University of Southern California Dissertations and Theses

An Exploratory Analysis Of Two Dimensions Of Family Separation

(USC Thesis Other)

An Exploratory Analysis Of Two Dimensions Of Family Separation

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content I
i
\
f 70-16,859
i
| COHEN, Stanley Norman, 1930-
| AN EXPLORATORY ANALYSIS OF TWO DIMENSIONS
' OF FAMILY SEPARATION.
!
i University of Southern California, Ph.D., 1970
Sociology, family
University Microfilms, Inc., Ann Arbor, Michigan
© Copyright 1970
by
Stanley Norman Cohen
THXS DISSERTATION has been microfilmed exactly as received
AN EXPLORATORY ANALYSIS OF TWO DIMENSIONS
OF FAMILY SEPARATION
by
Stanley Norman Cohen
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)
January 1970
UNIVERSITY O F SO U TH ERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 9 0 0 0 7
This dissertation, written by
.......St an ley Noraan Cohen.......
under the direction of h.i s . . . Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Gradu
ate School, in partial fulfillment of require
ments of the degree of
D O C T O R O F P H I L O S O P H Y
Dean
Date.
DISSERTATION COMMITTEE
TABLE OF CONTENTS
Chapter Rage
I. THE PROBLEM AND DEFINITION OF TERMS 1
The Problem
The Conceptual Frame of Reference
Definition of Terms
The Hypotheses
The Significance of the Study
Plan for the Dissertation
II. REVIEW OF THE LITERATURE.................. 33
III. THE METHODOLOGY............................ 45
Research Design
Characteristics of the Sample
Measurement of the Independent, Control,
and Dependent Variables
Limitations of the Study
Summary of Methodology
IV. DISCUSSION OF THE FINDINGS................ 76
Results
Summary
V. SUMMARY AND CONCLUSIONS 107
Future Research
BIBLIOGRAPHY 115
APPENDICES 122
ii
LIST OF TABLES
Table Rage
1. Frequency and Efercentage Distribution of
Subjects According to Family Developmental
Stages.................................... 51
2. Frequency and Etercentage Distribution of
Subjects According to the Degree of Marital
Happiness at Time 1 and Time 2 (One Year
Later).................................... 54
3. Frequency and Rarcentage Distribution of
Subjects According to the Number of
Children.................................. 55
4. Frequency and Rarcentage Distribution of
Subjects According to Age of Subjects ... 56
5. Frequency Distribution of Subjects According
to Age and Developmental Stages.......... 57
6. Frequency and Etercentage Distribution of
Subjects According to Income ............. 58
7. Frequency and Etercentage Distribution of
Subjects According to Education ......... 59
8. Frequency and Efercentage Distribution of
Subjects According to Race or Ethnic
Group.................................... 60
9. Selected Variables and Their Raarsonian
Correlations to the Theoretical
Variables................................ 78
iii
Table Eage
10. Selected Variables That Are Associated with
Separation of Frequency and length with
Bearsonian Correlations Greater Than .20 . . 79
11. Frequency Distribution of Subjects According
to Marital Happiness at Time 1 and
Number of Separations .................... 81
12. Frequency Distribution of Subjects According
to Marital Happiness at Time 2 and
Number of Separations .................... 82
13. Frequency Distribution of Subjects According
to Changes in Marital Happiness and
Frequency of Separation........... 83
14. Frequency Distribution of Subjects According
to Time 1 and Time 2 Measures of Marital
Happiness and Frequency of Separation ... 84
15. Frequency Distribution of Subjects According
to Developmental Family Stages and Number
of Alternatives .................... 89
16. Frequency Distribution of Unhappy Subjects
According to the Number of Alternatives
and Number of Separations................ 92
17. Frequency Distribution of Unhappy Subjects
According to the Number of Alternatives
and length of Separations................ 94
18. Frequency Distribution of Developmental Stage
One Subjects According to Number of Alterna
tives and Length of Separation............ 95
19. Frequency Distribution of Developmental Stage
Two Subjects According to Number of Alterna
tives and Length of Separation............ 96
iv
Table Ihge
20. Frequency Distribution of Developmental
Stage Three Subjects According to Number
of Alternatives and length of
Separations.............................. 97
21. Frequency Distribution of Developmental
Stage Four Subjects According to Number
of Alternatives and Length of
Separations.............................. 98
22. Stepwise Regression: Predicting Number of
Separations.............................. 101
23. Stepwise Regression: Predicting Length of
Separations.............................. 103
v
LIST OF FIGURES
Figure Rage
1. The Ideal Nuclear Family Conceptualized in
the Context of the Developmental
Framework................................ 9
2. The Operational Version of Developmental
Stages Used in This Study to Analyze the
Frequency and Length of Separation........ 19
vi
CHAPTER I
THE PROBLEM AND DEFINITION OF TERMS
This study deals with two aspects of family separa
tion behavior'*' within a specific population of married
couples.^
The Problem
Marital happiness, religion, education, occupation,
income, number of children, and deviance are factors that
have been found to have some bearing on the occurrence of
separation in the contemporary family. Another factor—
alternatives to the marriage— has been considered by some
family sociologists (Udry, 1966; Levinger, 1965) to be
another variable that is related to separation.
This investigation has two primary objectives. The
first objective is to examine the occurrence and frequency
of separation and the extent to which alternatives to living
3 /
together accounts for this phenomenon. The contributions
1
of marital happiness, religion, deviance, number of chil
dren, and socioeconomic variables of income and education
in explaining separation also will be examined.
The second aim of this study is to examine the
length of separation with particular emphasis on the regu
larity with which no separation, short separation, and long
separation occurs within the context of the family develop
mental stage with which a maladjusted marital couple is
identified.
The Conceptual Frame of Reference
Changes in the American family structure have been
noted over the past sixty years that appear to be associ
ated with an increasing frequency of family separation
(Goode, 1965 :94). The primary change— a shift from an
essentially agrarian and rural society to a highly indus
trial one--has been accompanied by a shift from an extended
family structure to a nuclear system that is isolated and
independent of its kinship base. The extent to which the
isolated nuclear family pervades various levels of American
society remains a sociological controversy (Litwak, 1959:
177, 187; Sussman, 1953:27), but most people considering
3
marriage expect to establish such a household. The iso
lated western style nuclear family, which is independent
from its kinship base, has become the idealized and desired
family system for persons considering marriage (Mogey,
1964:509).
As an integral part of a rapidly changing society,
the value orientation and functional utility of the modern
family have been modified greatly. The characteristics
that differentiate the extended and nuclear family systems
have been summarized in the following manner by Burgess and
Locke (1960):
Of the historical and existing types of families the
large-patriarchal type . . . approximates the ideal
construction of the institutional family with its . . .
powerful sanctions of the mores, religion, and law, and
. . . subordination of . . . members of the family to
the authority of the patriarch. The modem American
family residing in . . . the city approximates . . .
the ideal type of companionship family, in which
. . . members enjoy . . . self-expression and . . . are
united by . . . bonds of affection, congeniality, and
common interests.
A summary comparison will indicate the point-by
point outstanding differences between the small
democratic-family unit of husband, wife, and children
and the extended-patriarchal family.
The . . . family is authoritarian and autocratic
with power vested in the head of the family and with
the subordination of his wife, sons, and their wives
and children, and his unmarried daughters to his au
thority. The modem family is democratic, based on
equality . . . with consensus in making decisions and
. . . increasing participation by children. . . .
4
Marriage is arranged . . . patriarchal family with
emphasis upon prudence, upon economic, . . . social
status, and . . . adjustment of the son-in-law,
daughter-in-law to the family group. In the modem
family, marriage is in the hands of young people, and
selection is on the basis of romance, affection, and
personality adjustment to each other. Compliance with
duty and the following of tradition are guiding prin
ciples of the patriarchal family. The achievement of
personal happiness and the desire for innovation are
watchwords of the modem family. The chief historic
functions of the family— economic, educational, recre
ational, health, protective, and religious— are found
in the fullest development in the extended-patriarchal
family. These historic functions have largely departed
from the modem urban family. (Burgess and Locke,
1960:25-27)
Udry (1966) describes the process of functional
differentiation which he feels is a fundamental transaction
that occurs in the development of complex modem societies
(Udry, 1966:17-18). This process refers to the need for
the development of large specialized coordinating institu
tions as a society grows in complexity and scale. In
simple societies, social institutions, e.g., the family,
are very generalized, performing a wide range of functions.
The modem family has been modified by the func
tional differentiation process. Many of its former
functions have been incorporated into large specialized
educational, occupational, and religious institutional
systems. The contemporary family system is now a small
specialized society unity that is the primary agent for the
performance of functions associated with child rearing and
companions hip.
The essence of the contemporary American marriage
and family, then, is the development of a relationship that
is based on companionship, and nuclear role relationships
have been developed in order to provide husbands and wives
with marital happiness or satisfaction. later on, with the
introduction of children, a marital couple also is expected
to be primarily responsible for the socialization and en-
culturation of their child(ren).
Any analysis of family separation is a complex
conceptual and empirical task. One feasible approach, how
ever, is to examine the extent to which the functional
characteristics of nuclear role relationships generate dif
ferential degrees of marital interdependence. This husband-
wife interdependence essentially is an extra-reliance of
one marital partner upon the other for the gratification of
biological, psychological and social needs at certain
periods during the marriage. Consequently, the ability of
a dissatisfied or unhappy marital partner to live apart
from his (her) spouse for an appreciable length of time
varies and is related to the degree of interdependence
manifested in the marriage at the time when a separation is
being considered.
The marital partner in a nuclear family, then, must
have access to viable alternatives to the marriage if a
separation is to be a lasting one. Unlike the extended
family, whose structure provides built-in alternatives to
absorb the responsibilities of all of its dissident mem
bers, no such opportunities are readily available to a
marital partner in the isolated nuclear system. Alterna
tives to a marriage are external to the nuclear family
system.
Some general notions regarding the relationship of
alternatives to family separation have been discussed
directly and indirectly by family sociologists (Udry, 1966;
Pitts, 1964; Mercer, 1965). However, it appears that there
have not been any systematic conceptual or empirical
attempts to analyze the explanatory value of alternatives.
The influence of alternatives on family separation
is described in the following manner:
Looked at from a secular point of view, the decision
whether or not to end an unhappy marriage will usually
be resolved around the question, what are the alterna
tives to the marriage? If public opinion will not be
too oppressive, if economic conditions look favorable,
if single life seems attractive, or if the likelihood
of a good replacement marriage seems high, and if
divorce is not a moral issue, then marital frustration
tolerance will be low, and people will dissolve mar
riages often. (Udry, 1966:516)
Aside from normative constraints regarding separa
tion and/or divorce, alternatives must provide opportuni
ties for an unhappy marital partner to establish economic
and residential independence from his (her) spouse. In
this dissertation alternatives to a marriage are viewed as
those social situations that enable a dissatisfied or tin-
happy marital partner to live apart from his (her) spouse,
and at the same time fulfill obligations and duties
previously the responsibility of the other marital partner.
Analyzing a contemporary family over time provides
an excellent opportunity to investigate the extent to which
a dissatisfied marital partner has access to alternatives
that can be used to neutralize the varying degrees of
marital interdependence occurring in the nuclear family.
The developmental approach (Kenkel, 1966:391-481; Hill and
Rodgers, 1964:171-211; Stroup, 1966) is the conceptual
framework that has been used in this study because it:
1. Conceptualizes the nuclear family as a more or
less stable social system enduring and changing
through time;
2. Enphasizes the structural-functional aspects of
contemporary family life;
3. Provides an opportunity to focus on the degree
of interdependence encountered by marital
partners as a family progresses through its
developmental stages; and
4. Provides an opportunity to develop a research
design that can include and control for many
variables that have been demonstrated to have
some bearing on the occurrence of separation.
According to proponents of the developmental
approach, the family is organized around clusters of cul
turally determined role positions, e.g., husband-wife,
fat her-mother, mother-father-son, mother-daughter. Norms
prescribing role behavior for these clusters specify the
social and personal patterns for, a family interactional
system.
The predictable developmental history of an ideal
nuclear family is illustrated in Figure 1. The left hand
column describes variables associated with family structure,
i.e., family size, years in stage, years married by stage,
Stage
(Phase A)
Establishment
Stage 1
(Phase B)
Expectant
Stage 2
Childbearing
Stage 3
Preschool
Family
Age H 22 H 23 F 24-26 F 26-30
W 20 W 21 M 22-24
Sj^ 0-2
M 24-28
Sjl 2-6
52 0-4
53 0-2
Size 2 2 3 3-5
Years Married 0 1 3 7
Positional
Developmental
Tasks
(Age and
Relatedness
Positions)
Young Adult
DT
Young Adult
DT
Parent DT
Infant DT
Parent DT
Infant DT
Child DT
Major Family
Goal
Adjusting to
living as
married pair
Adjusting to
pregnancy
Reorganization of unit
around needs of infants
and pre-school children
THE IDEAL NUCLEAR FAMILY CONCEPTUALIZE
Stage 3
Preschool
Family
Stage 4
School-Age
Stage 5
Adult Trainees
Stage 6
Young Adult
Launching
Stage 7
Middle Years
AS'
F 26-30 F 30-38 F 37-44 F 44-52 H 52-65 ]
M 24-28 M 28-35 M 35-42 M 42-50 W 50-64 1
Sx 2-6 Sx 6-13 Sx 13-20 CSX 20-28, f)
S2 0-4 S2 4-11 S2 H-18 (S2 18-26, m)
S3 0-2 53 2-9
54 0-7
S, 9-16
s' 7-14
(S3 16-24, f)
(S^ 14-22, m)
3-5 5-6 6 6-2 2
7 14 21 29 43
Parent DT
Infant DT
Child DT
Parent DT
Infant DT
Child DT
Preadoles. DT
Parent DT
Child DT
Preadoles. DT
Teenage DT
Grandparent DT
Parent DT
Teenage DT
Young Adult DT
Postparental
Adult DT
Grandparent DT
Gra
Agi
on of unit
of infants
ol children
Reorganizat ion
of family to
fit into ex
panding world
of school-agers
Loosening of
family ties to
permit greater
freedom and
heavier respon
sibility to
members
Reorganization
of family into
equalitarian
unit and releas
ing of members
Reorganization
of family
around
marriage pair.
Strategy of
disengagement
Dis
FIGURE 1
,Y CONCEPTUALIZED IN THE CONTEXT OF THE DEVELOPMENTAL FRAMEWORK
9
Stage 6
Stage 5 Young Adult Stage 7 Stage 8
ult Trainees Launching Middle Years Aging Years
F 37-44 F 44-52 H 52-65 H 66-72
M 35-42 M 42-50 W 50-64 W 64-77
Si
13-20 (Si
20-28,
f)
s 2 11-18 (S2 18-26, m)
S3
9-16
(s 3
16-24,
f)
s 4
7-14
< S4
14-22, m)
6 6-2 2 2-1
21 29 43 43-50
rent DT
ild DT
eadoles. DT
enage DT
Grandparent DT
Parent DT
Teenage DT
Young Adult DT
Postparental
Adult DT
Grandparent DT
Grandparent DT
Aging DT
osening of
mily ties to
emit greater
eedom and
avier respon-
bility to
nbers
Reorganization
of family into
equalitarian
unit and releas
ing of members
Reorganization
of family
around
marriage pair.
Strategy of
disengagement
Disengagement
3 DEVELOPMENTAL FRAMEWORK
10
and family positions. Positional developmental tasks and
major family goals represent behavioral expectations that
are associated with the primary functions of the contempo
rary nuclear family.
The discussion that follows is a conceptual ex
planation of family separation. This phenomenon may be
explained, to a great extent, by the absence or presence of
alternatives to a given marriage. The fluctuating manner
by which a dissatisfied or unhappy marital partner has
access to alternatives is analyzed in the context of the
developmental framework.
Phase A of the first family stage in Figure 1
i
represents a relatively simple husband-wife structural con
figuration. The primary importance of establishing a
satisfying or happy relationship is emphasized with ad
justment to living together being the major family goal.
Phase B of this stage also focuses on the adjustment of a
husband and wife; and major family goal is adjustment of a
couple to pregnancy. Consequently, family stability essen
tially relies on the extent to which marital partners per
ceive their personal and social relations with each other
to be satisfying or happy.
11
The nature of the contemporary society offers a
latitude of activities to young married people who do not
have any children. A wide range of alternatives that can
be used for separation purposes are available to a dis
satisfied marital partner.
An unhappy husband or wife potentially is more
mobile during Stage 1 than at any other time in the history
of a family primarily because there are no children. There
fore, there are ample opportunities for a dissatisfied per
son to find employment, meet someone of the opposite sex
for romantic or remarriage purposes, and move geographi
cally.
Marital interdependence is relatively minimal in
family stage 1, and it is not very difficult for persons to
find suitable alternatives to an unhappy marriage. Conse
quently, one could expect to find separation to be more
prevalent in this developmental stage than in any of the
others.
The preeminence of child rearing tasks associated
with the function of socialization is emphasized in stages
2, 3, 4, 5, and 6. Jfejor family goals center on the re
organization of the family unit around the needs of infants
12
and pre-school children (stages 3 and 4), the reorganiza
tion of the family around the expanding world of school-
agers (stage 4), the loosening of family ties enabling
children to have more freedom and greater responsibilities
(stage 5), and the reorganization of the family unit for
the release of the adult children (stage 6).
A family with children is more complex structurally
than a childless family. Family interactional patterns are
expanded considerably with the arrival of children, ihren-
tal role patterns are developed to provide for the depen
dency needs of children. Children, of course, are com
pletely dependent upon their parents when they are very
young. As they mature, children become more independent
and eventually establish themselves apart from their
parents.
A "mothering" parent, during those family stages
that deal with child rearing tasks, becomes extra-dependent
upon the other marital partner, because of the confining
duties associated with the care of children. In other
words, marital interdependence is increased because con
fining child rearing tasks, along with the isolated nature
of the nuclear system, sharply reduces the mobility of
13
the ’ ’ mothering" parent. A dissatisfied parent, unlike a
couple who has no children, must have access to alterna
tives that include opportunities for the care of children
in addition to situations that attend to the social and
personal needs of the separating parent.
Marital interdependence is assumed to be greatest
in family stages 2, 3, and 4 because of the ages of the
children. Separation would be expected least likely to
occur during these developmental stages because the alter
natives to a marriage are likely to be few. If separations
do occur, they are expected to be of a shorter duration
than those occurring in family stage 1.
Stages 5 and 6 represent those periods in a family
when children mature and become less dependent upon their
parents. The mobility of a dissatisfied marital partner is
increased, during these stages, as child rearing obliga
tions become less confining. The range of alternatives to
an unhappy family situation person is expanded because the
alternatives are not restricted only to those limited
opportunities that must provide for both the demanding
needs of young children and those of the separating marital
partner. Separations are expected to occur more frequently
14
and for longer periods of time in stages 5 and 6 than those
occurring in stages that are characterized by the presence
of younger children.
The older nuclear family is represented in develop
mental stages 7 and 8. All of the children now have estab
lished independent life styles and the major family goal is
reorganizing the family around the marital couple during
the period of disengagement from the children. An emphasis
on youthfulness by the contemporary society and realistic
limitations brought on by the aging process appreciably
decrease opportunities for older people to find employment
and establish new personal and social relationships. Such
restrictions circumscribe and limit social mobility pat
terns of unhappy husbands and wives in developmental stages
7 and 8, and marital interdependence is similar to the type
exhibited in stages 2, 3, and 4. Therefore, unhappy or
dissatisfied marital partners whose children have grown up
and established themselves independent of their parents are
expected to exhibit family separation patterns that closely
parallel the type that appears in stages 2, 3, and 4.
An attempt has been made in the preceding discus
sion to use the developmental approach as a conceptual
15
framework that logically explains differential patterns of
separation occurring in the contemporary American family.
Recognition is given in this dissertation to the difficulty
of translating this statement into operational terms.
The developmental approach includes a range of
abstractions; several concepts refer directly to observable
behavior while many are one, two, and three steps removed
from such behavior. Family size, the age of family mem
bers, the number of years married, the positions within a
family are examples of easily translatable material. On
the other hand, marital adjustment is not observable di
rectly; it must be inferred from the interactions of the
couple or from established opinion or attitude scales. The
behavior includes at least two levels of abstraction— the
concept of norms, and the behavior of the marital partners
emanating from their expectations and performance of pre
scribed roles.
The establishment of clear-cut family stages poses
yet another operational difficulty for anyone using the
developmental approach to investigate family separation
behavior. The family stages overlap. For example, a
family can include children of different ages, a family
i
16
unit may include persons other than the husband, wife and
children, and children of different ages may be living in
or out of the family home. An operational set of family
stages must deal effectively with these practical problems,
and at the same time provide a discriminating base if a
meaningful analysis of the phenomenon is to be accomplished.
It is presently impossible to define measure, and
control all of the variables that probably are related to
family separation. However, the following assumptions have
been used as premises in establishing an operational format
of this analysis of separation.
1. The families that have been selected as subjects
for this study are analyzed in the context of
the nuclear family system.
2. Marital dissatisfaction or unhappiness as per
ceived by a spouse, and alternatives to a
marriage are assumed to be necessary but not
sufficient conditions of separation. Further
more, marital dissatisfaction or unhappiness
is considered to be the antecedent condition
of separation. Family separation is not ex
pected to occur when these two conditions
are not present in the situation. Separation
can be expected to occur to a varying degree
when these factors are present in a family
system.
All other factors that have a bearing on family
separation are treated as contingency variables
in this dissertation. They are neither neces
sary nor sufficient conditions of separation.
Religiosity, the socioeconomic variables of
income and education, deviance, the number of
children, and the presence and age of children
have been selected as variables that have some
bearing on the occurrence of separation. These
variables are controlled in this study.
Child rearing responsibilities affect the ex
tent to which a dissatisfied or unhappy marital
partner has access to alternatives. Family
developmental stages are defined by the pres
ence and age of children since the age of a
child affects his dependency on his parents,
which in turn is assumed to affect a parent’s
access to separation alternatives.
18
The operationalized version of the developmental
approach used in this study to investigate family separa
tion is illustrated in Figure 2. Four developmental stages
have been devised: families with no children (developmental
family stage 1); families with young children (develop
mental family stage 2); families with old children (de
velopmental family stage 3); and old families with children
not living in the family home (developmental family
stage 4).
Definitions of Terms
Maladjusted family. This terra refers to any family
in which one of the marital partners formally seeks the
services of the Los Angeles County Conciliation Court.
Nuclear family. This term refers to a residential
social unit composed of a legally married husband, wife,
and their natural or legally adopted children.
Extended family. This term refers to a residential
social unit composed of a nuclear family and relatives of
the husband and/or wife.
19
Stage 1 Stage 2 Stage 3 Stage 4
No Age of Age of Age of
Children Children Children Children
0-11 12-18 18 and over
FIGURE 2
THE OPERATIONAL VERSION OF DEVELOPMENTAL STAGES
USED IN THIS STUDY TO ANALYZE THE FREQUENCY
AND LENGTH OF SEPARATION
20
Happiness of marital partner. This term refers to
the perception of the degree of satisfaction of a marital
partner, as measured by a self-rating on a fivefold scale.
Alternatives. This term refers to those residen
tial, support, and occupational situations that enable a
dissatisfied marital partner to live apart from his (her)
spouse, as measured by single items and a summative scale
devised in this study.
Family separation behavior. This tern refers to
those situations when a dissatisfied marital partner in
tentionally lives apart from his (her) spouse.
Short separation. This term refers to a situation
when a dissatisfied marital partner lives apart from his
(her) spouse for less than two months.
Long separation. This term refers to a situation
when a dissatisfied marital partner lives apart from his
(her) spouse for more than two months.
Religiosity. This term refers to the degree of a
marital partner’s involvement in the activities of the
religion with which he (she) is identified. A series of
21
single Items and a scale have been devised In this study
to measure this phenomenon.
Deviance. This tern refers to a marital partner’s
negative perception of the emotional, alcoholic, drug, and
criminal behavior of his (her) spouse. Deviance is mea
sured by a series of single items and a scale devised in
this study.
Developmental family stage 1. This term refers to
those legally married couples who do not have any natural
4
or legally adopted children.
Developmental family stage 2. This term refers to
those legally married couples who have natural or legally
adopted children the majority of whom are younger than
twelve years of age.
Developmental family stage 3. This term refers
to those legally married couples the majority of whom are
between the ages of twelve and eighteen.
Developmental family stage 4. This term refers to
those legally married couples who have children older than
eighteen years of age, all of whom are not living in the
22
family home.
Independent variable and dependent variable. These
terms refer to the "cause'1 and "effect" relationship of one
phenomenon upon another. The independent variable is the
presumed "causal" factor; in this study, marital dissatis
faction and alternatives are defined as the independent
variables. The dependent variable is the factor being
affected; the occurrence, frequency, and duration of sepa
ration are defined as the dependent variables in this study.
The Hypotheses
Eight propositions specifying the relationship of
marital happiness and alternatives to marriage to frequency
and length of separation have been derived from the de
velopmental framework. They have been put in hypothesis
format, but they should be considered in the context of
exploratory postulates. The hypotheses to be examined are
as follows:
Hypothesis I. Marital partners who perceive them
selves to be "unhappy" will separate more frequently than
those marital partners who perceive themselves to be
"happy."
23
Hypothesis II. Marital partners in developmental
family stage 1 have more alternatives to their marriage
than do marital partners in developmental family stages
2, 3, and 4.
Hypothesis III. Marital partners in developmental
family stage 3 have more alternatives to their marriage
than do marital partners in developmental family stage 2.
Hypothesis IV. Marital partners in developmental
family stage 3 have more alternatives to their marriage
than do "unhappy" marital partners in developmental stage 4.
Hypothesis V. There is no difference in the number
of alternatives between marital partners in developmental
family stages 2 and 4.
Hypothesis VI. There is no difference in the num
ber of alternatives between marital partners in develop
mental family stages 1 and 3.
Hypothesis VII. "Unhappy" marital partners who
have alternatives will separate more frequently than those
"unhappy" marital partners who do not have alternatives.
24
Hypothesis VIII. "Unhappy" marital partners who
have alternatives will separate longer than those "unhappy"
marital partners who do not have alternatives. It will be
recalled that length of separation has been defined in the
context of long and short patterns. Specifically, long
separations are characteristic among those "unhappy" mari
tal partners in developmental family stages 1 and 3. Short
separations prevail among those "unhappy" marital partners
in developmental family stages 2 and 4.
The Significance of the Study
This research is a social "experiment" that is con
ducted in a relatively uncontrollable and very complex
field situation. It attempts to account for a widely
recognized contemporary social problem— the occurrence and
duration of family separation. This research design uti
lizes operational procedures that have been derived from a
sociological conceptual model of family interaction— the
developmental approach.
The primary significance of this study lies in an
attempt to conduct a research investigation that is both
scientifically meaningful and useful to the society at
25
large. A controversy among social scientists has ensued
for several years regarding the ability of a person to
maintain a scientific stance and at the same time be able
to conduct an objective investigation that can lead to
social action. The intrusion of values has been the under
lying issue in this controversy. Christensen (1964) cate
gorizes three positions relative to this subject: value
rejection, value espousal, and value separation. Value
rejection is the position that advocates a complete dis
sociation of the scientist from his value world. Lundberg
(1939) is an illustrative example of those supporting value
rejection in the scientist*s world of work.
Those persons who are partisan to value espousal
conclude it is virtually impossible for the social scien
tist to compartmentalize his life into separate areas of
"science" and "action." They call for the scientist to
make his position explicit, and since he is better informed
than the layman, assume active leadership in helping
society decide what society should do about social prob
lems. Mills (1959), Popper (1952), and Hanson (1963) ad
vocate value espousal.
Most sociologists adhere to the position of value
separation. The legitimacy of values of a scientist is
26
recognized. It is desirable to speak out on issues, but a
social scientist should not permit his values to influence
the methods of his research. The selection of a research
problem that is based on a subject that the scientist per-
ceives to be important to the welfare of his society is
legitimate; manipulating the research process in order to
confirm the personal values of a scientist is quite ille
gitimate.
The investigator of this study holds with the value
separation position. It seems clear that resolutions to
persisting and new social problems emanating from rapidly
changing social conditions in the contemporary society are
within the responsible domain of the scientific community
generally, and social science specifically. On the other
hand, any illegitimate attempt "to make the data fit" is
as dysfunctional to the resolution of problems as those
efforts based on emotions and preconceived notions about a
given phenomenon of concern to the society. A twofold task
of social science has been described as follows:
Social science has a two-in-one task. On the one hand,
its responsibility as a science is to develop a body
of principles which make possible the understanding
and prediction of the whole range of human interaction.
On the other hand, because of its social orientation,
27
it is increasingly being looked to for practical guid
ance with immediate problems of improving human inter
relationships. (Selltiz and Cook, 1948:454-459)
The following summary describes what Selltiz and
Cook (1948:454-459) specify as the factors that contribute
to the scientific relevance of a research investigation:
1. A study must include a systematic formulation
and verification of a set or sets of hypo
theses.
2. A research must afford the opportunity to
replicate the investigation in order to check
the findings.
3. Research must result in the presentation of a
generalization or principle that is not limited
to the immediate setting of a particular in
vest igat ion.
This study is a beginning effort in the direction
set forth by the preceding discussion. As such, this re
search effort, when judged according to the ideal criteria,
has several limitations. However, it does approach some of
these criteria when limitations imposed on this study are
considered.
First, an attempt has been made to develop and
verify systematically several hypotheses that have been
28
derived logically from the conceptual framework of the
developmental approach. Furthermore, the hypotheses have
been developed to explain the phenomenon of family separa
tion in the context of causality; marital satisfaction and
alternatives to a marriage are treated as necessary but not
sufficient conditions of separation. This research also
is one of the few attempts to analyze family separation
behavior using a longitudinal type of research design (Hill
and Rodgers, 1964:204).^ These positive attributes, how
ever, pose the greatest problems. Translating abstract
concepts such as family stage and alternatives into valid
operational terms perhaps has been difficult and too sim
plistic. The problems associated with the use of a longi
tudinal type of analysis also have had a limiting effect in
this study, particularly in the area of case attrition.
Notwithstanding these limitations, this analysis is con
sidered a necessary first step in the verification of the
utility of the developmental framework in explaining and
predicting family separation.
Second, this study has been carried out and de
scribed in terms that provide for a series of replications.
This research is considered to be the first of many studies
29
investigating family separation behavior. Finally, this
research effort does result in some tentative findings
with implications that are not limited to the setting of
this investigation.
The extent to which a research investigation has
practical social utilitarian significance is summarized in
the following manner.
1. A research investigation must deal with a prob
lem that has present and future social conse
quences ;
2. The results of a research project must have
possibilities of application in concrete social
situations; and
3. A research investigation must be carried out in
a manner that stimulates the application of its
results in a practical social situation.
These criteria for the social usefulness of a study
are met to some extent by this investigation. A qualita
tive assessment of family life in terms of "success" or
"failure" is not emphasized in this study. However, it is
fair to say that the general society perceives family sepa
ration to be a major social problem in need of immediate
remedial attention.
This study attempts to spell out, as clearly as
possible, the types of separation patterns that occur in
developmental family stages. Consequently, the results of
this study are applicable to those couples who seek the
services of public and private marriage and family counsel
ing agencies.
Finally, since the investigation was conducted in
a prominent public marriage and family counseling agency,
some of the personnel have been stimulated to modify pre
viously established attitudes regarding family separation
behavior. For example, there has been a greater emphasis
to focus on and to make available counseling services for
families with young children and older marital couples than
for those marital couples who have no children.
Plan for the Dissertation
The remainder of this dissertation is organized in
the following manner:
Chapter II presents an overview of the literature
relevant to family separation.
Chapter III describes the methodology used includ
ing the research design, characteristics of the sample,
31
the means by which the independent, dependent and control
variables were measured, and the limitations of the study.
Chapter XV discusses the findings. The indices
used and devised in this study have been analyzed, along
with the possible effects of bias.
Chapter V has summarized the conclusions of the
study.
32
Chapter I: Notes
"^Family separation behavior occurs when a legally
married partner considers living apart from his (her)
spouse because of conflict. A variety of terms have been
used to depict family separation. Dissolution, insta
bility, disintegration, divorce, desertion, and separation
are the most frequently used terms. These terms are used
interchangeably in this study.
2
The sample used in this study was taken from those
couples that formally requested the services of the Los
Angeles County Conciliation Court. Sincere appreciation is
expressed to this Court without whose active cooperation
this study would have been impossible. Thanks are also due
those couples who enthusiastically participated in this
project.
3
Alternatives are those social situations that
enable a marital partner to live apart from his (her)
spouse. Hereafter this term may be referred to as alterna
tives.
4
The criteria for determining the operational
family developmental stages are based on the comments of
several teachers, marriage and family counselors, Domestic
Relations judges and commissioners.
5
While the study design is longitudinal in the
general sense, it could be described best as a panel or
cohort design with a two point time sequence.
CHAPTER II
REVIEW OF THE LITERATURE
The material that follows is not presented as an
exhaustive exposition of studies related to family separa
tion. Rather, it is representative of the kind of work
that has been done in this area.
Most studies having a direct or indirect bearing on
family separation have been undertaken by sociologists and
clinicians, namely psychiatrists, psychologists, and social
workers. Clinically oriented investigators have focused
primarily on the relationship of personality development to
marital disharmony in their studies that have dealt with
family separation. Occurrences of divorce, separation, and
desertion are viewed typically as resulting from person
ality: problems of a husband or wife. Consequently, many
marriages are contracted for neurotic reasons, and family
stability is forfeited almost automatically.
Separation has been treated tangentially in most
of these family studies (Greene, 1965; Ackerman, 1958;
33
34
Eisenstein, 1956; and Chance, 1959) because the primary
interest has been to assess and interpret the psychological
dynamics and dimensions that produce disruptive relations
in family life. The work of Bergler (1946, 1948) is illus
trative of the few studies that deal directly with aspects
of family separation.
Studies by practitioners also are characterized by
a limited application of scientific methods to the analysis
of their respective research problems. Typically, the study
samples have been small in number and generally have been
drawn very selectively from the caseload of a therapist.
Often, definitional clarification of the subject matter has
been ambiguous, and minimal efforts have been made to con
trol any intervening variables that might influence the
findings.
In summary, the majority of studies conducted by
clinically oriented researchers have blended insightful
speculations with research designs that impose restrictions
on the measurement of the phenomena being investigated,
opportunities to replicate the studies and the validity of
the conclusions derived from the studies.
The major research investigations of family life
have been contributed by sociologists. Historically, this
35
interest was prompted, to a great extent, by the changing
characteristics of the family unit. Some sociologists
2
(Sorokin, 1937, 1941; Zimmerman, 1947), in fact, conjec
tured that the family was in jeopardy as a social unit.
Sociologists have contributed the major portions of
research investigations dealing with family life. Histori
cally, sociologically oriented studies were prompted as
rapid changes in society began to influence what was per
ceived to be changes in the structure and function of the
family unit. Increasing frequencies of divorce, declining
birth rates, increasing individuation of family members,
and role changes of women resulting in activities other
than homemaking were primary reasons for this concern and
more investigations on the family have been reported than
in almost any other substantive field in sociology (Goode,
1965:179).
Two basic types of research exemplify sociological
interest in family separation. The first type deals with
marital adjustment, while the second kind of research has
been concerned with analyses of a number of demographic and
social psychological variables and their relationship to
divorce, separation, and desertion.
36
Major attention has been directed to the adjustment
of family members to each other by sociologists, and mari
tal adjustment often has been viewed as a key variable in
marital success or failure. As such, marital adjustment
has been the object of considerable research. During the
past generation sociologists and psychologists have tried
to define and measure the phenomenon. The major marital
adjustment studies are represented by the work of Burgess
and Cottrell (1939), Terman (1938), Locke (1951), Karlsson
(1951), and Burgess and Wallin (1953). Recent studies of
marital adjustment have been outgrowths of these efforts.
Beginning with Ernest W. Burgess and his associate
Leonard S. Cottrell, many investigators have developed
marriage adjustment instruments by ascertaining which back
ground traits, e.g., age at marriage, religious affili
ation, occupation, and which current patterns, e.g.,
community of interests, expressions of affection are
associated with marital satisfaction. It was assumed that
with the development of objective, reliable and valid ad
justment instruments, the accurate prediction of success or
failure in marriage would be enhanced greatly.
Marital prediction studies have been the subject of
much discussion, and their positive and negative attributes
37
have been elaborated upon by Goode (1946:72-73), Bowerman
(1964:231-236), and Kirkpatrick (1963 :385-405 ). Those
factors directly relevant to family separation will be dis
cussed briefly in this section.
That marital adjustment has some bearing on the
occurrence of family separation has been demonstrated in
all of the major prediction studies generally, and specifi
cally by those studies (Locke, 1951; Karlsson, 1952) that
have analyzed samples of divorced couples. Marital pre
diction studies, thus far, however, have not contributed to
any comprehensive understanding of the dynamics of the
family separation process. First of all, family stability
has not been a primary research concern in these investiga
tions. Bowerman (1964:237), in reviewing over sixty
studies published since 1940 that employed some measure of
marital success, has observed that less than one-fifth of
the investigations used divorce or separation as the cri
terion for marital success. The data generally represent
a ’ ’ middle-class,’ ’ fairly well-educated population with a
possible bias toward higher adjustment. Separated or
divorced couples were not included in the samples of most
of the major studies. The major interest of these
38
investigators has been to develop reliable and valid mea
sures of marital adjustment, and Locke (1951) and Karllson
(1952) analyzed divorced, divorcing, or separated couples
with this purpose in mind.
The problems involved in translating complex social
phenomena into operational terms have hampered these
studies. It is obvious the levels of abstraction inherent
in concepts such as adjustment and success or failure in
marriage are several steps removed from observable behavior.
Consequently, these terms have been defined in many ways.
Self ratings on happiness scales (Davis, 1929; Landis,
1946; Terman, 1938), divorced and non-divorced couples
(Schroeder, 1939; Hart and Shields, 1926; Locke, 1951;
Karllson, 1952), and scores on devised marital adjustment
tests (Burgess and Cottrell, 1939; Terman, 1938; Terman and
Oden, 1950; Locke, 1951; Karllson, 1952; Burgess and Wallen,
1953) are illustrative examples of different ways of de
fining marital success.
The difficulties encountered in defining marital
success or failure are evidenced by the unimpressive pre
dictive power of marriage adjustment tests (Goode, 1964:
72-73; Kirkpatrick, 1963:381-383). Rather than focus on
39
family stability as the criterion of marital success,
permanence, happiness, fulfilling expectations of the com
munity, personality development, companionship, satisfac
tion with a marriage, integration in family life, and
adjustment have been used in a variety of combinations to
define the phenomenon. It is obvious that by using this
kind of definition of marital success, it is impossible to
discern the explanatory influence of marital adjustment on
family separation. One can argue, for example, that hap
piness, satisfaction, and companionship are independent
variables rather than being a part of the dependent
variable--success or failure in marriage. In fact that is
the position taken in this study. If this is the case,
continued analyses of marital success using methods de
veloped in prediction studies will yield minimal additional
knowledge regarding the relevance of marital adjustment in
the decisions of couples to separate from each other.
Many descriptive and correlational studies have
dealt more directly with marital instability than has
marital adjustment research. Occupation, income, educa
tion, age, and religious affiliations of marital couples,
along with childlessness, the duration of the marriage,
40
remarriage, in-laws, race, and the "third" party are illus
trative examples of variables that have been used to
describe and explain differential rates of divorce, sepa
ration, and desertion. The investigators, typically, have
used the cross-sectional information, and the primary
sources of data have been census and official records from
state, county, and municipal agencies.
Education, occupation, income, and other socio
economic variables, used singularly or in combinations,
have been demonstrated to be inversely related to divorce
and separation. Goode (1956), Glick (1957), Kephart (1955),
Udry (1966, 1967), Monahan (1955, 1961), and Bernard (1966)
are representative of these kinds of studies.
The investigations of Jacobson (1950, 1959), Mona
han (1955), and Goode (1956) have shown varying degrees of
association between the absence or presence of children and
divorce "proneness." The duration of marriage also has
some bearing on the occurrence of divorce or separation as
Glick (1959) and Jacobson (1959) have demonstrated.
The religious affiliation and the patterns of
religious behavior of marital partners and their relation
ship to family stability have been studied by several
41
researchers. Chancellor and Monahan (1955, 1961), and
Bernard (1966) are representative of these kinds of studies.
The study of four hundred women by Goode (1965) has
been recognized by most family sociologists as the best
empirical study of the subject. Many dimensions of family
disorganization were examined. Goode did make an attempt
to analyze the separation process although he did recognize
the limitations of his ex. post facto data. Five types of
separation were derived. One of these types— the rational
model--contained many notions that were similar to the con
ceptual framework underlying this study. In particular,
the concept of alternatives as a pivotal variable in the
separation process was given serious consideration. Un
fortunately, there has been minimal subsequent research of
some of Goode*s exploratory work. This may be due, in part,
to the difficult problems associated with following the
behavior of separating couples over time.
These research investigations, that have focused
primarily on aspects of desertion, separation, and divorce,
have provided valuable descriptive information regarding
many conjunctive and disjunctive factors that influence
family stability. The continued application of the
42
cross-sectional type of research using census and other
kinds of official data, undoubtedly, will yield additional
descriptive information about divorce and separation. The
findings of such studies, however, probably will not result
in a comprehensive explanation and prediction of marital
and family instability.
The need for new directions of research that would
focus on marital stability has been recognized by many
family sociologists, e.g., Hill (1964), Nye (1964), Goode
(1964), and Kirkpatrick (1963). There appears to be
general agreement that more comprehensive understanding,
explanation, and prediction of family separation depend on
the development and/or use of family interactional theory,
along with analyses of maladjusted couples using longitudi
nal research methods. The analysis of family separation,
as developed in this dissertation, is a small step in this
direction advocated by these social scientists.
This study of family separation differs from the
majority of previous research investigations dealing with
the subj ect matter in that:
1. A conceptual model of family interaction (the
developmental approach) has been used as an
explanatory base for the occurrence of family
separation;
Several variables having some relationship to
family separation have been incorporated into
the analysis as control variables; and
A longitudinal-type research design has been
used to analyze the process of family separa
tion.
44
Chapter II: Notes
1
The use of terms such as marital stability and
cohesiveness as applied to the family unit refers to the
unity of the system. These terms are the converse of those
terms in footnote 1 of Chapter I.
2
Zimmerman later changed his mind about the family
unit totally disintegrating. See Zimmerman (1960:11-13).
CHAPTER III
THE METHODOLOGY
This research is an exploratory, longitudinal field
study. It involve^ a systematic, empirical analysis of the
extent to which the degree of marital unhappiness and
alternatives to living together (the independent variables)
affect the occurrence, frequency, and the length of separa
tion (the dependent variable) among a selected group of
maladjusted married couples.
Such an investigation requires accurate classifica
tion, definition, and measurement of relevant variables, as
well as an appropriate selection of subjects and observa
tion of data. An accurate description of the procedures
used in this dissertation is presented in this chapter so
that other investigators of family separation may attempt
similar studies if they so desire.
Field studies have been discussed by many methodolo
gists (Rarlinger, 1966; Katz, 1953; French, 1953; Greenwood,
45
46
1945; Nye, 1964). This study combines both exploration and
"hypothesis testing," and it takes the form of a natural
"experiment" (Katz, 1953:78-79). This research is explora
tory because it is a beginning attempt to find out if
meaningful relationships exist between marital happiness,
alternatives and family separation. In addition, this re
search is one of the few studies (Hill, 1964:208) that
analyzes family separation using the conceptual framework
of the developmental approach.
Several propositions describing the nature of the
relationship between marital unhappiness, alternatives,
family developmental stages, and separation have been
derived from the conceptual framework underlying this
study. The propositions are presented in hypothesis form.
Because of the exploratory nature of this research, the
hypotheses have been developed primarily in order to
clarify the expected direction and form of the inter
relationships of the above variables. Tests of statistical
significance will not be made because the findings are not
generalizable due to sampling restrictions.
This chapter describes the research design, the
characteristics of the sample, measurement of the indepen
dent, dependent, and control variables, and the limitations
47
of the research.
Research Design
There are three basic types of field research
design— cross-sectional, ex. post facto, and projected. The
latter two are longitudinal in that they follow social
behavior over a specific period of time. Ex. post facto
research analyzes behavior after it has occurred; projected
research examines a research problem from the "present1 1 to
the "future." The projected approach has been used in this
study because it provides an opportunity for an observation
of the "natural" variations of the independent variables
(dissatisfaction or unhappiness and alternatives) over a
period of time as they relate to variations in the depen
dent variable (length of separation).
Three factors can create methodological problems
when a projected field study of this type is attempted.
They are: (1) the difficulty of controlling the innumerable,
complex variables intrinsic to any field setting; (2) the
need to develop means to observe accurately the ongoing
changes in the field setting; and (3) an accurate measure
ment of both the independent and dependent variables.
48
The design used in this exploratory study requires
the following elements: (1) a sample that will provide a
basis for an adequate investigation commensurate with the
goals of the study; (2) a procedure for observing the
"natural" variation of the independent variables (marital
happiness and alternatives) to a marriage; and (3) measure
ment over a sufficient period of time for the dependent
variable to be affected by the independent variables.
In keeping with this design, one hundred and twenty-
two couples were selected as subjects for the study. These
subjects were observed over a period of one year, during
which time marital unhappiness and alternatives to a mar
riage were measured on a before-and-after basis by a ques
tionnaire (Appendix A) that was developed for this study.
Implementing the Study
This field research was implemented by selecting a
sample of marital partners who were brought to the atten
tion of the Los Angeles County Conciliation Court. These
subjects were asked to complete a questionnaire when they
first came to the Court. Approximately one year later,
these marital partners were again required to complete a
follow-up questionnaire.
Selecting the Subjects
The marital partners used as subjects in this
analysis of family separation were selected from the popu
lation of married couples brought to the attention of the
Los Angeles County Consolidated Domestic Relations and
Conciliation Court. This public agency provides voluntary
short term marriage and family counseling services to any
legally married couple residing in Los Angeles County. The
counseling service is available to one or both marital
partners prior to and/or during their legal domestic rela
tions proceedings; it is not available to those couples
whose marriage has been terminated by a final divorce,
legal separation and/or annulment.
Conciliation Court couples are generally repre
sentative of an urban population that appears to exhibit
a wide diversification of demographic and sociological
characteristics. They include families in various phases
of the separation process, ranging from those marital
couples living together who have not legally filed for
divorce or conciliation services to those who have been
awarded interlocutory divorce decrees and are waiting for
final decrees.
50
No formal statistical informat ion is kept by the
Conciliation Court regarding the universe of couples who
contact the Conciliation Court; only selected descriptive
information is available on those couples who have filed a
conciliation petition for marriage and family counseling
services (Appendix B).
In order to make the study sample homogeneous, the
subjects were selected from that group of couples married
for the first time who had never been to the Conciliation
Court. Further, couples were selected on the basis of
their developmental family stage, in order to insure that
all stages of family development would be represented in
the sample.
The Conciliation Court counselors were advised of
the nature of the study, and its sampling procedures. The
counselors were then asked to administer a questionnaire to
any couple that met the criteria set forth when they served
their weekly stint as duty counselor.'*'
Within the procedures described previously, one
hundred and twenty-two sample subjects were selected for
this study. Table 1 presents the distribution of these
couples according to developmental family stages. Charac
teristics of the Conciliation Court population and practical
51
TABLE 1
FREQUENCY AND PERCENTAGE DISTRIBUTION OF SUBJECTS
ACCORDING TO FAMILY DEVELOPMENTAL STAGES
Family-
Developmental Frequency Ber cent
Stages
I 15 12.3
II 83 68.0
III 17 13.9
IV 7 5.8
Total 122 100.0
52
limitations in the field setting of this study resulted
in small sample sizes in stages one, three, and four.
Economic and time limitations prevented this investigator
the opportunity to increase the number of sample subjects
in these stages.
A brief comment regarding the procedure used in
this study to obtain the sample is in order. Recognition
is made that the ideal method for the selection of the
study sample would have been a stratified random procedure.
Again, limitations of cost and time, along with those pres
sures related to the Conciliation Court*s administrative
and routine business made stratified random sampling pro
cedures impractical. Consequently, given these field
situation circumstances, a purposive sampling method was
utilized in order to obtain a sufficient number of sample
subjects.
Characteristics of the Sample
In addition to the independent variables of marital
happiness and alternatives to a marriage, the number of
children, the age of the marital partners, income, edu
cation, and race are variables considered to have some
53
bearing on the occurrence of separation in the contemporary
American family. These characteristics are presented for
the sample subjects in Tables 2 through 8.
Table 2 presents information regarding the degree
of marital happiness measured when the subjects came to the
Conciliation Court and approximately one year later. The
sample, as would be expected, is biased toward marital
unhappiness.
Note that approximately 30 per cent of the subjects
report happy, above average, and very happy perceptions
about their marriage. This distribution might appear to be
unusual for a group of maladjusted couples; however, these
subjects most probably represent those marital partners who
were not dissatisfied with their marriage. It is also pos
sible that these subjects did not respond accurately to the
statement in the questionnaire regarding their perception
of happiness.
Of the subjects, 68 per cent perceived themselves
as unhappy or very unhappy when first coming to the Con
ciliation Court. Approximately one year later, the number
of unhappy subjects increased to slightly over 70 per cent.
The number of normal, happy, and very happy subjects de
creased accordingly.
54
TABLE 2
FREQUENCY AND PERCENTAGE DISTRIBUTION OF SUBJECTS
ACCORDING TO THE DEGREE OF MARITAL HAPPINESS
AT TIME 1 AND TIME 2 (ONE YEAR LATER)
Time 1 Time 2
Marital Frequency Per Frequency Iter
Happiness cent cent
Very Unhappy 45 36.9 56 45.0
Unhappy 38 31.1 31 25.4
Happy 31 25.4 24 19.5
Above Average
Happy 5 4.0 8 6.5
Perfectly Happy 3 2.6 3 2.6
Total 122 100.0 122 100.0
55
TABLE 3
FREQUENCY AND PERCENTAGE DISTRIBUTION OF SUBJECTS
ACCORDING TO THE NUMBER OF CHILDREN
Number of Children Frequency Iter cent
None 15 12.3
One 30 24.6
Two 35 28.7
Three 23 18.9
Four or More 19 16.5
Total 122 100.0
56
TABLE 4
FREQUENCY AND IERCENTAGE DISTRIBUTION OF SUBJECTS
ACCORDING TO AGE OF SUBJECTS
Age of Subjects Frequency Iter cent
Younger than 19 5 4.1
20-29 51 47.7
30-39 38 31.1
40-49 21 17.2
50-59 4 3.3
Older than 60 3 2.6
Total 122 100.0
57
TABLE 5
FREQUENCY DISTRIBUTION OF SUBJECTS ACCORDING TO AGE
AND DEVELOPMENTAL STAGES
Age of Subjects
Developmental Familv Staaes
I II III IV Total
Younger than 19 0 5 0 0 5
20-29 9 41 1 0 51
30-39 3 30 2 2 37
40-49 2 7 7 2 18
50 and Older 1 0 1 2 4
Total 15 83 11 6 115
G = .51
58
TABLE 6
FREQUENCY AND PERCENTAGE DISTRIBUTION OF SUBJECTS
ACCORDING TO INCOME
Income Monthly Frequency Iter cent
None 6 5.0
Less than $200 4 3.4
$201-$300 11 9.0
$301-$400 9 7.4
$401-$500 25 20.5
$501-$600 25 20.5
$601-$700 11 9.0
$701-$800 14 11.5
More than $801 17 13.7
Total 122 100.0
59
TABLE 7
FREQUENCY AND PERCENTAGE DISTRIBUTION OF SUBJECTS
ACCORDING TO EDUCATION
Grade Level Frequency Iter cent
0-6
7-9
10-11
High School
Business or
Vocational College
College 1 Year
College 2 Years
College 3 Years
College Graduate
College Post-Graduate
3
10
23
32
2
8
14
5
9
6
2.6
8.2
18.9
26.2
1.6
6.6
11.5
4.0
7.4
5.0
Total 122
60
TABLE 8
FREQUENCY AND PERCENTAGE DISTRIBUTION OF SUBJECTS
ACCORDING TO RACE OR ETHNIC GROUP
Race or
Ethnic Group Frequency Per cent
Negro 27 22.1
Mexican-American 10 8.1
Oriental 1 0.1
Caucasian 81 67.0
Not Stated 3 2.6
Total 122 100.0
61
Data regarding the number of children reported by
the sample subjects are presented in Table 3. The modal
and median numbers of children are two, with 28.7 per cent
of the sample having that number. These data are exactly
the same as Conciliation Court information (cf. 1964) re
garding the number of children in families seeking its
services. The sample data also compare favorably with in
formation concerning marital partners filing a complaint
for divorce in California (1967:46-47), with slightly over
20 per cent of these people reporting two children in the
family.
Table 4 presents information regarding age of the
sample subjects which closely parallels Conciliation Court
data regarding this attribute. For example, nearly 42 per
cent of the study sample were between the ages of twenty
and twenty-nine; Conciliation Court records (cf. 1964)
provide information that slightly over 44 per cent of their
female clients and approximately 31 per cent of the male
clients are in this age category. For persons filing for
divorce in California, the median age for women is thirty-
one, while slightly more than thirty-four is the median age
for men.
62
The relationship between age and developmental
stages is shown in Table 5. As would be expected, the
pattern of progression through the stages according to age
prevails, although there are a few cases of older subjects
in stage one and a small number of younger subjects in
stage four.
Monthly income and education level of the study
sample are given in Tables 6 and 7, respectively. These
distributions of attributes again roughly approximate in
come and educational information gathered by the Concili
ation Court. The educational level also is not too dis
similar from the educational information reported for
people filing divorce complaints in California (1967:36).
Finally, racial characteristics of the subjects
are shown in Table 8. This attribute indicates an over
represent at ion of Negro subjects and an underrepresenta
tion of Mexican-Americans and Caucasian subjects when the
data are compared to Conciliation Court and State of Cali
fornia information (1967:26) regarding race.
In summary, this study sample compares favorably,
for the most part, with a variety of social attributes
associated with the Conciliation Court population of
63
maladjusted marital partners. Further, the sample, to a
lesser extent, reflects broad characteristics of those
persons filing domestic relations complaints in California.
The sample subjects, when classified according to develop
mental family stages, are similar to the Court population
in a manner that resulted in small numbers of subjects in
every stage except stage two. The sample, although having
been selected by a non-random procedure, appears to be
fairly representative of a population that has been defined
in this study as maladjusted.
Measurement of the Independent. Control,
and Dependent Variables
The independent variables in this study of family
separation have been defined as marital unhappiness and
alternatives to a marriage. The control variables include
religiosity, deviance, number of children, and the socio
economic attributes of education and income. The dependent
variable is measured by frequency of separations and the
length of separation defined in patterns of no separation,
short separation, and long separation.
A five-point scale articulating a continuum of
marital happiness ranging from very unhappy to perfectly
64
happy was used to measure the degree of marital happiness.
This scale has been used in previous studies and has been
shown to have a reasonable degree of predictive validity
(Locke, 1951:391). Three measures of the degree of marital
happiness were taken--when the subjects first came to the
Conciliation Court (time 1), approximately one year later
(time 2), and the change in marital happiness over the
year.
Several questions regarding the presence of alter
natives to a marriage, and religious and deviant behavior
were asked in both the initial and follow-up question
naires. These statements were categorized and three sum-
mative type scales (Selltiz et al.. 1961:366-370) were de
veloped; religiosity, deviance, and alternatives were
measured by these scales.
Summative type scales are relatively simple in
their construction. Generally, subjects are asked to
respond to a series of items assumed to be related to the
phenomenon being measured. Each response is scored numeri
cally with all of the responses being summed; the summation
of the scores yields a subject*s total score regarding a
particular referent. The score is interpreted as repre
senting the favorable or unfavorable position of a subject
65
toward the referent. Selltiz et al. (1961:366) describe
the rationale of summative scales in the following manner:
The probability of agreeing . . . or . . . disagreeing
. . . varies directly with the degree of favorable -
ness . . . one could expect an individual to respond
favorably ... to many items favorable to the object
. . . disagree with many unfavorable items . . .
depending on the subject’s position regarding the ob
ject.
Five sets of statements regarding the religious
behavior were asked in the initial questionnaire. These
statements referred to the religious nature of the marriage
ceremony, the frequency of church attendance, the church
and Sunday school activities of the children, the attend
ance of children in parochial schools, and the desire on
the part of subjects for their children to attend parochial
schools.
The items for the religiosity scale were scored in
the following manner:
Respondents indicating they had a religious marital
ceremony were given a score of one; those not having such
a ceremony were given a score of zero. The range relative
to frequency of church attendance was from zero (no attend
ance) to four (bi-weekly church attendance); responses
regarding children’s church and Sunday school attendance
66
were scored in the same way. Items about parochial school
attendance ranged from zero (no children attending) to two
(all children attending). The desire for children to
attend parochial schools was judged to be zero when no
desire was indicated and one when the response was affirma
tive. The religiosity scale devised for this study is
listed in Appendix C.
Appendix C also presents the scale developed to
measure deviance. It is comprised of four items: the
respondent’s perception of personal problems; the respon
dent’s perception that the spouse has problems; the number
of non-traffic criminal arrests; and the respondent’s per
ception of problems associated with their children.
The statements that deal with deviant behavior were
scored in the following manner: perception of personal
problems ranged from a score of zero (no problems) to two
(any combination of yes responses to alcoholic, drug, and
emotional problems); responses regarding problems of the
spouse were scored in like fashion. The item about crimi
nal behavior was measured by giving a score of one for
every reported non-traffic arrest. Finally, perceived
behavioral problems of the children ranged from a score
67
of zero (no problems) to three (any combination of delin
quent and school behavioral difficulty).
Alternatives to a marriage were measured by de
veloping a scale comprised of items referring to residen
tial, occupational, and income and wealth situations that
would enable an unhappy marital partner to live apart from
his (her) spouse. These items categorize with whom a
separated person is residing, residential possibilities
perceived to be available to a separating person, and the
manner by which a separated person is being supported.
A negative response to the items was scored as zero, while
a positive answer was given a score of one. Appendix C
presents the alternatives scale.
The dependent variable— family separation— was
measured in two ways. The occurrence and frequency of
separation were measured by tallying the number of times
a subject reported living apart from his (her) spouse over
a time period of one year.
Three patterns based on the length of separation
were the second dimension of separation behavior analyzed
in this research. A no-separation pattern is self-
explanatory. A short separation pattern indicated that
68
a subject lived apart from his (her) spouse less than two
months, while a long pattern was indicative of a separation
exceeding two months.
Limitations of the Study
Ideally, adequate testing of the several explora
tory hypotheses regarding aspects of separation— frequency
and patterns— should include the following conditions:
(1) the selection of a sample derived through a stratified
random procedure large enough to insure a meaningful sta
tistical analysis; (2) adequate controls to insure that all
subjects entered the time period in the same condition
(not separated) with regard to those factors directly
pertinent to separation outcome; (3) adequate conceptuali
zation and measurement of the independent variables in
fluencing the dependent variable— the frequency and occur
rence of separation patterns; and (4) a sufficient period
of time to maximize the reliability of results. The
limitations of this research will be discussed in the light
of these major criteria.
Practical limitations of cost and time, along with
administrative and functional pressures peculiar to the
Los Angeles County Conciliation Court, an agency unaccus
tomed to research, necessitated the use of non-random
sampling techniques to select the sample. Usually, such a
sampling procedure, relying on a subjective procedure of
"hand picking" the subjects according to needs of the
study, results in a sample that is not representative of
the population being analyzed. In this case, a fair degree
of representativeness was achieved using a non-random
sampling method. An attempt was made to include all sub
jects that were identified as stage one, three, and four
cases during the period of time the data were being col
lected. A major limitation, however, was an under-sampling
of subjects representing developmental stages one, three,
and four which indirectly was the result of demographic
characteristics peculiar to the Conciliation Court popula-
2
tion. The small numbers of subjects in these stages
diminished the confidence with which statistical procedures
used to investigate the findings could be interpreted and
reported.
In general, the total sample size was not adequate.
Again, the practical considerations discussed previously,
along with the exploratory nature of the study, governed
70
the size of the sample selected for this analysis. How
ever, this was a limitation and future investigators
ideally should select a stratified random sample with pro
visions made to oversample subjects that represent family
stages one, three, and four. Each stage should include a
minimum of 150 subjects for this type of analysis.
An attempt was made to select only subjects mar
ried for the first time who also had never lived apart from
their spouses. For the most part, this condition was
satisfied although the control for such a condition was not
as rigorous as it should have been. The reliance on Court
counselors to obtain sample subjects again was a decision
dictated by practical considerations. A more efficient and
feasible method should tie the research to some aspect of
the Court procedure.
Some comment should be made regarding the effects
of sampling bias. The study sample probably is fairly
representative of those maladjusted marital partners inter
ested in maintaining the marriage. However, the character
istics of the Conciliation Court population (see Appendix B)
did not provide the opportunity to obtain many stage one,
three, and four cases during the time the data were
71
collected. In essence, the sampling bias contributed to
a fairly representative sample of Conciliation Court
couples. On the other hand, the bias probably was a con
tributor to the small number of cases in all of the family
stages except stage two.
Two important limitations regarding the content and
techniques for measuring the alternatives to a marriage
should be dealt with in any future study of family separa
tion. First, with regard to content, alternatives, as
defined in this study, did not include those situations
associated with an alternative to the spouse. Because of
the sensitive nature of this aspect of marital conflict,
information is very difficult to obtain. It becomes un
realistic to expect to obtain such information from a group
of subjects coming to the Conciliation Court, which is an
adjunct of the Domestic Relations Judicial system. In
order to obtain extra-marital information related to the
concept of alternatives, a method should be devised to
utilize information gleaned from the counseling inter
views. Such a procedure was not developed for this study.
Secondly, the type of scaling technique used to develop
the alternative scale does not provide an opportunity
72
to distinguish between universalistic and particularistic;
the former refer to alternatives present in all develop
mental stages, the latter cite those alternatives peculiar
to specific family stages.
Finally, all of the scales developed in this study
lack specificity. They do not provide the means to dis
criminate differential weightings of characteristics within
an item. For example, one of the alternative items in the
scale dealt with wealth. A subject was given a score of
zero if the response was negative and a score of one if the
answer was. affirmative. It is apparent there are degrees
of wealth, many of which provide greater opportunities for
a person to separate than do others. This qualitative
aspect of the alternative items was not measured in this
research.
The discussion, in essence, points to the limita
tions of summative type scales. The scales seem appropriate
because of the exploratory nature of this research, but
sophisticated techniques (such as a Ukert or Gutman Scale)
might have provided a better means to measure alternatives,
religiosity, and deviance.
In summary, a broader range of alternatives should
be included in future studies of the process of family
73
separation. While summative scaling procedures were of
value in this exploratory study, subsequent research should
utilize more sophisticated techniques to insure more valid
indices of the phenomena being measured and analyzed.
One year appears to provide sufficient time to
observe the process of family separation. However, before-
after measurements prevent a comprehensive observation of
the influence of marital happiness and alternatives (the
independent variables) on separation. For example, a sub
ject reporting happiness at time 1 and time 2, along with
a separation, might very well have been unhappy at the
time of the separation. A more continuous observational
system, i.e., quarterly measurements would afford a more
adequate assessment of the separation process than the
before-after technique used in this study.
Summary of Methodology
A projected design was used in this field study to
examine the influence of marital happiness and alternatives
to a marriage on the occurrence, frequency, and length of
family separation. A purposive sampling procedure was used
to select one hundred and twenty-two subjects married for
74
the first time who were referred to the Los Angeles County
Conciliation Court.
Data were collected from these subjects when they
first came to the Court and approximately one year later.
The items in the questionnaires, used in the collection
of data, were coded, the relevant variables were measured,
and data were analyzed.
The limitations of the study were discussed. Not
withstanding many practical considerations against a com
plete implementation of an ideal design, the study did
provide an opportunity to analyze the separation process
and provide tentative conclusions regarding the importance
of marital happiness and alternatives in determining the
frequency and length of separation.
75
Chapter III: Notes
^The primary function of the duty counselor is to
explain the conciliation service to marital couples coming
to the Conciliation Court for the first time.
9
These stages are underrepresented in the Concili
ation Court population.
CHAPTER IV
DISCUSSION OF THE FINDINGS
One hundred and twenty-two marital partners seeking
marriage and family counseling from the Los Angeles County
Conciliation Court were observed over a period of one year
with reference to their separation behavior. This chapter
will report and evaluate the findings relative to the fre
quency and length of separations among this group of sub
jects.
The findings were analyzed from three perspectives.
In the first instance, a preliminary analysis was made by
examining the zero order relationships of all the major
variables to each other and to the frequency and length of
separation. This type of analysis was made primarily be
cause of the exploratory nature of the study. Such an
evaluation of the data also provided an additional base in
the interpretation of some of the relationships set forth
in eight exploratory hypotheses.
76
77
These hypotheses describing the relationship between
marital happiness, alternatives to a marriage, and develop
mental stages to the frequency and length of separation
were examined using three measures of association; namely,
lambda, Gamma, and Baarsonian r. The primary focus of
these analyses is to confirm the magnitude and forms of the
relationships, rather than the statistical significance of
the relationships. Tests of significance were unwarranted,
not only because of the exploratory nature of the study,
but also because the sample was not selected through random
procedures.
The third approach to the data was an examination
of the major variables with reference to their "power” in
predicting the frequency and length of separation. A re
gression analysis was the method used in this kind of
evaluation.
Results
Two sets of data in Tables 9 and 10 present various
aspects of the preliminary correlation analysis. Table 9
shows the relationship of selected variables to the inde
pendent and dependent variables of this study: frequency
TABLE 9
SELECTED VARIABLES AND THEIR PEARSONIAN CORRELATIONS TO THE THEORETICAL VARIABLES
Number of
Separations
Length of
Separation
Developmenta1
Stages
Number of
Alternatives
Marital
Happiness
Religious Ceremony -.048 -.003 .155 .070 .020
Frequency of Church Attendance .029 -.059 -.092 -.070 -.077
Number of Months Pregnant -.007 .003 .110 -.074 -.077
Church School -.157
1
•
* - »
00
.170 -.200 .026
Children Attending Parochial
School -.082 -.082 .033 -.058 -.128
Would Send Children to
Church School -.094 -.125 -.059 -.077 .082
Summative Religiosity Scale -.061 -.131 .072 -.151 -.070
Do You Have Problems .099 .117 .193 .106 -.169
Spouse Problems .057 .155 .062 .022 .034
Number of Arrests -.107 -.043 -.018 -.024 .023
Children Trouble .015 -.059 .415 -.247 -.098
Summative Deviance Scale .014 .056 .287 -.047 -.086
Family Size Before .032 .062 -.137 .146 .028
Age -.032 -.037 .222 -.201 -.095
Schooling -.013 .04 -.213 .057 .023
Employment -.170 -.144 -.037 .008 .086
Income -.089 .112 -.050 -.118 -.013
'vl
00
TABLE 10
79
SELECTED VARIABLES THAT ARE ASSOCIATED WITH SEPARATION OF
FREQUENCY AND LENGTH WITH PEARS ON IAN CORRELATIONS
GREATER THAN .20
Frequency of Length of
_______________________________Separations_______Separation
Mental Adjustment After -.328 -.317
Change in Mental Adjustment -.294 -.316
Schooling
Employment
Income
If Apart: Living With You
Children .280 .271
Relatives .204
Pregnant
If Apart: Children Support
Voluntary Ihrtner
Court Order . 247
Rib lie Assistance
Source Income if Not Working
Spouse's Income -.224
If Apart: How Supported
Self .207 .213
Partner Voluntary
Court Order .209
Riblie Assistance
Summative Alternative Scale .281
80
of separation, length of separation, the number of alterna
tives, developmental family stages, and marital happiness.
Variables associated with the frequency and length of sepa
ration are found in Table 10.
These data are presented for informational purposes
only insofar as the majority of relationships are so small.
Using a correlation of .20 as an arbitrary cutoff point,
there are only six sets of relationships in this category
in Table 9 data, and the sets increase to thirty-six in
Table 10 data.
The first hypothesis deals with the relationship of
marital happiness and frequency of separation. Concep
tually, the degree of marital happiness has been viewed as
an antecedent condition of separation. This relationship
has been analyzed four ways, and the data are presented in
Tables 11,12, 13, and 14.
Data on the relationship between marital happiness
at time 1 and the frequency of separation are found in
Table 11. It is clear that the relationship is very weak
with a Gamma coefficient of .099 and an r of .038. Based
on this information, one could question the importance of
marital happiness to the frequency of separation. As such,
TABLE 11
FREQUENCY DISTRIBUTION OF SUBJECTS ACCORDING TO
MARITAL HAPPINESS AT TIME 1 AND
NUMBER OF SEPARATIONS
Number of
Alternatives
Very
Unhappy Unhappy
Above
Normal Average
Very
Happy Total
None 14 11 11 1 1 38
One 28 21 15 1 2 67
Two 0 5 3 2 0 10
Three 2 1 1 0 0 4
Four 0 0 1 1 0 2
Five 0 0 0 0 0 0
Six 1 0 0 0 0 1
Total 45 38 31 5 3 122
Gamma = .099
r = .038
00
TABLE 12
FREQUENCY DISTRIBUTION OF SUBJECTS ACCORDING TO
MARITAL HAPPINESS AT TIME 2 AND
NUMBER OF SEPARATIONS
Number of
Alternatives
Very
Unhappy Unhappy
Above
Normal Average
Very
Happy Total
None 8 9 14 5 2 38
One 40 15 8 3 1 67
TWo 4 4 2 0 0 10
Three 2 2 0 0 0 4
Four 1 1 0 0 0 2
Five 0 0 0 0 0 0
Six 1 0 0 0 0 1
Total 56 31 24 8 3 122
Gamma = -.515
r = -.328
83
TABLE 13
FREQUENCY DISTRIBUTION OF SUBJECTS ACCORDING TO
CHANGES IN MARITAL HAPPINESS AND
FREQUENCY OF SEPARATION
Frequency of
Separation
Negative
Change
No
Change
Positive
Change Total
None 10 9 19 38
One 25 27 15 67
Two 8 2 0 10
Three 2 0 2 4
Four 2 0 0 2
Five 0 0 0 0
Six 0 1 0 0
Total
47 39 36 122
Gamma = -.387
r = -.294
84
TABLE 14
FREQUENCY DISTRIBUTION OF SUBJECTS ACCORDING TO
TIME 1 AND TIME 2 MEASURES OF MARITAL HAPPINESS
AND FREQUENCY OF SEPARATION
Marital
Happiness
Time Time
1 2
Frequency of Separation
Total 0 1 2 3 4 5 6
U U 8 39 5 3 0 0 1 56
H U 6 13 3 1 2 0 0 25
U H 13 7 0 0 0 0 0 20
H H 7 6 2 0 0 0 0 15
Total 34 65 10 4 2 0 1 116
U = Unhappy
H = Happy
Gamma = -.41
85
this set of data confound both conceptual and common sense
notions regarding the importance of marital happiness to
frequency of separation.
Table 12 presents the data at time 2, regarding
the relationship of the variables being analyzed. Note
that fairly sizable correlations are present (Gamma = -.515
and r = -.328). They suggest, however, that marital hap
piness is affected by separation rather than causing it,
since time 2 measurement, rather than time 1, is predic
tive.
Table 13 data, which show the relationship between
changes in marital happiness and separation frequency,
again appear to refute the causal importance of marital
happiness. In other words, the direction of the relation
ship is contrary to the notions developed in the conceptual
framework in this dissertation. However, with subjects who
have had a change from marital unhappiness to happiness,
over one-half (53 per cent) did not separate. This is not
true for persons who did not manifest a change in marital
happiness or for those subjects who had a negative change
in marital happiness.
Finally, Table 14 presents data that consider be
fore and after measures of marital happiness and frequency
86
of separation. Four types of respondents are analyzed:
unhappy-unhappy; happy-unhappy; unhappy-happy; and happy-
happy. Again, a relationship exists (G = -.41), and the
hypothesis is supported but it is difficult to establish a
causal order of the variables.
An interpretation of these data regarding the re
lationship between marital happiness and frequency of
separation is difficult. On the other hand, there are
several issues which might clarify the ambiguity of the
data. First of all, the low correlation involving the
"before" measure of marital happiness does not rule out its
relationship to separation. The fact that someone is un-
happy with his (her) marriage does not mean that separation
is imminent. It is plausible that, over time, an unhappy
marital partner will attempt to separate from his (her)
spouse. In fact, this is suggested when the before and
after measures of happiness are considered simultaneously
(Table 14). Second, the data suggest that happiness is an
effect rather than a cause of separation. This ambiguity
might spring from two sources: either the fact that a per
son happy with the marriage has been separated which makes
him unhappy, or that a person is unhappy with the marriage
and also is unhappy with the separation. A third problem
with interpreting these data is the fact that the measures
of happiness were taken at a one year interval. It could
very well be that the separation reported occurred during
a period of time when subjects were not measured for hap
piness. Finally, the nature of the sample subjects may be
such that a true reading of the relationship may be diffi
cult. Conciliation Court cases, for example, represent,
for the most part, voluntary clients; those couples that
do not come to the court may be disproportionately those
that are representative of the extremely unhappy couples.
In other words, the sample of this study, because of its
nonrandomness, may not have permitted an adequate assess
ment of the relationship between marital happiness and
separation because it excluded important cases.
In summary then, when before and after measures of
marital happiness are considered, confirmation of the
hypothesis is suggested. On the other hand, the temporal
ordering of the relationship is cloudy and a more adequate
sample of subjects and a better measurement design is
needed to clear up ambiguities in this area.
Five hypotheses specify the relationship of the
number of alternatives to family developmental stages.
88
The relationships are examined by comparing the mean number
of alternatives occurring in the stages. The findings are
shown in Table 15.
It has been postulated in hypothesis II that stage
one marital partners have more alternatives than those
persons in any of the other developmental stages. The data
partially confirm the direction of the stated relationship.
The number of alternatives is greater in stage one than in
stages three and four. The exception is stage two which
manifests the largest mean number of alternatives.
Hypotheses III and IV state that stage three mari
tal partners will have more alternatives than those persons
in stages two and four. The mean number of alternatives
in stage three (1.27), in fact, is lower than the means of
alternatives in stage two (2.12) and stage four (1.55).
In hypotheses V and VI, it has been assumed that
no difference exists in the number of alternatives between
marital partners in stages two and four, and that the same
condition prevails between marital partners in stages one
and two. The data "speak1 1 in opposition to these postu
lates. A difference of .80 in the mean number of alterna
tives occurs in comparing stage one and three subjects,
89
TABLE 15
FREQUENCY DISTRIBUTION OF SUBJECTS ACCORDING TO
DEVELOPMENTAL FAMILY STAGES AND
NUMBER OF ALTERNATIVES
Number of
Alternatives
None
One
Two
Three
Four
Five
Six
Seven
Developmental Stages
4
3
2
2
2
2
0
0
11
22
20
7
12
11
7
1
3
III
5
2
1
2
1
0
0
0
TV
2
1
3
0
0
0
0
0
Total
33
26
13
16
14
9
1
3
Total
Mean
15
2.07
83
2.12
11
1.27
7
1.55
116
90
while the difference between stage two and stage four mari
tal partners is .57.
As was the case with the first hypothesis, the data
are not easy to interpret. The small sample sizes in
stages two, three, and four should be a sufficient warning
to approach any interpretation with caution, and the find
ings should be understood in the context of a suggestive
trend rather than confirmed evidence.
The findings, while not confirming the hypothesized
relationships, did suggest that a relationship exists be
tween alternatives and developmental family stages. Since
these stages are developmental, the age variable may have
some meaningful influence on the number of alternatives
available to a marital partner. A review of the relation
ship between age and categories of alternative variables
provided information that suggests the importance of rela
tives serving as alternative situations that can be
utilized by younger persons considering separation. As a
person becomes older, of course, the availability of rela
tives becomes increasingly improbable. The findings could
be an indication that young persons considering separation
have more access to an alternative than do older persons.
This aspect of family life— a modified extended family
system--has perhaps been overlooked by family sociologists.
However, the interpretation given to these findings would
tend to confirm notions by Litwak (1959) and Sussman (1953)
about contemporary family life. These sociologists main
tain that the interpersonal aspects of the extended family
system have been maintained and, in times of crisis such as
separation, the kinship group comes to the aid of its mem
bers.
Finally, the youth of a separating person is an
asset, particularly in the American society, when one con
siders alternatives to his marriage. For example, employ
ment of a potential husband or wife is more accessible to
young people, while these opportunities become increasingly
difficult for older individuals to obtain.
A direct relationship between the number of alter
natives among unhappy marital partners and the number of
separations is expressed in hypothesis VII. In this case,
the relationship is confirmed and the hypothesis is sup
ported. Both Gamma (.31) and ifearsonian r (.28) indicate
the existence of a moderately positive association between
these two variables. Table 16 presents the findings.
92
TABIE 16
FREQUENCY DISTRIBUTION OF UNHAPPY SUBJECTS ACCORDING TO
THE NUMBER OF ALTERNATIVES AND NUMBER OF SEPARATIONS
Number of
Separations
Number of Alternatives
0 1 2 3+ Total
None 19 7 1 0 27
One 1 9 9 40 59
Two 5 1 2 0 8
Three 1 3 0 0 4
Four 1 1 0 0 2
Five
Six
0 0 0 1 1
Total 27 21 12 41 101
Gamma = .31
r = .28
93
The last hypothesis (VIII) specifies that the num
ber of alternatives is associated directly with the length
of time a marital partner will live apart from his (her)
spouse. The relationship is examined generically, as well
as in the context of family developmental stages. The
findings are listed in Tables 17, 18, 19, 20, and 21.
Strong support exists for the generic hypothesis.
The Gamma coefficient is .88. An inspection of Table 17
reveals that with no alternatives, the modal category is
no separation among unhappy subjects. On the other hand,
with one or more alternatives, the modal separation pattern
is a long one. The relationship is most evident with sub
jects having three or more alternatives; no one in this
latter category reported no separation, three indicated a
short separation, while thirty-eight subjects separated for
more than two months.
When the relationship is examined within develop
mental stages, support also is maintained for the hypo
thesis. Unhappy subjects within each of the stages tended
to separate for long periods of time when they had alter
natives. Note, that when one controls for family stages
(see Tables 19 through 21), the Gamma and lambda coeffi
cients measuring the relationship between alternatives and
94
TABLE 17
FREQUENCY DISTRIBUTION OF UNHAPPY SUBJECTS ACCORDING TO
THE NUMBER OF ALTERNATIVES AND LENGTH OF SEPARATIONS
length of
Separation
Number of Alternatives
0 1 2 3+ Total
None 19 7 1 0 27
Short 7 4 1 3 16
long 1 10 9 38 58
Total 27 21 12 41 101
|Gamma = .88
Lambda = . 45
95
TAB IE 18
FREQUENCY DISTRIBUTION OF DEVELOMENTAL STAGE ONE SUBJECTS
ACCORDING TO NUMBER OF ALTERNATIVES AND
LENGTH OF SEPARATIONS
Length of
Separation
Number of Alternatives
Total 0 1 2 3 +
None 2 1 0 0 3
Short 0 0 0 0 0
Long 0 1 2 6 9
Total 2 2 2 6 12
Gamma = +1.0
lambda = .67
96
TABLE 19
FREQUENCY DISTRIBUTION OF DEVELOPMENTAL STAGE TWO SUBJECTS
ACCORDING TO NUMBER OF ALTERNATIVES AND
LENGTH OF SEPARATIONS
Length of
Separation
Number of Alternatives
Total 0 1 2 3+
None 14 6 1 0 2
Short 6 3 1 3 13
long 0 8 4 28 40
Total 20 17 6 31 74
Gamma = .81
Lambda ' = .46
97
TABIE 20
FREQUENCY DISTRIBUTION OF DEVELOPMENTAL STAGE THREE SUBJECTS
ACCORDING TO NUMBER OF ALTERNATIVES AND
LENGTH OF SEPARATIONS
Length of
Separation
Number of Alternatives
0 1 2 3+ Total
None
Short
long
0 2
1
5
Total 8
Gamma = +1.0
Lambda = . 67
98
TABLE 21
FREQUENCY DISTRIBUTION OF DEVELOPMENTAL STAGE FOUR SUBJECTS
ACCORDING TO NUMBER OF ALTERNATIVES AND
LENGTH OF SEPARATIONS
Length of
Separation 0
Number of
1
Alternatives
2 3+ Total
None 1 0 0 0 1
Short 0 1 1 0 2
Long 1 0 2 1 4
Total 2 1 3 1 7
Gamma = .64
Lambda = .33
99
separation length in stages one (1.0 and .67) and three
(1.0 and .67) are identical, and a similarity exists be
tween the coefficients of stage two (.81 and .46) and
four (.64 and .33). These findings suggest a similarity in
the viability of alternatives in stages one and three, as
well as stages two and four. This trend may be indicative
of the kinds of separation patterns that may be character
istic of the different developmental stages. Of course,
the interpretations associated with these data must be
viewed with caution because of the very small number of
subjects that comprised the family stage samples generally,
and stages one, three, and four specifically.
Notwithstanding the sampling limitations, the evi
dence in support of hypothesis VIII is strong, particularly
the generic association between alternatives and separation
pattern. The gamma coefficients range from .64 to 1.0.
It seems clear that the influence of alternatives in the
separation process has been demonstrated.
A stepwise regression technique also was used to
examine the relationships among the variables. Essentially,
this method enables one to select out those variables that
are most associated with a dependent variable. The
100
variables are selected in the order of their predictive
priorities. The most predictive variable is then selected
out and controlled, at which time the items again are
ordered and the second most predictive variable is selected
out. The remaining variables are again processed in this
manner, controlling for those variables already selected.
This procedure continues until the predictive utility of
all the variables have been analyzed, along with their
statistical contributions to an explanation of the variance
of the criterion variable being examined.
Two ten-variable stepwise regression models were
developed. The first one was used to predict frequency of
separation, and the second was used to predict length of
separation. The findings of the frequency of separation
regression analysis are listed in Table 22.
The relationships of fifty-four variables'*' to num
ber of separations were examined in the stepwise regression.
The findings are very supportive of the conceptual frame
work underlying this study. The three most predictive
variables are the measurement of marital happiness (taken
at time 2), alternatives as measured by the summative
scale, and developmental stages. These variables accounted
101
TABLE 22
STEEWISE REGRESSION: PREDICTING NUMBER OF SEPARATIONS
Step
Number Variable Entered
Cumulative Increase
RZ in
1 Marital Happiness Time 2
2 Summative Alternatives
Scale
3 Developmental Cycle
4 Number of Children
5 Children Supported
Voluntarily
6 Relatives in Home
7 Employment
8 Relatives in Home,
if Apart
9 Court Order for Support
10 Marital Happiness Time 1
.1066
.1409
.1710
.2054
.2277
.2427
.2607
.2832
.3023
.3225
.1066
.0343
.0301
.0344
.0224
.0150
.0180
.0225
.0192
.0202
Variance Added by
29 Additional Steps .4163 .0938
102
for 20 per cent of the variance, while the remaining seven
variables explained another 20 per cent. These findings
are encouraging evidence of the validity of the conceptual
explanation of separation in that all of ten most predictive
items are dimensions of marital happiness, alternatives to
a marriage, and developmental cycle. None of the control
variables— e.g., deviance, religiosity, age, etc.— emerged
as predictive in the first ten steps of the regression
analysis.
The findings for the stepwise regression analysis
of length of separation are presented in Table 23. The ten
most predictive variables explain 70 per cent of the vari
ation in length of separation. The variable most predic
tive of separation length is the ability of a person to be
self-supporting. This item explains 55 per cent of the
variance. The contributions of the remaining nine most
predictive variables are approximately 11 per cent. Of
the ten most predictive variables, four are dimensions of
alternatives, and the remaining six are items associated
with religion, pregnancy, and developmental stages. Five
of the items are related to economic support, suggesting
that this variable is a very powerful variable influencing
103
TABLE 23
STEIWISE REGRESSION: PREDICTING LENGTH OF SEPARATIONS
Step Cumulative Increase
Number Variable Entered R^ in R
1 If Apart, Self-Supported .5555 .5555
2 Public Assistance Support .5883 .0328
3 Voluntary Support .6069 .0186
4 Welfare .6235 .0166
5 Relatives Support .6398 .0163
6 How Far Pregnant Before
Marriage .6453 .0055
7 Children Attend Ihrochial
School .6493 .0040
8 Children in Trouble .6580 .0087
9 Developmental Stage .6629 .0049
10 Pregnancy .6668 .0039
Variance Added by
29 Remaining Steps .7049 .0381
104
the length of time an unhappy person lives apart from his
(her) spouse. Note that the degree of marital happiness
has no predictive utility in the explanation of length of
separation; it was the most predictive item when the fre
quency of separation was analyzed. The findings of this
analysis of separation length provide suggestive evidence
that alternatives to a marriage make up perhaps the most
important factor contributing to separation durability.
Self-support appears to be the critical alternative since
its contribution to the explained variance was much larger
than any other factor.
Summary
This chapter has presented the major findings of
this exploratory study of family separation behavior.
Eight exploratory hypotheses were evaluated. In addition,
a stepwise regression analysis was used to predict the
frequency and length of separation.
The empirical findings suggest that support be
given for the theoretical importance that has been placed
on the role that both marital happiness and alternatives
play in explaining the frequency of separation. Variables,
such as religiosity and deviance, were not found to be
105
of major significance in the examination of this dimension
of the separation process of the subjects of this study.
On the other hand, economic alternatives appear to
be the most important factors that contribute to the length
of time maladjusted marital partners live apart from each
other. There is a suggestion that the presence and age of
children (developmental stages) are factors to consider in
the occurrence and frequency of separation, but this
variable diminishes in predictive importance when length
of separation is considered.
Variables such as religiosity and deviance then
were not found to be of major significance in the separa
tion process. They did not demonstrate any predictive
usefulness with reference to frequency of separation.
There was only minimal suggestive evidence that aspects of
religiosity and deviance contribute to the length of time
an unhappy subject separates from his (her) spouse.
106
Chapter IV: Notes
^Illustrative items included the number and age of
children, before and after perception of happiness, indi
vidual and scaled alternatives, religious and deviant
characteristics, and demographic attributes, e.g., age,
race, income, and education.
CHAPTER V
SUMMARY AND CONCLUSIONS
Two dimensions of family separation behavior were
analyzed in this exploratory field study. The develop
mental approach was used as the conceptual base of this
research. The degree of marital happiness and alternatives
to a marriage were defined as the independent variables,
while the frequency and length of separation were viewed as
the dependent variables. Eight exploratory hypotheses were
developed to examine the extent to which marital happiness
and alternatives affected the frequency and length of time
an unhappy marital partner lived apart from his (her)
spouse.
The study was carried out with the approval and
cooperation of the Los Angeles County Conciliation Court
which is a part of the judicial system of the Superior
Court of the State of California. One hundred and twenty-
two subjects who were married for the first time were
107
108
selected on a non-random basis to serve as subjects for
the study. They were selected on their initial visit to
the Conciliation Court.
Ttoo questionnaires were developed to collect the
data for the study. An initial questionnaire was adminis
tered to the subjects when they first came to the Concili
ation Court. They completed a follow-up questionnaire
approximately one year later.
Indices of the number and length of separations
occurring in the sample of subjects were then compared.
The findings were analyzed in three ways. First, a pre
liminary correlational analysis was made which examined the
zero order relationships of all the variables to each other.
The second method of analyzing the findings evaluated the
eight exploratory hypotheses using Lambda, Gamma, and
Baarsonian r measures of association. The third approach
was a regression analysis of the findings. Two, ten-
variable stepwise regression models were developed to ex
plain the frequency and length of separation.
Generally, the empirical findings of this study are
supportive of the conceptual framework explaining the sepa
ration process, as it occurs in the contemporary American
109
family, when consideration is given for the sampling and
methodological limitations of the research design. The
findings are summarized in the following manner.
1. The contention that marital happiness and
alternatives to a marriage are primary factors affecting
the separation process is supported. The degree of marital
happiness appears to be an antecedent necessary condition
that must exist if a person is to separate from his (her)
spouse. On the other hand, marital unhappiness does not
appear to be a necessary contributor to the length of time
maladjusted marital partners live apart from each other,
once a separation has occurred; marital unhappiness may be
considered a sufficient but not necessarily a condition of
separation length.
2. The findings suggest that alternatives to a
marriage assume a sufficient but not necessary position in
the explanation of the frequency of separation. When the
phenomenon of separation length was examined, however,
alternatives appear to be a necessary condition that must
be met if the separation is to be a durable one. Those
hypotheses dealing with separation patterns and the regres
sion analysis of length of separation are strongly sup
portive of the primary influence of alternatives. Marital
110
unhappiness, on the other hand, was not one of the ten most
predictive variables to emerge in the regression analysis
of separation length.
3. The findings of this study did not support
those hypotheses that were developed under the assumption
that a relationship existed between alternatives and the
stage of family development. It was assumed that the
isolating characteristics of the nuclear family would ham
per the opportunity of marital partners responsible for the
care of young children to acquire alternatives. The find
ings do not support this contention. Rather, it is sug
gested that the age of the marital partner is related to
the availability of alternatives. Although this study was
not designed to assess the kinds of alternatives occurring
in the different family stages, the opportunity to find
employment and rely on the support of relatives is weighted
heavily on the side of young persons considering separation.
While the number of alternatives favor younger rather than
older marital partners, there is suggestive evidence that
alternatives are more viable or stronger in stages one and
three, rather than stages two and four. The statistical
relationships between alternatives and separation length
Ill
are identical in stage one and three, and similar in stages
two and four. The evidence must be looked upon as sug
gestive, however, because of the small samples that com
prised the developmental stages.
4. The influence of variables such as religiosity,
deviance, income, and education was not found to be of
major predictive importance in this analysis of separation
behavior. The correlational analysis did not indicate an
association of these variables to either frequency or sepa
ration durability. Further, they were not considered in
the analysis of the eight exploratory hypotheses. It was
assumed that their importance would be demonstrated when
the findings were examined using a regression analysis.
None of these variables demonstrated any predictive utility
when frequency of separation was analyzed. The involvement
of children in parochial school (a religiosity item) and
trouble with children (a deviance item) were predictive of
separation length, but their combined predictive utility
explained only approximately .02 of the variance.
Future Research
This exploratory research effort attempted to
examine the explanatory effects of the degree of marital
112
unhappiness and alternatives to a marriage on the occur
rence, frequency, and length of separation among a group of
maladjusted couples. In addition, an effort was made to
evaluate the contributions of religiosity, deviance, and
other variables assumed to contribute to these two dimen
sions of family instability. The implementation of this
study brought into focus many conceptual and methodological
problems. These problems, of course, are not peculiar to
the study of family separation behavior. Further research
is indicated in the following six areas.
1. A more specific clarification of the develop
mental stage framework is needed to improve the reliability
and validity of the major and subsidiary variables that
have been assumed to be primary factors in explaining fre
quency and length of separation. Future research should
attempt to define family stages in a more mutually exclusive
typology than was done in this study.
2. The limited operational definitions of the inde
pendent variables should be expanded in future research
that would provide an opportunity to include their multi
dimensional characteristics. The definition of alterna
tives, for example, should include factors that are
113
associated with a spouse substitute, as well as structural
items. The same comments are pertinent to the other vari
ables that were defined, scaled, and measured in this study.
Future efforts also should include a social position
measurement, in order to evaluate the contribution of this
attribute in the decision-making process of persons con
sidering separation.
3. A stratified random technique, if at all pos
sible, should be used to select a sample in future studies.
The small number of subjects, as well as the purposive
method by which this sample was selected, not only pre
cluded any generalizability of the findings, but neces
sitated many of the findings to be categorized as ’ ’ trends."
Future research efforts should obtain samples that would
permit one to accept or reject the findings with more
confidence. The sample also should include subjects drawn
from a group of married couples that have been judged to be
"happy," in addition to a maladjusted group.
4. The projected research design should be uti
lized in future studies, and the one year time period seems
adequate for this type of analysis of separation behavior.
However, future efforts should include at least three and
probably four measurement opportunities.
5. There is a need to develop more sensitive
instruments to "measure the major independent variables and
other factors affecting separation. A scaling method
should be used that would provide an opportunity to assay
differential characteristics with items that were used to
define and measure variables such as alternatives, reli
giosity, and deviance.
6. This type of study should be duplicated, and
the findings checked by an independent investigation. They
also should be conducted in a different locale using the
natural field setting. These studies, however, should be
done incorporating the suggested improvements that emanated
directly from this exploratory effort.
BIBLIOGRAPHY
BIBLIOGRAPHY
Ackerman, Nathan W.
1958 The Psvchodynamics of Family Life. New York:
Basic Books, Inc.
Bauman, Karl E.
1967 ’ ’ The Relationship Between Age at First Marriage,
School Dropout, and Marital Instability: An
Analysis of the Glick Effect.” Journal of
Marriage and the Family 29 (November): 672-681.
Bergler, Edmund.
1946 Unhappy Marriage and Divorce. New York: Inter
national Universities Press.
1948 Divorce Won*t Help. New York: Harper.
Bernard, Jessie.
1966 "Marital Stability and Patterns of Status
Variables." Journal of Marriage and the Family
28 (November): 421-440.
Bowerman, Charles E.
1964 "Prediction Studies" in Harold T. Christensen
(ed.), Handbook of Marriage and the Family.
Chicago: Rand McNally and Company. Pp. 215-247.
Burgess, Ernest W., and Leonard S. Cottrell.
1939 Predicting Success or Failure in Marriage. New
York: Prentice-Hall.
Burgess, Ernest W., and Paul Wallen.
1953 Engagement and Marriage. Philadelphia: lippin-
cott.
116
117
Chance, Erika.
1959 Families in Treatment. New York: Basic Books,
Inc.
Chancellor, Loren E., and Thomas Monahan.
1955 "Religious Preference and Interreligious Mixtures
in Marriages and Divorces in Iowa." American
Journal of Sociology 61 (November) : 233-239.
Chesser, Eustice.
1957 The Sexual. Marital, and Family Relationships of
the English Woman. New York: Roy.
Clements, William H.
1967 "Interaction and Marital Stability: A Point of
View and a Descriptive Comparison of Stable and
Unstable Marriages." Journal of Marriage and the
Family 29 (November): 697-703.
Eisenstein, Victor W. (Ed.)
1956 The Neurotic Interaction in Marriage. New York:
Basic Books, Inc.
French, John R. P. Jr.
1953 "Experiments in Field Studies" in Leon Festinger
and Daniel Katz (Eds.), Research Methods in the
Behavioral Sciences. New York: Kbit, Rinehart
and Winston, Inc. Pp. 98-136.
Glick, Raul.
1957 American Families. New York: John Wiley.
Gold, David.
1969 "Statistical Tests and Substantive Significance."
The American Sociologist 4 (February): 42-47.
Goode, William J.
1964 The Family. Englewood Cliffs: Prentice-Hall,
Inc.
1956 Women in Divorce. New York: The Free Press.
1965 "Sociology of the Family" in Robert K. Merton
et al. (Eds.) Sociology Today. New York and
Evanston: Harper and Row. Pp. 178-191.
118
Greene, Bernard L. (Ed.)
1965 The PSychotherapies of Marital Disharmony. New
York: The Free Press.
Greenwood, Ernest.
1945 Experimental Sociology. New York: King*s Crown.
Hill, Reuben L., and Roy H. Rodgers.
1964 "The Developmental Approach" in Harold T.
Christensen (Ed.), Handbook of Marriage and the
Family. Chicago: Rand McNally and Company.
Pp. 171-211.
Hawkins, James L.
1968 "Associations Between Companionship, Hostility,
and Marital Satisfaction." Journal of Marriage
and the Family 30 (November): 647-651.
Jacobson, Paul H.
1940 "Differentials in Divorce by Duration of Marriage
and Size of the Family." American Sociological
Review 15 (April): 235-244.
1959 American Marriage and Divorce. New York: Rine
hart.
Karllson, George.
1951 Adaptability and Communication in Marriage:
A Swedish Study of Marital Satisfaction.
Uppsala, Sweden: Almqvist and Wiksells.
Katz, Daniel.
1953 "Field Studies" in Leon Festinger and Daniel
Katz (Eds.), Research Methods in the Behavioral
Sciences. New York: Holt, Rinehart and Winston,
Inc. Pp. 56-98.
Kephart, William M.
1955 "Occupational Level and Marital Disruption."
American Sociological Review 20 (August): 456-465
Kerlinger, Fred N.
1966 Foundations of Behavioral Research. New York:
Holt, Rinehart and Winston, Inc.
119
Kirkpatrick, Clifford.
1963 The Family— as Process and Institution. Second
edition. New York: The Ronald Press Company.
Labovitz, Sanford.
1968 "Criteria for Selecting a Significance level:
A Note on the Sacredness of .05." The American
Sociologist 3 (August): 220-222.
Iandis, Judson T.
1949 "Marriages of Mixed and Non-mixed Religious
Faith." American Sociological Review 14 (June):
401-406.
Lev inger, Ge orge.
1965 "Marital Cohesiveness and Dissolution: An Inte
grative Review." Journal of Marriage and the
Family 27 (February): 19-29.
Locke, Harvey J.
1951 Predicting Adjustment in Marriage. New York:
Holt and Company.
Mercer, Charles V.
1967 "Interrelations Among Family Stability, Family
Composition, Residence, and Race." Journal of
Marriage and the Family 29 (August): 456-461.
Mogey, John.
1964 "Family and Community in Urban-Industrial Socie
ties" in Harold T. Christensen (Ed.), Handbook
Marriage and Family. Chicago: Rand McNally and
Company. Pp. 500-534.
Monahan, Thomas P.
1955 "Is Childlessness Related to Family Stability?"
American Sociological Review 20 (August): 446-
456.
1961 "Educational Achievement and Family Stability."
Journal of Social Psychology 55 (December): 253-
263.
1955 "Divorce by Occupational Level." Marriage and
Family Living 17 (November): 322-324.
120
Monahan, Thomas P., and William M. Kephart.
1954 ’ ’ Divorce and Desertion by Religious and Mixed-
Religious Groups.” American Journal of Sociology
(November): 454-465.
Morrison, Denton E., and Ramon E. Henkel.
1969 ’ ’Significance Tests Reconsidered.” The American
Sociologist 4 (May): 131-140.
Nye, F. Ivan.
1964 ’ ’Field Research” in Harold T. Christensen (Ed. ),
Handbook of Marriage and the Family. Chicago:
Rand McNally and Company. Pp. 247-273.
Rang, Henry, and Sue Marie Hanson.
1968 ’ ’ Highest Divorce Rates in Western United States.”
Sociology and Social Research 52 (January): 228-
236.
Skipper, James K., et al.
1967 "The Sacredness of .05: A Note Concerning the
Uses of Statistical Levels of Significance in
Social Sciences.” The American Sociologist 2
(February): 16-19.
Sorokin, Pitrim.
1941 The Crisis of Our Age. New York: Dutton.
1937 Social and Cultural Dynamics. New York: Harper
and Row.
State of California
1967 Divorce in California. Berkeley, California:
Department of Public Health.
Terman, Lewis M.
1938 Psychological Factors in Marital Happiness.
New York: McGraw-Hill Book Company.
Udry, Richard J.
1966 "Marital Instability by Race, Sex, Education,
and Occupation Using 1960 Census Data." American
Journal of Sociology 72 (September): 203-209.
121
1967 "Marital Instability by Race and Income Based on
1960 Census Data." American Journal of Sociology
72 (May): 673-674.
1966 The Social Context of Marriage. Philadelphia and
New York: J. B. Lippincott Company.
Zimmerman, Carle.
1947 Family and Civilization. New York: Harper and
Row.
APPENDICES
APPENDIX A
QUESTIONNAIRE USED TO COLLECT DATA
123
124
THE CONCILIATION COURT
OF THE
SUPERIOR COURT, COUNTY OF LOS ANGELES
Room 241
111 North Hill Street
Los Angeles 12, California
625-3414, Ext. 6-1331
William E. MacFaden
Judge, Superior Court
Presiding
The Conciliation Court always continues to seek
better methods to help people deal with family problems.
In order to increase our knowledge about family life and
improve our counseling service, I again am asking for your
help, by requesting that you complete this questionnaire.
Your experience will help us understand more fully, not
only your situation, but also future families that will
come to otar Court for help in dealing with family problems.
In no way can any of this information be used in
any Domestic Relations case. The information is confiden
tial and privileged.
Please try to be objective in your answers and com
plete the questionnaire without the assistance of your
marital partner. Your prompt return of the questionnaire
will be of great help in improving our services.
Thank you again for your cooperation.
Sincerely,
William E. MacFaden
Presiding Judge
125
PLEASE DESCRIBE YOUR FEELINGS AT THIS TIME
Circle the dot on the scale line below which best
describes the degree of happiness, everything con
sidered, of your present marriage. The middle point,
"happy," represents the degree of happiness which most
people get from marriage. The scale gradually ranges
on one side to those who are very unhappy in marriage,
and on the other, to those who experience extreme joy
in marriage.
* Sc Sc * Sc
Very Happy Perfectly
Unhappy Happy
WHAT IS YOUR WORK SCHEDULE?
Full-time housewife (not employed)
Employed part-time
Employed full-time
Other: Describe_________________________________
ARE YOU WORKING CURRENTLY? __ Yes __ No
IF YOU ARE WORKING, WHAT IS YOUR OCCUPATION?__________
WHAT IS YOUR MONTHLY INCOME?
126
IF YOU ARE WORKING, WHO TAKES CARE OF YOUR CHILDREN?
Does not apply to me (no children)
Other parent
Relatives
Day Nursery
Baby sitter
They are older and take care of themselves
Other:________________________________ _
Does not apply (I do not work)
IF YOU ARE NOT WORKING, HOW ARE YOU SUPPORTED?
Unemployment benefits
Marital partner*s income
Relatives
Disability benefits
Public and/or private welfare
Savings
Other (describe)_________________
None
Does not apply (I am working)
IF YOU ARE NOT WORKING, HOW LONG HAVE YOU BEEN
UNEMPLOYED?
127
DO YOU HAVE ANY WEALTH THAT DOES NOT BELONG TO YOUR
MARITAL PARTNER (NOT COMMUNITY PROPERTY)?
Stocks and bonds
Inheritance
Savings
Real estate
Valuable objects (jewelry, art, rare coins, etc.)
Other (describe)_________________________________
None
WHOSE SALARY DO YOU LIVE ON WHEN YOU AND YOUR MARITAL
PARTNER LIVE TOGETHER?
Husband only
Wife only
Husband and wife together
Other (explain)__________________________________
DO YOU SAVE MONEY REGULARLY? __ Yes No
PLEASE CHECK IN THE BOXES BELOW ALL PEOPLE WHO LIVE
WITH YOU PERMANENTLY- -WHEN YOUR FAMILY IS LIVING
TOGETHER.
Husband
__ Wife
Children
Mother
Father
128
Mother-in-law
Father-in-law
Friends or relatives
Housekeeper
Other (describe)
IF YOU WANTED TO, COULD YOU AND YOUR CHILDREN GO TO
LIVE PERMANENTLY WITH YOUR PARENTS OR RELATIVES?
Yes __ No __ I now am living with
parents or relatives.
ARE YOU AND YOUR MARITAL PARTNER NOW LIVING APART
BECAUSE OF MARRIAGE OR FAMILY TROUBLES?
Yes __ No
If "yes," how long have you lived apart ?______________
IF YOU ARE NOT LIVING WITH YOUR MARITAL PARTNER,
HOW ARE YOU SUPPORTED?
I support myself by working
Marital partner supports me voluntarily
Court order--alimony and/or child support
Relatives
Children are older and they support me
Friends (not including family members)
Public Assistance or other type welfare
Other: (describe )________________________________
Does not apply (I am living with marital partner)
129
IF YOU ARE NOT LIVING WITH YOUR MARITAL PARTNER,
CHECK ALL THOSE PEOPLE NOW LIVING WITH YOU.
Children
Mother and/or father
Mother-in-law and/or father-in-law
Relatives
Housekeeper
Friends
Other (describe)_________________________________
Does not apply (I am living with marital partner)
ARE YOU PREGNANT AT THIS TIME?
Yes __ No __ Does not apply
(I am husband)
HOW MANY CHILDREN DO YOU HAVE FROM THIS MARRIAGE?_____
HOW MANY OF YOUR CHILDREN ARE LIVING WITH YOU NOW?____
IF SOME OR ALL OF YOUR CHILDREN ARE NOT LIVING WITH
YOU, WITH WHOM DO THEY LIVE?
With other parent
Relatives (including mother)
Adopted parents
Foster home
Public or private organizations
They are older and live by themselves
130
Does not apply to me (no children)
Does not apply (children live with me)
IF YOU DO NOT NOW LIVE WITH YOUR MARITAL PARTNER,
HOW ARE YOUR CHILDREN SUPPORTED?
I support them by working
Marital partner supports them voluntarily
Court order— alimony and/or child support
Relatives
Friends (not including family members)
Public assistance or other type welfare
Other (describe)_________________________________
Does not apply (no children)
They are older and they support themselves
HAVE YOU AND YOUR MARITAL PARTNER LIVED APART "BECAUSE
YOU COULD NOT GET ALONG WITH EACH OTHER” SINCE FIRST
COMING TO THE CONCILIATION COURT?
Yes __ No
IF YES, HOW MANY TIMES?_______
HOW LONG WERE THESE SEPARATIONS? LIST ONLY THOSE
SEPARATIONS WHICH HAVE OCCURRED SINCE YOU FIRST CAME
TO THE CONCILIATION COURT IN 1967 OR 1968.
First separation ____________
Year
Less than one week
Less than two weeks
—
Less than one month
-- -
Less than two months
-
More than two months
Second separation
Year
----
Less than one week
Less than two weeks
-
Less than one month
-
Less than two months
_
More than two months
Third separation
Year
- -
Less than one week
Less than two weeks
Less than one month
- -
Less than two months
_
More than two months
Fourth separation
Year
Less than one week
■-
Less than two weeks
Less than one month
---
Less than two months
More than two months
132
DO YOU HAVE A PROBLEM WITH DRINKING?
Yes __ No
DOES YOUR MARITAL PARTNER HAVE A PROBLEM WITH DRINK
ING?
Yes __ No
DO YOU HAVE A PROBLEM WITH DRUGS ?
Yes __ No
IF YES, WHAT TYPE?
Tranquilizers, barbiturates
Heroin, morphine, demerol, etc.
Marijuana
LSD
Other (describe)_________________________________
DOES YOUR MARITAL PARTNER HAVE A PROBLEM WITH DRUGS?
Yes __ No
IF YES, WHAT TYPE?
Tranquilizers, barbiturates
Heroin, morphine, demerol, etc.
Marijuana
LSD
Other (describe)_________________________________
HAVE YOU EVER HAD SEVERE EMOTIONAL PROBLEMS?
Yes No
IF YES, WERE YOU:
__ Treated in a hospital
__ Treated out of hospital
__ Not treated
Other (describe)
133
HAS YOUR MARITAL PARTNER EVER HAD ANY SEVERE EMOTIONAL
PROBLEMS?
___ Yes __ No
IF YES, WAS YOUR MARITAL PARTNER:
__ Treated in hospital
__ Treated out of hospital
__ Not treated
Other (describe)
HAVE YOU EVER BEEN ARRESTED FOR BREAKING THE LAW
(OTHER THAN FOR TRAFFIC TICKETS)?
__ Yes __ No
IF YES. HOW MANY TIMES?
HAVE ANY OF YOUR CHILDREN BEEN IN TROUBLE AT SCHOOL?
__ Yes __ No
IF YES, WHAT KIND OF TROUBIE?
__ Failing grades
__ Truancy
__ Behavior problems
Other (describe)
HAVE ANY OF YOUR CHILDREN BEEN IN TROUBLE WITH
JUVENILE OFFICERS?
Yes __ No
IF YES, WHAT KIND OF TROUBIE? (PLEASE EXPLAIN)
APPENDIX B
SELECTED CHARACTERISTICS OF THE CONCILIATION
COURT POPULATION
135
136
LENGTH OF MARRIAGE
1964
Length January-June
0-4 months 20
(1.9)
5-8 months 20
(1.9)
9-11 months 23
( 2. 1)
1-3 years 276
(25.6)
4-6 years 187
(17.4)
7-10 years 166
(15.4)
11-15 years 172
(16.0)
16-20 years 120
( 11. 1)
Over 20 years 93
( 8.6)
RACE AND ETHNICITY
137
Husband
1964
January-June
Wife
Caucasian 646
(72.5)
650
(73.1)
Negro 130
(14.6)
131
(14.8)
Oriental 7
(.8)
5
(.6)
Other 8
(.9)
11
(1.2)
Mexican 100
(11.2)
92
(10.3)
RELIGION
Husband
1964
January-June
Wife
Protestant 535
(51.5)
561
(51.5)
Catholic 282
(27.1)
307
(27.1)
Jew 63
(6.1)
60
(6.1)
Other 51
(4.9)
60
(4.9)
None 108
(10.4)
50
(10.4)
138
0-6th grade
7th-9th grade
10th-12th grade
High school graduate
College: 1 year
2 years
3 years
4 years
Graduate
Post-Graduate
EDUCATION
1964
January-June
Husband Wife
36 28
(3.5) (2.7)
117 88
(11.4) (8.6)
218 249
(21.2) (24.4)
274 400
(26.7) (39.2)
90 105
(8.8) (10.3)
109 55
(10.6) (5.4)
48 36
(4.6) (3.5)
0 0
(.0) (.0)
82 36
(8.0) (3.5)
53 25
(5.2) (2.4)
PRESENT AGE
139
Husband
1964
January-June
Wife
0-17 1 14
(.1)
(1.3)
18-20 23 79
(2.1) (7.4)
21-24 159 189
(14.8) (17.8)
25-29 175 202
(16.3) (19.0)
30-39 375 350
(34.9) (33.0)
40-49 230 167
(21.4) (15.7)
50-59 88 53
(8.2) (5.0)
60-69 19 7
(1.8) (.7)
Over 70 4 1
(.4)
(.1)
140
NUMBER OF CHILDREN
1964
Number January-June
0 151
(13.9)
1 277
(25.5)
2 319
(29.3)
3 178
(16.4)
4 92
(8.5)
5 42
(3.8)
6 15
(1.4)
7 8
( . 8)
Over 7 4
(.4)
141
INCOME
1964
January-June
Husband Wife
$0 - 2,000 22 142
(2.3) (27.7)
2,000 - 4,000 104 186
(10.6) (36.3)
4,000 - 5,000 155 92
(16.0) (18.0)
5,000 - 6,000 185 58
(19.1) (11.3)
6,000 - 7,000 122 17
(12.5) (3.3)
7,000 - 8,000 138 8
(14.3) (1.6)
8,000 - 9,000 53 6
(5.7) (1.2)
9,000 - 10,000 79 0
(8.2) (.0)
Over 10,000 109 3
(11.3) (.6)
APPENDIX C
RELIGIOSITY, DEVIANCE, AND ALTERNATIVE SCALE
142
143
Religiosity Scale
A. Items comprising the scale, and their responses.
(1) Religious Ceremony?
Yes - 84 No - 38
(2) Frequency of Church Attendance
Never - 34
Few a year - 48
Once a month - 9
Once a week - 29
Twice weekly - 2
(3) Sunday or Bible School Attendance of Children
Never - 71
Few a year - 10
Bi-monthly - 8
Once a week - 27
Twice weekly - 6
(4) Number of Children Receiving Ihrochial Education
None - 111
Some - 5
All - 6
(5) Would You Send Kids to Church School?
Yes - 42
No (or doesnft apply) - 80
B. Scores on the Cumulative Religiosity Scale
(0) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10 or higher)
0 9 20 24 14 18 11 7 6 9 4
144
Deviance Scale
A. Items comprising the scale, and their responses.
(1) Does respondent have problems?
No - 94
One problem -25
Two or more - 3
(2) Does spouse have problems?
No - 85
One problem - 29
Two or more - 8
(3) Number of arrests on the part of respondent.
None - 105
One - 12
Two - 3
Five - 2
(4) Do children have problems?
No (or doesn’t apply) - 105
One problem - 16
Two or more - 1
B. Scores on the Cumulative-Deviance Scale
(0) (1) (2) (3) (4) (5) (6) (7)
22 42 29 19 6 2 0 1
145
Alternatives Scale
A. Items comprising the scale and their responses.
(1) If apart, who is living with you?
Mother and/or Father No - 103 Yes - 19
Other relatives No - 115 Yes - 7
Horaekeeper No - 120 Yes - 2
Friends No - 121 Yes - 1
(2) Sources of Income if not working
Unemp loyment No - 121 Yes - 1
Welfare No - 109 Yes - 13
(3) Relatives in the home? No - 109 Yes - 13
(4) Could you live with relatives if necessary?
No - 67 Yes - 55
(5) If apart, how supported?
Self No - 67 Yes - 55
Ehrtner, voluntary No - 119 Yes - > 3
Bartner, court order No - 99 Yes - 23
Relatives No - 115 Yes - 7
Rib lie No - 119 Yes - 3
B. Scores on the Cumulative Alternatives Scale*
(0) (1) (2) (3) (4) (5) (6) (7)
33 26 13 16 15 9 1 3
*N = 116 due to incodable responses

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An Investigation Of Complementary Needs Between Marital Partners

PDF

A Study Of The Relationships Between Religious Affiliation, Religious Practices And Marital Adjustment

PDF

Elementary Children'S Preferences For Ethnicity And Sex Of Teachers

PDF

An Analysis Of The Changing Focus Of Marriage Counseling

PDF

Personality Characteristics: Ideal And Perceived In Relation To Mate Selection

PDF

Power Relationships In Marital Discord

PDF

Academic Specialization And The Construction Of Social Reality: Ideologies Regarding Deviance

PDF

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

PDF

An Analysis Of Systematic Planning And Development In The Organizational Patterns Of The Intermediate Units In California

PDF

Communication Between Engaged Couples

PDF

Role Perception, Empathy, And Marital Adjustment

PDF

The Relationship Of A Collaborative Counseling Technique To The Patterns Of Enrollment, Curricular Choice And Curricular Achievement At The Community College

PDF

An Exploration Of Interpersonal Behavioral Possibilities And Probabilities

PDF

The Effects Of A Self Shock Procedure On Hallucinatory Activity In Hospitalized Schizophrenics

PDF

S.T.P.: A Simulation Of Treatment Processes

PDF

The Relation Of Parental Behavior To Adolescent Church Participation

PDF

A Study Of The Attitudes Of Mothers Of Mentally Retarded Children As Influenced By Socioeconomic Status

PDF

Some Behavioral Consequences Of Career Success: A Synthesis Of Reward Andbalance Approaches

PDF

Group Factors And Individual Internalization Of A Value

PDF

Status consistency and its effects on marital adjustment and stability

Asset Metadata

Creator Cohen, Stanley Norman (author)

Core Title An Exploratory Analysis Of Two Dimensions Of Family Separation

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 Lasswell, Thomas E. (committee chair), Brackenbury, Robert L. (committee member), Larson, William R. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-402119

Unique identifier UC11361266

Identifier 7016859.pdf (filename),usctheses-c18-402119 (legacy record id)

Legacy Identifier 7016859.pdf

Dmrecord 402119

Document Type Dissertation

Rights Cohen, Stanley Norman

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