages .................... 71
4. Reported Education of Men and Women in the
Stable and Unstable Marriage Groups, by
Percentages................................ 71
5. Comparison of Mean of Reported Years of
Schooling in Three Recent Marital Studies . . 72
6. Occupations of Men and Women, Stably Married
and Unstably Married, by Percentages „ . . . . 74
7. Annual Family Income of Stably and Unstably
Married Couples, by Percentages ....... 75
8. Distribution of Stable and Unstable Marriages. . 91
9. Distribution of Homogamous Couples by Type
and by Marital Stability,.................... 94
10. Distribution of Couples who are Homogamous
in L or H and who are Complementary D/S
or S/D, by Type and Marital Stability .... 99
iv
11. Distribution of Stable and Unstable Couples
in which one Spouse is Affiliative (Friendly)
and the other is Non-Affilia.tive (Hostile)
and where Dominance and Submissiveness are
Complementary ................................ 105
12. Distribution of Stable and Unstable Couples
in which one Spouse is Affilia,tive (Friendly)
and the other is Non~Affilia.tive (Hostile)
and who are either Both Dominant or Both are
Submissive......................... . . . . 108
13. Summary Chart of Findings .......... 115
1H. Distribution of Couples, Stable and Unstable
with only LOV as a Factor, by Frequency . 116
15. Distribution of Couples, Stable and Unstable
with only DOM as a . Factor, by Frequency .... 117
v
Figure
1.
LIST OF ILLUSTRATIONS
Diagram of ICL by Octants and by Quadrants „ .
Page
. . 85
vi
1
; CHAPTER I
THE PROBLEM AND DEFINITIONS
! OF TERMS USED
This is a study of the relationship between global
personality factors of marital partners and their marital
| stability. Winch^ asserted that spouses select each other
i on the basis of complementary needs. Many and varied studies
I
| of marital couples have shown that some partners indicate on
i
i
i tests and through interviews that they have complementary needs;
I that some have similar or same needs; and that some have
| opposite or conflicting needs. These studies were interested
in the question whether marriage partners choose one another
; in order to fill particular personality needs. Since almost
i every combination has been found among marriage partners, the
1
; question arises whether the need patterns of couples might be
i
; an important factor in the stability of the marriage. It is
| the purpose of this study to determine whether there is
i
| statistical evidence for the assertion that marital stability
I
| ________________________
lR. F. Winch, Mate Selection (New York: Harper and
Row, Inc., 1958)
I 2
j
: depends to a significant degree on whether the spouses have
; complementary needs, that is, on whether each fulfills the
i
I otherTs needs.
i
With the growing concern for marital harmony, counselors
t
j are becoming more aware that there are few validated instruments
which can be used to diagnose the difficulties in a marriage,
I to help in planning valid courses that will further a recipro-
| cally beneficial relationship between spouses, to plan
counseling programs for the faltering marriage, and to assess
therapeutic progress. The Locke Marital Adjustment Test is
; helpful In placing couples somewhere' on the continuum of
j
| successful marriage, but it Is not very helpful in pinpointing
the difficulty and Indicating to the counselor what areas of
; living need attention. Couples in distress are not very
: helpful to the counselor In finding the basic difficulties
| quickly enough to warrant short term marriage counseling.
They present too many rationalizations, excuses, and counter-
; attacks. Some couples will frequently assert that their sex
| life is perfect, when there is both frigidity and Impotence.
I
i
j The couple’s personalities may be completely devastating to
I
i
| one another, with the cause so skillfully guarded that neither
! spouse considers it part of the marital difficulties. Thus
i
I there is need for tests to pinpoint the area or areas where
i
the marital relationship is suffering.
3
The Problem
Statement of the problem
It was the purpose of this study:
1. To ascertain whether complementarity, as stated by
Robert Winch, is a factor in marital stability.
Winch maintained that complementarity is character
istic of mate-selection. However, numerous studies,
including his own, have shown that it is found in j
some marriages hut not in others. This may
indicate that complementarity may be related to
some other factor in marriage. It was the
hypothesis of this study that complementarity is
a factor in the stability of a marriage, as Robert
Winch suggested when he said that a marriage lack
ing complementarity could not survive ten years.
I
2. It is impossible at this stage in. the development
of tests and inventories to explore the total range
of complementarity. Basically, this study was
limited to the collation of personality traits
explored in LearyTs ICL. With the formula developed
by Leary, each person can be placed in a global
personality category of dominance-submission and
of love-hate. The latter is frequently called
affiliativeness-non-affiliativeness. Many of the
studies reported in the following review of
4
literature have confused complementarity with
oppositeness. An opposite personality trait such
as friendly to unfriendly cannot be called
complementary. With the ICL we can assess both
complementarity and oppositeness. Although there
can be no doubt that dominance and submissiveness
are complementary, it is equally obvious that
affiliative and non-affiliative are not. A person
who is non-affiliative is not likely to fulfill
the affiliative needs of his spouse. It is the
purpose of this study to assess the relationship of
global personality characteristics, rather than
single personality traits, to marital stability.
In addition to the aim of ascertaining whether the
lack of complementarity is indicative of lack of
stability in a marriage, this study tried to
ascertain whether the opposing needs of affiliative
versus non-affiliative indicate lack of stability.
This research explored the relationship of three
personality interactional patterns to marital
stability. These three are: homogeneity of person
ality needs, complementarity of personality needs,
and oppositeness of personality needs. It is
assumed at the outset that in complementary
relationships, either affiliativeness or non-
affiliativeness will be a homogeneous factor,
h. Another purpose of this study was to give the
marriage counselor additional tools for the
diagnosis of the marriage relationship, to supply
an indicator of areas of therapy, and to aid both
in planning and assessing the marriage counseling
program. If complementarity is found to be sig
nificantly related to marital stability, progress
toward complementarity can be used as a measure
of counseling progress.
Major hypotheses
Hypothesis 1: When husband and wife both have the
same global personality character
istics and, therefore, the same
personality needs, the marriage will
tend to be characterized as unstable.
Hypothesis 2: When the personality needs of a
husband and wife are complementary,
the marriage will tend to be stable..
Hypothesis 3: Even though the couple’s Dominance-
Submission needs are complementary,
a marriage will tend to be unstable
when one spouse is affiliative and
the other is non-affiliative.
Hypothesis 4 - r When the Dominance-Submission per-
6
sonality needs of a couple are the
same and their affiliative needs are
opposite, the marriage will tend to
be unstable. This is a marriage in
which both the necessary homogeneity
and need complementarity are lacking.
The Hypotheses— Operationally
Hypothesis 1: When the LOV score and the DOM score
of both spouses border on the same
quadrant of the ICL indicating like
global personality needs by inter
secting in the same quadrant, the
couple will tend to fall In the group
of spouses who are either divorced
or whose Marital Stability Score
exceeds thirty-eight. This is called
an homogamous relationship.
Hypothesis 2: (A) When the intersection of the DOM
and LOV scores of one spouse falls in
the quadrant called LD (affectionate-
dominant) and the other spoused
DOM-LOV Intersection falls in the
quadrant LS (affectionate
submissiveness) , the couple will tend
Hypothesis 2
Hypothesis 3
7
to be in the group expressing
satisfaction in being together with
a score of less than thirty-two on
the Marital Stability Test- This is
called a . complementary relationship-
in Dominance and a heterogeneous
relationship in Affiliativeness.
(B) When the intersection of the
DOM and LOV scores of one spouse
falls in the quadrant HD (hostile-
dominance) and the other spouseTs
DOM-LOV intersection falls in the
quadrant HS (hostile-submissiveness),
the couple will tend to be in the
group scoring less than thirty-two
on the Marital Stability Test. This
is also a complementary relationship
in Dominance and an homogamous
relationship in. Affiliativeness.
(A) When the intersection of the DOM
and LOV scores of one spouse falls
in the quadrant LD (affectionate-
dominance) and the other spouse's
falls in the quadrant HS (hostile-
submissiveness) , the couple will tend
I
Hypothesis 3
Hypothesis H
8
to be in the group of spouses who
are either divorced or whose Marital
Stability Score exceeds thirty-eight.
This is a relationship between
people who are opposites in
Affiliativeness and complementary
in Dominance.
: (B) When the intersection of the
DOM arid LOV scores of one spouse falls
in the quadrant HD (hostile-dominance)
and the scores of the other spouse
intersect in the quadrant LS
(affectionate-submissiveness), the
couple will tend to be in the group
of spouses who are either divorced
or whose Marital Stability Score is
higher than thirty-eight. This is a
relationship between people who are
opposites in Affiliativeness and
complementary in Dominance.
: (A) When the intersection of the DOM
and LOV scores of one spouse falls in
the quadrant HD (hostile-dominance)
and the scores of the other spouse
intersect in the quadrant LD
9
(affectionate-dominance), the couple
will tend to be in the group of
spouses who are either divorced or
whose Marital Stability Score is
higher than thirty-eight. This is a
marriage in which both homogeneity
in Affiliativeness need and the
complementarity between Dominance and
Submissiveness are lacking.
Hypothesis M - : (B) When the intersection of the
DOM and LOV scores of one spouse
falls in the quadrant HS (hostile-
submissiveness) and the scores of the
other spouse intersect in the
quadrant LS (affectionate
submissiveness) , the couple will tend
to be in the group of spouses who are
either divorced or whose Marital
Stability Score is higher than
thirty-eight. This, too, is a
marriage in which both homogeneity
in Affiliativeness and complemen
tarity between Dominance and
Submissiveness are lacking.
j 10
!
I Definitions Of The Terms Used
i ----------------------------
i
j & need is a goal-oriented drive and in this study refers
i
I specifically to responses which are sought in interpersonal
j
irelationships.
Interpersonal needs in this study refer to scores that
:a person receives on a need instrument, the Interpersonal Check
|
|List.^ This instrument, through a formula developed by Leary
and others, places interpersonal needs on the lines of two axes.
One axis is a continuum of ascendency ranging from dominance to
i submission (called the DOM scale). The other is a continuum of
i
iaffiliation, ranging from love to hate (called the LOV scale).
The intersection of the verticals from each score determines the
score for the overall personality trait. This score is located
in one of four possible quadrants and characterizes the subject
j
as Dominant-Affiliative (LD), Submissive-Affiliative (LS),
|Non-Affiliative-Dominant (HD), or Non-Affiliative-Submissive (HS).
\ Homogamous needs are those interpersonal needs that are
|found between partners who have similar characteristics. When
both spouses score above or below 50 on a need axis of the ICL,
I they are said to have the same or homogamous needs.
i
i
i
f
I
^Leary, Timothy, The Interpersonal Diagnosis of
Personality (New York: The Ronald Press Company, 1957).
11
Complementary needs3 are those interpersonal needs
that spouses fulfill for each other. In this study it is con
fined to the global personality characteristics of Dominance and
Submissiveness. When a person's DOM score is below 50, he is
called Dominant. If it is above 50, he is called Submissive.
When one spouse is classified as dominant and the other as sub
missive, the relationship is complementary.
Opposite needs are those needs which interfere with the
need satisfaction of the spouse. When one is affiliative and
the other is non-affiliative their needs oppose one another,
rather than complement one another. Operationally they are
identified as affiliative when the LOV score is higher than 50
and non-affiliative when the LOV score is below 50.
Marital instability for this study is the expressed
intention of separating or divorcing, or the expressed desire
to be separated or divorced. This is indicated by applying
for divorce or separation and by high scores (over thirty-eight)
on the Conflict Questionnaire, which includes the wish to be
separated or divorced. Divorced people are included in the
unstable sample without being tested with the Marital Stability
Test.
Marital stability is the probability of a lasting, non
crippling relationship. Operationally this is determined by a
3r. F. Winch, Mate Selection (New York: Harper and Row,
Inc., 1958)
12
I
score of less than thirty-two on "the Marital Conflict Question
naire and the expressed intention of the spouses to stay
together in what each considers to be a good relationship.
Personality in this study is the pattern of interpersonal
! responses expressed by the individual. When scored, these
; responses place the persons personality score in one of four
| global personality categories. These categories are indicated
! by the quadrant of the ICL in which the DOM and LOV scores
intersect. The person is then said to have a global personality
which is either LOVE-DOMINANT, LOVE-SUBMISSIVE, HATE-DOMINANT,
i or HATE-SUBMISSIVE.
!
I
The Conceptual Framework And The Variables
There are many lists of factors that are presumed to be
I
important for marital success. Some of the factors have been
i
more or less validated, some have been taken for granted. The
j present study was not interested in predictors, such as harmony
' of the parental marriage, which may have nothing to do with the
immediate relationship. The framework for this study is in
personality theory.
According to Winch^ and other sociologists, homogamy
operates at the level of social characteristics in mate selec
tion. Winch claimed that social characteristics operate T T to
limit the field of eligibles."
^Winch, op. cit., pp. 88 & 89.
13
I ~ At the psychological level, Winch emphasized the
importance that personality needs play in mate selection and
in the continuance of a stable marital relationship. Winch
stated that maximal need gratification is based upon complemen-
j
j tarity (heterogamy) , as contrasted to similarity (homogamy) of
I the personality needs of the couple. Winch posits complemen-
| tarity when one partner is high on a personality need and his
| mate is low on this same need, or when there is a relationship
between individuals with different needs.
This study begins with the assumption that some per-
| sonality needs are best met by complementary needs in the
partner and some by homogeneous needs in the partner. The
independent variable in this study is personality. The depen
dent variable is marital stability. Leary’s Interpersonal
1
I
| Check List was used to take inventory of the personality traits
and needs of the spouses in the study. The study was designed
to ascertain whether there is a significant relationship between
! marital stability and personality needs of the spouses. This
study hypothesized that if the spouses are alike in both
affiliativeness and dominance, the couple will more likely be
found in the unstable group. If they are unlike (or opposites)
in affiliativeness, the marriage will tend to be unstable, even
though there Is complementarity in that one spouse is dominant
and the other Is submissive. This study, then, hypothesized
that the marital relationship will have the highest probability
14
of being stable when the spouses are alike in affiliativeness
(LOV) and unlike in dominance (DOM).
CHAPTER II
A REVIEW OF THE LITERATURE
i
I
j
I Some Basic Studies On Marital Adjustment Tests
Thus far very few tests are available that specifically
aid the marriage counselor. The tests that are available seem
i
jto be for prediction of marital success, happiness, or stability.
The test most frequently used is Leary's Interpersonal Check
List, referred to in this study as the ICL. Discussions with
counselors who use this test in marriage counseling indicate
that it is used as an attention getter, as a diagnostic tool,
and as an instrument for testing the progress and efficiency in
I
icounseling. It is often used to make the respondents aware of
1 the incongruency between the way they see themselves and the way
the other sees them. It is. also used to assess the role of
romantic idealization in the marriage relationship. The
counselor is able to point out that the way the spouses see
each other coincides very little with the way they really are.
The earliest attempt to construct a marital adjustment
test was a thirteen question test by Gilbert V. Hamilton,
16
published in 192 6.5 This study was based on one hundred married
men and one hundred married women. The thirteen questions pur
ported to measure marital satisfaction and at the same time
equated marital satisfaction with positive answers to the ques-
jtions of satisfaction. Some of the test items have been used by
\ subsequent investigators.
| The next of the tests devised to assess the success or
'failure in marriage was that by Burgess and Cottrell.^ Marital
adjustment was measured by a 26-item test on five areas of
behavior. These areas are: (1) agreements and disagreements,
i
I (2) common activities and interests (3) demonstration of
I
affection and confiding, (9) dissatisfaction with the marriage,
and (5) feelings of personal isolation and unhappiness. Areas
two and three could indicate areas for educational counseling.
Terman^ devised the next test to measure marital
happiness. He used most of the Burgess-Cottrell items, but
| added some of his own. His sample consisted of 792 husbands
! and their wives. It was a college population exclusively. The
results were published in 1938.
^G. V. Hamilton, A Research in Marriage (New York:
Albert and Charles Boni, 1929), Chapter 3, "Kinds and Degrees
of Spousal Satisfaction and Dissatisfaction."
^W. Burgess and P. Wallin, Predicting Success or Failure
in Marriage (New York: Prentice-Hall, Inc., 1939).
Terman and others, Psychological Factors in Marital
Happiness (New York: McGraw-Hill Book Co., 1938).
17
Kirkpatrick1 s criterion of marital adjustment was
"Community of Interests." His findings were based on a study
of 58 well-adjusted and M-7 poorly adjusted couples as rated by
student friends. This study was completed in 1937.® His
| adjusted couples showed more community of interests than did his
|poorly adjusted couples.
| In 1951 Harvey J. Locke^ published the results of his
i
! study based on 201 divorced couples plus 123 persons where only
one ex-spouse cooperated, and 200 happily-married couples plus
four with only one spouse cooperating,
i In 1953 Burgess and Wallin^® published the results of
i
I
their study, which was based on one thousand engaged couples.
Six hundred and sixty-six of these couples were, interviewed
after three years of marriage. Engagement adjustment was meas-
I ured by a multiple index composed of nine criteria. These cri
teria were: permanence, happiness, general satisfaction,
i
specific satisfaction, consensus, love, sex adjustment, compan-
1 ionship, and compatibility of personality and temperament. Like
I
; Terman, Burgess and Wallin viewed personality as a factor in
i______________________
| ®C. Kirkpatrick, "Community of Interest and the Measure-
j ment of Marriage Adjustment," The Family, vol. 18 (1937),
!pp 133-137.
9h. J. Locke, Predicting Adjustment in Marriage: A
Comparison of a Divorced and a Happily Married Group (New York:
Henry Holt & Co. , Inc., 1951).
lOErnest W. Burgess and Paul Wallin, Engagement and
Marriage (Philadelphia, Penn.: J. B. LIppincott Co., 1953).
18
| marital adjustment. However, they did not make comparisons
| between personalities and temperaments of the spouses. They
I
jtried to ascertain only whether the mates found each other’s
i
| personalities and temperaments acceptable or to their liking.
i
j 1 1
| Using previously published data, Wallace and Lockex ,
: through factoring, found five general factors significant in
jmarital adjustment: companionship, consensus, affectional
j
j
\ intimacy, life accomodation, and euphoria. The purpose of this
test was to validate a test that was shorter and therefore more
likely to receive cooperation from respondents than was the 246
item Burgess and Wallin test. Through means of a coding tech
nique scoring the items circled, the respondents get a marriage
adjustment score. Although these are significant factors, these
tests do little or nothing to indicate those areas where improve
ment or change are called for and can be brought about. For
example, "Do you and your mate agree on right, good, and proper
;behavior?" may be answered ’ 'always agree," because it never
! comes up. Or it may be answered "always disagree," because
i
!there is such a deep personality conflict that they just wouldn’t
like what the other likes. They might not agree on demonstra
tions of affection, but the test does nothing to indicate where
the agreement or disagreement centers.
H h. J. Locke and R. C. Williamson, "Marital Adjustment:
A Factor Analysis Study,” American Sociological Review, vol. 23
(1958), pp 562-569.
j 19
|
| In the third edition of The Family by Locke, Burgess,
and Thornes-^ eight criteria of marital success are listed. These:
criteria sum up the rationale of most marital tests to date.
These criteria are: (1) permanence of the marriage, (2) happi-
i ness of the husband and wife, (3) satisfaction with the marriage,
(M-) sexual adjustment, (5) marital adjustment, (6) integration
j of the couple, (7) consensus, (8) companionship. Yet, stated
! negatively, these are the very complaints with which couples
i
come to the marriage counselor. He needs no tests to discover
them. He does need tests to discover why the marriage is not
i likely to be permanent, if something is not done; why the hus
band and wife cannot find happiness together; why there seems to
be no satisfaction in the marriage; what stands in the way of
sexual adjustment; what personality traits or communication
difficulties stand in the way of marital adjustment; why they
cannot be integrated enough to agree on common objectives; what , j
deep-lying conflicts impede consensus; why these two people who
'want companionship cannot find it with each other.
Most of the research in testing for marriage success,
happiness, or stability has been done by sociologists and social
psychologists. The aim has been to establish criteria for judg
ing the probability of marital success and thus to aid or advise
•^E. W. Burgess, 1 - 1 J. Locke, and M. Thornes, The Family,
From Institution to Companionship (American Book Co., 1963)
| those contemplating marriage. The use of tests to diagnose and
i
i
assist in planning counseling of faltering marriages is more
recent. The steps in the procedure for predicting success or
| failure in marriage include the following:
1. Choice of a criterion or criteria of
marital success.
| 2. Selecting predictive items.
j
I 3. Selection of experimental group.
M-. Determining the predictive value of the
items, the scores to be assigned to them
i and to their sub-categories.
I
5. Classification of predictive items under
factors.
6. Correlation of scores on premarital tests
I with marital scores, construction of an
expectancy table, and applying the
I
expectancy table.
! Marital success has been defined through the establish
ment of criteria such as: (1) permanence of the union, (2) ad
justment of the couple, (3) happiness of husband and wife,
(M-) satisfaction of the couple with the marriage and with the
spouse, (5) composite criteria, and (6) multiple criteria.
The fact that a marriage is permanent does not prove
that it is a success or that it is stable. Marriages that have
persisted may have failed in coital satisfaction, emotional
21
compatibility, and the growth of individual potentialities. A
marriage may have persisted only to protect a neurotic, destruc
tive relationship. It may be held together in mutual destruc-
|tiveness through religious constraints or through fears.
I Permanence is a mechanical criterion. The relevant problem is
I how and why the marriage has persisted.
Adjustment of the couple as the sole criterion is
' fictitiously dynamic. The estimation is usually made by an out
sider. Sometimes it is the researcher, but sometimes it is by
acquaintances, friends, or relatives of the pair. The pair may
ibe considered "adjusted" according to the values of the person
making the judgment. "Adjustment" could be a withdrawal from
active relatedness to the spouse.
| Happiness as a criterion of marital success is so sub-
!
|jective and broad that it is somewhat suspect. Of all the single
criteria of marital success Burgess and Wallin seem to be most
I
ifavorably inclined toward "satisfaction with the marriage."
jThey use Jessie Bernard’s method of measuring satisfaction with
I
|marriage through what she calls the index of the degree of
marital satisfaction. Burgess and Wallin’s favorable inclination
toward this criterion comes from their belief that it makes
possible tests of satisfaction and development of an index of
relative satisfaction. However, they also assert that this
single criterion for measuring marital success places too much
stress on one line of response from the married persons who are
22
being interviewed and who are taking the tests.
The composite index of Burgess and Cottrell consists of
the following categories: (1) agreement and settlement of dis
agreements; (2) common interests and activities; (3) demonstra
tion of affection and confiding; (B) satisfaction with marriage;
!(5) absence of feelings of unhappiness and loneliness. Questions
jwere asked in the Burgess and Cottrell study in each one of these
i
!
jcategories. Each question has a series of alternative answers,
and each alternative is given a score value. The score-values
are added together and become the marital adjustment score of the
i
individual taking the test. Locke judged marital success in his
comparison of a happily married group with a divorced group by
means of relatives, friends, and acquaintances who picked out
the most happily married couples known .to them. Marital failures
were judged through divorce.
The kinds of items selected and the basis for selection
|can be illustrated by the tests constructed by Ernest W. Burgess
jand Paul Wallin, and'by Harvey Locke. On the basis of the nine
criteria mentioned above Burgess and Wallin constructed nine
schedules. The seale on permanence consists of eight multiple-
choice questions, the one on general satisfaction of twenty-six
such questions, the one on happiness of twenty-one. The schedule
on specific satisfactions has questions under several general
headings with many subheadings as does the one on consensus and
the one on love. The inventory on sexual satisfaction has
23
seventeen items, the one on companionship twelve multiple-choice
! questions, the one on compatibility fifteen self-rating cate-
: gories. These items came in part from earlier workers in this
: field and in part from the three test constructors. Concerning
the items on sexual satisfaction Burgess and Wallin say: "The
judgment of the writers determined the selection of the questions
and their numerical weighting.” Locke’s Marital Adjustment Test
includes nineteen items from the Burgess-Cottrell Marital Adjust-
Iment Test, two adaptations of Terman’s items, and eight which
were formulated by the author. This test is useful for assessing;
the adjustment of the couple, but it does not uncover the basic
iproblem of the couple.
! The researchers were never able to get a representative
sample of an entire population, such as the population of the
United States as a whole. Locke has pointed out that the items
i he used for marital prediction are neither claimed nor assumed
to be applicable to all groups in the United States, not to
mention groups in other cultures. Waller and Hill have criti-
; cized the studies as essentially middle-class, with a bias
i favoring conventionality and conservatism that they think is
better suited to the Victorian bourgeois family situation of a
; past era than to those of today. Burgess and Wallin have also
; described limitations in their sampling. Moreover, certain
! lower-class and lower-education groups may not be able to take
the tests, not being literate enough to understand the questions
j 24
and make the multiple-choice alternative most suitable to them.
Kirkpatrick found five general influential factors in
the literature on marital adjustment. He listed them roughly
|in order of decreasing substantiation by evidence. In this
iorder they are: (1) early and adequate orgasm capacity,
(2) confidence in the marriage affection and satisfaction with
;the affection shown, (3) an equalitarian rather than a patriar
chal marital relationship with special reference to the husband’s
role, (4) mental and physical health, (5) harmonious companion-
|ship based on common interests and accompanied by a favorable
attitude toward the marriage and the spouse.
Burgess and Locke posited six basic background factors
;which enter into marital adjustment: (1) personality character
istics, (2) cultural backgrounds, (3) social participation,
(4) economic status, (5) response patterns, (6) sex desires.
■ They find six personality characteristics which make for unhappi
ness in marriage. They are: (a) an unhappy temperament indicated
:by a predisposition to be pessimistic rather than optimistic,
:(b) neurotic tendencies expressed by being touchy, grouchy,
'lonesome, easily hurt, and bothered by useless thoughts, (c) dom
inating and domineering behavior characterized by determination
i to get one’s way and by disregard for the feelings of thers,
i (d) critical and inconsiderate attitude toward others, (e) lack
I
iof self-confidence on the part of the husband, (f) self-
sufficiency as indicated by usually facing troubles alone and
25
avoiding asking other's advice. This part of the test can be
! valuable for marriage counseling. Burgess and Locke find
j
; similarity of cultural background favorable and dissimilarity of
I cultural background unfavorable to marital adjustment. A mar-
iriage in which the man marries below his cultural level has a
better chance for success than a marriage in which the woman
; marries below her cultural level. According to Terman, religious
differences are not crucial to marital failure. Burgess and
I Locke conclude that the sociable and conventional person seems
i
]
i to be better prepared for achieving what they call adjustment
in marriage. According to Burgess and Locke economic status is
; a resultant of noneconomic influences such as stability and
; socialization of the individual, a participative personality,
: and a capacity to respond to others. The economic items which
the various studies found to be significantly associated with
marital success were moderate income at the time of marriage,
some savings, occupations characterized by stability and social
control, and regularity and continuity of employment.
Going through the lists of criteria which may be
indicative of marital success and stability, we find five which
might be affected by counseling. That some of them are in turn
i determined or influenced by other variables is obvious. They
■have, however, the quality of being therapeutic for the marriage
relationship once the couple consciously cultivates them. How
| much they have this quality will have to be determined by other
26
studies. These five factors are: (1) common interests arid
activities, (2) sex adjustment and education, (3) compatibility
of personalities, (4) ethical and personal values, (6) communi
cation.
Common interests and activities may be neglected, because
the couple, feels they have nothing in common. However, when
couples do engage in common activities and cultivate common
interests, they usually find more satisfaction in their relation
ship. However, deep personality differences and differences in
values can make this almost impossible to accomplish.
Sex adjustment may be marred because of deep-lying
neuroses or personality distortions. No effective substitute
has been found for assessing the sex history and sex attitude of
each spouse. The counselor’s open and realistic attitude toward
sex may be the first therapeutic step towards a better sexual
relationship of the couple.
Some Studies of Role, Personality, Empathy,
and Marital Adjustment
A. R. Mangus used the Interpersonal Check List to test
role and role perception. He checked the performance of one
spouse with the expectations of the other. According to
Mangus1^ :
13a . R. Mangus, "Role Theory and Marriage Counseling,"
Social Forces, vol. 35 (March, 1957), pp 200-209
27
. . .the integrative quality of a marriage is reflected
in degrees of concordances and discrepancies among the
; partners1 qualitative role perceptions and expectations
| as reciprocally reported by them. Role perceptions of
self and others are viewed as organized against a back
ground of role expectations including the current self-
concept of each marriage partner. The self-concept, or
social self, is defined as the generalized role expec
tations that one reports as part of a self description.
It is taken for granted that the husband's and wife's
perceptions of themselves and of each other are closely
associated with the adaptive or maladaptive character
; of their marriage. It is assumed that role performance
in the marriage will be appropriate and integrative to
the extent that there is a genuine sharing of role
; expectations by the partners to the marriage.
; Mangus stated that the ICL provides descriptions of (1) the sub-
ject's perception of himself as a spouse, (2) the subject's
perception of the partner as a spouse, (3) the subject's role
| expectations of a husband, and (*4) the subject's role expecta-
I
[ tions of a wife. The items in the final check list are so
arranged that they represent and provide scores for sixteen types
of role descriptions. These are labeled autocratic, responsible,
docile, cooperative, conventional, dependent,- self-effacing,
modest, distrustful, skeptical, blunt, aggressive, competitive,
; exploitative, and managerial. These, however, are personality
: traits, not roles. Mangus made no attempt to compare the per-
; sonalities of the spouses. There is no indication that he was
looking for significant differences, or for correlations,
; One of the first studies relating specific role factors
i
• to marital adjustment was the doctoral dissertation of'. Rathan
j
jHurvitz. In a study of functional and cdntrol roles in marriage
28
Hurvitz hypothesized:
1. Husbands and wives whose control roles scores
indicate companionship attitudes within the family
are happier in their marriage than husbands and wives
whose control scores indicate traditional attitudes
within the family.
2. Husbands and wives whose role performances
and role expectations of the functional roles of hus
bands and wives are similar are happier in marriage
than husbands and wives whose role performances and
role expectations are divergent.
3. Husbands and wives whose performance roles
correspond with the typical pattern of performance
roles in their subculture are happier in marriage
than those husbands and wives whose performance roles
do not correspond with the typical pattern of perform
ance roles in their subculture.
H. Husbands and wives whose role expectations of
the other spouse correspond with the typical pattern
of role expectations of the other spouse in their .
subculture are happier in marriage than those husbands
and wives whose role expectations of the other spouse
do not correspond with the typical pattern of role
expectations of the other spouse in their subculture. ^
In this study functional roles define the activities of
husbands and wives, and control roles define the source of
authority in the family. Hurvitz describes his study as an
attempt to elicit those roles which are available to conscious
exploration, the functional and control roles. It is, further,
an exploration of the relationship between the functional and
control roles to each other and to happiness in marriage.
The first hypothesis was rejected, since the correlation
of husbands’ scores was zero. However, the findings indicate
l^Nathan Hurvitz, ’ ’ Marital Roles and Adjustment in
Marriage in a Middle-Class Group,” Unpublished Doctoral
Dissertation, Libraries, University of Southern California,
Los Angeles, 1958.
; 29
:that the opposite is true for the wives. Wives who hold
f traditional attitudes regarding the source and kind of authority
I
^ in the family are happier than wives who hold companionship
attitudes.
The second hypothesis was accepted fov the husbands on
the basis of the association between the husband’s index of
: strain and the happiness scores of the spouses. This hypothesis,
however, was rejected for the wives on the basis of the associa-
i
i tion between the wives’ Index of strain and the happiness scores
: of the spouses.
Hypothesis three was rejected on the basis of the
' association between the spouses’ index of deviation of perform-
; ance roles and their happiness scores. Hurvitz deduces that
this suggests that happiness in marriage is associated with; an
Individual pattern of marital interaction and not with
; similarity with or difference from the typical pattern of role
performances In the subculture with which one identifies.
Hypothesis four was accepted for the husbands, but not
for the wives. Hurvitz found that as the husbands’ expectations
i
of their wives differ from the modal pattern of the husbands'
expectations of their wives, their happiness scores decrease.
| He believes this finding suggests that unhappy husbands make
I role demands upon their wives which are different from the
j
| typical pattern, but that this is not true for wives.
j . 30
Eleanor Luckey-^-3 used the Interpersonal Check List to
compare marital satisfaction with Congruence of Role Perception.
Equating personality trait perception with role perception, she
assessed congruence of role concepts in relation to self-
sperception and spouse-perception, self-concept with ideal self,
congruence of ideal self and concept of spouse, concept of parent
of the same sex and self-concept, and congruence of concept of
spouse and parent of the opposite sex.
; Luckey found significant relationships in all the per-
I ceptions she studied to marital satisfaction, except in the
• relationship between self-concept and concept of the ideal self.
j In addition, Luckey found that men who identified with their
! fathers were found more in satisfactory marriages than in un-
i
satisfactory ones; hut women who identified with their mothers
: were found almost equally in satisfactory marriages and in un
satisfactory ones.
Luckey1s findings that the ideal husband role-concept
for all groups is in the managerial-autocratic (AP) octant and
the ideal wife concept for all groups, with the exception of
the unadjusted husbands, is in the responsible-generous (NO)
octant are pertinent to the present study. However, not even
the majority agreed on what traits place a spouse in the "ideal"
l^Eleanore Luckey, "Marital Satisfaction and Its Asso
ciation with Congruence of Perception," Marriage and Family
Living, vol. 22 (February, 1960) , pp 9-9-54-.
31
category. We need to ask, for example, whether the "unadjusted"
; husband would have been In the "adjusted" column, If the wife
! had been AP and had been meeting his passive and dependent needs.
. If the theory of complementarity is valid, he may have seen his
marriage as satisfactory.
LuckeyTs results led her to accept a revised hypothesis:
: • There is a cultural definition of the ideal marital
role for the husband and for the wife, and the
greater the similarity between the spouses’ per-
i ception of their self, mate, and ideal marital roles,
' and the cultural ideal marital roles, the greater
will be the degree of marital adjustment.
Luckey used the Interpersonal Check List, the Role Attitude
Survey, and the Wallace Adjustment Scale.
! Kotlar-*-® also used the Interpersonal Check List and the
I
! Wallace Adjustment Scale to test role perception and role
: expectation as factors in marital adjustment. Her major hypoth
eses were:
1. The greater the degree of similarity between
the mate’s self perception and the spouse's percep
tion of that self, the greater will be the degree of
marital adjustment.
2. The greater the degree of divergence between
role expectations and role fulfillment, the greater
will be the degree of marital maladjustment.
3. The greater the degree of the spouses’ con
currence on role expectations, the greater will be
the degree of marital adjustment.
4. The greater the degree of similarity between
the individual’s self perception and the definition
' of his ideal marital role, the greater will be the
! -L6ga] _ ; i _ y Kotlar , "Middle-Class Marital Roles— Ideal and
i Perceived in Relation to Adjustment in Marriage," Unpublished
jDoctoral Dissertation, Libraries, University of Southern
! California, Los Angeles (1961). _ __________________
32
degree of marital adjustment.
5. There is a cultural definition of the ideal
marital role for the husband and for the wife, and
the greater the similarity between the spouses1
definitions of their ideal marital roles and the
cultural ideal marital roles, the greater will be
the degree of marital adjustment.
6. The ideal marital role of the husband con
tains more instrumental role attitudes than does
the marital role for the wife.
7. The ideal marital role of the wife contains
more expressive role attitudes than does the ideal
marital role for the husband.
■8. Where there is a reversal of role in terms
of instrumental versus expressive role attitudes
for the husband and wife, whether this is with
respect to role expectation or role perception,
it is correlated with marital adjustment.
Statistical evidence supported hypotheses one and two.
The direction of the findings on the other hypotheses were in
the predicted direction, but they were not significant.
The study most nearly related to the present study is
i that of Blazer.17 in 1963 he published the findings of his
' investigation on whether complementariness of married couples
' was positively correlated with marital happiness. There was a
significant negative correlation between the wife’s happiness
; score and the degree, of complementarity. There was also a
; negative correlation for the husbands, but this was not sig-
i nificant. Blazer concluded that both husbands and wives are
happier with people of similar needs than with people of opposite
| or complementary needs, and that this is more true of wives than
I 17jdhn A. Blazer, "Complementary Needs and Marital
; Happiness," Marriage and Family Living, vol. 25 (February, 1963).
|pp 89-95.
j 33
of husbands. He concluded that his findings not only failed to
; support WinchTs theory of complementary needs, but that the
jreverse was more likely— people with similar needs tend to marry.
Taylor1® studied fifty adjusted couples and fifty
unadjusted couples to compare the role perceptions of husbands
and wives and relate discrepancies of these role perceptions to
marital adjustment. This aspect of Taylor’s project is a cross
validation of the study of role perception by Kotlar. An attempt
was made in Taylor’s study to secure a sample similar to the one
| obtained by Kotlar in her study of marital roles and marital
satisfaction. Church-club members having high marital success
; scores on the Wallace Marital Success Test made up the sample of
i
! adjusted couples. Couples with extremely low marital success
scores and just entering marriage counseling made up the sample
of unadjusted couples. The hypothesis that greater similarity
;between self-perception and the spouse’s perception of that self
would be related to marital adjustment was supported by the
: comparisons. Empathy was operationally defined as the correct
prediction of mates responses on the Interpersonal Check List.
;Taylor’s findings suggest that accuracy of empathy is more sig
nificant with respect to perception of the husband than in
I perception of the wife. They also suggest that accuracy- of
' 18a . B. Taylor, "Role Perception, Empathy, and Marital
jAdjustment,” Unpublished Doctoral Dissertation, Libraries,
I University of Southern California, Los Angeles (1965).
; 34-
empathy in marriage is related to adjustment in the interpersonal
i
-relationship; however, a strong consistent pattern was not found.
There are several tests designed to assess subjectTs
I values. The Verden-MIchael’s Cultural Orthodoxy Test is a test
of middle class values. This instrument can he used to test for
a significant difference between the Cultural Orthodoxy scores
! of harmoniously married couples and the scores of couples in
! marital distress. It can also be used as a test of the commun-
! ication between the couple. The rationale of the communication
I aspect is that husband and wife should want to know what values
; each has. If there is good communication between husband and
■wife, we assume that each would know how the other thinks and
j
jfeels about such things as religious beliefs, political beliefs,
'economic traditions, interpersonal mores, and cultural lore.
These are the subjects explored by the Verden-Michaels Cultural
; Orthodoxy Test. A study by Lossner-^ found that happily married
couples have communicated their values to each other signifi
cantly better than have unhappily married couples. There was no
; significant difference between the actual cultural orthodoxy
| scores of the happily married and the scores of the unhappily
married.
| Some of the above researches have examined the signifi-
i
i
^Walter Lossner, "Cultural Values and the Communication
of Cultural Values as a Factor in Marital Happiness," Unpublished
Research Project, Libraries, University of Southern California,
Los Angeles (1962).
cance of personality and personality traits for marital satis
faction and stability. Substituting role for personality, some
;have directed their studies to the question of what kind of
roles lead to a happy, stable, satisfactory marriage. Others
have studied the importance of congruence between role expecta
tions and role performance. Stated in terms of personality,
I the question asked is whether the spouses1 perceiving each other
differently from the way they see themselves is a factor which
;tends to make the marriage relationship unsatisfactory. Blazer
;studied the correlation between complementary personality needs
; and marital happiness.
Some Studies on Homogamy and Heterogamy
! It is the purpose of this study to find whether there is
a significant indication that spouses with complementary person
alities are more likely to have a satisfactory and stable marital
relationship than are spouses with like or with opposite
personality needs. Winch^O contended that marriage partners
.tend to be alike culturally, because the field of eligible
|partners is limited by conventional barriers. However, within
this field of eligibles each person is attracted to that person
who gives the greatest promise of providing him or her with maxi-
|mum need gratification. This could be true, of course, only in
20r . F. Winch, Mate Selection, (New York: Harper and Row,
Inc., 1958). Also the publications listed in the Bibliography.
3 6
societies in which young people are given an opportunity to
1 choose their own mates. In a preliminary study of twenty-five
; couples he found considerable support for this theory. He iso-
i lated four frequently found conbinations: (a) families that re
semble the conventional relationship of a mother and son, where
a dominant and capable woman takes care of a husband who needs
someone on whom to lean, (b) families in which a strong, capable
husband looks after a wife who is passive and compliant, (c) fam-
I ilies that resemble the conventional relationship of a master
1 and-servant girl, in which a condescending husband is served by
a capable wife, and (d) families in which an efficient woman
; dominates a frightened and frustrated husband.
| Winch defined two kinds of complementariness, intensity
: and kind, which he called Type I and Type II respectively.
Type I-~The same need is gratified in both (a) and (b)
!but at very different levels of intensity. A negative inter-
spousal correlation is hypothesized (intensity) .
Type II--Different needs are gratified in (a) and (b).
The interspousal correlation may be hypothesized either to be
I positive or negative, contingent upon the pair of needs involved
(kind).
The study used forty-four sub-variables. With this many
j sub-variables it is possible to compute 1936 interpersonal
i
|correlations. Out of this total Winch, and others, made hypoth
eses about the signs of 388 interpersonal correlations on the
’ 37
;basis of the theory of complementary needs. For example, they
i
;hypothesized that selected nurturance-nurturance correlations
i
; would be positive and that selected nurturance-succorance cor-
I relations would be negative. In accordance with the two kinds
:of complementariness noted above, they hypothesized (a) that
34-4- interspousal correlations, each of which involved two differ
ent needs or traits, would be positive, and (b) that-44 inter
personal correlations involving the same need or trait would be
I
:negative.
I
; The researchers discovered that the data diverged so
much from chance in the direction hypothesized that the prob-
;ability of the correctness of the null hypothesis is virtually
[infinitesimal.^-*-
P P
Thomas Ktsanes ^ tested the hypothesis that the need
pattern of persons involved will be different rather than similar
!No attempt was made to specify the nature of the difference, i.e.
to specify that the difference would be complementary according
to either of the two possible types, intensity and kind.
The findings of this study, based upon a sample of twenty
[five recently married, college age, middle class couples, indi-
;cate that for the population sampled the tendency for an individ-
;ual to select a spouse unlike himself in total emotional make-up
F. Winch, Mate Selection (New York: Harper and Row,
|Inc. , 1958), p 112.
^T. Ktsanes, "Mate Selection on the Basis of Personality
Type," American Sociological Review, vol. 20 (1955) pp 547-555.
38
j
far exceeds the tendency for him to select a spouse like himself
!in that respect. At the .05 level of confidence the probability
I
for heterogamous choice on these grounds appears to be .83 or
more, and the probability for homogamous choice only .17 or less.
:This finding tends to support the basic premise of Winch’s com
plementary need hypothesis, that the patterns of marriage part
ners differ and, further, specifies the limits within which this
hypothesis may be correct.
I Bowerman and Day^3 in another test of Winch’s theory,
I
I
! found that a study of sixty courtship couples failed to support
predictions made on the basis of complementary needs. In fact,
in the case of only two of the fifteen variables used were there
I negative interspousal correlations of the same need, and neither
:was statistically significant.
2U
Rosow ^ criticized Winch’s original approach for dealing
;with many different need ratings on an individual as if they were
discrete and independent variables. He recommended analyzing
need complementarity in a framework of global personality types.
'This is the methodology of this study. It uses Leary’s formula
! for placing the person in one of four personality quadrants.
Bowerman and B. Day, "A Test of the Theory of Comp
lementary Needs as Applied to Couples During Courtship,”
American Sociological Review, vol. 21 (1956) pp 602-605.
^LRosow, ’ ’Issues in the Concept of Need-Complementar
ity_Socinmetry, vol. 20 (September, 1957) pp 216-233.
39
Winch‘ d published his support of his theory more elabor-
;ately in 1955. In this work, and repeated in his book, Mate
I Selection (1958), Winch lists and defines his list of needs:
NEEDS
VARIABLE
Abasement
Achievement
Approach
Autonomy
Deference
Dominance
Hostility
Nuturance
Recognition
DICHOTOMY
Overt-covert
within-without
Overt-covert
only without
Overt-covert
only without
Overt-covert
DEFINITION
Overt-covert
within-without
Overt-covert
within-without
Overt-covert
Overt-covert
within-without
Overt-covert
within-without
To accept or invite blame,
criticism, or punishment.
To blame or harm the self.
To work diligently to
create something and/or to
emulate others.
To draw near and enjoy
Interaction with another
person or persons.
To get rid of the con
straints of other persons.
To avoid or escape from
domination. To be un
attached and Independent.
To admire and praise a
person.
To Influence and control
the behavior of others.
To fight, Injure, or kill
others.
To give sympathy and aid to
a weak, helpless, ill or
dejected person or animal.
To excite the admiration
and approval of others.
-*R. Winch, "Empirical Elaboration of the Theory of Comple
mentary Needs in Mate Selection,” Journal of Abnormal and Social
)Psychology, vol. 51 (1955) pp 508-513.
40
Status
Aspiration
Status
Striving
Succorance
None
None
Anxiety
Overt-covert
within-without
Within-without
To desire a socio-economic
status considerably higher
than one has (a special
case of achievement).
To work diligently to alter
one’s socio-economic status
(a special case of achieve
ment) .
To be helped by a sympa
thetic person. To be
nursed, loved, protected,
indulged.
Fear, conscious or uncon
scious, of harm or misfor
tune arising from the
hostility of others and/or
social reaction to one’s
own behavior.'
The show of affection in
behavior.
The gratification of a
need derived from the per
ception that another
person is deriving grati
fication
Winch grouped these general traits under four trait
names: striving (Ach), independent (Aut), controlling (Dom),
and hostile (Hos), which he called the "assertive” cluster.
This 1955 paper_was concerned with a further analysis of
j
;the data on the twenty-five husbands and twenty-five wives for
the purpose of formulating more specific hypotheses concerning
|the nature of "complementariness."
i
i It is hypothesized that high "assertive’ ^’tend not to
i
marry persons who are like themselves in this respect, but rather
persons who are high "receptives." Winch claimed that his find-
Emotionality
Vicariousness
Within-without
Within-without
< 4 - 1
Ings demonstrated satisfactory evidence for his hypotheses four
and five. They are:
Hy. H. Men who are high in Ach tend to select as
wives women who are high in Aba, Def, Sue, Nur, and
Vic, and who are low in Dom, Hos, Ach, Aut, SS, and
Rec.
Hy. 5. Women who are high in Ach tend to select as
husbands men who are high in Rec, App, Sue, Vic, and
who are low in Aut, Ach, and Nur.
Winch^ further elaborated his theory. He continued to
use the data from his first study. He bases his entire theory ■
upon the argument that we love persons who provide gratification
of our needs and who thereby bring us pleasure. As a beginning
argument for his theory of heterogamy Winch says:
It seems dubious wisdom to extrapolate the principle of
homogamy from social characteristics, where it has been
demonstrated, to motivational or psychodynamic variables,
where it Is not.
Winch states that the purpose of his study is to specify
in concise, naturalistic terms what is meant by love, to set up
a procedure for quantifying love, and to test a general hypoth
esis about love. Basing his first Inference onuFreud and Bergler,
Winch states his proposition that homogamy operates in the area
of choice and heterogamy in the psychodynamics of the mates. He
interprets the homogamy reported by the sociologists ad defining
the class of persons in which we find our loved ones and the
heteronomy reported by the psychoanalysts as suggesting what
2&R„ F.' Winch, Mate Selection (New York: Harper and Row,
Inc., 1958) pp 15-16.
12
love is. More precisely, he says It is a useful way of thinking
about love.
? 7
His postulates: '
Postulates Of The Theory Of Complementary Needs
I. Postulates concerning behavior, needs, organization
of needs, and the gratification of needs.
A. All human behavior may be viewed as activity
oriented to the gratification of needs.
B. Some human needs are innate; others are
learned. We are particularly interested in
those which are learned and expressed in
interpersonal relationships.
C. Certain important needs are organized by
the formation of the ego-ideal.
D. The organization of needs gives pattern
to behavior, i.e., it makes behavior
(including perception) selective.
E. Needs may be experienced consciously or
unconsciously. Hence, a person may be
completely aware, partially aware, or quite
unaware of the motivation of his behavior.
F. An individual's "set" or expectation of
gratification of one or more of his needs may
be situation specific, i.e., there may be
situations in which he regularly receives
gratification of that need or needs and
other situations in which he regularly does
not.
II. Postulates concerning culture.
A. Every culture contains a principle of
preferential mating which is the result- •
ant of the emphases given to the incest
taboo and to ethnocentrism.
B. Each person's field of eligibles is the
resultant of the principle of preferential
mating and the principle of differential'
association under which he lives.
^Winch, Mate Selection, pp 87-88
43
C. In all cultures the congeniality group
is an important source for the gratifi
cation of social needs.
D. In the subculture of the American middle
I class the man-woman dyad is viewed as a
singularly gratifying conteniality group.
From these postulates Winch states (1) a definition of
love and (2) a principle of mate-selection in middle-class
America. Love is the positive emotion experienced by one person
(the person loving, or the lover) in an interpersonal relation-
iship in which the second person (the person loved, or love-
object) either (1) meets certain important needs of the first or
f
: (2) manifests or appears (to the first) to manifest personal
^attributes (e.g., beauty, skills, or status) highly prized by
|the first, or both.
j
' Mate selection may then be stated as taking place in
terms of needs. In mate-selection each individual seeks within
his or her field of eligibles for that person who gives the
:greatest promise of providing him or her with maximum gratifi
cation.^®
| The study was conducted by interviewing twenty-five
|couples with open-end questions. An example of a question
; assessing aggression is "How do you feel when someone steps in
front of you in a queue in a crowded restaurant?" A second
source of scoring was an eight-card TAT. Neither could be con-
i
j sidered an objective test.
2®Winch, Mate Selection, p 14.
44
The researchers compiled five sets of data on the twenty-
! five couples. They conclude that three of these support the
; general hypothesis of complementary needs in mate-selection.
; The other two do not support their hypothesis, but neither do
they show a counter-trend. Winch sums up his findings:
Two special tests for Type I complementariness on
the same set of subjects yield confirmatory results
and the matching operation is that the bulk of
evidence supports the general hypothesis with the
tentativeness usual in any scientific conclusion.
S c h e l l e n b e r g ^ studied sixty-four married couples,
i thirty-six pre-married couples, and a number of artificial
: couples. He chose dominant personal values as the focus of his
I study of homogamy. He used convergence scores to measure the
| extent to which the characteristics of the two members of a
couple are similar. He got the following statistics:
Mean of Conver-
gence Scores S. D. 5EM
Married 134.5 4-4.0 5.5
Premarried 132.0 28.1 4.7
Artificial 115.3 43.8 4.9
■ Schellenberg concluded that there can be no doubt as to the gen-
j eral finding of homogamy, but that the crucial question remains
unsettled concerning the extent to which this homogamy simply
reflects characteristics of social background which a person
: might be expected to have in common with most of his associates.
A. Schellenberg, T , Homogamy in Personal Values and
the Field of Eligibles," Social Forces (Dec., 1960) pp 157-162.
45
In August 1960 Schellenberg published another study,30
,T. . .to reconcile the contrasting findings of the studies of
Winch and those of Bowerman and Day." Winch used married couples
as subjects, while Bowerman and Day used courtship couples. In
order to allow for any difference in results which might be due
to marital status the Schellenberg study used both married and
premarried couples as subjects. These groups were analyzed
separately. A total of sixty-four recently married couples (all
married less than ten years, with the median length of marriage
less than two years) and thirty-six premarried couples (eighteen
engaged and eighteen going steady) were included.
The Instrument used for the study of needs was Edwards
Personal Preference Schedule, the same instrument used by
Bowerman and Day. Winch had used Murray’s list of needs used
in scoring the TAT. The basic hypothesis of the Initial part
of the study was that the need patterns of recently married
couples and of courtship couples are more dissimilar than would
be expected by chance.
Sixty-nine of the one hundred couples showed positive
correlation of need patterns, including seventy-three percent
of the premarried sample. The brunt of the evidence thus leads
to the rejection of the theory of complementary needs, if it is
30j. Schellenberg and L. Bell, "Re-examination of the
Theory of Complementary Needs in Mate Selection," Marriage and
Family Living (Aug., 1960) pp 227-232.
: 46
' Interpreted "to mean that need patterns of spouses or prospective
i spouses tend to be dissimilar rather than similar. Schellenberg
; drew no conclusions about intensity of personality needs. No
: significant difference between samples of married and samples of
premarried were observed in this study.
B. Day, who in 1956 collaborated with Bowerman in a test
' of Winch’s theory of complementary needs made another study^l
independently. As subjects of his study Day used sixty volunteer
■ university courtship couples and the respective same-sex friends
ifor both male and female members. Each subject completed a
questionnaire containing some background items, a few control
I
jitems and the Edwards Personal Preference Schedule, which gives
! measures of the 15 personality variables associated with 15
I
manifest needs (see page 39 for items).
Comparing personality-need patterns of courtship couples
'and their respective same-sex friends, there appears to be no
general, systematic pattern of either complementariness or homo
geneity of personality needs as related to selection of either
1 courtship partners or same-sex friends. However, the specific
|needs and need combinations significantly related to selection
;of courtship partners were considerably different from the needs
and need combinations related to selection of same-sex friends.
31b Day, "A Comparison of Personality Needs of Courtship
Couples and Same-Sex Friendships," Sociology and Social Research,
vol. 45 (1961), pp 435-440.
47
op
, B. Mursteirr used the Edwards Personal Preference
jSchedule, Wallace's Marital Adjustment Scale, and Bass Famous
I Sayings Test in a study. His subjects were twenty newlywed
;couples and forty-eight couples who had been married ten years or
imore. Murstein selected the latter because Winch thought persons
: who do not exhibit complementary need patterns would presumably
;be likely to terminate their marriage before ten years had passed.
In his discussion of the results of the study Murstein
j
concludes that the heterogeneity-homogeneity dichotomy seems to
| be a gross simplification of the actual marital situation. For
example, the strength of a need Is not an omnipotent unvarying
■ characteristic of a person, but varies according to the Individ-
i
I ual concerned, where he is, and who is present. Further, some
persons may exhibit many needs, while others manifest only a few.
Also, needs change as a function of time. He concludes that the
data is far from conclusive. They are, however, in the directior
of a homogamous theory of need pattern choice for non-newlyweds.
The evidence for newlyweds is entirely Inconclusive In the sense
! that neither the homogamous nor the heterogamous theory is
' supported.
In 1962 Kerckhoff published his findings. on a sample
Q P
; B. Murstein, "The Complementary Need Hypothesis in
[Newlyweds and Middle-Aged Couples," Jr. of Abnormal Social
Psychology, vol. 63 (1961) pp 194-197.
33a C. Kerckhoff, "Value Consensus and Need Complemen
tarity in Mate Selection," American Sociological Review, vol. 27
(1962) pp 293-303.
* 4 - 8
of volunteer Duke University Women who were engaged, pinned, or
| "seriously attached" and their fiances or boy friends. He had
I
'two hypotheses: (1) the degree of value consensus is positively
; related to progress toward a permanent union and (2) the degree
! of need complementarity is positively related to progress toward
: a permanent union.
In Kerckhoff1s study value consensus was related to
: progress toward permanence for the couples as a whole. When the
I sample was divided into long-term and short-term couples, value
i
| consensus was related to progress for the short-term couples,
i and two of three measures of complementarity were related to
| progress for the long-term couples. These findings are inter-
| preted as indicating that a series of "filtering factors" operate
; in mate selection at different stages of the selection process.
Kerckhoff believes that his data generally support the idea that
i social status variables (class, religion, etc.) operate in the
early stages, consensus on values somewhat later, and need com-
plementarity still later on. His interpretation of the delay in
' the operation of complementarity factors is that such personality
1 linkages are often precluded by the unrealistic idealization of
the loved one in the early stages of courtship.
Kerckhoff evaluates his study as a positive step in
| defining the factors which lead to a lasting relationship between
i
f
! a man and a woman. His conclusions are that individuals who are
|similar to each other are most likely to choose each other as
1+ 9
mates and are most likely to be successful in the relationship.
Similarities have been noted in a large number of characteristics
such as area of residence, socioeconomic level, religious affil
iation and activity, and many kinds of attitudes and values.
Winch indicated that those variables normally associated with the
theory of homogamy in mate selection simply define the "field of
eligibles" from which each individual then chooses a mate who is
likely to complement himself on the personality level.
KerckhoffTs study introduces a longitudinal perspective
during the selection period so that further knowledge of the
actual selection process is gained. He attempts to examine the
relationship between progress in the mate-selection process in
the premarital period and measure of homogamy and complementarity.
Looking over the work that has been done to discover the
determinants of mate-selection, C. D. Bolton, writing in Marriage
:and Family Living, says:
We know very little, scientifically, about mate-
selection either as a process or as a relationship as
such--that is as a love or intimate relationship.
After decades of correlation studies the theoretical
dispute between the theories of homogamy and heterogamy
i (or theory of complementary needs) is unsolved. While,
: on the surface, homogamy appears to have the stronger
case, the evidence actually is indecisive. The large
and diversified Burgess-Wallin sample produced for
personality traits a very small ratio in the homogamous
direction, while Winch's sample drawn entirely from
Northwestern University undergraduates, and, therefore,
more likely to have a more homogamous set of field of
| eligibles, exhibited a generally complementary relation
! in personality characteristics.
i Winch's study, though widely acclaimed, can hardly
j be said to offer convincing evidence for complemen-
50
tarity as a sufficient explanation of mate selection.
i In view of the facts (1) that Winch uses MurrayTs set
of personal need and (2) that the TAT was devised for
1 getting at just those needs, the fact that the TAT
i ratings were not only the least favorable of three
sources of data used by Winch to test his complemen
tarity hypothesis, but were actually more than chance
contrary to his hypothesis, casts considerable doubt
on the findings. Barely half the correlations based
on case history data were in the hypothesized direction.
Even then the number of significant correlations out
of 388 Is only 71. Even taking the contributions of
both the homogamy and complementarity theories, much
if not most of the variance is mate selection continues
to be unexplained.^
Although Winch was not the first to propose a theory of
i
j complementarity in mate selection, he gave the study of comple
mentarity its Impetus. Most of the recent studies of this factor
jhave been in relation to his hypothesis. Perhaps he was the
I first to set up testable criteria, even though it appears that he
Imay have interpreted his data in such a way that they supported
his hypothesis. Summing up the studies on complementarity, it is
!obvious that it Is by no means a universal in marriage. Since
some marriages are characterized by personality complementarity
:and some are not, the door Is left open for some other Interpre-
'tation. It could mean that complementarity of personality needs
iIs a factor in marital satisfaction and stability. Winch himself
stated that a marriage without complementarity of personality
needs could not last ten years.
' ^ U -
i ^C. D. Bolton, T T Mate Selection as the Development of a
|Relationship," Marriage and Family Living, vol. 23 (1961)
j pp 2 3 4- 2 M-0.
51
|
One recent study which recognized that complementarity
i
j of personality needs might be an indication of some other factor
I
| in marriage is the study by Louis Miller. ^ Miller thought that
I since complementarity did not characterize all marriages, it
might be the characteristic of equalitarian-role marriages.
Homogamy might then characterize the authoritarian role. In a
: study of 210 couples who were approached only minutes after they
i had obtained a license from the marriage bureau in the Los j
i
; Angeles Courthouse, Miller found no relationship between person-
jality needs as measured by the ICL and familial role attitudes.
:He divided his sample into three groups: both traditional
■ (authoritarian) , both equalitarian (companionable) , and one
j traditional and one equalitarian. He also found no relationship
; between perceptual accuracy or role taking and interpersonal
needs. This again affirms the findings of the studies reviewed
i in this paper that in some marriages we find interpersonal comp
lementarity and in some we do not. It was the purpose of the
present study to ascertain whether complementarity is a' function
|of stable marriages. Many writers on marriage have asserted that
jit is, but the theory has not been tested.
Perhaps the first study to propose personality factors
' ^L. Miller, "Familial Role Typology, Accuracy of Per-
jception, and Mutual Needs Among Pre-Nuptial Partners," Unpub-
jlished Doctoral Dissertation, Libraries, University of Southern
jCalifornia, Los Angeles (1966).
52
as being significant in marital happiness or stability was the
study by Terman in 1937.He had nothing to say about comple
mentarity, however. He barely hinted that there are certain
personality types that have a better chance at marital happiness
than other types have.
Schooley3^ stressed the personality resemblance among
married couples. . He believed that like picks like, and the two
then become more alike the longer they live together.
30
Kirkpatrick ° seems to hold that if couples have opposing
or complementary personalities it is an indication of neurotic
interaction.
H. M. Richardson38 studied personality traits in relation
to marriage rather vaguely and loosely. What he did,.actually,
was to suggest that it is a field for study. Calling personality
traits mental traits, he found no significant relationship be
tween personality traits in marital partners. He was sure that
38L. M. Terman, and others, Psychological Factors in
Marital Happiness, (New York: McGraw-Hill Book Co., 1938)
“ 3 7
3/M. Schooley, "Personality Resemblance Among Married
Couples," Journal of Abnormal Psychology, (October, 1936)
pp SHO-^M-l.
38C. Kirkpatrick, "Statistical Investigation of the
Psychoanalytic Theory of Mate Selection," Journal of Abnormal
Psychology, (October, 1937) pp 427-M-30.
38H. M. Richardson, "Studies of Mental Resemblance
Between Husbands and Wives. . .," Psychological Bulletin, vol. 36
(1939) pp 104-120.
53
with the introduction of valid and reliable instruments some
thing significant could be demonstrated about marriage partners
and personality traits.
One of the early studies was that by E. L. Kelley.^ As
part of a long-time genetic study of factors underlying marital
compatibility, three hundred engaged couples, most of whom later
married, were interviewed and administered an extensive battery
of psychological tests. The course of each marriage was studied
by means of an annual follow-up for a period of seven years.
Kelley’s analysis of assortative mating was based on the Otis,
S. A., Bernreuter, Bell, Strong, Allport-Vernon scales, attitudes
toward marriage, church, divorce, housekeeping, entertaining,
care of lawns, gardening, and rearing of children, and a-number
of physical traits and anthropomorphic indices. Correlation of
mates on these forty-nine measures range from -.05 for Bernreuter
F-2 scores to +5 4 - for Allport-Vernon ’ ’ religion" scores. The
median correlation is +2M-, as compared with a standard error of
.06 for an.r of .00 based on three hundred cases. No signifi
cantly negative coefficients of assortative mating were found.
Kelley concludes that where assortative mating in humans occurs,
homogamy, rather than heterogamy, occurs.
L *"^E. L. Kelley, "Psychological Factors in Assortative
Mating," Psychological Bulletin, vol. 37 (19M-0) p M-73.
54-
H. T. Christensen^ studied views on mate selection based
upon two surveys at Brigham Young University (194-1-194-4-) . The
survey was administered to a number of classes in courtship and
marriage. Christensen's conclusions were that (1) males and
females tend to emphasize about the same things when it comes to
selecting a partner for marriage, and they both pay most atten
tion to such personality traits as dependability and emotional
maturity; (2) males, however, look somewhat more to such things
as youthfulness in relation to self, attractiveness and popular
ity, homemaking ability, non-smoking, and agreeableness of dis
position; (3) females, on the other hand, show themselves to be
generally more selective and to give greater emphasis to such
things as financial stability, education, ambition, similarity
of backgrounds, prospects for normal children, and chastity.
Anselm Strauss1 "^ of Indiana University perceived two
kinds of linkage between marital choice and personality needs
formulated by psychologists. (1) Personality needs are a direct
outgrowth of early childhood affectional relationship with one or
both parents. The mate fills either the same needs for the in
dividual as were previously filled by a parent, or he fills those
needs that a parent left unsatisfied. (2) Needs are not merely
^H. T. Christensen, "Student Views on Mate Selection,"
Marriage and Family Living, vol. 9 (194-7) pp 85-88.
Li?
^tA. Strauss, "Personality Needs and Marital Choice,"
Social Forces (March, 1947) pp 332-335.
55
an outgrowth of early affectional relationships with parents.
Experience in school is another source of needs. Thus the per
son may choose a mate to fill certain needs whose development
may be less directly linked with early childhood experiences
with parents.
Strauss gathered statistical data from a group of 373
engaged or recently-married persons (200 women, 173 men).
Strauss says that statistical evidence supported the hypothesis
that the person’s major personality needs may influence his
choice of a marriage partner. As major needs Strauss presents
these:
For someone.... to love me; to confide in; who shows
me a lot of affection; who appreciates what I have
to achieve; who understands my moods; who helps me
in making important decisions; to stimulate my
ambition; whom I can look up to very much; who gives
me self-confidence in my relations with people.
Coming somewhat closer to our interest in complementar
ity is the study by Horace Gray.1 ^ • Gray made a study of Jung’s
psychological types: introversion and extraversion; sensation
and intuition; thinking and valuing (feeling function) in 271
couples. In regard to these types, Gray concluded that spouses
are unconsciously attracted to complementary mating. He summar
ized four relationships and the proportions of each that he
found in his study:
no
_ ^Horace Gray, ’ ’ Psychological Types in Married People,"
Journal of Social Psychology, vol. 29 (1949) pp 189-200.
(a) Three complementary powers and no like
power for forty-one couples or fifteen percent
of the 271 pairs.
(b) Two complementary powers and one like
power for thirty-two percent of the 271 couples.
(c) One complementary power and two like
powers for thirty-eight percent of the 271
couples.
(d) All three powers alike for forty pairs,
or fifteen percent.
A transitional study was the one by R. J. Corsini.^
Corsini calls this an investigation of some correlates of happi
ness in marriage of twenty university student couples. The
specific objective was to determine the relationship between cer
tain special perceptions and marital happiness. Corsini drew
five conclusions:
1. There is no evidence that understanding the
mate is a function of similarity of the selves of
mates.
2. There is no evidence that happiness in mar
riage is a function of understanding the mate (social
perception) . z
3. The evidence indicates that happiness in
marriage is associated with similarity of self
perception of the mate.
H. Husbands and wives are no more similar in
their self-perception than randomly paired men and
women.
S. Happiness in marriage is related to the con
formity of men to self-perception of their selves.
This relationship does not hold for women.
There must be many factors which contribute to marital
stability. Whatever other factors go into determining the satis
faction and the stability of a marriage, satisfaction of person-
^R. J. Corsini, "Understanding and Similarity in
Marriage," Journal of Abnormal and Social Psychology, vol. 52
(1958) pp 327-332.
57
ality needs would seem to be one of them. Would only neurotic
people unconsciously select a mate who complements or satisfies
his needs, neurotic or otherwise, as Bergler posits? Healthy
people also have unconscious needs and conscious ones. Perhaps
the most recent statement from a clinician on this subject is
that of Rheingold. He states:
The presence or absence of complementarity as the
key to adjustment is a major tenet of the "neurotic
interaction" theory .... Not the personalities
of the partners but the way in which they interact
determines the stability of the union; thus, healthy
personalities may interact discordantly, while
neurotic personalities may form a stable "compound."
In the latter case, Piotrowski and Dudek declare:
"The marriage remains permanent because the partners
consciously or unconsciously fulfill each other's
neurotic needs--those needs on which the equilibrium
of their personalities and their social adjustment
depends." And complementarity does make for stability,
whether mutual adaptation of roles or interlocking of
neurotic needs. It comprises only part of the total
interaction, however, and it may be a phase rather than
a settled adjustment. In the neurotic form it entails
suffering and ill-effect upon the children. Consider
ing the multiplicity of needs and desires and aspir
ations on both sides, the contradictory strivings in
the individual, both at the conscious and at the
unconscious level, and the incompatibility of uncon
scious and ego goals, a priori one would rule out
the very possibility of complementation. . . . The
more important contradictory needs in both husband
and wife lie deeper and involve fundamental polarities
such as dependence-independence, manliness-fear of
being a man, and womanliness-fear of being a woman.
It is with this theory of complementariness that the
present study is concerned. In the studies reviewed above there
^-“ J. C. Rheingold, The Fear Of Being A Woman, (New York
and London: Grune & Stratton, 196 4 - ) .
58
f
is no. agreement as to its universality in marriages. Even Winch,
who posits the theory of complementary mating where the choice
is free, believes that marriages not characterized by complemen
tarity will not last. Many psychoanalysts are agreed that
neurotic complementarity is characteristic of unions in which the
partners cling destructively to each other, not being able to
tolerate or to leave the mate. Again, it seems unlikely that two
people can long tolerate one another in a union such as marriage,
unless both receive some need gratification. This study, using a
validated and reliable limited instrument, was an attempt toward
finding usable relationships between marital stability and per
sonality needs. As a means of measuring the personality traits
of the spouses and the divergences that may mean complementarity
or dissonance, the study used Leary’s Interpersonal Check List.
By means of this instrument the respondents received
four classifications of a global personality type, which Rosow
considered more valuable than many different need ratings. The
■four types are: Affiliative-Dominant, Affiliative-Submissive,
Non-affiliative-Dominant, and Non-affiliative-Submissive. In the
literature on the Interpersonal Check List Love is used synony
mously with Affiliative, and Non-affillative is synonymous with
Hate and Hostile. This study also uses these terms interchange
ably. With the division into four personality types, three types
of marital relationships emerge. These are the homogeneous, the
complementary, and the opposite types. Other studies seemed to
59
equate difference with complementarity. Differences may be
either complementary or oppositional. This study hypothesizes
that homogeneous needs may contribute both to stability and to
instability, depending on the need in question. Opposite needs
are more characteristic of unstable marriages than of stable
;marriages. Complementary personality needs are more character-
!
jistic of stable marriages than of unstable marriages.
[
I
j Summary
I
j Marital adjustment tests began with direct self-
jassessment of satisfaction. On the Burgess-Cottrell Test respon-
I
dents rated themselves on five areas of behavior (p 16).
Kirkpatrick stressed "community of interests." Burgess and
Wallin used nine criteria, (p 18) . Wallace and Locke related five
factors to marital adjustment. These are: companionship, con
sensus, affectional intimacy, life accomodation, and euphoria.
The ra.tiona.le of most marital adjustment tests ca.n be summed up
by the following criteria.: (1) permanence of the marriage,
!
(2) happiness of the husband a.nd wife, (3) satisfaction with the
marriage, ( 4-) sexual adjustment, (5) marital adjustment, (6) in
tegration of the couple, (7) consensus, and (8) companionship.
Mangus introduced the study of role and role perception,
relating them to ma.rita.l adjustment. He used Leary’s Inter
personal Check List. Using the same instrument Luckey a.ccepted
the hypothesis that there is a . cultural definition of
60
the ideal marital role for the husband and for the wife, and the
greater the similarity between the spouses’ perception of their
self, mate, and ideal marital roles, the greater will be the de
gree of marital adjustment.
Kotlar found statistical support for her hypothesis that
marital-adjustment is positively related to mate’s self percep
tion and the spouse’s perception of the self. She also found
that marital maladjustment is greater, when the degree of diver
gence between role expectations and role fulfillment is greater.
Taylor’s findings supported those of Kotlar.
The above studies contributed to the understanding of
the importance of the communication of role and the fulfillment
of role expectations to marital adjustment and stability. Winch
stressed the importance of complementarity of personality needs
in mate selection and in the endurance of the marriage.
Most of the literature on mate selection failed to
support Winch’s theory of complementarity. The study of
Kerckhoff and Davis did support Winch. Miller studied the rela
tionship between need complementarity and familial role typology.
The project reported in this study looked for significant
relationships between personality needs and marital stability.
The hypotheses were:
Hypothesis 1: When husband and wife both have the same
global personality characteristics and,
therefore, the same personality needs, the
Hypothesis 2
Hypothesis 3
Hypothesis 4 -
marriage will tend to be characterized
as unstable.
When the personality needs of a husband
and wife are complementary, the marriage
will tend to be stable.
Even though the coupleTs Dominance-
Submission needs are complementary, a
marriage will tend to be unstable, when
one spouse is affiliative and the other
is non-affiliative.
When the Dominance-Submission personality
needs of a couple are the same and their
affiliative needs are opposite, the mar
riage will tend to be unstable. This is
a marriage in which both the necessary
homogeneity and need complementarity is
lacking.
62
: CHAPTER III
;
THE SAMPLE AND THE METHODOLOGY USED
i
l
3
3
I The Sample
The sample of the present study
| In order to obtain samples for the stable married popu-
j
jlation, various church clubs were contacted. On condition that
members complete the three questionnaires the researcher offered
to speak and lead discussion on various aspects of family life
and problems. Before beginning the main lecture, the speaker
described the study briefly and stressed its importance and the
importance of filling out the questionnaires carefully and
accurately. The husbands and wives were told to sit on opposite
jsides of the room so that they could not see'each other's answers
I
j
or discuss them. The first questionnaire was a description of
"self” on the ICL. As soon as each person raised his hand as a
sign of completion, he was given another ICL sheet with instruc
tions to check those items which described his or her spouse.
This made it impossible to compare answers with others or to com
pare "self" and "spouse descriptions. The last questionnaire
was "The Marital Stability Test." No attempt was made to match
couples. After all the questionnaires were completed and gather
ed, the lecture and discussion followed. Eight church groups
participated in the study: two Presbyterian, two Lutheran.; one
Methodist, one Congregational, one Church of Religious Science,
and one Community Church, claiming no other denominational
alliance. The tests were administered to sixty-seven men and
seventy-three women in the church groups. Although these were
couples' clubs, there were three in which members came without
their spouses. In one club the men were preparing refreshments
and did not complete the questionnaires. From the church sample
forty-six men and forty-one women scored low enough on the
Marital Stability Test to be placed in the Stable Sample. Thir
teen men and twenty-one women fell in the High Score-Unstable
Sample. For various reasons eight sets of men's questionnaires
and eleven sets of women's questionnaires had to be eliminated.
The remaining stably married sample was gathered at a neighbor
hood bowling alley. Over the loud speaker the nature of the
project was described in its simplest terms. Here the question
naires were stapled together with clear instructions on each
one. The bowlers were permitted to complete the questionnaires
and return them that evening or to mail them in stamped envelopes
provided by the researcher. Out of a possible sixty-four men
and sixty-four women this source produced twenty-four men and
twenty-nine women, who had scores falling in the stably married
range. It produced only five men and three women for the un
«
64-
stable sample. There were eight incomplete questionnaires re
turned by women and four by men. The rest did not return the
questionnaires. Four stably married men and two stably married
women were drawn at random and dropped from the stable sample to
make a sample of seventy stably married men and seventy stably
married women.
The unstable group was obtained by three different meth
ods. The twenty-four women and thirteen men who scored high
(high-unstable) in the Marital Stability Test in the sample
gathering described above made up part of the sample. Three
xtfomen and five men were from the groups at the bowling alley.
This process added twenty-four women and seventeen men to the un
stable sample. The researcher exchanged speaking for completing
questionnaires to get samples from two singles’ groups. Although
all were asked to fill out the questionnaires, only those filled
out by divorced people were used. A few men and women did not
return the questionnaires. The same procedure used in the church
groups was used here. The questionnaires were distributed and
gathered one at a time before the start of the lecture. There
was much commotion— tittering, wise cracks, etc., during this
procedure, which was not noticeable at the church groups. There
were more women present than men. Since divorce was used as the
criterion for marital instability the Marital Stability Test was
not given. After two meetings the unstable marriage sample
stood at seventy-three women and forty-one men. Three women
65
were drawn at random by number and dropped from the sample. The
remaining twenty-nine divorced men were added by going to places
where single people gathered for discussions, dances, etc. Only
divorced men were interviewed and asked to help in beginning to
bring stability to marriage relationships. Only two men refused
to participate in the study. Four left the meetings before com
pleting the questionnaires.
The basic format of both the Marital Stability Test and
the Interpersonal Check List are described in the methodology
section below. In addition to the basic points to check, sub
jects were asked to give information on: age, years married,
family income, occupation, and years of schooling. Race and
religion were not included. Except for those gathered at the
bowling alley the subjects in the stable sample were all protes-
tants. No effort was made to find out the religious preferences
of the divorced people'making up the unstable sample. In the
church-club groups there were one Negro couple and two Japanese
couples.
The final sample for the main hypotheses of this study
contains a total of 280 sets of questionnaires, seventy men and
seventy women in stable marriages and seventy men and seventy
women in and out of unstable marriages. No attempt was made to
obtain a random sample of the general population.
66
Social Characteristics
Eight social characteristics are presented to describe
the sample and to compare the stably married and the unstably
married groups: religion, age, number of years married, number
of times married, number of children, yearly family income,
occupation, years of education. No question was asked about
religious preferences. However, the sample of the stable group
came largely from Protestant churches. Those from the bowling
alley were not identified as to religious preference; however,
all were in church sponsored leagues. In the church population
were one Negro couple and two Japanese. The rest were Caucasian.
All subjects in the bowling alley were Caucasian. In the un
stable sample all were Caucasian. None were identified as to
religious preference. However, the make up of the singles groups
have low Catholic membership. Protestants, Jews, and non-church
affiliated make up the members of the group.
The ages of the men in the stably married group ranged
from 26 to 56, with a mean of 37.7. The ages of the women in the
stably married group ranged from 23 to 56, with a mean of 35.3.
The men and women were not necessarily married to each other so
that the differences in their ages is not meaningful in indicat
ing differences in ages of husbands and wives.
Men in and from unstable marriages ranged from 23 to 51
with a mean of 35.3. Women in this group ranged from 20 to 53,
with a mean age of 32.3. The two groups are not too dissimilar
67
in age. How they fell in the different age groups can be seen
from Table 1. Men and women under twenty-five were so rare that
they were included in the first grouping of subjects thirty or
under.
TABLE 1
PERCENTAGE OF MEN AND WOMEN OF EACH GROUP
(STABLE AND UNSTABLE) IN VARIOUS AGE GROUPS
STABLE UNSTABLE
Men Women Men Women
N=70 N=70 N=70 N=70
Under 30 27 47 27 45
31 to 35 25 12 22 28
36 to 40 12 11 23 15
41 4o 45 14. 13 10 10
46 to 50 10 6 10 1
Over 5 0 10 11 8 1
Previous Marriages
In the male sample seven percent of the men in stable
marriages and six percent of the men in the unstable marriage
group had been married more than once. For the women, in stable
marriages four percent had been married previously and eight per
cent of the women in unstable marriages had been married more
than once.
68
Number of Children
The average number of children in the stable marriage
was 2.3!+, the average number for the unstable marriages, ongoing
and terminated, was 2.80. Table 2 shows the distribution of the
number of children for each group. This distribution does not
substantiate Taylor's^® findings that the unadjusted couples
tended to have more one-child families than the adjusted. How
ever, thirteen percent of the stable couples had four to six
children and twenty-eight percent of the unstable had four or
more children.
TABLE 2
NUMBER OF CHILDREN OF STABLE AND UNSTABLE
MARRIAGES, BY PERCENTAGES
Number of Children Stable Marriages
N=lM-0
Unstable Marriages
N=1«*0
0 6.0 2.0
1 17.0 15.0
2 39.0 30.0
3 25.0 25.0
4 - 8.0 17.0
5 3.0 5.0
6 2.0 6.0
'
100.0 100.0
^Alexander B. Taylor, "Role Perception, Empathy, and
Marital Adjustment," Unpublished Doctoral Dissertation,
Libraries, University of Southern California, Los Angeles, (1965)
p 108.
Number of Years Married
The average number of years married for the stable group
was 13.8 and 11.6 for the unstable group, as shown on Table 3.
This is near the average of Taylorsample. Taylor had more
in his 0-5 than this study did in its under 5. His averages
were 10.62 for the adjusted group and 10.86 for the unadjusted.
KotlarTs^ sample had an average length of marriage of M-.72
years for the adjusted group and 7.36 for the unadjusted group.
The unstable sample of the present study is quite different from
the unadjusted sample of Kotlar and of Taylor. Their unadjusted
couples were still together. In the present study the majority
of the unstable sample had already been divorced. It must be
noted that among the unstable the respondents were describing a
relationship which had been terminated for an unknown length of
time. Table 3 shows the number of years married for both the
stably and unstably married couples.
I
[Educational Level
[ — --- ; -- ----
j On the face sheet of the Marital Stability Test the
respondents were to fill in "years of schooling." The mean of
years of education for men in stable marriages was 13.8. . For
L | '^Taylor, op. cit. , p 105.
H8
Sally Lee Kotlar, "Middle Class Marital Roles--Ideal
and Perceived in Relation to Adjustment in Marriage," Unpublish
ed Doctoral Dissertation, Libraries, University of Southern Cali
fornia, Los Angeles (1961), pp 65-66.
70
the women in stable marriages the mean of years of education was
12.5. The mean for men in and out of unstable marriages was
13.4 and for women it was 11.7. The stably married men had an
average of .4 years more education than the unstably married.
The stably married women had .8 more years of education than the
unstably married women had. The discrepancy in Taylor's^ sample
of adjusted and unadjusted was greater. His adjusted men had an
average of 1.82 more years of education than did the unadjusted
men. The adjusted women had an average of 1.22 more years of
education than did the unadjusted women. Table 4 shows percent-
|
ages of reported education in six categories of education.
Taylor's findings of a greater number of adjusted men than un
adjusted men who had graduate level education was not borne out
in this study. A comparison of the females indicates that the
greatest difference was in the 13 to 16 (college) level. The
unstable had a higher percent in the first two years level, but
the stably married had sixteen percent in the 15 to 16 year level
! compared with 4.5 for the unstable.
The level of education of the present sample is lower
than Taylor's sample and closer to what would be expected from
a random sample of Los Angeles or Orange County, but is higher
than would be expected. The mean'of three studies by Kotlar,
Taylor, and Lossner are shown in Table 5.
^Taylor, op.cit., p 100.
71
TABLE 3
LENGTH OF MARRIAGE OF STABLE AND
UNSTABLE COUPLES, BY PERCENTAGES
Years Stable
N=lM-0
Unstable
N=1H0
Under 5 2.0 13.0
5 to 9 38.0 33.0
10 to IM 21.0 23.0
IS to 19 1M.0 16.0
20 to 2M 12.0 8.0
Over 25 13.0 7.0
100.0 100.0
Averages 13.8 11.6
■
TABLE i +
REPORTED EDUCATION OF MEN AND WOMEN IN
THE STABLE AND UNSTABLE MARRIAGE
GROUPS, BY PERCENTAGES
Years of
Education
Men Women
Stable
N=70
Unstable .
N=70
- Stable
N=70
Unstable
N=70
No High School M-.O 3.0 7.0 M.O
9 to 10 3.0 9.0 13.0 13.0
11 to 12 32.0 29.0 M-2.0 52.0
13 to 1M 28.0 25.0 17.0 22.0
15 to 16 16.0 1M.0 16.0 M.5
Over 16 17.0 20.0 5.0 M. 5
100.0 100.0 100.0 100.0
72
TABLE 5
COMPARISON OF MEAN OF REPORTED YEARS OF SCHOOLING
IN THREE RECENT MARITAL STUDIES
Kotlai^^
Taylou
Lossner
Stable
Men
Unstable
Women
Stable Unstable
15.0 15.0 13.7 13.7
16.6 Ik.8 14-.7 13.5
13.8 13.1 4 - 12.5 11.7
Occupational Level
Eleven occupational classifications were used for the
present study: professional, business, managerial, supervisory,
engineering, technical, crafts, clerical, sales, service, and
housewife. Under professional were those in medical practice,
medical arts, ministry, teaching, and the like. Business includ
ed subjects in the decision making policies of banks, lending
agencies, real estate, brokers, retailers, etc. Managerial were
those in top administration positions. Subjects who listed
ithemselves as manager and showed incomes under $10,000 were
i
!
placed in the supervisory category. Supervisory were those
people who were responsible for the performance of production and
sales personnel. Engineers were those who were in the planning,
design, etc. of production and production products. Those who
^Kotlar, op. cit., p 7M-.
^Taylor, op. cit., p 100.
73
i
were classified as technicians were those who had to have an
understanding of the processes involved, more than assembling
according to plans. They are the trouble shooters, repairmen,
mechanics„ Under crafts were listed all who worked primarily
with their hands, following plans, sueh as carpenters, sheet
metal workers, assemblers. Those listed as doing clerical work
were the people who carry out such office tasks as filing, typ
ing, etc. Under sales came what is commonly called the sales
clerk in retail stores, the car salesman, insurance salesman, and
the like. Service covered people in filling stations, hair
dressers, milk and bread delivery, and the like. The term house
wife is self-explanatory, but it should be noted that it was not
a term selected arbitrarily by the researcher, but was used by
97% of the respondents, stable and unstable. Only three percent
listed themselves as "homemaker." None of the women in the
unstable group did so. Table 6 shows the occupational distribu
tion of the subjects.
| Just as the population used in this study has a lower
educational level than has Taylor’s study, so there are fewer in
v the professional-technical, proprietor-managerial. Eighty-four
percent of Taylor’s adjusted sample and sixty-two percent of his
unadjusted sample fell in this category. In this study, combin
ing the first six categories, which correspond to Taylor’s first
two, sixty-three percent of the stable and forty-nine percent of
the unstable fall in the top category. Taylor’s study found
74
TABLE 6
OCCUPATIONS OF MEN AND WOMEN STABLY MARRIED
AND UNSTABLY MARRIED, BY PERCENTAGES
Occupation
Men
Stable Unstable
Women
Stable Unstable
1. Professional 13.0 6.0 7.0 6.0
2. Business 6.0 5.0 0.0 0.0 -
3. Managerial 8.0 6.0 0.0 0.0
4. Supervisory 10.0 5.0 0.0 0.0
5. Engineering 13.0 11.0 0.0 0.0
6. Technical 13.0 16.0 0.0 0.0
7. Crafts 13.0 ■ 5.0 0.0 0.0
S. Clerical 8.0 ■ 11.0 9.0 13.0
9. Sales 1.0 5.0 2.0 3.0
10. Service 15.0 21.0 6.0 0.0
11. Housewife 0.0 0.0 76.0 78.0
100.0 100.0 100.0 100.0
twenty percent more of the adjusted men in the professional-
technical and proprietor-managerial occupations than of the un
adjusted. This study found fourteen percent more of the stabiy
married men In this classification than there were unstably
married men. Seventy-six percent of the stably married women and
seventy-eight percent of the unstably married women were house
wives, with the remaining of both groups in the professional
(mostly teacher), clerical, sales, and service.
Family Income
The family income^levels of the stable and unstable mar
riages is consistent with some of the differences in occupation
and education. The responses of the subjects were divided into
five categories, which are shown in Table 7. Thirty-one percent
TABLE 7
ANNUAL FAMILY INCOME OF STABLY AND UNSTABLY
MARRIED COUPLES, BY PERCENTAGES
Annual Income
By Thousands
Stable Unstable
Under 8 18.0 29.0
8 to 10 25.0 27.0
10 to 12 26.0 25.0
12 to 15 20.0 7.0
Over 15 11.0 12.0
100.0 100.00
of the stable couples had incomes over $12,000, while nineteen
peraent of the unstable couples had incomes in this range. In
the $10,000 to $12,000 category the percentages were almost
identical, twenty-six percent for the stable marriages and
twenty-five percent for the unstable marriages. The same per
centage closeness was found for the $8,000-$10,000 category;
twenty-five percent for the stable, twenty-seven percent for the
unstable. Only eighteen percent of the stable marriage partners
!
I reported incomes under $8,000, while twenty-nine percent of the
i
partners in the unstable marriages reported an income of less
than $8,000. The Taylor and Kotlar studies found more adjusted
couples having incomes over $12,000 than did unadjusted couples
and more unadjusted couples with income under $8,000 than there
were adjusted couples in the low level income group. This pres
ent study found the same to be true of the stably and unstably
married: more stably married couples having incomes over $12,000
and fewer having incomes under $8,000 than was indicated for the
76
unstably married couples. However, the differences in this study
were not as great as in the two studies mentioned. In general,
this study is more closely matched in education, occupation, and
income than are the studies of Kotlar and Taylor.
The Methodology Used
In addition to the selection of the sample and the gather
ing of the data, we need to examine the other methodological
aspects of the study. These are the statistical procedures,
integrity of the subject’s responses, The Marital Stability Test
and its validity for this study, The Interpersonal Check List
and its validity for this study, the limitations of the study,
and a general summary.
Statistical Procedures
This research project is a study of the predictability
of marital stability from global personality characteristics.
This is predicting attributes from other attributes. Although
prediction can be made in both directions, this study Is inter
ested only in predicting stability from global personality
factors. In this study we are dealing with two groups of people--
stably married and unstably married and with sixteen possible
personality relationships. Since Chi Square Increases with in
creasing variation from the expected frequency, It is a reason
able and informative measure of the deviation of the observed
sample numbers from the expected sample numbers. It also enables
77
us to judge whether the sample ratio itself departs much or
little from the hypothetical population value. It also makes a
convenient table for viewing the distribution of cases in the
many categories. We can compute a Chi Square for the entire con-
tingency table involved in this prediction problem. However,
some predictions are better than others within the table. By
breaking Chi Square down into its components, or rather, by
examining the contributions to Chi Square from the different
categories, we obtain a more analytical picture of each one's
contribution to the prediction.
In the present study, it was decided to accept as sig
nificant a difference between frequencies of stable and unstable
groups found in a given personality combination, if there were
five chances or less in a hundred that it might be the result of
sampling. For Chi Square with one degree of freedom this is
3.8M-1. A Chi Square value of 6.635 indicates there is only one
; ■ chance in a hundred that the difference might be the result of
i
|sampling and would indicate a very significant difference; How
ever, there is always the possibility that a very small probabil
ity of an obtained difference is due to factors other than
chance. For example, a small sample may yield a Chi Square that
is not significant whereas a larger sample yielding the same
ratio might be very significant. It may also be important to
find that a hypothesis of difference is not true.
78
Integrity of the Subjects Responses
Subjects may intend to give factual and truthful
responses, hut may depart from fact because of misunderstanding
of the terms used in the questionnaire, or misunderstanding of
instructions, forgetting, or slanting toward emotional invest-
Jment. It is possible that the answers of the divorced sample
jsuffered a negative influence from both forgetting and emotional
I
jinvestment; since some had been separated from a few months to
eight or ten years. As Kotlar points out, however, Locke has
indicated that what is important is not whether the reported be
havior actually occurred but rather the meaning of the behavior
for the subject.
If there is contamination of the sample, it is most
likely to be by unstable couples being placed in the stable
sample. Some individuals find it difficult to give answers which
they know will present a picture of poor marital relations. How
ever, this was somewhat minimized by having a gap of discards
Ibetween the cutting score of the stable "low” score and the un
stable "high" score. Most of the unstable sample was made up of
divorced people, an outside criterion, not depending on any sub
jective replies to the questionnaire.
^Sally Lee Kotlar, "Middle-Class Marital Roles— Ideal
and Perceived in Relation to Adjustment in Marriage," Unpublish
ed Doctoral Dissertation, Libraries, University of Southern
California, Los Angeles (1961).
79
Leary’s Interpersonal Check List
Since the Interpersonal Check List is the instrument
chosen for this research, it is necessary to justify its selec
tion. Its first public appearance was in 195M-, and it is being
used increasingly in personality research and in the marriage
counseling profession.
The ICL is a list of 128 adjectives and adjectival
phrases, which are used to describe personality at five different
levels of behavior. It was designed by Leary, a psychologist,
for three purposes: (1) to assess personality in a psychiatric
i
setting; (2) to determine the type of psychotherapy indicated;
(3) to predict the course of treatment and its outcome. It
emerged in its present form in 19S5, the culmination of six ■
years of collaborative effort by a team of psychologists and
psychiatrists connected with the Kaiser Foundation in Oakland,
California.
The theoretic background underlying the rationale of the
ICL can be traced to cultural anthropology, social psychology,
and neopsychoanalysis--disciplines which regard personality as
socially and culturally determined. Within this socio-cultural
framework, personality is behavior in interpersonal relationships
The study of personality thus becomes the study of interpersonal
behavior with reference to its social stimulus value, that is,
with reference to its impact on others. The data of personality
is not what a given individual is or what he does, but the effect
80
of his behavior as interpreted, by those with whom he interacts.
In the tradition of Jung, Horney, Sullivan, and Fromm,
personality is regarded as embracing a continuum of behavior,
one extreme being regarded as normal (adaptive), and the other
extreme designating abnormal (maladaptive) activity. Within this
frame of reference normal behavior is not qualitatively, but
quantitatively, distinguised from abnormal behavior.
One of the first major problems which confronted Leary's
team of investigators was the complexity of personality, its
many variabled, and the task of systematizing them either on a
continuum or on a scale. Early in its work the team decided that
the existing formal systems of personality classification would
not be suitable; they were either too broadly or too narrowly
conceptualized. Therefore, the group approached the problem of
determining personality variables by studying a variety of inter
personal situations. Several scores of individuals— male and
female, neurotic, psychosomatic, normal— were brought into
interpersonal relationships in small groups. Some of the groups
were studied in a psychiatric setting, and others, small dis
cussion groups--were observed in different situations. Hundreds
of interactions of each subject were observed, recorded, and
studied. Other types of behavior were obtained from them, in
cluding descriptive statements of themselves and others. Their
dreams, fantasies, and autobiographies were recorded, and their
responses to batteries of psychological inventories and projec-
81
tlve testrs were scrutinized. Thousands of sta.tements of activ
ities and emotion were thus gathered and analyzed. As a final
result of this procedure of sifting and combining data, sixteen
generic interpersonal themes of behavior emerged. These were
systematically rela.ted in such a way that the va.ria.bles could be
I ordered along a continuum, and thus the ICL grid emerged.
These interpersonal trends all referred to a power or
affiliation factor. By designating the vertical axis as the
a.ffilia.tion-hostile dimension, it was discovered that all the six>
teen interpersonal themes could be expressed within these four
f
nodal points. Leary, in discussing the construction of the grid,
says that considerable theoretical formulation justified the
selection of the two main dimensions and the four quadrants. For
example, he notes the four nodal points fit closely the classical
humors theory of Hippocrates. The upper left quadrant (hostile
strength) equates with the choleric temperament, the lower left
: (hostile weakness) with the melancholic, the lower right (friend-
|ly weakness) with the phlegmatic, and the upper right (friendly
j
strength) with the sanguine. This same fourfold classification
appears in Freudian analysis of the individual’s dynamics in
which Freud stresses two ba.sic motives--love and hate. In his
theories of social phenomena, he emphasized group interaction as
composed of domination, power, and the interaction of the strong
versus the weak. Similarly, Ross Stagner presents the hypoth
esis that the directions of variability in human behavior can be
82
conceived, as operating along two continua: approach or with
drawal from a stimulus object, and increased or decreased organ-
ismic activity with reference to the object. Thus four direc
tional types emerge: those of aggressiveness and withdrawal on
the alienative side (corresponding to the left half of the ICL
grid) and compulsive performance and compulsive acceptance on
the affiliative side (corresponding to the right half - of the
grid).
During the five years following collection of the orig
inal interpersonal data, the instrument was subjected to contin
uous empirical study. The checklist of 128 items currently in
use is considered of sufficiently high validity and reliability
to be used in clinical settings and in personality research. It
was used in the role studies by Mangus, Luckey, Kotlar, Miller,
and Taylor. The statements comprising the current form of the
instrument were selected on the basis of frequency of the re
sponses to each item determined its empirical weight. Thus, a
built-in intensity dimension emerged, represented by a four-point
scale. Each of the interpersonal themes contains two items
having intensity number one, six items of intensity two, six
items of intensity three, and two items of intensity four.
Intensity one items, which reflect a mild or necessary amount of
the trait, are checked by ninety percent of the population.
Intensity three items reflecting a marked or inappropriate amount
of the trait are checked by sixty-seven percent of the population
83
Intensity three items reflecting a marked or inappropriate amount
of the trait are checked by thirty-three percent of the respon
dents. Intensity four items, expressing an extreme amount of the
trait, are checked by ten percent of the population.
The instrument as now developed can be used in either of
two ways. The raw scores for the variables can be profiled and
shaded on the ICL grid for clinical purposes, or those scores
can be treated with special formulae to ascertain the locus of
personality description. (The latter method will be used in the
present study.) In both treatments the variables, are paired,
using adjacent variables to yield eight octants. Justification
for combining the sixteen variables into eight octants was
empirically determined by the method of correclation coefficient.
Inter-sixteenth correlations were obtained on several samples,
and In each sample the result showed a higher correlation between
adjacent sixteenths than between those two, three, or more steps
apart. Thus, A (forceful dominance) is combined with P (respect
ed success) into a power octant, and B (self-confident.Indepen
dence) with C (competitive self-seeking) into a narcissism octant^
and so forth, around the circle.
When standard scores are desired (as in the present study
to determine the boundaries of the quadrants), the following
formulae are used:
Domlnant-Submission axis = AP-HI + .7(NO+BC-FG-JK)
Hostility-Love axis = IM-DE + .7 (NO-BC-FG+JK).
84
The standard scores thus yielded can be plotted on the
ICL grid, the center of which is determined by the intersection
of the means of the horizontal and vertical distribution of a
normative psychiatric clinic admission sample of 537 persons.
Standard scores derived from this sample thus define the limits
of normal (adaptive) and abnormal (maladaptive) behavior. Scores
from forty to sixty inclusive thus enclose the normal continuum
and scores lower than forty and higher than sixty the abnormal
continuum. In this way the ICL dichotomizes normal and abnormal
behavior.
Empirical validity of the ICL has been determined by
comparing diagnoses obtained from it with the results yielded by
the Minnesota Multiphasic Personality Inventory. The procedure
involved Is taking two hundred clinic patients' ICL diagnoses and
comparing them with their MMPI diagnoses as determined by pooled
ratings of three psychologists. Six of the most common types of
neurotic and character disorders were employed as the diagnostic
criteria. They included psychopathic, schizoid, obsessive,
phobic, hysteric, and psychosomatic labels. The results showed
a. high degree of correspondence between the diagnoses obtained
by the two different instruments.
Reliability of the ICL was determined by the method of
test-retest correlation. A somewhat homogeneous all-female
group of seventy-seven obese patients was used for this investi
gation. The average octant correlation was .78 and the average
FIGURE 1
DIAGRAM OF ICL BY OCTANTS AND BY QUADRANTS
DOM
SUB
AP— Managerial-Autocratic
BC--Competitive-Exploitive
DE— Blunt-Aggressive
FG— Skeptical-Distrustful
HI— Modest--Self-Effacing
JK--Docile-Dependent
LM--Cooperative-Overconventional
NO— Responsible-Overgenerous
DOM--High score on DOM-SUB axis
SUB--Low score on DOM-SUB axis
LOV— High score on LOV axis
HOS— Low score on LOV axis
LD— Friendly Dominance
HD--Hostile Dominance
HS--Hostile Submission
LS— Friendly Submission
86
sixteenth correlation was ,73. These correlations are quite high
in view of the fact that the self-concept is subject to changes
in mood and perspective. This, therefore, is an excellent in
strument for the present study for the following reasons:
1. It is validated, and its reliability has
been established.
2. It is decriptive of behavior which determines
personality types.
3. The personality type can be given a derived
standard score for purposes of statistical
analysis and inference.
i + . The combined standard scores of DOM and LOV
can be used to determine the global person
ality type in four major categories, falling
in four quadrants.
5. The standard scores can be used both in find
ing correlations between marital stability
ratings and personality scores and in testing
for significant differences between the means
of two populations.
6. The location In personality quadrants permits
the use of Chi Square to determine whether
there is a significant difference between
complementarity and non-complementarity In
relation to marital stability.
7. It had already empirically established normal
and abnormal scores so that examination can
be made of complementarity in both normal and
abnormal divergence, and of abnormal paired
with normal scores in spouses (if they appear).
! The Marital Stability Test
In order to reduce contamination in the stable sample as
much as possible, a short discriminating test was needed. The
available tests of marital adjustment and of marital stability
were in themselves not indicative of particular stress in the
marriage. It was felt that a test was needed that would also
indicate areas of conflict which were typical of unstable mar
riages and not characteristic of stable marriages. With this in
mind the researcher went through the literature on marriage and
selected all statements that pointed out causes of stress and
instability or that made claims of being significant for marital
lhappiness and success. Then the researcher informally inter
viewed two psychiatrists having an extensive marriage counseling
practice, three psychologists, five marriage counselors, and two
lawyers in an attempt to ascertain the importance and discrimin
ating ability of the various statements. In this way sixty-
three statements were reduced to thirty-six. A questionnaire
with these thirty-six statements was administered to over one
hundred divorced men and women at a meeting and to about the
same number in a large church gathering, while they were waiting
88
to eat dinner. The purpose of the questionnaire was explained
to each gathering. This questionnaire was in the form of a true
and false inventory, with the instructions used in the MMPI.
"This inventory consists of numbered statements. Read each
j statement and decide whether it is more nearly true as applied
I
|to you or more nearly false as applied to you." The words, "more
|nearly," were added in the belief that it would make the forced
i
|choice easier. The subjects were told in the explanation period:
"Do not leave any unanswered, please!" The two groups were then
balanced in number for easier computation by randomly eliminat
ing some of the subjects from the divorced group, since it was
larger. Before selecting the final items to appear in the
Marital Adjustment Test, it was decided that only those items
would be retained which were checked TRUE five times oftener by
i
i
jthe divorced people than by the still-married people at the
church dinner. In actuality, the items finally used in the
Marital Adjustment Test were marked TRUE about ten times oftener
by the divorced people than by the still-married people. There
were sixteen of these items. Question number three was retained,
because it had preceded question number four in the pilot study.
It was felt that it influenced the answer to four. To keep
question four this set-producing question was retained. However,
this question number three is not scored.
The seventeen-item inventory was then used at the
churches, bowling alleys, and singles groups. Instead of a
i
89
forced TRUE or FALSE answer, the subjects could check one of
three answers which they considered to be "the most nearly
accurate." The three choices were: (1) never, (2) now and then,
(3) frequently.
i
i In order to establish a . cutting score that would leave a
l
Ipurer sample of stably married subjects, the questionnaire was
|
iadministered to people coming to a marriage counseling center and
]
i
jwhose marriage the director considered to be critica.lly unstable.
There were thirty couples, or sixty subjects. In addition, the
inventory was filled out by thirty divorced people at a PWP
(Parents Without Partner) seminar, with instructions to answer
it in a . way that would have been true prior to the divorce. The
scores on the Inventory can range from 16 to 80. No one divorced
or coming for counseling scored lower than thirty-six. Few in
[the general population scored over thirty. Thirty-two was
accepted as the high cutting score for the stable sample. Those
;wlth scores higher than thirty-eight were Included in the un
stable sample. There is good reason to believe tha.t this test
does help to eliminate stable marriages from the unstable sample.
There is less confidence that some of the "stable" may not be ■
"unstable." However, they make up a minor part of the unstable
sample.
A sample of the Marital Stability Test is in the
Appendix. Answers are scored 1, 3, or 5. Item three Is not
scored, but is used to establish a set for item number four.
90
Items two and fifteen are scored opposite to the other items in
the test.
Summary
A final sample of 280 couples was selected from several
^sources. Forty-six men and forty-one women in the stable sample
iwere selected from church groups. Twenty-four men and twenty-
inine women came from church-sponsored bowling leagues. In the
i
unstable sample thirteen men and twenty-one women were selected
from the church clubs by high scores on the Marital Stability
jTest. Five men and three women came from the church-sponsored
i
bowling leagues through high scores on this test. Forty-six
divorced women from singles clubs completed the sample of seventy
unstable marriages reported by women. Twenty-six divorced men
were gathered from the same source. The remaining twenty-nine
men needed for the sample of seventy unstable marriages reported
by men were from other gatherings of single people. Where the
Interviewer had them fill out the questionnaires singly and in
i
very small groups.
The instrument used to measure the personality variables
was the Interpersonal Check List. A Marital Stability Test
was used to select the stable sample and part of the unstable
sample.
Chi Square was used to test all the hypotheses. Two by
four tables were used to test for differences in each major
category between the stably and the unstably married. For each
91
of the sixteen personality dyads a two-cell Chi Square was used,
The .05 level was accepted as significant.
Distribution of the sample is found in Table 8.
TABLE 8
DISTRIBUTION OF STABLE AND UNSTABLE MARRIAGES
LD LD LD LD HD HD HD HD HS HS HS HS LS LS LS LS
LD HD HS LS LD HD HS LS LD HD HS LS LD HD HS LS
Total
s 24 13 4 17 26 21 3 18 3 0 0 2 7 0 2 0 140
u 8 21 8 2 28 28 13 9 4 2 3 3 2 6 2 1 140
32 34 12 19 54 49 16 27 7 2 3 5 9 6 4 1 280
92
I
1
! CHAPTER IV
I
I THE FINDINGS OF THE STUDY
Hypothesis 1
When husband and wife both have the same global person
ality characteristics and, therefore, the same personality needs,
jthe marriage will tend to be characterized as unstable.
!
Hypothesis 1--Opera.tiona.lly
When the LOV score and the DOM score of both spouses
border on the same quadrant of the ICL, indicating like global
i
personality needs by intersecting in the same quadrant, the
couple will tend to be found in the group of spouses who are
leither divorced or whose Marital Stability Score exceeds thirty-
!
jeight. This is called an homoga.mous relationship.
The Null Hypothesis
When both spouses fall in the same ICL quadrant, indicat-
ing like personality needs, they are more likely to be found
in the stably married group than in the unstably married group.
This is a . test of homogamous relationships, where the
need scores of both spousesT DOM and LOV scores intersect and
93
place them in the sa.me personality quadrant. There are four wa.ys
in which they can be alike. Both ean be Affilia.tive-Domina.nt
(LD), Hostile-Dominant(HD), Hostile-Submissive (HS), or Friendly-
Submissive (LS) .
| Some literature on marriage sta.tes that cultura.l and
; social values are optimally homogeneous, while personality needs
I are almost necessarily complementary. Many researchers found
|
‘ results which contradicted this theory. This study examines the
possibility of predicting marital stability from three basic
dyadic combinations of global personality needs: homogeneous,
i
jcomplementary, and opposite. The homogeneous personality combin
ation is regarded by Winch as being inimical to marital stability.
The test of the null hypothesis was first done with a .
2 x 1 Chi Square. The distribution of couples under the four
jcombinations: LD/LD, HD/HD, HS/HS, and LS/LS is shown in Table 9 .
As can be seen, very few couples seem to combine in HS/HS or
LS/LS. It may be that two submissive people find it difficult to
bake the necessary step to get married to each other. The Chi
Square for this test of no difference with three degrees of
freedom is 8.56 (Chi Square = 8.56). This is significant
beyond the .05 level of confidence. We, therefore, reject the
null hypothesis. To find where and in what direction the differ
ence is, we divide the homogamous couples into the four ca.te-
^ories in two-cell squares.
94-
TABLE 9
DISTRIBUTION OF HOMOGAMOUS COUPLES
BY TYPE AND BY MARITAL STABILITY
LD/LD HD/HD HS/HS LS/LS . Totals
Stable 24- 21 0 0 4 - 5
Unstable 8 28 3 1 4-0
Totals 32 4-9 3 1 , 85
Chi Square = 8.56 df = 3
1 3 V
1 1 I I
v/\
.05
.02
Examining the total of eighty-five couples whose mar
riage is homogamous in both the love and in the dominance cate
gories, we find forty-five in the stable group and forty in the
unstable. From this we might have taken for granted that the
jnull hypothesis would have to be accepted. However, there are
[different kinds of homogamy.
I
Testing by dyads where both spouses are Dominant-
Affiliative, or Friendly-Dominant, with a two cell Chi Square,
we get a Chi Square of 7.03 which, with one degree of freedom,
is significant beyond the .01 level of confidence. This differ
ence is, however, in the opposite direction from the prediction
that homogamy would be a characteristic of the unstable marriages.
9.5
The Hostile-Dominant quadrant contained forty-nine
couples. Twenty-one of these were in the stably married group
and twenty-eight of them were in the unstably married or divorced
group. Although the Chi Square indicates a tendency toward plac
ing these in the unstable group, the difference is quite insign-
nificant. To be significantly different at the .05 level with
one degree of freedom, Chi Square of 3.891 would be needed. The
Chi Square for these two cells is only .73. We, therefore,
accept the null hypothesis for this combination, and state that
knowing that both spouses fall in the Hostile-Dominant quadrant
does not help us to predict whether the marriage would be stable
or unstable.
Too few couples were found in the Hostile-Submissive dyad
to draw any meaningful conclusion pertaining to marital stability.
A Chi Square of 1.33 was obtained, where a 3.891 is needed to
indicate a significant difference upon which one would care to
base predictions. We can, however, accept the indication that
very few Hostile-Submissive people tend to marry each other.
The same must be said about the Friendly-Submissive.
This study yielded only one such marriage. It was in the un
stable group. This was a marriage terminated by divorce.
Although the tests of three of the homogamous dyads are
in the direction of the hypothesis that a marriage where both
spouses have the same global personality would tend to be un
stable, the only significant test was in the opposite direction.
96
The Friendly-Dominant dyad was very significantly indicative of
a stable marriage. It is evident, however, that we cannot base
any predictions about marital stability on a factor called homo
geneous global personality. We have to ask, for one thing, in
what way the spouses’ personalities are alike. Although the
j
jprofiles did not indicate it, it may be that there are different
kinds of LDTs . , HD's, etc. For instance, there may be a Hostile-
Domineering and a Non-Affiliative-Assertive.
On the basis of the findings on marriages in which the
spouses tend to be alike in global personality characteristics
(homogamous), we would predict with considerable confidence that
when both spouses are affiliative and dominant (assertive ?) , the
marriage would tend to be stable. If both are hostile and dom
inant there is slightly more chance that the marriage will be un
stable than that it will be stable. However, another study might
by chance sampling show slightly in the opposite direction.
lAbout Hostile-Submissive and Friendly-Submissive, we can only
say that such people seem not to marry one another, but that
there is also little indication that they can expect a stable
marriage, if they do marry.
Hypothesis 2
■ When the personality needs of a marriage couple are comp
lementary, the marriage will tend to be stable.
97
Hypothesis 2 (A)--Operationally
When the intersection of the DOM and LOV scores of one
spouse falls in the quadrant called LD (Affectionate-Dominant)
and the other spouse's DOM-LOV intersection falls in the quadrant
LS (Affectionate-Submissiveness), the couple will tend to be in
the group expressing satisfaction in being together by scoring
less than thirty-two on the Marital Stability Test, This is
called a complementary relationship in Dominance and a hetero
geneous relationship in Affiliativeness.
Hypothesis 2 (B)--Operationally
When the intersection of the DOM and LOV scores of one
spouse falls in the quadrant HD (Hostile-Dominance) and the other
spouse's intersection falls in the quadrant HS (Hostile-
Submissiveness), the couple will tend to be in the group scoring
less than thirty-two on the Marital Stability Test. This is a-
complementary relationship in Dominance and an homogamous
relationship in Affiliativeness.
The Null Hypothesis 2 (A)
When both spouses are affiliative, and one is Dominant
while the other is Submissive, the marriage is no more likely to
be classified as stable than it is likely to be classified as un
stable.
The Null Hypothesis 2 f B ~ )
When both spouses are Hostile, and one is Dominant while
the other is Submissive, the marriage is no more likely to be
classified as stable than it is likely to be classified as un
stable.
For the purpose of this study complementarity has been
defined as a combination of dominance and submissiveness, with
the affiliation homogeneous. It has been stated that we cannot
speak of true complementarity when the affiliation aspect of the
personalities is in opposition. Complementary global personal
ities would, for one, be those in which either the husband or the
wife is Affiliative-Dominant and the other is Affiliative-
Submissive and, for the other, where one spouse is Hostile-
Dominant and the other is Hostile-Submissive. In the identifi
cation symbols the husband’s personality quadrant is indicated by
the first combination of letters and the wife’s is indicated by
the second. LD/LS, for example, indicates that the husband is
Affiliative-Dominant and the wife is Affiliative-Submissive.
Winch and others have indicated that it doesn’t matter in a
complementary relationship who is on top. This study does not
take that for granted.' In addition to the Chi Square for the
combined complementarity, a 2 x 4 - cell table, a Chi Square has
been ascertained for each of the four possible combinations:
LD/LS, HD/HS, HS/HD, and LS/LD, and again combining LD/LS with
LS/LD and HD/HS with HS/HD. Table 10 shows the distribution of
99
complementary personalities in the stable and unstable groups.
TABLE 10
DISTRIBUTION OF COUPLES WHO ARE HOMAGAMOUS IN
L OR H AND WHO ARE COMPLEMENTARY
D/S OR S/D, BY TYPE AND
MARITAL STABILITY
LD/LS HD/HS HS/HD . LS/LD Totals
Stable 17 3 o: 7 27
Unstable 2 13 3 2 20
Totals 19 16 3 9 H7
Chi Square = 16 .79; df = 3; 11.3 Hi = p = < . 01
Testing the null hypothesis by means of a 2 x H Chi
Square that there is no difference in complementarity between
couples In stable marriages and couples in unstable marriages
produces a Chi Square of 16.79. With three degrees of freedom
a Chi Square of 11.3HI Is significant at the 0.01 level. If we
combine the two affiliative complementarities with each other and
the two hostile complementarities with each other, we get a Chi
Square of 18.; 82. A four-cell Chi Square has one degree of free
dom. 6.635 indicates a significant difference at the .01 level
of confidence. With almost one hundred percent confidence we
reject the hypothesis of no difference. To find which combi-
100
nations are predictive of significant differences in stability,
we again resort to the two-cell Ghi Square for each possible
combination. There are six possible combinations to test with
the Chi Square. .Ignoring who is on top, the LD/LS can be com
bined with the LS/LD, and the HD/HS can be combined with the
HS/HD. Then we separate into cells with the husband LD and the
wife LS, the husband HD and the wife HS, the husband HS and the
wife HD, the wife LD and the husband HS.
When we ignore who is on top, the prediction of stability
as related to complementarity can be stated in two operational
hypotheses. In this way we can determine the direction of the
difference indicated in Chi Square. The hypotheses to be tested
will then indicate whether the difference is in one direction
only, in two directions, or whether a big difference on one
direction has been reduced by the other relationship.
With LOV, or affiliativeness, or friendliness, as the
homogeneous personality characteristic and DOM, dominance-
submissiveness, as the complementary global personality character
istic, we find that this combination is very significantly favor
able to marital stability. It yields a Chi Square of 12.89,
where a Chi Square of 6.635 Is significant at the .01 level of
confidence with one degree of freedom. This indicates that in a
marriage in which both are affiliative and one is dominant while
the other is submissive, the marriage will tend to be character
ized as stable. This coincides with the prediction of the
101
hypothesis.
With HOS as the global personality characteristic shared
by both spouses and dominance and submissiveness as the comple
mentary relationship, we find a difference in frequency between
stable and unstable marriages. A Chi Square of 6.72 goes beyond
the .01 level of confidence, which is 6.635 for a two cell table
with one degree of freedom. On the basis of this figure, we con
clude that where both spouses are Hostile and one is Dominant
while the other is Submissive, the marriage is more likely to be
characterized as unstable than as stable. This is very signifi
cantly in the direction opposite from the prediction of the
hypothesis.
The above statistics furnish very strong evidence that
there is more probability of a stable marriage than of an unstaHe
marriage when both spouses are alike in affiliativeness and are
complementary in the dominanoe-submissiveness factor. They also
furnish strong evidence that there is more probability:! of an
unstable marriage than of a stable marriage if both spouses are
characterized as hostile and one is dominant and the other is
submissive. However, there is no way to make any inferences
about who is in the dominant and who is in the submissive posi
tion. This may make a difference. To find this we use a two-cell
Chi Square for each possible combination. This means finding four
Chi Squares: both affiliative with the husband dominant and the
wife submissive (LD/LS) , both affiliative with the wife dominant
102
and the husband submissive (LS/LD) , both hostile with the hus
band dominant and the wife submissive (HD/HS), both hostile with
the wife dominant and the husband submissive (HS/HD).
When both are affiliative with the husband dominant and
the wife submissive (LD/LS), the Chi Square of 10.32 indicates a
very high probability that the marriage will be a stable one.
When both are affiliative with the wife dominant and the husband
submissive (LS/LD), the distribution is in the direction of sta
bility. The Chi Square, however, falls far short of the neces
sary 3.84-1 needed to be significant at the .05 level of confi
dence. It is possible that a larger N would have yielded a' sig
nificant Chi Square, but we cannot make any confident prediction
either of stability or of instability.
When both are hostile with the husband dominant and the
wife submissive (HD/HS), the resultant Chi Square of 5.06 is sig
nificant between the .05 and the .01 level and is predictive of
an unstable marriage. Again, the reverse position, where the
wife is dominant and the husband is submissive (HS/HD), is-in
the direction of instability, but the Chi Square of 1.78 falls
short of the 3.84-1 needed to be significant at the .05 level of
confidence.
We find, then, that with affiliativeness homogeneous, a
marriage will tend to be characterized as stable when it is
complementary In dominance and submissiveness. When n o n-
affiliativeness, or hostility, Is the shared personality charac-
103 '
teristic, the marriage will tend to be characterized as unstable,
even though dominance and submissiveness are complementary. When
the distribution is further separated to show who is on top, the
marriage will strongly tend to be characterized as stable when
the husband is affiliative and dominant while the wife is affil
iative and submissive (LD/LS). This cannot be said when the re
lationship is reversed LS/LD), although the indication of the
data is in that direction. When the wife is dominant and the
husband is submissive, the prediction of stability becomes very
uncertain.
When the husband is hostile and dominant and the wife is
hostile and submissive (HD/HS), the marriage will tend to be
characterized as unstable. Again, there is insufficient data: to
make any prediction when the order is reversed. A larger N might
be very significantly predictive of instability, if the differ
ence trend remained the same as in the present sample. Hostility
emerges very strongly as a factor that is disruptive of the mar
riage relationship. The theory was entertained by this research
er that two non-affiliative people might.have a stable relation
ship by "uniting against the world." This possibility is not ex
cluded by the evidence of the data, but it is strongly contra
indicated. Hostility, or non-affiliativeness can be seen only
as a highly negative factor in the marital relationship.
The next step is to ascertain the predictability of
marital stability from unions between opposites. These are de
fined in this study as, marriages in which the spouses are oppo
sites in the LOV category, i.e., where one is friendly or affil
iative, and the other is hostile or non-affillative„ This cate
gory must, then, also be divided into two main subcategories.
In one the dominance-submission needs of the spouses are comple
mentary, in the other they are homogeneous.
Hypothesis 3
Even though the couple’s Dominance-Submission needs are
complementary, a marriage will tend to be unstable, when one
spouse is affiliative and the other is non-affillative.
Hypothesis 3 (A)--Operationally
When the intersection of the DOM and LOV scores of one
spouse falls in the quadrant LD (affectionate dominance) and the
' other’s falls in the quadrant HS (hostile submissiveness), the
couple will tend to be in the group of spouses who are either
divorced or whose Marital Stability Score exceeds thirty-eight.
This is a relationship between people who are opposites in
Affiliativeness and complementary in Dominance.
Hypothesis 3 (B)--Operationally
When the intersection of the DOM and LOV scores of one
spouse falls in the quadrant HD (hostile dominance) and the scores
of the other spouse intersect in the quadrant LS (affectionate
submissiveness), the couple will tend to be in the group of
spouses who are either divorced or whose Marital Stability Score
105
is higher than thirty-eight. This is a relationship between
people who are opposites in Affiliativeness and complementary in
Dominance.
The first test is the test of the combined dyads, the
2x4- cell table. We are again testing the general hypothesis,
i. e., singling out the four possible combinations. The distri
bution of couples in this category is shown in Table 11.
TABLE 11
DISTRIBUTION OF STABLE AND UNSTABLE COUPLES IN
WHICH ONE SPOUSE IS AFFILIATIVE (FRIENDLY)
AND THE OTHER IS NON-AFFILIATIVE (HOSTILE)
AND WHERE DOMINANCE AND SUBMISSIVENESS
ARE COMPLEMENTARY
LDAlS HS/LD HD/LS LS/HD Totals
Stable 4 18 3 0 25
Unstable 8 9 4 6 27
Totals 12 27 7 6 52
Chi Square = 7 . 47 ; df = 3
I I
V
■
05; 7.815 = p = ..05
For this 2 x 4 table in which fifty-two of two hundred
and eighty couples fall, a Chi Square falls just short of a sig
nificant difference between stable marriages and unstable mar
riages, 7.47, where 7.815 indicates a difference which is sig
nificant at the .05 level of confidence.
106
The first personality quadrant is that of the husband,
and the second is that of the wife. When we combine the LD/HS
with the HS/LD, and combine the HD/LS with the LS/HD, the Chi
Square almost completely disappears to a Chi Square of .88, with
one degree of freedom. This indicates that in some of these, in
reversing the personality relationship, there is a change in the
direction of prediction. However, it is not significant.
When the role positions are identified through the two
cell table for Chi Square, three Chi Squares are not indicative
of any significant difference between the stable and the unstable
groups. One is significant at the .05 level of confidence. When
the husband is affiliative-dominant and the wife is hostile-
submissive, a very small Chi Square of .75 is in the direction of
instability. When the husband is hostile-submissive and the wife
is affiliative-dominant, the Chi Square is zero. When the hus
band is hostile-dominant and the wife is affiliative-submissive,
a Chi Square of 2.39, though short of the 3.851 necessary for a
.05 level of confidence with one degree of freedom, is in the
direction of predicting a stable relationship. With the husband
affiliative-submissive and the wife hostile-dominant, a Chi
Square of H.17 indicates a significant difference in the direction
of instability. This is the only significant difference in this
category of spouses which were alike on the LOV scale and comple
mentary on the DOM scale.
The last major category is the marriage in which the *
107
spouses have opposite global LOV personality characteristics and
the same, or homogeneous, global personality characteristics on
the DOM axis. One spouse is affiliative-dominant, while the
other is hostile-dominant; or one spouse is affiliative-
submissive, while the other is hostile-submissive.
Hypothesis M -
When the Dominance-Submission personality needs of a
couple are the same and their affiliative needs are opposite, the
marriage will tend to be unstable. This is a marriage in which
both the necessary homogeneity and need complementarity are lack
ing.
Hypothesis (A)— Operationally
When the intersection of the DOM and LOV scores of one
spouse falls in the quadrant HD (hostile-dominance) and the
scores of the other spouse intersect in the quadrant LD
(affectionate-dominance), the couple will tend to be in the group
of spouses who are either divorced or whose Marital Stability
Score is higher than thirty-eight. This is a marriage in which
both homogeneity in Affiliative need and the complementarity be
tween Dominance and Submissiveness are lacking.
Hypothesis H (B)— Operationally
When the intersection of the DOM and LOV scores of one
spouse falls in the quadrant HS (hostile-submissiveness) and the
scores of the other spouse intersect in the quadrant LS
108
(affectionate-submissiveness), the couple will tend to be in the
group of spouses who are either divorced or whose Marital Stabil
ity Score is higher than thirty-eight. This, too, Is a marriage
in which both homogeneity In Affiliativeness and complementarity
between Dominance and Submissiveness are lacking.
The first' test is again the test of the combined dyads,
the 2 x 4 cell table. This Is a test of the undifferentiated
general hypothesis. This distribution of couples is shown in
Table 12.
TABLE 12
DISTRIBUTION OF STABLE AND UNSTABLE COUPLES IN
WHICH ONE SPOUSE IS AFFILIATIVE (FRIENDLY)
AND THE OTHER IS NON-AFFILIATIVE (HOSTILE)
AND WHO ARE EITHER BOTH DOMINANT
OR BOTH ARE SUBMISSIVE
LD/HD HD AD LSAS HSAS Totals
Stable 13 26 2 2 43
Unstable 21 28 2 3 54
Totals 34 54 4 5 97
Chi Square = 1.18; df = 3; p = > .05; 7.815 = p = . 05
More than one third of the two samples fell in this cate
gory. There are four major divisions. Although there are more
unstable couples in this category than there are stable couples,
the difference Is not significant. The 2 x 4 cells produce a
109
Chi Square of only 1.18, where a Chi Square of 7.815 is needed to
be significant with three degrees of freedom.
Combining LD/HD with HD/LD yields a Chi Square of .92.
Combining HS/LS with LS/HS yields a Chi Square of .00. Therefore,
when one spouse is affiliative-dominant and the other is hostile-
dominant, no prediction regarding the likelihood of stability or
instability can be made. It is just as likely to go one way as
the other. The same is true, when one spouse is hostile-
submissive and the other is affiliative-submissive.
When the husband is affiliative-dominant and the wife is
hostile-dominant, the direction of the difference is in the direc
tion of an unstable marriage, but the Chi Square of 1.4-1+ with one
degree of freedom falls short of the 3.84-1 needed to be signifi
cant at the .05 level of confidence. When the husband is hostile-
dominant, the incidence of stable and of unstable marriages is
almost equal. The Chi Square is zero. The Chi Square is also
zero for those marriages in which the husband is hostile-
submissive, and the wife is affiliative-submissive. Again, when
the husband is affiliative-submissive and the wife is hostile-
submissive j the Chi Square is zero. The LD/HD and HD/LD findings
are based on fairly large N’s, while the zeros of the HS/LS and
the LS/HS are on the strength of only very small NTs.
Summary
There are four different major combinations of personal
ity needs possible between married couples, when they are class-
110
ified according to LOV and DOM on the ICL. These are: the pair
may be homogamous both in affiliative (LOV) and in dominance (DOM).
They may be homogamous in LOV needs and complementary in DOM
needs. They may be opposites in LOV and complementary in DOM.
They may be opposites in LOV and alike in DOfcl needs. They can
also be opposite in both LOV and DOM.
There are sixteen different personality combinations
possible for spouses when classified by the ICL quadrant accord
ing to LOV/DOM needs.
Hypothesis 1 predicted that when a married couple are
alike both in DOM and in LOV, the marriage would tend to be
characterized as unstable. For a marriage that is homogamous
in both respects, there are four different ways in which the
homogeneity may appear. Both can be affiliative-dominant,
affiliative-submissive, hostile-dominant, or hostile-submissive.
Only the first yielded a significant difference, between stable
and unstable, and It was in the direction opposite to the hypo
thesis. The hypothesis Is rejected for the hostile-dominant, the
hostile-submissive and the affiliative-submissive. The opposite
subhypothesis Is accepted for the affiliative-dominant: when
husband and wife are both affiliative and dominant, the marriage
will, tend to be characterized as stable. This is in line with
KotlarTs findings. Her composite adjusted couple was LD/LD.
This was true both for the husband's report and for the wife's.
Luckey also found this combination characteristic of the adjusted
Ill
couples.
Hypothesis 2 predicted that when the personality needs of
a couple were complementary on the DOM axis and homogeneous on
the LOV axis, the marriage would tend to be stable. Again, there
are two kinds of LOV homogeniety: affiliative and hostile. When
both spouses are affiliative and one is dominant while the other
is submissive the findings favor the hypothesis very significant
ly with a Chi Square of 12.89. When this is broken down to the
husbandTs being dominant and the wife submissive the Chi Square
is 10.32. With the wife dominant and the husband submissive the
Chi Square is only 1.78. A larger N, holding the same ratio
would be significant, but there is no assurance that the ratio
would stay the same.
When both spouses are hostile, and one is dominant while
the other is submissive, the Chi Square of 6.72 is very signifi
cantly in the direction of instability, the opposite direction
from the hypothesis. With the husband in the dominant role and
the wife in the submissive .role, the prediction is still sig
nificantly toward an unstable marriage. There were only two mar
riages in which the wife was dominant and the husband was sub
missive. Both were divorced.
On the basis of the above evidence we need to modify
hypothesis 2, separating the affiliative and the hostile and mak
ing two statements in place of the one. We conclude then that:
(1) If both husband and wife are affiliative and one spouse is
112
dominant while the other is submissive the marriage will tend to
be stable. ( 2) When both husband and wife are hostile and one is
dominant while the other is submissive the marriage will tend to
be unstable. Kotlar and Taylor found that a marriage of any HS
was unstable.
Hypothesis 3
Even though the couple’s dominance-submissive needs
are complementary, a marriage will tend to be unstable.when one
spouse is affiliative and the other is non-affiliative. For
either spouse being affiliative-dominant and the other being
hostile-submissive, there was no significant difference between
stable and unstable marriages. When the husband is affiliative-
submissive and the wife is hostile-dominant, the marriages sig
nificantly tended toward unstable. When the husband is hostile-
dominant while the wife is affiliative-submissive there were
twice as many in the stable marriages as from the unstable mar
riage (18/9), but the difference fell short of being significant.
When role position is ignored, the difference cancels out.
Hypothesis M -
When the dominance-submission needs of a couple are the
same and their affiliative needs are opposite, the marriage will
tend to be unstable. This is a marriage where neither the homo
geneous needs nor the complementary needs are met.
Although there were fifty-four unstable marriages and
113
forty-three stable marriages in this classification, none of the
Chi Squares reached the necessary .05 level of significance. No
predictions regarding stability or instability would be better
than chance.
114-
! CHAPTER 'V
!
; EMERGENT FINDINGS
i -
l
]
| This resea.rch began with the proposition that in the main
couples who were homogeneous in LOV and complementary in DOM
would have the most stable marriages. It was found that this
i
proposition was accepted when the LOV factor wa.s "affiliative"
for both and that the opposite was accepted when the LOV fa.ctor
was "hostile" for both. However, it was also found that when
both pa,rtners were affiliative and both also were dominant, the
probability for marital stability would be high. When both were
hostile and dominant, there was no significant difference in pre
dicting whether the marriage would be stable or unstable,
j By- separating the LOV and DOM factors, the distribution
of combina.tions is a.s shown in Table 13. In the four possible
LOV combinations ea.ch is well represented. The most frequent
combination is that of a . hostile husband and affiliative wife.
However, the greatest discrepancy is between this most frequent
combination and the combination of a hostile wife and an affili
ative husband. This difference is not significant, so it would
not be significantly different from the other two.
115-
TABLE 13
SUMMARY CHART OF FINDINGS
With one df. Chi Square = 3.891 = p = .05
With one df. Chi Square = 6.635 = p = .01
Significantly toward stability:
LD/LD: N = 32; Chi Square = 7.03
LD/LS: N = 19; Chi Square = 10.32
Significantly toward instability:
HD/HS: N = 16; Chi Square = 5.06
LS/HD: N = 6; Chi Square = 9.17
In direction of stability— not significantly:
HS/LD: N
=
27; Chi Square = 2.39
LS/LD: N =
9; Chi Square = 1.78
l of instability---not significantly:
LD/liD: N
=
3M-; Chi Square = 1.99
HS/HS: N =
3; Chi Square = 1.33
LD/HS: N = 12; Chi Square = .75
HD/HD: N = 99; Chi Square = .73
HS/HD: N = 2; Chi Square = .50
No direction:
LS/LS: N = 1; Chi Square = .00
LS/HS: N = 4 - ; Chi Square = .00
HD/LS: N = 7 - ; Chi Square = .00
HS/LS: N = 5; Chi Square = .00
HD/LD: N = 59-; Chi Square * = .00
116
When both partners a,re characterized by the ICL- as being
affiliative (column 1, Table 13) , without regard to DOM, the Chi
Square of 18.95 is very significantly indicative of marital in
stability. When the husband is affiliative and the wife is
hostile (column 3, Table 13), the Chi Square of 5.16 is signifi
cantly in the direction of unstable marriage. When the husband
is hostile and the wife is affiliative the Chi Square of .17 is
Insignificant. There seems to be a . fifty percent chance in
either direction.
TABLE lb
DISTRIBUTION OF COUPLES, STABLE AND UNSTABLE
WITH ONLY LOV AS A FACTOR,
BY FREQUENCY
L/L H/H L/H H/L Totals
Stable bs 2 b 19 b9 lbO
Unstable 13 b6 37 bb lbO
Totals 61 60 56 93 280
In the four possible DOM combinations (Table 1M-) , that
ma.rriages in which both spouses are dominants, number 169, both
submissive 13, husband dominant and wife submissive 7b, and wife
dominant and husband submissive 2b. In over half of the mar
riages, then, both spouses are dominant. In only thirteen
117
ma.rrla.ges, or four percent, were both submissive. In twenty-six
percent of the marriages the husband was dominant with the wife
submissive, and in just short of nine percent the wife was dom
inant and the husband wa.s submissive. Combining the homogamous
with each other and the complementary we find that in this study
|182 of the 200, or sixty-five percent were homogamous in DOM and
i
100, or thirty-five percent were complementary. The Chi Square
for this difference is 2M-.6. A Chi Square of 6.635 would be very
significant in the opposite direction of Winch1s theory of comp
lementarity.
|
TABLE 15
DISTRIBUTION OF STABLE AND UNSTABLE COUPLES
WITH ONLY DOM AS A FACTOR, BY FREQUENCY
D/D S/S D/S S/D Totals
Stable 84 9 - A2 10 19-0
Unstable 85 9 32 19- 19-0
Totals 16 9 13 79- 29- 280
118
CHAPTER VI
I
! ' DISCUSSION OF THE FINDINGS
|
| Although the findings of this study were not always in
the direction of the hypotheses, they should prove of va.lue to
the marriage counselor. There is very strong evidence that some
combina.tions of personality characteristics are more predictive
of marital stability than others. There is also strong evidence
that some combina.tions of personality characteristics are very
inimical to marital stability.
On the basis of this study the combination most predic
tive of marital stability is tha.t in which both spouses are
laffiliative and in which the husband Is dominant and the wife is
jsubmissive (LD/LS). This looks like strong evidence for comple-
i
mentarity in DOM as most conducive to marital stability, as the
main hypothesis predicted. However, when both spouses fell in
the affIllative-dominant category, this wa.s also very significant
ly predictive of marital stability. The odds in favor of a .
stable marriage are, the, very high for both LD/LS and for
LD/LD.
Two other combina.tions were in the direction of being
119
predictive of marital stability. Small N's may have been the
only reason LS/LD and HS/LD were not significant. Seven LS/LD
marriages were in the stable group and nine in the unstable. If
these N’s were doubled and the ratios remained the same they
would be significantly predictive of marital stability.
| Two global personality combinations were significantly
I
indicative of marital instability. These were the HD/HS and the
LS/HD with Chi Squares of 5.06 and 1.17 respectively. The chance
of X =5.06 being different due to sampling is greater than .02
and less than .05 as Is also 1.17. In the HS/HD there were only
three marriages, and these were all unstable. The HD/HD were
plentiful, twenty-one stable, and twenty-eight unstable, and
the Chi Square of .73 did not indicate a significant difference.
The HS/HS Chi Square was 1.33, but again there were only three
such marriages In the sample and all were in the unstable group.
|The LD/I-IS had four stable and eight unstable couples with Chi
'Square equals .75 (greater than thirty and less than fifty).
! lD/HD wa.s also a nonv-significant difference with Chi Square
i
equals 1.11 (less than .30 and greater than .20).
There were five personality need combinations for which
the Chi Square was zero. Four of these had very low NTs. These
were: LS/LS (S = 0, U = 1) ; HD/LS (S = 3, U = 1) ; HS/LS (S = 2,
U = 3); LS/HS (S = 2, U = 2). For the HD/LD there were twenty-
six in the stable group and twenty-eight in the unstable group.
In all of these there was just about a , fifty percent chance
120
either way. As can be seen by the figures, however, only one of
these combinations occurs with any frequency in our study.
From the emergent findings, however, it seems to be that
the deciding factor is the LOV scale. There was no significant
difference between stable and unstable in any of the DOM combi-
i
.nations. If the LOV factor is eliminated no prediction better
i
i
than chance could be made a.s to whether the marriage would tend
to be unstable or stable. However, when both partners are
affiliative on the ICL there is a very high probability that the
marriage will be characterized as stable. When both the husband
and wife are chara.cterized on the ICL as being hostile, and when
the husband is affiliative and the wife is hostile the marriage
will tend to be chara.cterized as unstable. When the husband is
hostile and the wife is a.ffilia.tive there seems to be a fifty
percent chance in either direction. It seems that a wife's being
hostile Is more inimical to marital stability than is the hus
band's being hostile.
i It would seem, then, that if marital stability can be pre
dict ed on the basis of the Interpersonal Check List, it can be
done with a . degree of confidence only If both partners are a.ffil
ia.tive or loving. The best that can be expected otherwise is a
fifty percent chance. It would seem, then, that the most effec
tive stabilizing influence a . marriage counselor could have would
be to have the hostile spouse or spouses work through and out of
their hostile personality characteristics. Homogeneity in
121
affiliativeness is the one factor most characteristic of marital
stability. Complementarity in Domina.nce-Submissiveness greatly
increases the confidence level in predicting marital stability.
The Marital Stability Test showed itself to be a conven-
j
jient and effective instrument in assessing the, stability of a
marital relationship. No difficulties in administration were
encountered. It is simple and direct. This fact also makes it
possible to give answers to make oneself look good rather than
honest answers. In some cases the husband1s answers indicated
more satisfaction with the marital relationship than did the
wifeTs answers. On the basis of this fact, the test might be
used to assess some areas in which the spouses disagree on the
acceptability of certain behaviors. This would need to be veri
fied by additional research.
122
CHAPTER VII
IMPLICATIONS FOR FUTURE RESEARCH
The fact that a person who Is assertive can receive the
same standard score as a person who is pseudo aggr'essive and
that a . person who is friendly can receive the same standard score
as a person who is obsequious shows the need for a more sophisti
cated personality inventory. Working out and-validating a dif
ferent way of scoring the ICL could also be a . positive contri
bution.
This study did show some highly significant rela.tion-
ships between personality needs and marital stability. However,
they actually show only how people who a.re stable or unstable a.t
the time of interview rate each other. A longitudinal study,
beginning a.t marriage and repealed about every five yeans would
be a much more reliable study. On the basis of such a . study, it
could be determined whether global personality need combina.tions
have any predictive value. This study really shows only the
diagnostic value.
To test the validity for using the ICL or other person
ality inventories to measure progress in marriage counseling,
123
another longitudinal study would be needed. This would be done
by administering the ICL to a sample of couples coming for mar
riage counseling. Those who had been terminated as having
stabilized their marriage would be interviewed a year or some
other predetermined time after termination to see if their per
sonalities are seen through the ICL as In the category which is
more characteristic of marital stability.
CHAPTER VIII
LIMITATIONS OF THE STUDY
I In order to approximate a . true dichotomy between stable
and unstable marriage samples, the majority of unstable marriage
subjects were already divorced. Since the lecture exchange
|
method produced more than enough women but not enough men, a dif
ferent technique, a . personal Interview, was used to get enough
males from unstable marriages. This ma.y have brought a . differ
ent kind of response.
The stable sample also was acquired by two different
methods. Part of the sample was through lecture-exchange, an-
iother part came from members of a bowling league. Some of these
Iturned 'in their questionnaires at the end of the bowling evening,
i
some mailed them in to the researcher later. In the lecture ex
change method, most of the people attending filled out the ques
tionnaire. Less than half of the bowling league members filled
out and turned In their questionnaires.
Another factor that ma.y have affected the reliability of
the ICL inventory was that of memory. One woman pointed this up
when she wrote on her questionnaire, "After eight years It’s
125
hard to remember." Some may have answered in a way to justify
their divorce, some may have forgotten the personality traits
that may have led to divorce.
It has been shown by Kotlar and Taylor that unadjusted
couples mark the ICL significantly differently from adjusted
couples. Although an equal number of men and women were used in
each ca.tegory, only one person was used from each marriage. The
other spouse may have filled out the questionnaire differently.
Other possible limitations of the study were: those who
refused to answer the questionnaires and participate in the study
may have answered significantly differently from those who did.
There was no wa.y of checking on the accuracy or truthfulness of
replies. Social desirability need could have influenced some
self and spouse descriptions. And there seems to have been a low
average rate of descriptive a.djectives checked by many of the
respondents.
CHAPTER IX
I CONCLUSIONS
I The findings of this study have added dimensions to both
the question of complementarity in mate selection and to the
question of marital stability. Other studies in complementarity
have started with the hypothesis that it is or is not a necessary
fa.ctor in mate selection. Winch went so far as to say that a
ma.rria.ge not characterized by complementarity could not last ten
years. This statement was the point from which this study began.
A new dimension added by this study was.the position that hetero
geneity is not the same as complementa.rity. Heterogeneity can
mean either oppositeness or complementarity. This study made the
assumption that heterogeneity in affiliativeness (LOV-HOS) is a
marriage of people with opposing needs, while homogeneity in
affiliativeness is a marriage of mutual needs. It was, therefore
hypothesized that homogeneity in affiliativeness would be con
ducive to marital stability and heterogeneity in affiliativeness
would be detrimental to marital stability. It was also hypothe
sized that homogeneity in dominance or submissiveness would be
detrimental to ma.rita.l stability, since it would hinder respon-
127
siveness between the spouses. Heterogeneity, on the other hand
would be an instance of personality need complementarity and
would be conducive to marital stability. On the basis of these
assumptions, hypotheses were made about four global personality
types in sixteen combina.tions. This added another dimension to
the studies on complementarity of personality needs. Tha.t is
that some personality needs are best met through complementarity
and others by homgeneity and that heterogeneous (opposing) needs
can impede the satisfaction of a . need. The study found this to
be true for affiliativeness but not for non-affiliativeness.
The LD/LS marriage wa.s the relationship most predictive of
marital stability. However, the HD/HS relationship was the most
predictive of instability.
This study also moved from the concern with marital ad
justment to marital stability. This difference is slight, but
it was felt that marital stability can be defined operationally
;more exa.ctly than can marital adjustment. The last dimension
!added was that this study moved from the study of the personality
of the adjusted or unadjusted spouse to the study of which com
bination of personalities is most conducive to the stable mar
riage. It found some to be highly conducive and some to be
highly detrimental. It found many others gave little or no basis
for predicting whether the ma.rria.ge would be stable or unstable.
However, even on this evidence one could say that one of these
given combinations would have a fifty percent chance either way.
■ 128
This also means something.
APPENDIX
Sample Set Of Forms Employed In This Study
MARITAL STABILITY TEST 130
Name_____________________________ Age____ Years married
Times married No. of children Yearly family income
Occupation______________. __________ Years of schooling
Check the answer which is most nearly accurate
Questions Never Now Often
and then
1. My spouse takes advantage of
me in some situations.
2. My marriage is a happy one.
3. My spouse makes the rules.
< 4 - . I think my spouse is
unreasonable.
5. I have one or more of the
following: headaches, anxiety
attacks, sleeplessness.
6. We disagree about the rules of
living together.
7. We argue about extra-marital
heterosexual attachments.
8. There is disagreement about
who is to make the rules for
our family
9. My spouse attempts to enforce
rules that I don't like.
10. Are there subjects which you
can't safely talk about?
11. Has divorce been openly put
forth as a solution between
you and your spouse?
12. Are debt payments such as to
leave you too little to live
on?
13. Did pressure from your spouse
determine your choice of job
or position?
14. Is there in-law trouble? That
is, is your spouse lined up
with in-laws against you?
15. Are you satisfied with your
sex relations?
16. Do you secretly wish you were
divorced?
17. Do you wish you were married
to someone else?
131
Please check those items which describe
well thought of
makes a good impression
able to give orders
forceful
self-respecting
independent
able to take care of self
can be indifferent to others
can be strict if necessary
firm but just
can be frank and honest
critical of others
can complain if necessary
often gloomy
able to doubt others
frequently disappointed
able to criticize self
apologetic
can be obedient
usually gives in
grateful
admires and imitates others
appreciative
very anxious to be approved of
cooperative
eager to get along with others
friendly
affectionate and understanding
considerate
encourages others
helpful
big-hearted and unselfish
often admired
respected by others
good leader
likes responsibility
self-confident
self-reliant and assertive
businesslike
likes to compete with others
hard-boiled when necessary
stern but fair
irritable
straightforward and direct
resents being bossed
skeptical
hard to impress
touchy and easily hurt
easily embarrassed
lacks self-confidence
easily led
modest
often helped by others
very respectfuT of authority
accepts advice readily
trusting and eager to please
always pleasant and agreeable
wants everyone to like him
sociable and neighborly
jwarm
_kind and reassuring
_tender and soft-hearted
_enjoys taking care of others
_gives freely of self
AP HI NO BC FG JK
NAME
DATE
always giving advice
acts important
bossy .
dominating
boastful
proud and self-satisfied
thinks only of himself
shrewd and calculating
impatient with other's mistakes
self-seeking
outspoken
often unfriendly
bitter
complaining
jealous
slow to forgive a wrong
self-punishing
shy
^passive and unaggressive
meek
dependent
wants to be led
lets others make decisions
easily fooled
too easily influenced by friends
will confide in anyone
fond of everyone
likes everybody
forgives anything
over sympathetic
generous to a fault
overprotective of others
tries to be too successful
expects everyone to admire him
manages others
dictatorial
somewhat snobbish
egotistical and conceited
selfish
cold and unfeeling
sarcastic
_cruel and unkind
frequently angry
hard-hearted
resentful
_rebels against everything
_stubborn
distrusts everybody
_t imid
_always ashamed of self
_obeys too willingly
_spineless
_hardly ever talks back
_clinging vine
_likes to be taken care of
will believe anyone
jwants everyone's love
_agrees with everyone
_friendly all the time
_loves everyone
_too lenient with others
_tries to comfort everyone
_too willing to give to others
_spoils people with kindness
LM DE DOM LOV
BIBLIOGRAPHY
133
A SELECTED BIBLIOGRAPHY
1. Arden, T. Z. Handbook for Husbands and Wives, n. p.:
Association Press, n. d.
2. Bee, L. S. Marriage and Family Relations. New York:
Harper Brothers, 1959.
Blazer, J. A. "Complementary Needs and Marital Happiness."
Marriage and Family Living, XXV (February, 1953), 89-95.
Bolton, C. D. "Mate Selection as the Development of a
Relationship." Marriage and Family Living, XXIII (1951),
239-290.
Bowerman, C., and Day, B. "A Test of the Theory of
Complementary Needs, as Applied to Couples During
Courtship." American Sociological Review, XXI (1965),
602-605.
Bricklin, B. "Prediction of Some Aspects of Marital
Compatibility by Means of the Rorschach Test."
Psychiatric Quarterly Supplement, XXXIV (1961), 281-303.
7. Burgess, W., and Cottrell, Leonard S. Predicting Success
or Failure in Marriage. New York: Prentice-Hall, Inc.,
1939, Chapters 9 & 5.
8. Burgess, Ernest W., ancT Wallin, Paul. Engagement and
Marriage. Philadelphia: J. B. Lippincott Co., 1953.
Burgess, Ernest W.; Locke, H. J.; and Thornes, M. The
Family, From Institution to Companionship. New York:
American Book Co., 1963.
Centers, R. "Marital Selection and Occupational Strata."
American Journal of Sociology, LIV (1999), 530-535.
11. Christensen, H. T. "Student Views on Mate Selection.”
Marriage and Family Living, IX (1997), 85-88.
12. Clark, E. L. "Educational Backgrounds in Assortative
Mating." American Psychologist, IV (1999), 287-288.
13. Corsini, R. J. "Understanding and Similarity in Marriage."
Journal of Abnormal and Social Psychology, LII (1956),
327-332.
9.
10.
| 9.
5.
|
134
14-. Day, B. "A Comparison of Personality Needs of Courtship
Couples and Same Sex Friendships." Sociology and
Soeial Research, XLV (1961) , 435-4-4-0.
15. Farher, B. "An Index of Marital Integration." Sociometry:
A Journal of Research in Social Psychology, XX (1957),
117-134.
| 16. Freeman, L. "Homogamy in Interethnic Mate Selection."
! Sociology and Social Research, (July, 1955), 369-377.
; 17. Gray, H. "Psychological Types in Married People." Journal
| of Social Psychology, XXIX (1949) 189-200.
| 18. Hamilton, G. V. A Research in Marriage. New York: Albert
and Charles Boni, 1929, Chapter 3, "Kinds and Degrees
of Spousal Satisfaction and Dissatisfaction."
19. Harte, T. J. "Trends In Mate Selection In a Tri-Racial
S Isolate." Social Forces, (March, 1959), 215-221.
i
l 20. Himes, J. S. "Mate Selection Among Negro College Students."
Sociology and Social Research,XXXIII (1949), 204-211.
21. Hurvitz, N. "Marital Roles and Adjustment in Marriage in
a Middle-Class' Group." Unpublished Ph.D. dissertation,
University of Southern California, Los Angeles, 1958.
! 22. Kallman, F. J. "Applicability of the Twin Study Method
in the Analysis of Variations in Mate Selection and
Marital Adjustment." American Journal of Human Genetics.
IV (1952), 209-222.
| 23. Katz, I., and others. "Need Satisfaction and Edwards PPS
• Scores in Married .Couples." Journal of Consulting
Psychology, (June, 1960), 205-208.
24. Kelley, E. L. "Psychological Factors in Assortive Mating."
Psychological Bulletin, XXXVII (1940), 473.
25. Kent, D. "Subjective Factors in Mate Selection."
Sociology and Social Research, XXV (1951), 391-398.
26. Kerckhoff,,A. C. "Value Consensus and Need Complementarity
in Mate Selection." American Sociological Review.
XXVII (1962), 293-303.
135
27.
28.
i
| 29.
i
!
30..
i
31.
i
32.
33.
34.
: 35.
i
I
35.
37.
Kernodle, W. "Some Implications of the Homogamy-
Complementary Needs Theories of Mate Selection for
Sociological Research." Social Forces. (December, 1959) ,
14-5-152.
Kirkpatrick, C. "Community of Interest and the Measurment
of Marriage Adjustment." The Family. XVII (1937),
133-137.
Kirkpatrick, C. "Statistical Investigation of the Psycho
analytic Theory of Mate Selection." Journal of Abnormal
Psychology. , - (October, 1937), 427-430.
Kotlar, Sally. "Middle-Class Marital Roles— Ideal and
Perceived In Relation to Adjustment In Marriage."
Unpublished Ph.D. dissertation, University of Southern
California, Los Angeles, 1961.
Ktsanes, T. "Complementary Needs In Mate Selection."
Dissertation Abstracts, XIV (1954), 882-883.
Ktsanes, T. "Mate Selection on the Basis of Personality
Type." American Sociological Review. XX (1955), 547-555.
Landis, J. T. and. Landis, M. C. Building a Successful
Marriage. Englewood Cliffs, New Jersey: Prentice Hall,
Inc., 1958.
Langhorne, M. C., and Secord, P. "Variations in Marital
Needs With Age, Sex, Marital Status and Regional
Location." Journal of Social Psychology, XLI (1955),
19-37.
Leary, T., and Coffey, H. S. "Interpersonal Diagnosis:
Some Problems of Methodology and Validation." Journal
of Abnormal and Social Psychology. L (January, 1955),
110-124.
Leary, T. The Interpersonal Diagnosis of Personality.
New York: The Ronald Press Company, 1957.
Locke, H. J., and Wallace, K. "Short Marital-Adjustment
and Prediction Tests: Their Reliability and Validity,"
Unpublished paper.
38. Locke H. J. Predicting Adjustment in Marriage: A Compari
son, of a Divorced and a Happily Married Group. New York:
Henry Holt & Co., Inc., 1951.
136
39. Locke, H. J., and Williamson, R. C. "Marital Adjustment:
A Factor Analysis Study." American Sociological Review,
XXIII (1958), 562-569.
40. Lossner, Walter. "Cultural Values and the Communication
of Cultural Values as a Factor in Marital Happiness."
Unpublished research project paper, University of
Southern California, Los Angeles, 1962.
41. Luckey, Eleanore. "Marital Satisfaction and Irs Associa
tion with Congruence of Perception.” Marriage and
Family Living, XXII (February, 1960), 49-54.
42. Manger, J. A. Art of Happy Marriage. Milwaukee, Wisconsin:
Bruce Publishing Company, 1961.
43. Mangus, A. R. "Role Theory and Marriage Counseling,"
Social Forces, XXXV (March, 1957), 200-209.
44. Miller, L. "Familial Role Typology, Accuracy of Perception,
and Mutual Needs Among Pre-Nuptial Partners." Unpub
lished Ph.D. dissertation, University of Southern
■ California, Los Angeles, 1966.
45. Murstein, B. "The Complementary Need Hypothesis in Newly
weds and Middle-Aged Married Couples." Journal of
Abnormal Social Psychology, LXIII (1961), 194-197.
46. Peterson, James A. Education for Marriage. New York:
Charles Scribner's Sons, 1964.
47. Popenoe, P. "Mate Selection." American Sociological
Review. , (October, 1937), 735-743.
48. Rheingold, J. C. The Fear of Being a Woman. New York:
Grune & Stratton, 1964.
49. Richardson, H. M. "Studies of Mental Resemblance Between
Husbands and Wives . . . ." Psychological Bulletin,
XXXVI (1939), 104-120.
50. Roos, D. "Complementary Needs in Mate-Selection."
Dissertation Abstracts. XVII (1957), 426-427.
51. Rosow, I. "Issues in the Concept of Need-Complementarity."
Sociometry, XX (September, 1957), 216-233.
137
52.
53.
54- .
55.
56. .
57.
58.
59.
60.
61.
62.
63.
Schelleriberg, J. A. "Homogamy in Personal Values and the
Field of Eligibles." Social Forces, (December, 1960),
157-162.
Schellenberg, J., and Bell, L. "Re-examination of the
Theory of Complementary Needs in Mate Selection."
Marriage and Family Living, (August, 1960), 117-232.
Schooley, M. "Personality Resemblances Among Married
Couples," Journal of Abnormal Psychology. (October,
1936), 31+0-34-1. ,
Simpson, G. People in Families. New York: Thomay Y.
Crowell Co., 1960, 213-218.
Smith, M. "Similarities of Marriage Partners in
Intelligence." American Sociological Review, (October,
194-1), 697-701.
Strauss, A. "Personality Needs and Ma.rita.l Choice."
Socia.l Forces, (March, 194-7), 332-335.
Stuart, I. R. "Complementary vs. Homogeneous Needs in
Mate Selection." Journal of Social Psychology, LVI
(1962), 291-300.
Taylor, A. F. "Role Perception, Empathy, and Marital
Adjustment." Unpublished Ph.D. dissertation, University
of Southern California., Los Angeles, 1965.
Terman, L. M. "Personality Fa.ctors in Marital Compatibil
ity." Journal of Social Psychology. (May, 1935),
14-3-171.
Terman, L. M., and others. Psychological Factors in
Marital Ha.ppfness. New York: McGraw-Hill Book Co.,
1938.
Wallace, K. M. "Construction and Validation of Marital
Adjustment and Prediction Scales." Unpublished Ph.D.
dissertation, University of Southern California., Los
Angeles, 194-7.
Winch, R. F. Mate Selection. New York: Harper and Row,
Inc., 1958.
138
64. Winch, R. F., and others. "Empirical Elaboration of the
Theory of Compelementary Needs in Mate-Selection."
Journal of Abnormal and Social Psychology, LI (1955) ,
508-513.
65. Winch, R. F. "Theory of Complementary Needs in Mate-
Selection . . . ." American Sociological Review. XVIX,
(1954), 241-249.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A Study Of Relationships Between Occupational And Marital Roles And Marital Adjustment

PDF

Middle-Class Marital Roles - Ideal And Perceived In Relation To Adjustment In Marriage

PDF

An Empirical Study On The Differential Influence Of Self- Concept On The Professional Behavior Of Marriage Counselors

PDF

Familism, Suburbanization, And Residential Mobility In A Metropolis

PDF

Group Factors And Individual Internalization Of A Value

PDF

Perception Of The Power Structure By Social Class In A California Community

PDF

Time And Man-Nature Value-Orientations As Related To Sex Identity, Ethnicidentity, And Socioeconomic Status And As Predictors Of Academic Success

PDF

Power Relationships In Marital Discord

PDF

An Investigation Of Complementary Needs Between Marital Partners

PDF

Some Social Factors Affecting The Power Structure And Status Of A Professional Association In Reference To Social Work

PDF

Reference Group Theory, Selection, And The Images Of Professions

PDF

Role Perception, Empathy, And Marital Adjustment

PDF

Consensus Of Role Perceptions In A Welfare Planning Council

PDF

The Parameters Of Singleness: An Inquiry Into Some Factors Influencing The Choice Of Singleness Over Marriage As A Way Of Life

PDF

Attitudes Toward Sex In Marriage And Patterns Of Erotic Behavior In Dating And Courtship Before Marriage

PDF

The Career Business Executive As A Definitive Occupational Type

PDF

Social Factors Related To Dentistry As A Career

PDF

Normative values of selected law enforcement officers and adult male offenders

PDF

A Critique Of The Concept Ethnocentrism As Set Forth In Selected Social Science Literature

PDF

Interpersonal Relations In Ethnically Mixed Small Work Groups

Asset Metadata

Creator Lossner, Walter Martin (author)

Core Title Complementarity, Homogeneity, Heterogeneity And Marital Stability

Degree Doctor of Philosophy

Degree Program Sociology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,sociology, individual and family studies

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Peterson, James A. (committee chair), McDonagh, Edward C. (committee member), Stoll, Earline (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-636018

Unique identifier UC11361047

Identifier 6900625.pdf (filename),usctheses-c18-636018 (legacy record id)

Legacy Identifier 6900625.pdf

Dmrecord 636018

Document Type Dissertation

Rights Lossner, Walter Martin

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA