University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS
This material was produced from a microfilm copy of the original document. While
the most advanced technological means to photograph and reproduce this document
have been used, the quality is heavily dependent upon the quality of the original
submitted.
The following explanation of techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1.The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)". If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting thru an image and duplicating adjacent
pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black mark, it
is an indication that the photographer suspected that the copy may have
moved during exposure and thus cause a blurred image. You will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
photographed the photographer followed a definite method in
"sectioning" the material. It is customary to begin photoing at the upper
left hand corner of a large sheet and to continue photoing from left to
right in equal sections with a small overlap. If necessary, sectioning is
continued again — beginning below the first row and continuing on until
complete.
4. The majority of users indicate that the textual content is of greatest value,
however, a somewhat higher quality reproduction could be made from
"photographs" if essential to the understanding of the dissertation. Silver
prints of "photographs" may be ordered at additional charge by writing
the Order Department, giving the catalog number, title, author and
specific pages you wish reproduced.
5. PLEASE NOTE: Some pages may have indistinct print. Filmed as
received.
Xerox University Microfilms
300 North Zeeb Road
Ann Arbor, Michigan 48106
74- 21,494
OLANDER, Edward Bernard, 1925-
THE PARAMETERS OF SINGLENESS: AN INQUIRY
INTO SOME FACTORS INFLUENCING THE CHOICE OF
SINGLENESS OVER MARRIAGE AS A WAY OF LIFE.
University of Southern California, Ph.D., 1974
Sociology, family
| University Microfilms, A XEROX Company, Ann Arbor, Michigan
©
Copyright by
Edward Bernard Olander
1974
THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED.
THE PARAMETERS OF SINGLENESS
AN INQUIRY INTO SOME FACTORS INFLUENCING
THE CHOICE OF SINGLENESS OVER MARRIAGE
AS A WAY OF LIFE
by
Edward Bernard Olander
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Sociology)
June 1974
UNIVERSITY OF SOUTHERN CALIFORNIA
T H E G R A D U A TE S C H O O L
U N IV E R S IT Y P A RK
LO S A N G E LE S . C A L I F O R N IA 9 0 0 0 7
This dissertation, written by
...5j ■r
under the direction of A.i.?... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Graduate
School, in partial fulfillm ent of requirements of
the degree of
D O C T O R O F P H IL O S O P H Y
f'jnnujo
Dean
DahdC)L&J±jU^^k.).l9 7 *
^ \ T —^ Chairman
..
ACKNOWLEDGMENTS
The author acknowledges with gratitude the support,
encouragement, patience, and technical suggestions of the
members of his Dissertation Committee, and particularly the
Chairman, Dr. CarIfred B. Broderick. He also wishes to ex
press his gratitude to Dr. Alan Acock of the U.S.C. Sociol
ogy Faculty, whose kind assistance in the formulation of
the multiple regression equations used in this study were
most influential in determining the direction the research
has taken.
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ....................................... ii
TABLE OF CONTENTS.......................................iii
LIST OF TABLES..........................................iv
LIST OF FIGURES .......................................................................................................................................... V
V I T A .....................................................vi
ABSTRACT............................................... vii
Chapter
I A STATEMENT OF THE RESEARCH PROBLEM .......... 1
II OPERATIONALIZING THE VARIABLES ............... 25
III THE METHOD OF DATA ANALYSIS......................40
IV THE SAMPLE AND ITS SUBSETS......................49
V STATISTICAL DESCRIPTION OF THE VARIABLES . . . 62
VI FINDINGS..........................................68
VII CONCLUSIONS,AND SUGGESTIONS FOR
FURTHER RESEARCH ............................ 87
REFERENCES ........................................... 1Q2
APPENDICES
A THE RESEARCH QUESTIONNAIRE • ................ 105
B DESCRIPTIVE STATISTICS FOR THE SAMPLE ......... H I
C DESCRIPTIVE STATISTICS FOR THE VARIABLES . . . 135
D STATISTICS SUPPORTING THE FINDINGS .......... 142
iii
LIST OF TABLES
Table Page
1. INTERPERSONAL CHECK LIST
TEST-RETEST CORRELATIONS ....................... 2 6
2. INTERPERSONAL CHECK LIST
AVERAGE INTERVARIABLE CORRELATION AS
A FUNCTION OF THEIR SEPARATION AROUND
THE CIRCLE.......................................26
3. SAMPLE DISTRIBUTION BY SEX ...............51
4. SAMPLE DISTRIBUTION BY A GE....................... 52
5. SAMPLE DISTRIBUTION BY INCOME ................. 53
6. SAMPLE DISTRIBUTION BY EDUCATION ............... 54
7. SAMPLE DISTRIBUTION BY RELIGIOSITY ............. 55
8. FACTOR ANALYSIS OF THE SOCIAL INFLUENCE
INDICATOR................................ 66
9. MULTIVARIATE REGRESSION OF THE SINGLENESS
ORIENTATION WITH ASSERTIVENESS, DEPENDENCY,
THE SOCIAL INFLUENCE INDEX, AND THEIR
INTERACTIONS .................................... 73
iv
LIST OF FIGURES
Figure Page
1. THE PROBABLE EFFECTS OF ASSERTIVENESS AND
DEPENDENCY UPON THE LIKELIHOOD OF
REMAINING SINGLE ............................ 7
2. PROBABLE INTERACTIONAL EFFECTS OF
ASSERTIVENESS, DEPENDENCY, AND
THE SOCIAL INFLUENCE INDEX ................. 15
3. GRAPHIC SUMMARY OF HYPOTHESES 1 - 4 ..... . 17
4. SUMMARY OF THE RESEARCH HYPOTHESES..............24
5. RESULTS OF AUTOMATIC INTERACTION DETECTOR . . . 75
6. AID PROGRAM FOR NEGATIVE AFFECT, AS
DEPENDENT VARIABLE ........... 82
7. AID PROGRAM FOR POSITIVE AFFECT AS
DEPENDENT VARIABLE .......................... 87
8. A REVISED "SINGLENESS" MODEL .................. 89
9. A REVISED "POSITIVE AFFECT" MODEL ............. 92
10. HYPOTHETICAL REPLICATION OF THE "SINGLENESS"
STUDY, WITH EXPANDED VARIABLES, AND A
MIXED SAMPLING THAT INCLUDES MARRIED
SUBJECTS AS WELL AS SINGEE...................101
V
EDWARD BERNARD OLANDER
1925
1942
1948
1951
1958
1960-61
1968-69
1969-74
1971-72
1972
1971-
present
1969-71
1960-69
1958-60
1953-54
Education
Born in Superior, Wisconsin
Graduated from Superior (Wis.) Central High
School
B.A. (Music, Social Science), Wisconsin State
University at Superior, ("With High Honors")
M.A. (Music) University of California, Berkeley
B.D. (Theology) Pacific Lutheran Theological
Seminary, Berkeley
Special Study (Theology) Church Divinity School
of the Pacific, Berkeley
Special Study (Sociology of the Family) Cali
fornia State University, Los Angeles
Graduate Student, U.S.C.
Research Trainee: ("A Study of Generations and
Mental Health"), Vern L. Bengston, Director,
U.S.C.
Teaching Assistant ("Education for Marriage")
Carlfred B. Broderick, Professor
Experience
Licensed Marriage, Family, and Child Counselor
Partner in Olander, Caddell, and Associates in
Counseling, Culver City, California
Formerly Assistant Director, Department of
Counseling, The American Institute of Family
Relations, Los Angeles
Clergyman, The Episcopal Diocese of Los Angeles
The United Lutheran Church in America, Synod of
the Southwest
Assistant Professor of Music: The Mission House
College, Plymouth, Wisconsin
vi
ABSTRACT
THE PARAMETERS OF SINGLENESS
AN INQUIRY INTO SOME FACTORS INFLUENCING
THE CHOICE OF SINGLENESS OVER MARRIAGE
AS A WAY OF LIFE
This study is concerned with two general questions
(1) What variables affect a person's choice for singleness
over marriage as a way of life? and (2) What variables
(other than general "mental health" factors) affect a per
son's overall happiness (i.e., his psychological well
being) in the single role?
It is hypothesized that significant amounts of
variance in the choice for singleness or marriage can be
explained in terms of a person's "Assertiveness" and his
"Dependency" upon others, in interaction with a cluster of
socially controlled factors operationalized in this study
as the "Social Influence Index" ("SII"). It is further
hypothesized that these same variables would explain at
least a portion of the person's "Psychological Well-Being.
This was operationalized in terms of Bradburn's scales for
measuring "Positive Affect" and "Negative Affect."
A questionnaire based upon "Assertiveness,"
"Dependency," and the "Social Influence Index" was admin-
istered to a sample of 281 single adults in the Los Angeles
area, representing both secular and church-sponsored or
ganizations for "singles," single clients in counseling,
and other single adults not connected with any of the
above groups. The hypotheses were tested by means of mul
tiple regression analysis.
The findings indicated that the most influential
source of variance in the choice for singleness or mar
riage was the "SII." Where the "SII" was high in value,
neither "Assertiveness" nor "Dependency" explained any
appreciable additional variance. However, when the "SII"
was moderate or low in value, "Dependency" and "Assertive
ness," in that order, made their effects apparent. (The
multiple regression coefficient was .61). The "new" stan
dards of sexual freedom for single adults, though openly
accepted and practiced by a majority of the subjects, seem
to have had no effect upon their choice for singleness
over marriage.
The hypotheses concerning the effects of "Asser
tiveness," "Dependency," and the "SII" on the subjects'
psychological well-being received tentative confirmation.
Their effects accounted for approximately 30% of the vari
ance in "Positive Affect" (multiple "R" = .54), but only
2% of the variance in "Negative Affect" (multiple "R: =
.13). While both of these results were statistically
viii
"significant" (p. < .01), the amount of variance explained
in the "Negative Affect" variable was negligible.
In conclusion, the main hypotheses of the study
appear to have been confirmed. However, much further work
is indicated, especially in the refining and expanding of
the variables. The "SII" in particular, deserves atten
tion, since it accounted for the lion's share of the vari
ance in the choice for singleness. For example, the so
cially-controlled variables, "Income" and "Education,"
which other studies have shown to be related to the choice
for marriage, were not considered because of excessive
sample skew. A factor analysis of the items contributing
to the "SII" nhowed that items related to clarity of per
ception of the singleness role, and acceptance of that
role accounted for 68% of the variance explained by the
"SII" (this suggests that some subjects today may be be
coming socialized to accept a "singleness" role as a via
ble alternative to the traditional "spouse" role); other
factors were related to role conflicts (20.5% of the "SII"
variance), and family pressures (11.5%).
Possible refinements to the intrapersonal varia
bles might include the effects of traumas in former mar
riages, and traumas in the families of origin.
Overall, the study indicated that the choice for
marriage is still pervasive in the single adult popula-
ix
tion, but the strength of the "Role Clarity and Accep
tance" factor of the "SII," described in the preceding
paragraph, may well portend the emergence of a new "Single
ness Role" in our society, as an accepted and even attrac
tive alternative to the traditional spouse role.
x
CHAPTER I
A STATEMENT OF THE RESEARCH PROBLEM
A necessary concomitant of the rising divorce rate
in our time and place is the existence of a growing body
of single adults of mature age— a few never married or
widowed but most of them divorcees, and many of them par
ents— who must deal with the manifold and perplexing prob
lems of life as a single person in a society structured
primarily around the married couple. A certain number of
unmarriageable singles (the spinster aunt, for example,
and the hired man of questionable mentality but unques
tionable loyalty) have been with our society for a longer
time than this paper cares to chronicle, but the present-
day "single" includes a far broader spectrum. Among the
divorcees, widowers, and never married in today's popula
tion of mature-age singles, one can see a broad distribu
tion on many dimensions, including competence and success
in the workaday world, socio-economic status, education,
and, apparently, in psychological well-being.1
1According to the U.S. Bureau of Census report for
1970, there were 1.1 million males and 1.9 million females
between the ages of 20 and 64 who were single. This com
prises 21% of the total males and 34% of the total females
in this age group. Of this number, 16% of the males and
1
2
One aspect of the life of the present-day single
adult— greater freedom and permissiveness in sexual behav
ior— has produced a voluminous contemporary literature,
ranging from scholarly to popular, from polemical to re-
portorial. However, with two notable exceptions, little
has been done to study other aspects of the single life,
such as pertain to relationships, friendships and inti
macy. Kirkendall's early (1961) study of the sexual be
havior of college-age males was an attempt to approach the
question of sex versus relationship, but little follow-up
was ever made of this important topic.
Another important area concerns those variables
that affect the prospects of a single person's marrying
(or his desire to do so). Cutright (1970) has done some
pioneering work in showing the relationship of income to
the tendency to marry.
The studies of Kirkendall and Cutright have raised
questions that remain largely unanswered. It is into
this relatively virgin and unexplored territory that the
present study is directed.
The purpose of this paper is to explore two broad
12% of the females were never married; 3% of the males
and 5% of the females were divorced; the remainder were
widowed. (These figures do not include persons who may
have divorced and subsequently remarried prior to the
census.)
3
questions, one of sociological interest, and one of pri
mary interest to the marriage and family therapist who is
likely to encounter single adults in counseling relation
ships.
Family sociologists today are concerned with the
stability and continuity of the family as an institution.
Related to this question is whether our present period is
witnessing the rise of another institution (or other in
stitutions) , counter-familial in nature (or at least
counter the "nuclear family") idea, composed of single
adults of an age and competence that in another time would
likely have been married, but who have elected to remain
single. Bernard (1968) has described some new approaches
to the family whose participants would belong to the in
stitution of singles, at least in the technical sense that
they are legally unmarried. Dreitzel (1972) has assembled
a collection of provocative papers describing recent
changes in even the "traditional" family structure, new
approaches to familism, and concluding with an argument
by Roy and Roy suggesting that the traditional monogamous
marriage has outlived its usefulness. A large number of
newspapers directed towards the single adult are now to
be found in many newsstands in cosmopolitan areas. While
these newspapers, like most of the popular "singles" lit
erature, are devoted primarily to the championing of new
4
freer forms of sexual expression, their existence suggests
a sizable readership of single adults.
The first question to be considered in this study
is whether a significant part of the single adult popula
tion sees their singleness as a way of life, or whether
the single role is viewed primarily as a temporary state
until a suitable marriage partner can be found. An at
tempt will be made to test some of the variables which
might determine whether a person decides to remain single
or to seek remarriage.
The second broad question, of more interest to the
counselor than to the sociologist, is whether the single
state affects the individual's happiness and sense of well
being, and if so to determine the variables that affect
the happiness of single adults, above and beyond the usual
"mental health" variables.
The paucity of research literature on these two
questions demands that the present study be exploratory in
nature. Therefore, its thrust will not be to test hy
potheses derived from previously existing theory, but to
generate a theory of the single adult role, which hope
fully might be productive of other hypotheses for future
research.
Question I: What variables affect the choice of the sin
gle role versus the marriage role?
5
It is obvious even to the casual observer that
some singles appear to be single by choice, apparently
viewing their single role as a complete and fulfilling
one, while others view their singleness as incomplete, and
actively seek a partner for marriage. In a pilot survey
that I recently conducted among a nonrepresentative sample
of single adults, (42 members of a "single parents" organ
ization) approximately 50% of the subjects stated that
they did not intend to remarry. Among the clinical popu
lation seen by the counseling organization in which I am
employed, the clients are about evenly divided among those
who desire marriage, those who desire to remain single,
and those who are undecided.
An important goal of this study is to identify
some of the variables that determine who elects to remain
single, who seeks a partner for (re)marriage, and who can
not decide what to do. In the absence of prior research
in this area, it will be necessary to assemble the first
rough categories of variables from bits and pieces derived
elsewhere: from clinical experience as a counselor of
single adults, from social learning theory, and from role
theory.
In a clinical sample of single adults there seem
to be two dimensions of personality that correlate highly
with decisions to marry or to remain single. For purposes
6
of this research, I choose to label them "Assertiveness"
and "Dependency." By "Assertiveness" is meant the capaci
ty in a person to know what he wants, and actively seek it
out. Among counseling clients, persons with high "Asser
tiveness" scores tend to see marriage as an impediment to
their career and recreational lives, and tend to prefer
"living together" arrangements without marriage, or to be
unencumbered altogether. Persons with extremely low lev
els of assertiveness tend to be unable to make the inter
personal contacts necessary either for marriage, or for
good relationships.
"Dependency," as used in this paper, refers to a
person's need for emotional support, affection, etc., from
other persons. It is an interpersonal dependency, not an
economic one. Generally speaking, one would expect de
pendency to be negatively correlated with an orientation
towards singleness. The probable effects of two variables,
"Assertiveness" and "Dependency," is illustrated in Figure
1.
The foregoing paragraphs provide the rationale for
the first two of the working hypotheses to be tested in
this study. Stated in formal terms, they are:
Hypothesis 1: The orientation towards remaining single is
positively correlated with personal asser
tiveness.
FIGURE 1
THE PROBABLE EFFECTS OF ASSERTIVENESS AND
DEPENDENCY UPON THE LIKELIHOOD OF REMAINING SINGLE
High
Low
High Low
Probability of remaining single
Assertiveness
Dependency
Hypothesis 2; The orientation towards remaining single is
negatively correlated with personal depen
dency .
A vast amount of recent research iH the field of
marital stability shows that a number of socially con
trolled variables, quite unrelated to any intrapersonal
need for assertiveness or dependency, are highly corre
lated with the survival or breakup of marriages, and it is
not unreasonable to suppose that similar social factors
might also influence the "stability" of the single state.
Hicks and Platt (1971) in reviewing the "marital stability"
research of the decade of 1960-1970 found that such social
ly oriented factors as income, socioeconomic status, simi
larity of educational background, and congruence of role
expectations— especially regarding the "instrumental" role
of the Husband— were all influential in affecting the sta
bility of marriages, regardless of whether they were
"happy," or not. Outfight (1970) has shown a strong cor
relation between income and the decision to marry. Such
literature leads one to hypothesize that the variables
"Assertiveness" and "Dependency" would be most effective
as predictors of singleness if due consideration were made
for the effects of social influences that also bear upon
the individual, possibly at times even to the detriment
of his intrapersonal needs and preferences.
However, the definition of this "Social Influence"
9
variable is complicated by the sheer number of social fac
tors that might possibly influence a person's choice for
singleness or marriage. Money, education, religion, age,
family influences, and peer group pressures are but a few
of the more likely candidates. Since the purpose of this
research at this present primitive stage is more to test
the feasibility of a theory (indeed, to formulate a theory
for testing!) than to isolate and measure every possible
and separate source of variance, a decision was made to
limit the "Social Influence" variable in scope, even
though this admittedly oversimplifies the sources of vari
ance that might be involved in the choice of singleness as
a way of life. Such a simplified variable would be able
to show whether the general theoretical orientation is
tenable, and whether exploration into additional social
sources of variance would be worth the effort.
A ready framework for a possible simplified index
of "Social Influence" variables exists in Burr's (1972)
axiomatic transformation of role theory. In this work,
Burr has presented a system of propositions that determine
when a role transition is likely to be permanent or tempo
rary. It is possible to view the single adult as a case
of role transition from the traditional "spouse" role.
(Even the never-married adult, who has not actually held
the spouse role, has had a considerable degree of sociali
10
zation towards it.) If this view is accepted, some of
Burr's propositions can be employed to organize a set of
social and interpersonal variables that indicate the prob
ability of a single person's remaining that way. Burr's
basic theory would suggest that that the easier the tran
sition into the single role, the greater the probability
that the person will choose to remain single, and con
versely, the more difficult the transition, the greater
the probability of (re)marriage.
The simple effect of the interpersonal-social di
mension (called the "Social Influence Index" ("SII") in
this study) is stated in formal terms as follows:
Hypothesis 3: The orientation towards remaining single is
positively correlated with those interper
sonal and social variables that facilitate
the transition from the married role to the
single role, or that support maintaining
the single role.
For the purpose of organizing the several items
that make up the "Social Influence Index," the following
topical categories have been extrapolated from Burr's
axiomatic system of role theory propositions:
1. Role clarity and understanding (Derived from
Burr's propositions 1-3). The more clearly the subject
perceives his singleness as a purposeful role in itself
(rather than a kind of waiting period before marriage, or
11
between marriages) the more likely he will elect single
ness as a way of life.
2. Role conflict and strain (Burr's propositions
4-6). The greater the amount of conflict between the sin
gle and married role, the less likelihood of the subjects'
choosing singleness as a way of life.
3. Role Incompatibility (Burr's proposition 8).
The greater the incompatibility among the several func
tions in a person's single role (e.g., the single parent,
who must coordinate dating and sex with child care, etc.)
the less likelihood of his electing to remain single.
4. Role Compartmentalization (Burr # 9), where
possible, tends to lessen the effect of role incompatibi
lity. (E.g., the parent who only sees his children on
designated visiting days will probably be able to compart
mentalize his parent and single responsibilities more
readily than the parent whose children are with him all
the time.)
5. Activity and involvement outside the family
(Burr #10) was predicted by Burr to increase role strain.
However, in the case of the single adult it seems more
likely that activity will facilitate compartmentalization,
and thus reduce role strain.
12
6. Goal Attainment (Burr's # 11-14). This cate
gory attempts to measure whether one's most valued goals
can best be attained within marriage and the family insti
tution, or outside of it.
7. Substitute Gratifications (Burr's #15) would
include such things as the availability of sex, intimacy,
and companionship outside of marriage.
Burr's remaining propositions deal with such cri
teria as the definitiveness of the transition process, and
normative changes involved in the role transition. In the
transition from the married to the single role, it seems
reasonable to presume that such macrosociological factors
would operate quite similarly upon all members of the sin
gle population, and therefore would not show a significant
amount of variance in the particular sample for which the
present study is intended.
The "SII" is constructed around the seven cate
gories described above. For most of the data analysis
(see Chapter 3) they will be considered as a single vari
able, measuring the total effect of interpersonal and so
cial factors upon the individual.
It is not enough to measure the simple one-to-one
effects of the three main independent variable sets (i.e.,
"Assertiveness," "Dependency," and "SII") described above.
13
As Acock and DeFleur (1972) have shown, the combined ef
fects of several variables operating concurrently in mul
tivariate models is often greater than the sums of the
separate effects of the variables taken separately. There
is considerable logical support for the assumption that
such will indeed be the case in the present study. For
example, it is to be expected that "Assertiveness" and
"Dependency" will tend to undermine one another. If a
person is highly assertive, he will tend to ignore his de
pendency needs in his press for goal-attainment. If such
a person's dependency score is low, interpersonal needs
will tend to be ignored altogether. On the other hand, a
person with a high degree of dependency may sacrifice what
personal assertiveness he may have in order to make a
needed interpersonal relationship. As another possible
example, certain components of the "SII," such as family
pressure or religious beliefs, tend to press an individual
in the direction of marriage. If such a person were high
ly assertive, he might tend to ignore and override these
pressures, thus suppressing the effect of the "SII." A
highly dependent person, on the other hand, might give the
"SII" factors greater weight than would be expected under
conditions of intrapersonal neutrality.
A way to conceptualize the "configurational ef
fect" of concurrent variables might be in terms of feed-
14
back loops (see Figure 2) . As the diagram indicates, the
loops that connect adjacent pairs of variables are all al
gebraically positive, since their signs are alike in each
pair, while the loops interconnecting all the variables
(arrows all going in the same direction) are negative,
since in each case there is one unlike sign. According to
Buckley (1967, p. 72) positive feedback loops are morpho
genetic, (i.e., they tend to accelerate one another and
move the system in the direction of change) while negative
feedback loops are morphostatic (i.e., they tend to check
one another to keep the system in balance). Thus, in
terms of feedback theory one expects the variables acting
in pairs to have a continously accelerating (or decelerat
ing) effect upon each other, pressing each other towards
greater exaggeration (e.g., in a high "Assertive"-low "De
pendency" condition, "Assertiveness" will tend to increase
and "Dependency" will tend to decrease). The interaction
of the three variables together, being in a negative feed
back condition, will tend to limit the system, however,
producing stability and consistency of behavior. The
interactive effect of the three variables is stated for
mally as follows:
Hypothesis 4: The concurrent interactive effect of Asser
tiveness, Dependency, and the SII upon the
Singleness Orientation is greater than the
sum of their individual effects.
15
FIGURE 2
PROBABLE INTERACTIONAL EFFECTS OF
ASSERTIVENESS/ DEPENDENCY, AND
THE SOCIAL INFLUENCE INDEX
Assertiveness
Social Influence Index
Dependency
16
Figure 3, shows the effects of the first four Hypotheses.
Ideally, a study of this type should be made long
itudinally, with a follow-up five or ten years hence to
determine which members of the sample have, in fact, re
mained single. Since such a follow-up is beyond the scope
of the present project, an attempt has been made instead
to get at the dependent variable indirectly, by measuring
the subjects' "Desire to Remain Single," and their "Expec
tation of Remaining Single." This approach is based upon
the assumption that the subjects are sufficiently in touch
with the reality of their own situation that the sum of
their expectations and desires should be a fair index of
their probable future behaviors. In any case, this ap
proach will give a measure of how the subjects feel now,
which is significant in itself. The chief dependent vari
able, called "Singleness Orientation," will be the sum of
the subject's preference for singleness, and his expecta
tion of it.
The pilot project referred to on page revealed
an additional possible source of variance in the likeli
hood of remaining single: "Age." While age is not ger
mane to the other hypotheses under consideration here, it
is well to take the possible effects of age into consider
ation, as a control, if nothing else.
The pilot study showed a curvilinear relationship
FIGURE 3
GRAPHIC SUMMARY OF HYPOTHESES 1 - 4
Hypothesis 2
Hyp. 1
Social
Influence
Index
Dependency
Assertiveness
Goal Attainment
Substitute
Gratifications
Role Conflict
Role Incompatibility
Outside Activity
Role Clarity
18
between age and the desire to remain single (the greatest
desire for singleness occurring during the middle years).
There is a possible rationale for this: Young people de
sire marriage because they want children, because of the
higher status they believe will accrue to them as married
persons, and possibly to fulfill ideals instilled in them
since childhood. People in their forties and fifties have
most likely had all the children they are going to have,
and have experienced the harsh realities of marriage, as
well as its joys. Such persons may well be interested in
meeting some of their own private unfulfilled needs. As
persons approach old age, however, the prospect of living
and dying alone is uninviting, at best, and it is probable
that many such persons seek marriage again, this time for
companionship. Stated in formal terms, the forgoing be
comes :
Hypothesis 5: The desire to remain single (but not the
expectation of it) bears a curvilinear re
lationship to age: the time of greatest
desire to remain single is during the mid
dle years.
Interestingly, the expectation of remaining single
seemed linearly and positively correlated with age in the
same pilot sample. This is likely due to the lack of a-
vailable partners as one grows older (especially for
women), and possibly also due to an increasing lack of
19
hope as one grows older without finding, in fact, a mar
riageable possibility.
Hypothesis 6; The expectation of remaining single (but
not the desire for it) is positively and
linearly correlated with age.
Question 2; What variables (other than "mental health"
factors that affect everyone affect a person's
happiness with the single role?)
The assessment of a person's general happiness
(i.e., sense of well-being) is of particular importance to
the professional counselor-therapist, who is interested in
helping his clients cope with their life situation in the
best possible way. While it is undeniable that a person's
sense of well-being may be affected by a host of factors,
physical as well as interpersonal, that have nothing to do
with his being single (indeed, some persons' pre-existing
psychological problems may have contributed to their being
single instead of married), yet my clinical experience as
a marriage and family counselor has suggested that varia
bles within the single role itself play a part in affect
ing an individual's sense of well-being, independent of
whatever general "mental health" variables might also be
impinging upon him. Some clients who seem to be coping
adequately with life in every other way are nevertheless
intensely unhappy with their single role; other whose
20
lives seem to be a neurotic shambles are able to cope with
their singleness rather well.
Bradburn (1969) has shown that psychological well
being (i.e., overall happiness) is a function of two fac
tors, positive and negative affect. These factors are
independent of each other, and are caused by different
sets of variables. ("Psychological well-being is defined
by Bradburn as the balance between the positive and nega
tive affect sets.") I believe it is possible to show that
the variables under consideration in the present study
contribute significantly to both the positive and the
negative affect sets.
One of the possible generators of negative affect
is conflict. One particular kind of conflict might be
that between the intrapersonal factors (i.e., "Assertive
ness" and "Dependency") and the interpersonal factors
(i.e., the "SII") that impinge hpon the individual. Where
these factors press for contradictory ends, the individual
is likely to experience conflict, which would be reflected
in a higher "Negative Affect" score. For example, the
person with high "Assertiveness" and low "Dependency"
would be expected to prefer the freedom afforded by the
single role. In this case, if the interpersonal factors
are loaded in favor of marriage, the individual will ex
perience conflict, and an increase of "Negative Affect."
21
Similarly, the person with high "Dependency" and low "As
sertiveness" is likely to prefer the security offered by
marriage. If the interpersonal factors in his case are
loaded in favor of singleness, he will experience conflict,
and elevated "Negative Affect." Stated in formal terms,
the proposition explaining "Negative Affect" becomes:
Hypothesis 7: Negative affect is increased to the extent
that there are conflicting relationships
among the three independent variable sets:
E.g., high "Assertiveness" with low "SII"
or high "SII" with low "Assertiveness";
high "Dependency" with high "SII" and/or
high "Assertiveness" with high "Dependency."
It is likely that "Positive Affect" will also be
influenced by some of the variables under consideration in
this study. One element pointed out by Bradburn is that
most kinds of social participation tend to increase "Posi
tive Affect." If this is true, then a number of items
within the "Social Influence Index" would likely contri
bute to "Positive Affect": involvement with group activi
ties, closeness with friends, and availability of partners
for sex and intimate relationships. By deduction, it also
seems likely that "Positive Affect" should be highly corre
lated with "Assertiveness." This is based upon the as
sumption that the highly assertive person is most likely
to involve himself in the kinds of activities, social and
otherwise, that will contribute to the fulfilment of his
22
needs, and to the attainment of his goals. High "Depen
dency," on the other hand, is likely to be correlated nega
tively with "Positive Affect," since such persons tend to
depend upon others for need fulfillment and goal attain
ment, and are thus less likely to be satisfied. Stated in
formal terms, the proposition governing positive affect is:
Hypothesis 8 : "Positive Affect" is positively correlated
with "Assertiveness" and with "Involvement
with Other People," but is negatively cor
related with "Dependency."
The dependent variables for hypotheses 7 and 8 will
be items from Bradburn's "Affect Balance" scale (Bradburn,
1969, p. 267). (See Chapter II for a listing of the
items.)
The hypotheses to be tested in this study are sum
marized as follows:
I. Under general research question 1: What variables
affect the choice of the single role versus the mar-
riajc role?
1. The orientation towards remaining single is posi
tively aorrelated with Assertiveness.
2. The orientation towards remaining single is nega
tively correlated with personal Dependency.
3. The orientation towards remaining single is posi
tively correlated with those interpersonal and
social variables that facilitate the transition
from the married role to the single role, or that
support maintaining the single role.
23
4. The concurrent interactive effect of Assertive
ness, Dependency, and the Social Influence Index
upon the Singleness Orientation is greater than
the sum of their individual effects.
5. The desire to remain single (but not the expecta
tion of it) bears a curvilinear relationship to
age.
6 . The expectation of remaining single (but not the
desire for it) is positively and linearly corre
lated with age.
II. Under general research question 2: What variables
(other than general "mental health" factors) affect a
person's happiness (i.e., psychological well being)
with the single role?
7. Negative Affect is increased to the extent that
there are conflicting relationships among the
three independent variable sets. E.g., high As
sertiveness with low SII or high SII with low As
sertiveness; high Dependency with high SII; high
Assertiveness with high Dependency.
8 . Positive Affect is positively correlated with As
sertiveness and with Invovlvement with Others, but
is negatively correlated with Dependency.
A graphic summary of the entire research project is pre
sented in Figure 4.
FIGURE 4
SUMMARY OF THE RESEARCH HYPOTHESES
Hypothesis 7
+
Hypothesis 1
Assertiveness
Assertiveness-
Dependency
Conflict*
Social
Influence
Index
Involvement
with others
Singleness
Orientation
w c n
H- H-
3 3
iQ iQ
• •
M
X h
ts CD
( D Hi
O (D
rf H
O ( D
rt 3
H- O
0 f i >
3
U’
•
t!
•
C T l U1
Age
rIntervariable conflicts (hypothesized to contribute to "Negative Affect") occur ^
where two variables tend to press the individual in opposite directions (e.g., "Asser
tiveness" presses for singleness and "Social Influence Indicator" presses for marriage).
CHAPTER II
OPERATIONALIZING THE VARIABLES
The purpose of this chapter is to describe the
specific items that were selected to represent the vari
ables described in Chapter I, along with a selection of
demographic variables to be used as controls, and in sample
description and subset comparisons.
"Assertiveness" and "Dependency"
These variables were operationalized in terms of a
selection of items from the Interpersonal Check List (Leary,
19 57). There are several advantages to this method: (1)
The items discriminating these variables are both drawn
from the same frame of reference. (2) The various person
ality categories (described in the literature of the ICL as
octants around a circle graph) have been thoroughly vali
dated, both in terms of test-retest correlations, and
intervariable separation. (See Tables 1 and 2. For a com
plete discussion of the validation results, the reader is
referred to Leary, 1957, pp. 461-462.) (3) The ICL items
are well suited to a pencil-and-paper type questionnaire,
since they are simple both to administer and to score.
25
26
TABLE 1*
INTERPERSONAL CHECK LIST
TEST-RETEST CORRELATIONS
Octants
AP BC DE FG HI JK LM NO
.76 .76 .81 .73 .83 .75 .80 .80
*Tests were made on a sample of 77 Obesity patients.
TABLE 2*
INTERPERSONAL CHECK LIST
AVERAGE INTERVARIABLE CORRELATION
AS A FUNCTION OF THEIR SEPARATION
AROUND THE CIRCLE
Intervariable Distance**
1 2 3 4
.51 .37 .22 .12
*Tests were made on a sample of 77 Obesity patients.
**The mean correlations between variables one step
apart, two steps apart, etc.
Source: Timothy Leary, The Interpersonal Diagnosis
of Personality (New York: Ronald Press
Company, 1957), Tables 41 and 42, pp. 461,
462.
27
11 Assertiveness. " "Assertiveness" is meant to
measure the capacity to be for oneself, i.e., to seek to
meet one's own needs, and to assume what leadership func
tions may be necessary to bring this about. For the pur
poses of the present questionnaire, a selection of 2 0 items
was made from the "Competitive-Narcissistic" octant (BC) of
the ICL which measure the capacity to be for oneself, and
from the "Managerial-Autocratic" octant (AP), which measure
the capacity to assume leadership (See Leary, 1957, pp.
170-171). These and all other octants of the ICL contain
items that are both socially adjustive and maladjustive;
an effort was made in this questionnaire to avoid selecting
the most flagrantly maladjustive items. The items selected
to measure the Assertive dimension are these:
Makes a good impression. ___ Self confident.
Able to give orders. ___ Self reliant and asser
tive.
Self respecting___________ ___ Businesslike.
Independent._______________ ___ Likes to compete with
others.
Able to take care of self. Dominating.
Proud and self satisfied. ___ Can be indifferent to
others.
Often admired.________________ Thinks only of him(her)
self.
Respected by others. _ Manages others.
28
Good leader. Somewhat snobbish.
Likes responsibility. Selfish.
"Dependency." The term "dependency" is used here
to describe a tendency to rely upon others for the meeting
of one's own emotional needs— to "need" other people.
Such a person tends to be trusting, clinging, agreeable
and congenial, and to bring this about, he will tend to
behave passively and indecisively. The 20 items selected
for the Dependency dimension were taken from octants HI
(self-effacing-masochistic), JK (docile-dependent), and LM
(cooperative-overconventional) (See Leary, pages 170-171
for a description of these octants). The complete set of
"Dependency" items is:
Usually gives in. Grateful.
Appreciative. ___ Admires and imitates
others.
Anxious to be approved ___ Eager to get along with
of. others.
Lacks self confidence. ___ Affectionate and under
standing.
Modest. ___ Accepts advice readily.
Trusting and eager ___ Wants everyone to like
to please. him (her).
Warm. ___ Shy.
Dependent._________________ Lets others make deci
sions.
29
Hardly ever talks ___ Too easily influenced by
back. friends.
Likes to be taken ___ Wants everyone's love.
care of.
Since a list of personality self-evaluation items
such as these might appear formidable to a shy or defen
sive person if they were placed at the beginning of a
questionnaire, the "Assertive-Dependency" items were bur
ied inside the questionnaire form (on page 3), where they
might seem less incriminating to a subject who had al
ready completed the interesting and less formidable "So
cial Influence Index" items. To reduce the likelihood of
forming response sets, and yet not produce too complex a
problem in scoring and coding, the two sets of items were
printed together, as a continuous 40-item list. Scoring
consists of tabulating the number of check marks in each
set; thus, each variable can have a score ranging from
zero to 2 0 .
The "Social Influence Index"
This index is compiled from a variety of inter
personal and social factors, as described more fully in
Chapter I. The items are displayed as four-point Likert
scales. There are 21 items in all, covering the range of
role-theory categories described in Chapter I, and derived
30
from Burr's (1972) axiomatic transformation of role theory.
Role Clarity and Understanding. These terms are
represented by items 1-4:
Strongly
Agree:
1. A person can have a
normal and happy life
without being married: ____
2. Being single has many
advantages over being
married:
3. Sometimes I'm not sure
what is expected of me
as a single adult:
4. There is a place and
function for single
persons in our society:
Role Conflict and Strain. These terms are repre
sented by the next four.items. Pressures brought about by
family, religion, moral strictures against extramarital
sex, and by the urgings of a companion who might desire
marriage are taken as typical conflicts in the lives of
single adults.
Strongly Strongly
Agree Disagree
5. Members of my family
would really like to
see me married:
Strongly
Disagree:
31
Strongly
Agree
6 . My religion teaches
it is better to be
married than to be
single: ____
7. Sex outside of
marriage is morally
wrong:
8 . I now have a com
panion who would like
to get married:
Role Incompatibility. This term is expressed in
terms of incompatibility of the single role with the par
ent role (item 9), and with married friends (item 10).
Yes: Some- No Not a
times: parent:
9. I find it hard to be
a parent and a single
person at the same
time:
10. Concerning friends:
a. Most of my good friends are married.
b. I have many good married friends, a few single
friends.
c. I have many good single friends, a few married
friends.
d. Most of my good friends are single.
Role Compartmentalization. This term is repre
sented by two situations: child custody (item 1 1 ), and
job demands (item 1 2 ):
Strongly
Disagree
32
11. Concerning minor children:
a. The children are with me most of the time.
b. My ex-spouse frequently takes the children
for extended visits.
c. The children are with my ex-spouse most of
the time.
d. There are no minor children.
12. Concerning my job:
a. For reasons of public attitudes, social
obligations, etc., it is far better to be
married than single on my job.
b. It is somewhat better to be married than
single on my job.
c. On my job it doesn't matter whether one is
single or married.
d. It's better to be single on my job.
Activity and Involvement Outside the Family. This
term is measured in terms of time spent at work (item 13),
time spent in group activities (items 14, 15), and time
spent on hobbies and recreation (item 16):
13. How much time do you spend at work each week?
a. Over 60 hours.
b. 40 - 60 hours.
c. 20 - 40 hours.
d. Under 20 hours.
14. How much time do you spend each week in group activi
ties with people of your own sex (i.e., National
Guard, volunteer organizations, etc.)?
a. Three times weekly, or more.
b. Once or twice weekly.
c. Less than once a week.
d. Don't belong to any such organizations.
33
15. How much time do you spend with mixed-sex organiza
tions for singles?
a. Three times weekly, or more.
b. Once or twice weekly.
c. Less than once a week.
d. Don't belong to any such organizations.
16. How much time do you spend each week on recreation
and/or hobbies?
a. Ten hours or more.
b. Five to ten hours.
c. One to five hours.
d. Less than one hour.
Goal Attainment. This term is arrived at indi
rectly in item 17, which asks how important it is to an
individual to live as a part of a family, and more direct
ly in item 2 0 , which asks how important marriage is as a
life goal:
17. Concerning your living arrangements:
a. I prefer to live as a part of a warm and
caring family.
b. I prefer a roommate type arrangement with
a relative.
c. I prefer living with a roommate who is not
a relative.
d. I prefer to live alone.
20. Concerning your life goals (career, home, etc.):
a. Being married is essential to my goals.
b. I would like marriage, but it's not essen
tial to my goals.
c. It doesn't matter to my goals whether I'm
married or single.
d. Being single is essential to my goals.
Substitute Gratifications. This term to the
34
closeness of marriage are found in the availability of
satisfying sexual relationships (item 18), and intimacy
(item 19):
18. Concerning your sex life as a single person:
a. My sex life is fully satisfying to me.
b. My sex life is only partially satisfying.
c. My sex life is not satisfying to me.
d. I do not engage in sexual activities.
19. Concerning your friendships as a single person:
a. I have friends with whom I feel perfectly
close and trusting.
b. I have friends, but I wouldn't feel free to
confide some things to them.
c. My friendships are mostly superficial.
d. I don't really have any friends.
The above 20 items are scored on a scale of 1 to
4, with the highest number favoring the single role. Thus,
the higher the "SII" score, the stronger the interpersonal
and social factors pressing towards singleness.
One additional item is included in the "SII" set.
While it is somewhat related to Burr's category of "ease
of transition" from one role to another, it is not identi
cal to it. Two items, 21 and 22, are included on the as
sumption that the more pain and unhappiness a person expe
rienced in a previous marriage, the less likely he will be
to try again, but that the memory of the unhappiness is
dimmed over time.
35
If you have never been married, check here:____ and go on
to Item 23.
If you are widowed or divorced, check the statement on
each of the following items that is most true for you:
21. My previous marriage, overall, was
a. Very happy.
b. Fairly happy, most of the time.
c. Fairly unhappy, most of the time.
d. Very unhappy.
22. My previous (most recent) marriage ended
a. Less than six months ago.
b. Between six months and one year ago.
c. From one to two years ago.
d. More than two years ago.
Item 21 is scored from 1 (a) to 4 (d), and item 22 from 0
(a) to 3 (d). The net item score is computed by substract-
ing 2 2 from 2 1 , with the proviso that the net score is
never entered at less than zero. (Persons who have never
been married receive zero for the item, since they have
had no first hand experience prejudicing them against mar
riage. )
When the item 21-22 combination is included, the
"SII1 * has a total saore range of 20 (strongly favoring
marriage) to 84 (strongly favoring singleness). (This was
later transformed mathematically into a 0 - 2 0 point scale,
to match the scales of the other variables. See Chapter
V.)
36
Involvement with Others
This term is a subset of the "SII" and is only used
in the testing of one hypothesis (#8 ). It is computed as
the sum of the scores of the items dealing with social
groups (items 14, 15), sexual relationships outside of
marriage (18), and friendships (19). The numerical score
range of this variable is from 4 (low involvement) to 16
(high involvement).
"Age"
This term is taken from the demographic section of
the questionnaire. In this section age is given in five
classes, from "under 30" to "over 60." The variable score
ranges from 1 (youngest) to 5 (oldest) . ,
The Variables, "Singleness Orientation,"
"Singleness Preference," and
11 Singleness Expectation"
These terms are all derived from questionnaire
items 2 3 and 2 4.
23. Please check the statement that best expresses what
you would like;
a. I would definitely like to remain single.
b. Mostly I would like to remain single, but
some things about singleness bother me.
37
c. Right now I'm about evenly divided: I
don't know whether I wantto remain sin
gle or get married.
d. Mostly I would like to get married, but
some things about marriage bother me.
e. I would definitely like to get married.
24. Please check the statement that best expresses what
you expect:
a. I expect to spend the rest of my life as a
single person.
b. I'll probably remain single, but there's a
small possibility that I might get mar
ried.
c. Chances are about 50-50 whether I get mar
ried or not.
d. I'll probably get married, but there's some
possibility that I may not.
e. I fully expect to get married.
Each item is scored on a scale ranging from 5 (favoring
singleness) to 1 (favoring marriage). The sum of the two
items is the "Singleness Orientation." The variables
"Negative Affect!1 and "Positive Affect" are taken directly
from the Bradburn "Affect Balance" scale (Bradburn 1969,
p. 267).
25. During the past few weeks, did you Yes No
ever feel----
a. Particularly excited or interested
in something?______________________ ____ ___
b. So restless that you couldn't
sit long in a chair?___________________ ___
c. Proud because someone compli
mented you on something you had
done? ____ ___
d. Very lonely or remote from other
people? ____ ___
e. Pleased about having accom
plished something? ____ ___
38
Yes No
f. Bored?_____________________________ ____ ____
g. On top of the world? ____ ____
h. Depressed or very unhappy? ____ ____
i. That things were going your
way? ____ ____
j. Upset because someone
criticized you?___________________ ____ ____
The statistical validation of the Affect Balance Scale is
discussed in Bradburn, chapter 5 (pages 71-89). Each
"yes" answer on items a, c, e, g, and i gives one point
to the Positive Affect Score; each "yes" on items b, d,
f, h, and j gives one point to the Negative Affect score.
For purposes of this research, the appropriate yeses will
simply be tallied to give a range of 0 to 5 for each
variable.
Demographic information is requested in the ques
tionnaire for the purpose of identifying various sample
subsets, and for running control statistics, if needed.
Categories include sex, age, income, and education, self-
rated degree of religiosity, and present living arrange
ments. The complete questionnaire, in the form in which
it was administered to the sample, is given in Appendix
A.
The completed questionnaire was pretested on a
sample of 22 members of a Single Parents' group, and on
12 single members of a therapy group. The pretest indi
39
cated that there was enough variance in all of the items
to suggest a high probability of interpretable results.
CHAPTER III
METHOD OF DATA ANALYSIS
NOTE: In the statistical formulas used in this chapter,
it is convenient to refer to the variables by code names.
The independent and dependent variables employed in this
study, along with their codes, are as follows:
Independent Variables Dependent Variables
XI: Assertiveness Y6 : Singleness Orientation
X2: Dependency Y7: Negative Affect
X3: Social Influence Index Y8 : Positive Affect
X4: Involvement with Others. Y9: Singleness Preference
X5: Age Y10: Singleness Expectation
The data to be gathered in this study can be ana
lyzed in terms of a multivariate linear regression model
(McNemar, 1962, p. 177), with modifications suggested by
Acock and DeFleur (1972) to allow for the effects of con
figurations of variables acting concurrently. However,
consideration must be taken of two possible violations of
the assumptions underlying such a model.
First, the multivariate linear regression model
demands that there be linear relationships between each
40
41
independent variable and the dependent variable(s). One
of the independent variables, "Assertiveness," will only
meet the linearity assumption in part. As indicated in
the theoretical chapter (see Figure 1), this variable is
hypothesized to be linear throughout most of its span, but
reverses itself at the very low end, near the zero point.
In nonstatistical terms, a person with extremely low as
sertiveness is not likely to get married, as the strict
regression line would predict, because he lacks the neces
sary aggressiveness to seek out a marriage partner. How
ever, I do not expect such extremely low "Assertiveness"
scores to appear in this sample in any significant numbers.
It is more likely that such persons will be self-excluded,
since the sample will be drawn primarily from organiza
tions for single people, and a certain minimum level of
assertiveness is required to join such an organization.
Therefore it does not seem unreasonable to treat the "As
sertiveness" variable as if it were fully linear.
Another assumption of the multivariate linear re
gression model is that the data be of an interval level.
Two of the dependent variables in the present study will
not meet this assumption: "Desire to remain single," and
"Expectation of remaining single." These variables will
be measured in terms of a 5-point ordinal scale. However,
Labowitz (1967) has shown that the error produced in ap-
42
plying product-moment correlational procedures to ordinal
data is very slight, even when gross differences in the
ordinal interval size are allowed. Since the multivariate
regression model offers important interpretive advantages
over comparable ordinal procedures, and since the statis
tics involved are sufficiently robust to allow for minor
violations of the basic assumptions without significant
errors, the use of the more powerful technique seems war
ranted, although due allowance will need to be made, when
the results of the analysis are interpreted, for the lib
erties that have been taken.
Question 1: What variables affect the choice of the sin
gle role versus the marriage role?
The first general proposition of this study is
that three variables— "Assertiveness," "Dependency," and a
variabled called "SII" (which is an aggregation of inter
personal and social factors)— will have an effect upon a
person's decision to remain single or to seek (remar
riage. The dependent variable, a "Singleness Orientation,"
is a summary of the subjects' desire to be single (ques
tionnaire item 23) and their expectation of remaining sin
gle (item 24).
Hypothesis 1: The orientation towards remaining single
(Yg) is positively correlated with Asser
tiveness (X]_) .
43
Hypothesis 2: The orientation towards remaining single is
negatively correlated with personal Depend
ency (X2 ) •
Hypothesis 3: The orientation towards remaining single is
positively correlated with those interper
sonal and social variables that facilitate
the transition from the married role to the
single role, or that support maintaining
the single role (X3 ).
These first three hypotheses can be analyzed by
means of the zero-order correlations between the indicated
variable pairs. Hypotheses 1 and 3 predict that the cor
relations rX^Yg and rX3Yg will both be positive, while
hypothesis 2 predicts that the correlation rX3Yg will be
negative. However, the square of the correlation coeffi
cients in the above cases will not give a true picture of
the amount of variance in Yg (i.e., "Singleness Orienta
tion") explained by each of the independent variables,
since they are assumed to be functioning concurrently, and
therefore their interactional effects must be taken into
account. The next hypothesis deals with the interactional
factor:
Hypothesis 4: The concurrent interactive effects of As
sertiveness (Xi), Dependency (X2 ) and the
SII (X3 ) upon Singleness Orientation (Yg)
is greater than the sum of their individual
effects.
The simple additive effects of the three inde
pendent variables upon the dependent variable can be ex-
44
pressed in terms of a three-dimensional model. If the
variables are first converted into standard scores, the
mathematical representation of the model is:
ZY g = 3l Zx 1+ e 2zx 2+ e 3zx 3
The term "beta" (3) in each case represents the factor by
which the corresponding variable must be multiplied to
give the amount of change in Y that is accounted for by
that particular variable. In other words, when multi
plied by Zx^, gives the amount of variation in Yg (ex
pressed in standard scores) that is accounted for by vari
able Xj. (3 is analogous to the "slope of the line" fac
tor in bivariate regression.) The absolute sum of the
betas gives the relative amount of variance in Yg that is
accounted for by variation in Xj, X2 / and X3 , operating
concurrently.
However, the above model does not enable one to
estimate the amount of variance in Yg that is directly
attributable to interaction among the several independent
variables. Chapter I has described in detail the basis
for supposing that a considerable amount of the variance
in Yg is brought about through such interaction. This is
stated concisely in Hypothesis 4. Acock and DeFleur
(1972) have presented a mathematical model that describes
45
these interactional effects.
In addition to the simple additive multivariate
model shown above, four terms are required:
3 5 ZX^ZX3 , and describe the interactional effects
of the variables taken in pairs; ^7 zx^zX2 ZX3 describes
the effect of the interaction of the three variables taken
together. (If there should happen to be no interaction in
any of the above variable combinations, its "beta" would
become zero, and the entire term would disappear.)
The complete regression model, showing both indi
vidual and interactional effects, is:
Zyg_ ^lZXjL+ ^2ZX2+ ^3ZX3+ ^4ZX j ZX2+ ^5Zx^ZX3+
^6ZX2ZX3+ ^7ZX],ZX2ZX3
Since each "beta" in the above equation represents the
relative amount of variance contributed by the term of
which it is a part, then Hypothesis can be considered dis-
proven if |3i+32+^31 1 1 ^4+^5+36+^7I•
Hypotheses 5 and 6 both consider the effects of
age as an independent variable:
Hypothesis 5: The desire to remaing single (Yg) (but not
the expectation of it) bears a curvilinear
relationship to age (X5 ).
46
Hypothesis 6 ; The expectation of remaining single (Yjq)
(but not the desire for it) is positively
and linearly correlated with age.
Both of these hypotheses can be analyzed by means of their
zero-order correlations. However, since Hypothesis 5 pre
dicts a curvilinear relationship between the variables,
the proper coefficient in this case will be "eta" rather
than Pearson's "r" (McNemar, 1962, page 202).
Question 2: What variables (other than general "mental
health" factors) affect a person's happiness
(i.e., his psychological well being) in the
single role?
The second major question under consideration in
this study predicts that the overall happiness of the sin
gle person (i.e., his psychological well-being) is related
somewhat, at least, to the three independent variable sets
of the present research. Since Bradburn (1969) has found
that psychological well-being consists of two independent
elements, positive affect and negative affect, separate
hypotheses must be generated to explain each element.
NOTE: Since many "mental health" factors that contribute
to psychological well-being are not being measured or con
trolled for (except by assumption that they are randomly
distributed throughout the sample), it is anticipated that
there will be a relatively high amount of unexplained
variance in the tests of Hypotheses 7 and 8 .
47
Hypothesis 7: Negative Affect (Y7 ) is increased to the
extent that there are conflicting relation
ships among the three independent variable
sets (X]_, X2 / and X 3 ) . E.g. : high Asser
tiveness with low SII, or high SII with low
Assertiveness; high Dependency-wiLh~high---
SII; high Assertiveness with high Dependeh-
cy.)
The mathematical model for Hypothesis 8 requires three
terms, one for each of the conflict situations described
above. However, it is not likely that all three terms
will have high values at the same time in any particular
case.
ZY7~^8 I zx 1~ zX3 l+ ^9zx2 ZX3+ ^10zX2ZX2*
While it anticipated that the "betas" will be small be
cause of the large amount of expected unexplained vari
ance, they should be statistically "significant" if the
hypothesis is to be credited with any vitality.
Hypothesis 8 describes the anticipated effect of
the three independent variables upon Positive Affect. For
this hypothesis only, a subset of the "Social Influence
Index" will be used, labeled "Involvement with Others"
(X4 ) .
Hypothesis 8 : Positive Affect (Yg) is positively corre
lated with Assertiveness (Xi) and with In
volvement with Others (X4 ), but is nega
tively correlated with Dependency (X2 ).
48
The rationale behind this hypothesis is discussed in Chap
ter I. The simplest way to describe this hypothesis math
ematically would be in terms of a simple three-dimensional
regression formula. However, if this were expanded to in
clude interactive effects among the independent variables
as well, in a manner like that used to test Hypothesis 4,
above, the interpretive potential of the model is greatly
enhanced.
Zy8“3llZx1+3l2Zx2+3l3Zx4+3l4Zx1zx2+3l5Zx2Zx4+
^16zx 2Zx 4+ ^17zX j Zx 2zx 4 •
Hypothesis 8 predicts that 3^ and 8 ^ 3 will be positive,
while 8^2 will be negative. Effects caused by the inter
action of variables will be measured by 8 1 4 f 3 i5 ,
317.
The procedures described in this chapter are all
within the capabilities of the computer program "Statisti
cal Package for the Social Sciences" (Nie, Bent, and Bell,
1970).
CHAPTER IV
THE SAMPLE AND ITS SUBSETS
One of the goals of this study has been to gain a
sampling of data from as representative a cross section of
the single adult population as possible, and from those
older than college age, in particular. Of the relatively
few recent studies that have been made of the ramifications
and parameters of singleness, most have been based upon
college student populations, (see, for example, Kirkendall,
1961) or on even more specialized subpopulations, such as
those engaging in group sex or communal living (see Brech-
er, 1969, Chapter 9). Little has been done up to this
time to tap the attitudes and goals of single adults from
the workaday world, and especially of those older than the
early 30's.
"Representativeness" can best be achieved in a re
search sample through the costly and time consuming proc
ess of random, or stratified random sampling. Since ade
quate funds and time necessary to make such a sampling
were not available, an attempt was made to simulate repre
sentativeness as much as possible within the limitations
of what could be done. This was accomplished through the
sampling of various subsets of the "singles" population in
49
50
the Los Angeles area. Four such subsets were chosen:
CLIENTS (persons now engaged in counseling), SECULARS
(members of secular organizations for singles), RELIGIOUS
(members of church-sponsored organizations for singles),
and OTHERS (persons not connected with counseling or with
singles organizations). This method has produced a sample
that is well distributed with respect to sex and age, but
which is skewed positively with respect to income, and
negatively with respect to education. The complete dis
tributions of the sample and its subsets by sex, age, in
come, education, and religiosity are given in Tables 3 to
7, on the following pages.
Differences Within the Sample: By Subsets
Significance tests broken down by subsets for the
first two of the predictor variables, "Assertiveness" and
"Dependency" showed no significant differences among any
of the subsets. When the "Social Influence Index" was
taken as the criterion, it was found that the sociologi
cal factors pressing towards singleness were significantly
higher for the subset OTHERS than for the subset SECULARS.
When the dependent variable "Singleness Orientation" was
taken as a criterion, the subset OTHERS was found to be
significantly different from the subsets CLIENTS, AND
51
TABLE 3
SAMPLE DISTRIBUTION BY SEX
Total N Male Female No Data
Total Sample: 281 109 (40%) 161(59%)
1 1 ( 1 %)
Subset CLIENTS: 18 8 (47%) 9 (59%) 1(4%)
Subset SECULARS: 157
6 6 (42%)
85 (54%) 5 (4%)
Subset RELIGIOUS: 69 24 (36%) 41 (61%) 3%
Subset OTHERS: 37 11 (30%) ‘ 26 (70%) 0
52
TABLE 4
SAMPLE DISTRIBUTION BY AGE
(Modal classes underlined)
Whole
Sample
Subset
CLIENTS
Subset
SECULARS
Subset
RELIGIOUS
Subset
OTHERS
30 & under 55 (20%) 9(50%) 19 (12%) 8 ( 1 2 %) 19 (51%)
31-40 82(30%) 8 (44%) 45 (29%) 19 (28%) 10 (27%)
41-50 L04(37%) 1 (6 %) 67 (43%) 30 (43%) 6(16%)
51-60 34(12%) 0 23 (15%) 9 (13%) 2(5%)
Over 60 3(1%) 0 1 (1 %) 2 (3%) 0
No Data 3(1%) 0 2 ( 1 %) 1 (1 %) 0
53
TABLE 5
SAMPLE DISTRIBUTION BY INCOME
(Modal Glasses underlined)
Whole
Sample
Subset
CLIENTS
Subset
SECULARS
Subset
RELIGIOUS
Subset
OTHERS
Under
1 0 , 0 0 0 113(42%) 6 (35%) 6 6 (44%) 25 (37%) 16 (43%)
1 0 -2 0 , 0 0 0 121(45%) 9(53%) 59 (40%) 33 (49%) 20 (54%)
20-30,000 24(9%) 2 (1 2 %) 15 (10%) 7 (10%) 0
Over
30,000 13(5%) 0 10 (7%) 2 (3%) 1(3%)
No Data* 1 0 1 7 2 0
*Cases with No Data were not considered in the com
putation of the percentages.
54
TABLE 6
SAMPLE DISTRIBUTION BY EDUCATION
(Modal classes underlined)
Highest
grade or
degree
Whole
Sample
Subset
CLIENTS
Subset
SECULARS
Subset
RELIGIOUS
Subset
OTHERS
Under 12 3 (IS) 0 3 (2%)
0 0
H.S.Grad 56 (21%) 4 (22%) 39 (27%) 9 (14%) 4 (11%)
Some Coll. 83(31%) 4 (22%) 54 (37%) 17 (26%) 8 (2 2 %)
Bachelor's72 (27%) 4 (22%) 30 (21%) 25 (38%) 13(35%)
Master's 39 (15%) 4(2 2 %) 16 ( 1 1 %) 9 (14%) 10 (27%)
Doctor1s 11 (4%) 2 ( 1 1 %) 3(2%) 4 (£%) 2(5%)
No Data 17 0 1 2 5 0
55
TABLE 7
SAMPLE DISTRIBUTION BY RELIGIOSITY
(Modal classes underlined)
Whole Subset Subset Subset Subset
Sample CLIENTS SECULARS RELIGIOUS OTHERS
Very
religious
2 2 (8 %)
6(33%) 10 (7%) 6(9%) 0
Moderately
religious 131 (478) 7 (39%) 74 (48%) 39 (57%) 11 (30%;
Not
-
religious 124 (45%) 5(28%) 71 (46%) 23 (34%) 25 (6 8 %;
No Data * 4 0 2 1 1
*Cases with No Data were not considered in the com
putation of the percentages.
56
SECULARS (See Appendix B, tables 10 to 13). However, when
multiple regressions were run on the effects of "Aggres
siveness," "Dependency," and the "SII" upon "Singleness
Orientation," the differences between the multiple corre
lation coefficients among the subfiles was insignificant
(See Appendix B, table 16).
When "Negative Affect" was taken as a criterion, a
significant difference was found between the subset OTHERS
and the subset SECULARS, and when "Positive Affect" was
taken as a criterion, significant differences were found
between subsets OTHERS and CLIENTS, and between CLIENTS
and SECULARS. However, when multiple regressions were run
using "Negative Affect" and "Positive Affect" as dependent
variables, again the differences between the correlation
coefficients of the several subsets were found to be insig
nificant. (See Appendix B, Tables 14, 15, 17 and 18). It
can be concluded, therefore, that for the purposes of this
study the four subsets can be treated as a single, homo-
genious sample.
Differences Within the Sample
By Demographic Categories
When the means of the main independent variables
were broken down by Sex, significant differences were en-
57
countered in every case: Males were found to be more
"Assertive" than the females (p. < .001), and females were
more "Dependent" than the males (p. < .01). The "SII"
factors conspired to dispose the males more towards sin
gleness than the females (p. < .001). However, when the
main dependent variable, "Singleness Orientation," was
broken down by sex, the difference between the means was
statistically insignificant, indicating that there is no
significant difference in the preference for singleness
by either sex (See Appendix B, Tables 19-22). Multiple
regressions broken down l?y "Sex" were similarly insignifi
cant (Appendix B, Table 2 3). These results were confirmed
by the Automatic Interaction Detector programs which
showed that when "Singleness Orientation," "Singleness
Preference," and "Singleness Expectation" were taken as
dependent variables, sex differences did not account for
any significant amount of variance.(See Figure 5; Appen
dix D, Figures 11 and 12; and discussion in Chapter VI.)
Where "Positive Affect" and "Negative Affect" were taken
as dependent variables, some variance was explained by
differences in sex, but only after the effects of "Assert
iveness," "Dependency," "SII," and "Age" had been re
moved. (See Figure 5, Figure 7, and discussion in Chapter
VI. )
When the Chief independent variables were broken
58
down by "Age," some differences were found, but on the
dependent variable, "Singleness Orientation," differences
between the means of the several age categories were in
significant (See Appendix B, Table 27). The Automatic In
teraction Detector showed that age contributed to variance
in "Negative Affect" (higher in persons under 40), and in
"Positive Affect" after the effects of "Assertiveness" and
the "SII" had been removed (higher in persons under 40).
However, when the multiple regressions on the same de
pendent variables were broken down by age, the difference
between the multiple correlation coefficients were found
to be insignificant. (See Appendix B, Table 28).
The category "Religiosity" yielded one unexpected
result: the group who designated themselves as "Very Re
ligious" were significantly more inclined towards single
ness than were the other classes, both in terms of the
mean "Singleness Orientation" broken down by "Religiosity,"
and in terms of the multiple regression coefficients on
the "Singleness Orientation," broken down by "Religiosity"
(See Appendix B, Table 29, 30). A closer examination of
the "Very Religious" category (Table 29, part II) reveals a
significant difference between the mean "Singleness Orien
tation" for the "Very Religious" persons under 40, and
those for persons 41 and older. The Mean "Singleness Ori
entation" scores for the two younger classes (5.60 for
59
those 30 and under, and 4.75 for those 31-40) do not dif
fer significantly from the Mean "Singleness Orientation"
of the total sample (5.10), but the means for the over-40
classes (7.60 and 7.00) were well beyond the 99% confi
dence interval around the sample mean. This may suggest
that among the very religious, age is a secondary factor
(possibly relating to beliefs that marriage is primarily
for childbearing, or that remarriage after divorce is a
sin). However, the total number of the "Very Religious"
class is too small to permit reliable inferences.
The categories "Income" and "Education" were not
tested for differences, since they did not show sufficient
variance to yield interpretable results. In the case of
"Income," 87% of the sample fell in the lower middle range
(from above the poverty level to $20,000 per year). Both
the very poor and the very rich were under-represented in
the sample. Another factor clouding the interpretability
of statistics based upon "Income" is the large number of
divorced mothers in the sample, who live totally or in
part upon support payments from their ex-husbands, and
whose income level would not reflect either their atti
tudes or their life style accurately.
"Education" was the most biased of all the demo
graphic categories, with 79% of the sample having had at
least some college, and only 1% who were not at least High
60
School graduates.
Summary of the Chapter
and General Comments
The sample consisted of 281 subjects drawn from
four subsets of the single adult population in the Los
Angeles area: clients in counseling, members of secular
organizations for single adults, members of church-spon
sored organizations for singles, and single persons drawn
at large, unconnected with the above groups. The sample
was drawn largely from the lower middle income range, and
from the better-than-average educated. It was well dis
tributed by categories of sex, age, and religiosity.
When the major variables under consideration were
broken down by subsets, no significant differences were
found. The demographic categories of sex, age, and reli
giosity each contributed small, but significant amounts of
variance on one or another of the variables. These dif
ferences will be considered in detail in Chapters VI and
VII.
Overall, the most interesting characteristic of
the sample was its uniformity with respect to the major
variables, in spite of efforts made to obtain variety.
This suggests that attitudes towards marriage and single-
61
ness are still rather widely and uniformly held across the
single adult population.
CHAPTER V
DESCRIPTION OF THE VARIABLES
The independent variables "Assertiveness" and
"Dependency" were measured by means of scores taken direct
ly from the questionnaire, as were the dependent variables
"Negative Affect," "Positive Affect," Singleness Prefer
ence," and "Singleness Expectation." Certain other vari
ables had to be computed from components taken from the
coded questionnaire. Among these were the independent
"Social Influence Index" and "Involvement with Others," and
the chief dependent variable, "Singleness Orientation."
The Directly Scored Variables
These variables need only brief mention here. The
two independent variables in this class, "Assertiveness"
and "Dependency," were each constructed from twenty items
taken from the Interpersonal Check List (See Chapter II,
and Appendix A). They had a possible range of zero to
twenty, and the full range was utilized. Mean "Assertive
ness',' score for the entire sample was 10.21, mean "Depen
dency" score was 8:31. Distributions are nearly normal
with respect to skew and kurtosis, except for a moderate
62
63
positive skew (.43) in the "Dependency" variable (See Ap
pendix C, Table 31).
The dependent variables "Negative Affect" and
"Positive Affect were derived from the Bradburn "Affect
Balance" scale (see Chapter 2). Each had a range of zero
to five. Mean "Negative Affect" for the sample was 2.34,
mean "positive Affect" was 3.46. The "Negative Affect"
distribution was mildly platykurtic (-1.24); the "Positive
Affect" distribution had a marked negative skew (-.78) . The
variables "Singleness Preference" and "Singleness Expecta
tion" each had a range of one to five. Mean Singleness
Preference was 2.54; mean Singleness Expectation was 2.67.
The "Singleness Preference" distribution showed a mild
positive skew (.48); the "Singleness Expectation" distribu
tion was nearly normal (See Appendix C, Table 32).
The Computed Variables
The independent "Social Influence Index" was con
structed from a set of twenty Likert-type items scaled
from 1 to 4, based upon topics derived from Burr's role
theory (See Chapter 2). Items were scored so that higher
scores indicated greater anticipated singleness. Of the
twenty original items, ten correlated with "Singleness
Orientation" at a significance level beyond .01; these
ten best items— which included representatives from every
64
one of the Burr-derived classes— were used to compute the
final "SII" (See Appendix C, Table 33). When the ten items
were summed, the "SII" scale would have a possible range of
zero to 40. In order to give it a twenty-point range simi
lar to that of the "Assertiveness" and "Dependency" vari
ables, the summed score was divided by two. In its final
computed form the "SII" had an actual range of 12, from a
minimum of 6 to a maximum of 18. Mean for the entire sam
ple was 12.27; the distribution was slightly platykurtic
(-.6 6 ) but nearly normal with respect to skew (See Appen
dix C, Table 34).
Since the "Social Influence Index" purports to in
dex the combined effects of the interpersonal factors that
affect one's orientation for singleness (in contrast to
the intrapersonal factors of "Assertiveness" and "Depen
dency," and since the components for this variable admit
tedly do not include the entire universe of social factors
that might possibly play a part in the interactions, (see
discussion in Chapter I) it is a matter of some interest to
determine whether the "SII" as presently constituted does,
in fact, have a single, unidimensional effect upon the
other variables, or whether it represents several indepen
dent sources of variance. Therefore, factor analyses were
run to test the degree of unidimensionality present in the
"SII."
65
The "Alpha" method of factoring was selected for
this task. This method is based upon the assumption that
the independent variables are a sampling of a larger uni
verse of possible variables, and the factors generated by
the "Alpha" method are meant to be representative of that
universe. The rotational method selected was "Quartimax,"
since this type of rotation is most likely to reveal the
characteristic of unidimensionality, if such is indeed
present (see Nie, et al., pages 220, 223).
The analysis broke the "SII" down into three or
thogonal factors (see Table 8 ). The first factor, account
ing for 6 8 % of the variance, loaded most highly on those
items that pertained to clarity of understanding of single
ness role, and to the general feeling that life as a single
person could have meaning and significance.
Factor 2, accounting for 20.5% of the total vari
ance, loaded most highly on those items pertaining to con
flicts within the single role, or to conflicts between
goals and the single role.
Factor 3, accounting for the remaining 11.5% of the
variance, loaded highly on only one item: family pressure.
(For further commentary on the significance of these fac
tors, the reader is referred to the full discussion in
Chapter 7).
The independent variable "Involvement With Others"
66
TABLE 8
FACTOR ANALYSIS OF
THE SOCIAL INFLUENCE INDICATOR
Factor 1 (6 8 % of the total Variance)**
Variable #
Factor
Loading
0 0 1 .71 A person can lead a normal... life without marriage,
0 2 0 .58 Being single (married) is essential to my goals.
0 0 2
r»
in
•
Being single has many advantages over...marriage.
003 .43 I understand what is expected of me as a single
person.
Factor 2 (20.5% of total Variance)**
009
017
016
0 1 0
.53
.48
.41
. 31
I find it hard to be a parent and a single person.
I prefer...a warm and caring family.
I spend (1 to 10 or more) hours weekly on recrea
tion.
Most of my good friends are single (married).
Factor 3(11.5% of the total Variance)**
005 .57 Members of my family would like to see me married.
*"Alpha" type factoring was used, with "Quartimax" rota
tion.
**Vajriables with factor loadings less than . 30 have been
omitted.
67
(used in Hypothesis 8 ) was computed by summing four of the
Likert-type role-theory items (See Chapter 2). The compu
ted variable had a theoretical range of zero to 16, and an
actual range of 12 in the sample, from a minimum of 3 to a
maximum of 15. Mean for the whole sample was 10.67. The
distribution was mildly leptokurtic (.63), and had a
marked negative skew (-.76). (See Appendix C, Table 34).
The dependent variable "Singleness" and "Expecta
tion of Singleness" items. It had a possible range of zero
to 10, and an actual range of 1 to 10. Mean for the sample
was 5.11. Skew was moderately positive (.46); kurtosis was
negligible (See Appendix C, Table 34.
CHAPTER VI
FINDINGS
In Chapter I of this paper, the purpose of the
present project was described under two general research
questions:
1. What variables affect the choice of the single role
versus the marriage role?
2. What variables (other than general "mental health"
factors) affect a person's happiness with the
single role?
These two general questions generated eight specific re
search hypotheses. In the interest of parsimony, the
findings of this research will be presented in the same
order that the general questions and hypotheses were pre
sented in the opening chapter.
Question 1: What variables affect the choice of the
single role versus the marriage role?
The position taken by this paper is that the
choice of singleness or marriage among adults of mature age
can be explained in terms of three variables: Assertive
ness, Dependency upon persons, and a "Social Influence
68
69
Index" comprised of a number of interpersonal and social
factors derived from role theory. The findings in general
confirm this hypothesis, though not strongly. The multi
ple regression coefficient for the correlation between
"Assertiveness," "Dependency," and the "SII" as predic
tors, upon Singleness Orientation, as criterion variable,
is .61. While this correlation coefficient is statisti
cally significant beyond the . 0 0 1 level, it is still only
explaining 37% of the total variance in the criterion
variable. This is an encouraging result in a pioneering
study such as this, but it shows that there is a large
amount of variance yet to be explained.
Hypothesis 1: The orientation towards remaining single is
positively correlated with Assertiveness.
The zero-order correlation of the "Assertiveness"
variable with the "Singleness Orientation" was found to
be .04 (Appendix D, Table 35) . This correlation coeffi
cient is in the expected direction, but it falls short of
statistical significance. However, the multiple regres
sions show that "Assertiveness," when in interaction with
the other predictor variables, plays a role much more sig
nificant than its zero order correlation coefficient,
taken alone, would indicate. The interactive effect of
"Assertiveness" will be discussed subsequently, under Hy-
70
pothesis 4.
Hypothesis 2: The orientation towards remaining single is
negatively correlated with Personal Depen
dency.
The zero-order correlation coefficient between
"Dependency" and the Singleness Orientation is -.26
(Appendix D, Table 35). This coefficient is in the ex
pected direction (i.e., negative), and is significant be
yond the .001 level. Like the "Assertiveness" variable,
"Dependency" is found to be even more effective when its
interaction with the other variables is taken into account
(see discussion under Hypothesis 4).
Hypothesis 3: The orientation towards remaining single
is positively correlated with those inter
personal and social variables that facili
tate the transition from the married role
to the single role, or that support main
taining the single role (SII).
The "SII" is the strongest of the predictor vari
ables. Its zero-order correlation with the Singleness
Factor is .52, significant beyond .001 (Appendix D, Table
35). It is also the dominating variable among the inter
actions to such an extent that when the "SII" is high, the
effects of "Assertiveness" and "Dependency" tend to be
negligible (see further discussion under Hypothesis 4).
72
is greater than the sum of the Betas of the simple vari
ables, it can be said that the effects of the interactions
are greater than those of the simple variables, when the
interactive effects are partialed out. As Table 9 indi
cates, this is the case with respect to the data of this
study. The sum of the Betas (absolute) for the inter
action variables is 2 .2 1 , while for the simple variables
it is 1.44. The effect of the variables in interaction
with each other is illustrated with particular clarity in
the behavior of the "Assertiveness" variable. While its
zero-order correlation coefficient (.04) with the "Single
ness Orientation" was the lowest of all the predictor
variables (and statistically "insignificant!"), yet in the
full multivariate regression equation its Beta was higher
than that of any other variable. A fuller explanation of
this behavior is given in the analysis provided by the
Automatic Interaction Detector.
The "Automatic Interaction Detector" (acronym "Ard’ )
is a computer generated statistical procedure devised by
John A. Sonquist and James N. Morgan, both of The Univer
sity of Michigan, for selecting from a group of predictor
variables that set of subgroups that will reduce the error
of predicting the dependent variable as much as possible
relative to the number of groups (Sonquist, 1970).
71
Hypothesis 4: The concurrent interactive effect of Asser
tiveness, Dependency, and the SII upon the
orientation towards remaining single is
greater than the sum of their individual
effects.
In Chapter III it was pointed out that the inter
active effects of the three predictor variables could be
represented mathematically in a multiple regression equa
tion by their cross products, which then become additional
variables in the equasion. Thus the number of predictor
variables in the regression equation is increased from
three to seven. In addition to the original three predic
tors, "Assertiveness" (Xi), "Dependency" (X2 ) and the
"SII" (X3 ), there will also be the product of "Assertive
ness" and "Dependency" (Xu) (representing the interaction
effect of the two variables), of "Assertiveness" and the
"SII" (X^2 )/ "Dependency" and the "SII" (X^3), and the
product of the three variables together (X1 4 ). Since each
"Beta" term in multiple regression represents the relative
amount of variance explained by its variable in relation
to the others, the ranking of the Betas gives the relative
order of importance of each variable in the equation, and
the sum of the Betas (in absolute values) of any combina
tion of variables gives the relative explaining power of
that combination. Therefore, if the absolute sum of the
Betas of the interaction (i.e., cross product) variables
73
TABLE 9
MULTIVARIATE REGRESSION OF
THE SINGLENESS ORIENTATION WITH
ASSERTIVENESS, DEPENDENCY, THE SOCIAL INFLUENCE INDEX,
AND THEIR INTERACTIONS
Multiple "R'f : 0.61
The Variables, listed in descending order of their Beta
weights:
Beta* Simple "r"
Assertiveness -1. 03 .04
Assertiveness-SII Interaction .93 .28
Dependency-SII Interaction - .54 -.04
Assertiveness-Dependency-SII Interaction .54 .04
Social Influence Index .24 .52
Assertiveness-Dependency Interaction - . 2 0 -.14
Dependency .17 -.26
Sum of the Betas (absolute), Interacting Variables: 2.21
(Assertiveness-Dependency, Assertiveness-SII,
Dependency-SII, Assertiveness-Dependency-SII)
Sum of the Betas (absolute), Simple Variables: 1.44
(Assertiveness, Dependency, SII)
*At the final step of the regression.
74
The technique is a step-wise application of a one
way analysis of variance model. Its objective is
to partition the sample into a series of non-over
lapping subgroups whose means explain more of the
variation in the dependent variable than any other
such set of subgroups (Sonquist, 1970, p. 20).
To qualify for inclusion in the AID algorithm, a variable
must account for more than a minimum level of reduction in
the unexplained sum of squares, and must contain more than
a prescribed minimum number of subjects. When the AID
program was applied to the data of the present study, the
demographic categories of "Age," "Sex," and "Religiosity"
were included in the program as predictor variables. Both
"Sex" and "Religiosity" failed to meet the minimum para
meters of the program, and were dropped. The final re
sults of the AID analysis are shown on Figure 5.
As indicated in the diagram, the "SII" is the most
powerful of the predictors. Where this variable is high,
indicating that social pressures are loaded in favor of
singleness (see the lower branch of the AID "tree" in Fig
ure 5), neither "Assertiveness" nor "Dependency" add any
further explanatory power. Interestingly, however, the
demographic category "Age" does add a further increment of
explanation. Two age classes, those 40 to 50, and those
60 and over, seem more inclined to singleness than do the
others. It is a moot question whether this irregular
splitting of the Age category is merely a quirk of the
FIGURE 5
RESULTS OF AUTOMATIC INTERACTION DETECTOR
Dependent Variable is Singleness
Orientation (Multiple R = .61)
Range of Singleness Orientation: 0-10
Note: For this program,
the Predictor Variables
(Assertiveness, Dependency
Total Sample
N = 281
Mean S.O.= 5.18
Age
40-50, 60 +
N = 39
Mean S.O.= 7.44
SII
Value 3, 4
N = 8 8
Mean S.0.= 6 . 73
As sert ivene s s
Value 2, 3
N = 50
Mean S.0.= 3.60
Assertiveness
Value 1, 3, 4
N = 55
Mean S.0.= 4.53
Assertiveness
Value 0, 2
N = 39
Mean S.O.= 5.56
Mean S-0.= 4.96
Dependency
Value 0, 1
Age
Under 40, 50-60
N = 49
Mean S.O.= 6.16
Dependency
Value 2, 3, 4
N = 99
Mean S»0.= 4.04 SII
Value 0, 1, 2
N = 193
Mean S.Q.= 4.48
N = 49
.Mean s.O
Assertiveness
4.49
SII) were recoded into five
classes of equal size, Value 0, 1, 2, 3, 4.
76
present sample. Since the "over 60" class only included 3
members in the sample, its behavior can be attributed to
sampling error. If this class is disregarded, the remain
ing classes form a curvilinear pattern, with the youngest
and oldest groups more inclined to favor marriage, and the
middle group favoring singleness. This is similar to the
rationale for Hypothesis 5 presented in Chapter I (p. 18)/
and may be considered a partial confirmation of it. How
ever, the effect is only apparent under the conditions
where the "SII" is high; it does not apply to the entire
sample.
When the "SII" is low in value, indicating that so
cial pressures are loaded more in favor of marriage, both
"Dependency" and "Assertiveness" explain additional vari
ance (see the upper branch of the AID "tree," Figure 5).
In this case, where "Dependency" is low the orientation to
singleness is increased somewhat, and is further increased
where "Assertiveness" is also low. High "Dependency" in
creases the orientation to marriage, as does a middle-
range score on "Assertiveness." However, on this branch
if "Assertiveness" is either very low or very high the
subject is more likely to remain single. This is similar
to the prediction made for the "Assertiveness" variable in
Chapter I (p. 6 )• The predicted curvilinearity of the
Assertiveness variable was in error on two points, how-
77
ever: in Chapter I it was stated that the extremely low
"Assertiveness" scores would not likely appear in the pre
sent sample, and the contrary proved to be the case; also,
the prediction failed to forsee that the curvilinearity
would be evident only in a restricted segment of the sam
ple, i.e., where the "SII" is low and "Dependency" is high.
When "Dependency" is low, the correlation of "Assertive
ness" with the "Singleness Orientation" is a negative
linear one. (The correlation coefficient of "Assertive
ness" and the "Singleness Orientation," which was .04 over
the entire sample, is increased to -30 when only those
cases are considered whose mean "SII" is 12.0, or less.)
In summary, the AID analysis indicates that the
persons most likely to remain single are those for whom
social pressures are highly in favor of singleness, and
who are from 40 to 50 years old. The persons most likely
to seek (re)marriage are those whose social pressures
favor marriage, whose Dependency needs are high, and who
are moderately Assertive. Interestingly, the AID analyses
using Singleness Expecation and Singleness Preference as
dependent variables produced results almost identical to
those of Higure 5 (Appendix D, Figures 11 and 12).
Hypothesis 5: The desire to remain single (Yg) (but not
the expectation of it) bears a curvilinear
relationship to age (X5 ).
78
Hypothesis 6 ; The expectation of remaining single (Yio)
(but not the desire for it) is positively
and linearly correlated with Age.
Hypotheses 5 and 6 are sufficiently related so
that they can be discussed together. In general it can be
stated that these hypotheses were not confirmed by the
data. When "Age" was tabulated against "Singleness Pref
erence" in a bivariate table, the correlation ratio, Eta
(a statistic that is sensitive to curvilinearity), for the
two variables was .01 (See Appendix D, Table 36). When
"Age" was tabulated against "Singleness Expectation," the
assymetrical correlation coefficient, Somer' s D, was .06
(Appendix D, Table 37).
However, the low correlations of these variables
over the entire sample must be reevaluated in the light of
the effect that "Age" was shown to have in the special
case where the "SII" was high (see discussion under Hypoth
esis 4, above). Here the behavior of "Age" seems indeed
to be curvilinear, not only for "Singleness Preference,"
But for "Singleness Expectation" and the "Singleness Ori
entation," as well.
Question II; What variables other than general "mental
health" factors affect a person's happiness
(i.e., psychological well being) with the
single role?
This general question raises the possibility that
79
the same variables that influence the choice of singleness
or marriage may also affect a person's psychological well
being. Since psychological well being is also affected by
a host of "mental health" variables for which no controls
were made in the present study, it was assumed that the
correlations would be low, but significant. Such proved
to be the case.
The multiple regression coefficient for the ef
fects of "Assertiveness," "Dependency," and the "Social
Influence Index" upon "Negative Affect" (Hypothesis 7) is
.13 (significance level, .05). The coefficient for their
effect upon "Positive Affect" (Hypothesis 8 ) was .38 (sig
nificance level, .0 0 1 ).
A pertinent question here is whether a person's
psychological well being itself might have some effect
upon his "Singleness Orientation." Perhaps whether a per
son is generally happy or unhappy with himself might have
some effect upon his choice for singleness or marriage.
When "Positive" and "Negative Affect" were introduced as
predictors in a multivariate regression with the "Single
ness Orientation" as the criterion, the "F" statistic for
"Positive Affect" (4.62) was barely significant (p<.05),
indicating that it does explain a small amount of the
variance, though less than that of the three principal
predictors. Interestingly, its Beta (-.12) indicates that
80
the variance it does explain is in the negative direction
(i.e., to the extent that "Positive Affect" has any effect,
it predicts a choice in favor of marriage, rather than of
singleness). (Appendix D, Table 38). The "F" statistic
for "Negative Affect" in the same regression was not sig
nificant (0.82) (Appendix D, Table 38). When "Positive"
and "Negative Affect" were introduced as predictors in an
AID program, neither could account for enough variance in
the "Singleness Orientation" to meet the program's mini
mum parameters, so they were both dropped. It can be
stated in summary, then, that while "Positive Affect"
might have some slight effect up on a person's choice in
favor of marriage, "Negative Affect" seems to have no sig
nificant effect upon whether a person elects to remain
single, or to (re)marry.
Hypothesis 7; Negative Affect (Y7 ) is increased to the
extent that there are conflicting relation
ships among the three independent variable
sets (Xi,X2 ,X3 ): E.g., high Assertiveness
with low SII or high SII with low Asser
tiveness, high Dependency with high SII,
high Assertiveness with high Dependency.
The low correlation between the variable configu
rations that would be expected to produce conflict and
"Negative Affect" was unexpected. The zero-order corre
lation of the "Assertiveness" "SII" difference and "Nega
tive Affect" was only .07, and those of the other vari-
81
ables under consideration were equally unpromising. Ap
parently the reason for this is that the "SII" itself cor
relates very poorly with "Negative Affect" (-.06), so its
interactions with "Assertiveness" and "Dependency" did not
matter much. The AID analysis of the "Negative Affect"
variable attributed the greatest amount of variance to the
demographic categories "Age," "Sex," and "Religiosity,"
none of which have anything to do with singleness, as
such. "Assertiveness" and "Dependency" were found to have
a secondary influence, after the effects of age were al
lowed for, but the "SII" failed to explain the minimum
variance required by the program (see Figure 6 ). There
fore, even though the multiple correlation for this hy
pothesis was barely significant (R=.13, sig. level .05),
this general line of thought does not appear to be fruit
ful. (See Appendix D, Table 39 for a summary of the mul
tiple regression.)
Hypothesis 8 : Positive Affect (Yg) is positively corre
lated with Assertiveness (Xj.) and with In
volvement with Others (X4 ), but is nega
tively correlated with Dependency (X2 ).
All three of the predictor variables produced cor
relations with Positive Affect that were in the expected
direction, and the correlations for "Assertiveness" (r=
.27) and "Involvement with Others"(r=.18) were both statis-
FIGURE 6
AID PROGRAM FOR NEGATIVE AFFECT, AS DEPENDENT VARIABLE
(MULTIPLE R = .13) RANGE n.a: 0 to 5
Dependency
mean n.a: 2.81
NQte: The Predictor ^
Variables (Assertiveness, De
pendency, and SII) were recoded
into five equal classes, Value
CO
to
Entire Sample
N: 278
mean n.a: 2.31
N: 52
mean n.a: 1.29
Assertiveness
Age
31 to 50
N: 137
Mean n.a: 2.83
Age
under 30, 51+
N: 141
mean n.a: 1.80
Age
31 to 40
mean n.a: 3.20
Religiosity
Not religious
mean n.a: 2.28
mean n.a: 1.40
Religiosity
Strong & Mod
Age
41 to 50
N: 82
mean n.a: 2.59
mean n.a: 2.60
Dependency
Val.: 2, 3
mean n.a: 3.55
Dependency
Val.: 0, 2
mean n.a: 1.80
Dependency
mean n.a: 2.82
Sex
Female
mean n.a: 2 . 2 2
Sex
Male
Assertiveness
Val.: 0, 1, 2 ,
N: 89
mean n.a: 2 . 1 1
83
tically significant (sig. level .01). The zero-order cor
relation of "Dependency" and "Positive Affect" (-.01) was
not significant, but its F statistic in the multiple re
gression was significant (F=6.84, sig. level .01), showing
that "Dependency" does have some effect in interaction
with the other variables (see Appendix D, Table 40 for a
summary of the multiple regression). The AID program with
"Positive Affect" as the dependent variable showed that
Positive Affect is affected by "Assertiveness," "SII," and
"Dependency," in that order. The demographic categories
"Age" and "Sex" also have a small influence. Only those
persons with the lowest "Assertiveness" scores were sig
nificantly lower than the entire sample mean "Positive Af
fect." Where "Assertiveness" is high, the "SII" becomes
important, the higher values (i.e., those pressing most
strongly towards singleness) accounting for the higher
mean "Positive Affect." When the "SII" is high, "Depen
dency" explains an additional increment of variance:
those with moderate amounts of "Dependency" show a higher
mean "Positive Affect" than can be accounted for by the
"SII" alone, while those whose "Dependency" is either very
low or very high show a lower mean "Positive Affect."
Within the moderate "Dependency" group, "Sex" explains a
further difference, with the males showing a higher mean
"Positive Affect" than the females (see Eigure 7).
FIGURE 7
AID PROGRAM FOR POSITIVE AFFECT AS DEPENDENT VARIABLE
(MULTIPLE R = .38) RANGE OF POSITIVE AFFECT: 0 to 5
Note: The Predictor Variables
(Assertiveness, Dependency, and
SII) were each recoded into five
classes of equal size, Value
CO
Entire Sample
N; 279
Mean p.a: 3.45
Sex
Male
Mean p.a: 4.29
k ______ r- _ _ _ _____
Mean p.a: 3.54
Sex
Female
Age
Und. 30, 41-50
Mean p.a: 3.73
Age
31-40, 51 plus
Mean p.a: 3.00
Dependency
Val.: 1, 2
N: 116
Mean p.a: 3.85
Mean p.a: 3.20
SII
Mean p.a: 3.43
Dependency
Val;: 0, 3
SII
Val.: 2, 3, 4
N: 179
Mean p.a: 3.70
Assertiveness
Val.: 0
Mean p.a: 2.50
N: 249
Mean p.a: 3.56
Assertiveness
Age
41 and over
Age
40 and under
Mean p.a: 2.95
Mean p.a: 3.53
85
A Note on Sexual Behavior
In the group of interpersonal items designed to
furnish data for the "SII," two of them (items 7 and 18)
dealt with facets of sexual behavior. It was supposed
that where persons felt that sexual intercourse among sin
gles was acceptable (item 7) and where satisfying sexual
relations were available outside of marriage (item 18)
that persons would be less inclined to seek (re)marriage.
In view of the volume of literature relating to the sexual
behavior of single adults, it was surprising to find that
these two items did not qualify for inclusion in the
"SII." (See Appendix B, Tables 22 and 23.) Item 7 ("Sex
outside of marriage is morally wrong") correlated with the
"Singleness Orientation" with a Pearson coefficient of
.10, (sig. level .047). Item 18 ("My sex life is [fully]
[partially] [not] satisfying...") correlated with the
"Singleness Orientation" with a Pearson coefficient of
less than .01; (sig. level .476). The relatively high
means for each of these variables was particularly inter
esting. In a possible range of 1 to 4, the mean for item
7 was 3.50, indicating a high degree of permissiveness
towards sex among singles across the sample. This accords
with Reiss's finding (1971), that singles were the most
permissive subset of the population. Mean for item 18 was
86
2.81, showing that for most of the sample, sex is at least
partially satisfying. Taken together, the data gathered
here on sexual behavior suggest that the members of the
present sample are well on their way towards a high level
of sexual permissiveness, in practice as well as in be
lief, but that this does not cause them to prefer single
ness over marriage.
CHAPTER VII
CONCLUSIONS AND SUGGESTIONS FOR
FURTHER RESEARCH
An expiatory study such as this one must neces
sarily employ a kind of "shotgun" approach, utilizing
hypotheses and variables derived more from experience and
intuition than from carefully reasoned, pre-existing theo
ries, in the hope that the pattern will be broad enough to
enable a few of the shots to strike the target. It is to
be expected that such a study will raise more questions
than it answers. We have not been disappointed in either
respect.
The data have shown that the basic approach to the
concept of "singleness" is correct: the choice for sin
gleness or marriage is a function of a set of social vari
ables, and of the intcapersonal characteristics "Depen
dency" and "Assertiveness" (Hypotheses 1-3). However,
the amount of variance that the "Social Influence Index"
explains (approximately 37%), and the finding that it
includes at least three orthogonal factors indicates that
more needs to be learned about the nature of the "SII," and
87
88
about other possible sources of variation.2 The hypothe
sis that the interaction among the variables is important
(Hypothesis 4) has received strong confirmation, and a
comparison of the AID analysis with the multivariate re
gressions has suggested a change in the original hypothet
ical model, as described in Chapter I (Figure 3). The
data suggest that the process can be represented more pre
cisely by means of a sequential type model that allows for
variance to take several directions, according to whether
the values of the key variables are high or low. Such a
model is represented in Figure 8 .
According to Figure 8 , the most important variable
acting upon the individual, and pressing him to opt for
singleness rather than marriage, is the "Social Influence
Index." Where this variable is high, it explains the
choice for singleness directly, without reference to ei
ther "Assertiveness" or "Dependency," and subject only to
some modification according to Age.
(As explained in the previous chapter, the direct
effects of "Age" upon singleness, hypothesized in Hypothe
ses 5 and 6 , were not confirmed by the data pertaining to
the entire sample. However, Age did have an effect upon
the "SII" when its value was high. In this case it acted
2see further discussion under "Suggestions for
further research.
89
FIGURE 8
A REVISED "SINGLENESS" MODEL
If SII is High
• H
>I/S
| Dependency
c o
• h £
If dependency is High
Assertiveness
Age
The Individual
Social Influence
Index
90
in a curvilinear way, with the middle-aged subjects show
ing a slightly greater preference for singleness than ei
ther the very young or the very old.)
Where the "SII" is moderate or low in value, the
intrapersonal variable "Dependency" has an effect. Where
"Dependency" is high, it affects the "Singleness Orienta
tion" directly, whether the value of "Assertiveness" is
high or low. Where the "Dependency" variable is moderate
or low, however, "Assertiveness" has an effect, acting in
a curvilinear way, with those of very high and very low
"Assertiveness" showing the greatest preference for sin
gleness.
As indicated in the preceding chapter, the princi
pal independent variables under consideration in this
study failed to account for a sufficient amount of vari
ance in "Negative Affect" to warrant further pursuit of
Hypothesis 7, at least in its present construction. It is
possible, of course, that when the independent variables
have been further refined, and when other sources of vari
ation in the singleness-marriage contuum have been iden
tified, the question of their possible effect upon "Nega
tive Affect" can be reconsidered.
"Positive Affect," on the other hand, did appear
to be affected significantly by "Assertiveness" and the
"SII," and to lesser extent by "Dependency," also. In
91
general, the data seemed to indicate that where a person1s
life scene is congruent with the press of social factors,
and with his need for self-fulfilment and intimacy, his
"Positive Affect" will be increased. More specifically,
the effects of "Assertiveness," the "SII," and "Dependen
cy" upon "Positive Affect" can be illustrated by another
sequential model. In this case, however, the primary
variable appears to be "Assertiveness" (Figure 9). The
next variables in turn are the "SII," and "Dependency."
"Age" has a modifying effect upon low-value "SII" (nega
tive) , and upon very high or very low "Dependency." In
the latter case, the effect of "Age" is curvilinear, with
those in middle age showing the greatest inclination to
"Positive Affect." "Sex" also enters the final stage of
the model, modifying the effects of "Dependency" when "De
pendency" values are in the moderate range. In this case,
males appear to have slightly more "Positive Affect" than
females.
General Observations
The most interesting overall observation was the
apparent persistence of marriage as a life goal, even
among a sample of single adults who have to all appear
ances solved the major problems involved in coping with
92
FIGURE
A REVISED "POSITIVE AFFECT" MODEL
Individual
| Assertiveness]"
If low
a )
- p
n J
U C
0)
P
i
± ±
O'
H
LC
Social Influence
Index
< D
-P
r e t&
5 H&>
< 1 >■H
Xi
O
S M
0
m
H
>
f
If low
Age
If very low,
Dependency i i f —■ ■ -
* or very high
I
U
If moderate
Sex
Male:
Female: -
EH
U
w
P m
P m
P = a
>
H
Eh
H
CO
o
PM
93
their singleness. With a mean "Singleness Orientation" of
5.11 on a possible scale of 10, the sample did not appear
highly dedicated to singleness as a way of life.
On the other hand, the strongest factor operating
within the strongest single independent variable (i.e.,
Factor One in the "Social Influence Index"— see Table 8 ,
and discussion in Chapter V) suggests the emergence of a
sense of a "single role" that has intrinsic meaning for
life, and even possible advantages over the traditional
spouse roles. Clearly this indicates an area for further
investigation. It may even adumbrate the coming of a new
"Social Institution" of indepdendent, single adults who do
not consider marriage for themselves. Other variables,
not operationalized in the present study, must also be
brought into the "singleness" model.
One such variable probably is seated in the cul
ture, itself. Ours has (traditionally, at least) been a
marriage-oriented culture, where most children are social
ized to become spouses and parents when they are grown.
It is possible, however, that these primitive socio-cul-
tural forces are being overridden by present trends that
favor singleness.
In the intrapersonal dimension, one variable that
was not operationalized in the present study, but which
seems important in a large number of cases that come to
94
the attention of the counselor, is the conflict between
the need for intimacy and primary relationships on the one
hand, and the fear of being hurt on the other. In the
case of divorced people this can be accounted for in terms
of the trauma of a previous unsuccessful marriage; in the
case of the divorced and the never-married alike, the pos
sibility of a traumatic childhood in a conflicted and bad
ly functioning family of origin is worthy of consideration
(Spreitzer, in press).
Another interesting and surprising observation
growing out of this study was the lack of difference be
tween males and females in their interest of marriage.
The myth that males are footloose and irresponsible with
respect to marriage and family life, and that females, on
the other hand, are propelled by some kind of "nesting in
stinct" to seek marriage (even if this involves the en
trapment of a male) does not seem to be supported by the
present data. On the contrary, the males in the present
sample were only slightly (and insignificantly) less in
terested in marriage than the females.
The lack of correlation between sexual behavior
ahd the desire for singleness appears to give the lie to
yet another myth often quoted by persons opposed to pres
ent day standards of sexual permissiveness: that the in
stitution of marriage is doomed if unmarried persons are
95
permitted free access to sexual relationships. The pre
ponderant majority of our sample, both males and females,
indicated that they feel quite free to engage in sexual
belations whenever and with whomever they please, but this
does not appear to prejudice their attitude towards the
possibility of marriage.
A final observation was the apparent lack of im
portance played by religiosity in the choice of marriage
or singleness. When controls were made for "Age," no sig
nificant variance in either the probability of singleness
or satisfaction with the single state was attributable to
religiosity.
Suggestions for Further Research
The overall results of this study were sufficient
ly satisfactory to warrant its being replicated on a
larger and more representative sample. It would be par
ticularly informative to have a sample that included mar
ried couples as well as singles, so that one could measure
the extent to which the opposites of the variables that
seem to press people towards singleness would press others
towards marriage. Such a study could also be tailored to
give information as to whether singleness, including non-
mar ital pairing relationships of whatever permanency, is
96
replacing legally sanctioned marriage to any significant
degree.
However, if such a study were to be of maximum
utility, it would be worth the effort beforehand to expand
and refine the present independent variables, in the light
of the present findings.
Possible refinements of the "Social Influence."
This variable should first take into consideration the
finding that the "SII" of the present study was really the
combined effect of three orthogonal factors. This is of
particular importance in view of the relative strength of
the "Social Influence" variables.
"Factor One," which seems to center around under
standing and acceptance of the singleness role as a mean
ingful way of life, needs further testing in at least two
ways: (1) Items should be retested and possibly recon
structed to eliminate any possibility of tautological re
lationships with the dependent variable. (There is at
least a logical suggestion that some degree of tautology
exists in some of the present Factor One items— e.g., Item
20: "Being married (single) is essential to my goals"—
although the zero-order correlation coefficients between
the Factor One items and the "Singleness Orientation" do
not suggest a strong degree of autocorrelation.) (2) Some
97
method of measuring the valence of Factor One items must
be included: i.e., does a person accept his single role
because this is what he really wants, or is he merely re
signed to it?
"Factor Two" items seem to measure two kinds of
conflict: (1) intrarole conflict, such as conflicts be
tween dating and child care responsibilities, and (2) con
flicts between the single role and goal attainment. Fur
ther exploration of these dimensions might reveal that
"Factor Two" is itself actually two separate factors.
"Factor Three" only loads heavily on one item:
Family Pressure. Other social pressures ought to be in
vestigated, however, such as that from peer groups, reli
gious groups, etc. (While most of the items designed to
measure peer group pressures did not correlate highly with
"Singleness Orientation" in the present study [See Appen
dix C, Table 33, items 6, 8, 10, 14, 15] this could be due
to idiosyncratic elements in the items themselves, or in
the sample.)
Another important area for further investigation
that falls within the aegis of "Social Influence" involves
the new standards of sexual permissiveness, particularly
as changes in this area affect the institution of marriage.
The findings of the present study indicate that while the
"sexual revolution" may indeed have already been won, at
98
least as far as the population of single adults is con
cerned, it seems to have had far less impact upon society
at large than the popular literature would suggest. How
ever, the limitations of our sample cause one to be cau
tious about generalizations here, until further study has
been made on a broader scale. If it is indeed true that
contemporary sexual standards are not affecting society
and its institutions (particularly marriage) in any radi
cal way, then what other possible sources can account for
the changes taking place in marriage today?
Two demographic factors, Education and Income,
were not considered in the present study because of ex
cessive skew in the sample. An effort should be made in
any replication to secure a greater variance in these cat
egories. As Cutright (1970) has indicated, Income, at
least, does affect both marriage and remarriage.
Finally, if the study is to be replicated on a
sample that includes both married and single persons, the
"Influence" variable (or variable cluster) should measure
not only the tendency towards singleness or marriage, but
the strength of its influence, as well. One can conceptu
alize a situation where the social forces were totally
oriented in a particular direction, but where their force
was too weak to have more than a token effect.
99
The refinement of the intrapersonal variables.
These variables ought to include the expansion of the "As
sertiveness" variable into a "Need for Self-Fulfilment"
dimension. Similarly, the "Dependency" variable might
well be expanded into a "Need for Intimacy." To these two
should be added a "Fear of Intimacy" dimension, which
seems to figure strongly into the problems of many single
adults, as revealed in personal interviews, but for which
no adequate measurement was devised in the present study.
Possible components of the "Fear of Intimacy" might in
clude trauma in a previous marriage, and unhappy childhood
experiences, such as excessive parental conflict. I be
lieve that the interaction of this triad of intrapersonal
variables: Need for Self-Fulfilment, Need for Intimacy,
and Fear of Intimacy, would explain a greater amount of
variance in the singleness-marriage continuum that was ex
plained in the present study.
Once the suggested refinements and expansions of
the socio-cultural and intrapersonal variables have been
made and pretested, a replication of the entire research
might well be expected to follow a pattern similar to that
suggested in Figure 10. "Positive Affect"-^ might then be
hypothesized as being directly correlated with a congru
^Positive and Negative Affect are not illustrated
in Figure 10.
100
ence between "Need for Self-Fulfilment" and the "Socio
cultural Variable," and "Negative Affect"^ will likely be
most highly correlated with conflict between the "Need for
Intimacy" and the "Fear of Intimacy."
^Positive and Negative Affect are not illustrated
in Figure 10.
FIGURE 10
HYPOTHETICAL REPLICATION OF THE "SINGLENESS" STUDY, WITH EXPANDED VARIABLES, AND
A MIXED SAMPLING THAT INCLUDES MARRIED SUBJECTS AS WELL AS SINGLE
Education
Income
Role Perception
and Acceptance
Role Conflicts
Conflicts with Goals
Family Pressure*
Other Social
Pressures
Social
Influences
Need for
Intimacy
*----
+
Traumas
in former
marriage
If
H
C D C f i
s: a
0
o
f i > o
•
P ) Qi
W * H
0
3
Hi
n H
>
•
( L ,
Dependency
Needs
Traumas
in family of
origin
Fear of
Intimacy
H- H
C O Hi
a O
0 C D
F d ■d
O C D C D
S H t s
Pi
ft C D
C D P
O
0 ' • <
H
> f
Desire for
Self-
Assertion
If Social Influences are
-------
Strong
U
If Dependency is High
Need for
Self-
Fulfillment
D
SINGLENESS - MARRIAGE ORIENTATION
REFERENCES
102
REFERENCES
Acock, Alan C. and Melvin DeFleur, "A Configurational Ap
proach to Contigent Consistency in the Attitude-
Behavior Relationship." American Sociological
Review, 37 (Dec., 1972), 714-726.
Bernard, Jessie, "Present Demographic Trends and Struc
tural Outcomes in Family Life Today." Marriage
and Family Counseling; Perspective and Prospect,
James A. Peterson, (ed.), New York: Assocxated
Press, 1969.
Bradburn, Norman M. The Structure of Psychological Well-
Being. Chicago: Aldine, 1969.
Brecher, Edward M. The Sex Researchers. Boston: Little,
Brown and Company, 1969.
Buckley, Walter., Sociology and Modern Systems Theory,
Englewood Cliffs, New Jersey: Prentice Hall,
1967.
Burr, Wesley R., "Role Transition: A Reformation of
Theory." Journal of Marriage and the Family, 34
(Aug., 1972), 407-416.
Cutright, Phillips, "Income and Family Events: Getting
Married." Journal of Marriage and the Family,
32:4 (Nov., 1970), 628-637.
Dreitzel, Hans Peter, Ed. Family, Marriage, and the
Struggle of the Sexes. New York: MacMillan,
1972.
Hicks, Mary W. and Marilyn Platt, "Marital Happiness and
Stability: A Review of the Research in the
Sixties." A Decade of Family Research and Action,
Carlfred B. Broderick, (ed.), National Council on
Family Relations. 1971, 59-78
Kirkendall, Lester A. Premarital Intercourse and Inter
personal Relations. New York: Julian, 1961.
Labowitz, Sanford, "Some Observations on Measurement and
Statistics." Social Forces, 46:2 (Dec., 1967),
151-160.
103
104
Leary, Timothy. The Interpersonal Diagnosis of Personality.
New York; Ronald, 1953.
McNemar, Quinn,. Psychological Statistics. New York:
Wiley, 1962.
Nie, Norman H., Dale H. Bent, and C. Hadlai Hull. Statis
tical Package for the Social Sciences. New York:
McGraw-Hill, 1970.
Reiss, Ira L. The Social Context of Premarital Sexual
Permissiveness. New York: Holt, Rinehart and
Winston, 1967.
Sonquist, John A. Multivariate Model Building. Institute
for Social Research: The University of Michigan,
Ann Arbor, Michigan, 1967.
Spreitzer, Elmer, "Factors Associated with Singlehood."
Journal of Marriage and the Family, (In press).
U. S. Department of Commerce: Bureau of Census: Census
of Population, 1970, PC (1) - D1 U. S. Summary,
Table 203.
APPENDIX A
THE RESEARCH QUESTIONNAIRE
105
RESEARCH QUESTIONNAIRE
Return to:
Ed Olander
10826 Venice Blvd, #1
Culver City, Ca. 90230
(213)836-6449
Dear Single Adult:
You are about to take part in a research project designed to measure
some of the attitudes and needs of single adults. Please answer all the items
simply and directly, as if you were telling yourself, rather than some other
person.
Please do not put your name anywhere on this Questionnaire. All information
you give will be treated confidentially, as a part of a larger group of data.
No individual questionnaire will be singled out for special emphasis.
Your cooperation is needed and appreciated very much. If the results of
the research are to.be accurate, it is important that the persons selected to
participate in the project make up a sample that is highly representative of the
total single adult population, and since you are a part of that select sample,
your answers are very important. Thank you.
§ § § § §
On each of the following multiple-choice items, make a check mark in
the blank that best represents your attitude towards the item:
Strongly
Disagree •
L . A person can have a normal and
happy life without being married: ___ ___ ___ ___
2. Being single has many advantages
over being married: ___ ___ ___ ___
3. Sometimes I'm not sure what is
expected of me as a single adult: ___ ___ ___ ___
4. There is a place and function for
single persons in our society:_______ ___ _ _ _ ___ ___
5. Members of my family would really
like to see me married: ___ ___ ___ ___
6. My religion teaches it is better to
be married than to be single:
7. Sex outside of marriage is morally
wrong:
8. I now have a companion who would
like to get married:
Strongly
Agree :
106
107
9. I find it hard to be a parent and a V*s: : U>t a pa
single person at the same time: ___ ___ _ __ ___
10. Concerning friends:
a. Most of my good friends are married.
b. I have many good married friends, a few single friends.
c. I have many good single friends, a few married friends.
d. Most of my good friends are single.
11. Concerning minor children:
a. The children are with me most of the time.
b. My ex-spouse frequently takes the children for extended visits.
c. The children are with my ex-spouse most of the time.
d. There are no minor children.
12. Concerning my job:
a. For reasons of public attitudes, social obligations, etc., it is far
better to be married than single on my job.
b. It is somewhat better to be married than single on my job.
^ c. On my job it doesn't matter whether one is single or married.
d. It's better to be single on my job.
9
13. How much time do you spend at work each week?
a. Over 60 hours.
b. 40 - 60 hours.
c. 20 - 40 hours.
d. Under 20 hours.
14. How much time do you spend each week in group activities with people of
your own sex (i. e. , National Guard, volunteer organizations, etc.)?
a. Three times weekly, or more.
b. Once or twice weekly.
c. Less than once a week.
d. Don't belong to any such organizations.
15. How much time do you spend with mixed-sex organizations for singles?
a. Three times weekly, or more.
b. Once or twice weekly.
c. Less than once a week.
d. Don't belong to any such organizations.
16. How much time do you spend each week on recreation and/or hobbies?
a. Ten hours or more.
b. Five to ten hours.
c. One to five hours.
d. Less than one hour.
17. Concerning your living arrangements:
a. I prefer to live as a part of a warm and caring family.
b. I prefer a roommate type arrangement with a relative.
c. I prefer living with a roommate who is not a relative.
d. I prefer to live alone.
-3-
108
18. Concerning your sex life as a single person:
a. My sex life is fully satisfying to me.
b. My sex life is only partially satisfying.
c. My sex life is not satisfying to me.
d. I do not engage in sexual activities.
19. Concerning your friendships as a single person:
a. I have friends with whom I feel perfectly close and trusting.
b. I have friends, but I wouldn't feel free to confide some things to them.
c. My friendships are mostly superficial.
d. I don't really have any friends.
20. Concerning your life goals (career, home, etc.):
a. Being married is essential to my goals.
b. I would like marriage, but it's not essential to my goals.
c. It doesn't matter to my goals whether I'm married or single.
d. Being single is essential to my goals.
Please make a check mark ( \S ) in front of every one of the following items
that, in general, applies to you as you see yourself now. There is no lim it to
the number of items you may check. If an item seems not to apply to you, simply
leave it blank.
Self confident.
Self reliant and assertive.
Businesslike.
Likes to compete with others.
Dominating.
Can be indifferent to others.
Thinks only of him(her)self.
Manages others.
Somewhat snobbish.
Selfish. _____
Grateful.
Admires and imitates others.
Eager to get along with others.
Affectionate and understanding.
Accepts advice readily.
Wants everyone to like him(her).
Shy.
Lets others make decisions.
Too easily influenced by friends.
Wants everyone's love.
Makes a good impression.
Able to give orders.
Self respecting.
Independent.
Able to take care of self.
Proud and self satisfied.
Often admired.
Respected by others.
Good leader.
Likes responsibility.
Usually gives in.
Appreciative.
Anxious to be approved of.
Lacks self confidence.
Modest.
Trusting and.eager to please.
Warm.
Dependent.
Hardly ever talks back.
Likes to be taken care of.
109
-4-
If you have never been married, check here: ______ and go on to Item 23.
If you are widowed or divorced, check the statement on each of the following
items that is most true for you:
21. My previous marriage, overall, was
a. Very happy.
b. Fairly happy, most of the time.
c. Fairly unhappy, most of the time.
d. Very unhappy.
22. My previous (most recent) marriage ended
. a. Less than six months ago.
b. Between six months and one year ago.
c. From one to two years ago.
d. More than two years ago.
Now go on to item 23:
23. Please check the statement that best expresses what you would like:
a. I would definitely like to remain single.
b. Mostly I would like to remain single, but some things about singleness
bother me.
c. Right now I1 m about evenly divided: I don't know whether I want to
remain single or get married.
d. Mostly I would like to get married, but some things about marriage
bother me.
e. I would definitely like to get married.
24. Please check the statement that best expresses what you expect:
a. I expect to spend the rest of my life as a single person.
b. I 'll probably remain single, but there's a small possibility that I
might get married.
c. Chances are about 50-50 whether I get married or not.
d. I 'l l probably get married, but there's some possibility that I may not.
e. I fully expect to get married.
25. During the past few weeks, did you ever feel Yes No
a. Particularly excited or interested in something? ___ ___
b. S o restless that you couldn't sit long in a chair? ___ ___
c. Proud because someone complimented you on
something you had done ? ___ ___
d. Very lonely or remote from other people? ___ ___
e. Pleased about having accomplished something? ___ ___
f. Bored? ___ ___
g. On top of the world? ___ ___
h. Depressed or very unhappy? ___ ___
i. That things were going your way? ___ ___
j. Upset because someone criticized you? ___ ___
-5-
110
Now we would like some personal information about yourself:
1. dex: Male Female___
2. Age: _ _ Under 30.
31 - 40.
41 - 50.
51 - 60-
Over 60.
3. What is your approximate annual income?
Under $ 10, 000.
$10,000 - 20,000.
$20,000 - 30,000.
O ver $30, 000.
4. Education: Give highest grade completed, and/or degrees: _
5. How religious do you consider yourself?
Very religious.
Moderately religious.
Not religious.
6. What are your living arrangements at present?
I live alone, or alone with my child(ren).
I live with my parents.
I live with relatives other than parents or children.
I live with a person of my sex, not a relative.
1 live with a person of the opposite sex, not a relative.
I live with more than one person, not relatives, in a household
consisting of _____ men, ______ women.
My child (children) do______ do not_____ live with me.
(If you have no children, check here:__________ )
Thank you very much for your cooperation in filling out this questionnaire.
APPENDIX B
SAMPLE CHARACTERISTICS
111
APPENDIX B
Contents
I. Sample Differences by Subfiles
Table
10. Variable Assertiveness
11. Variable Dependency
12. Variable Social Influence Index
13. Variable Singleness Orientation
14. Variable Negative Affect
15. Variable Positive Affect
II. Regressions by Subfiles
16. Singleness Orientation as Dependent Variable
17. Negative Affect as Dependent Variable
18. Positive Affect as Dependent Variable
III. Sample Differences by
Demographic Categories
19. Sample Differences by Sex
Variable: Assertiveness
20. Sample Differences by Sex
Variable: Dependency
21. Sample Differences by Sex
Variable: Social Influence Index
22. Sample Differences by Sex
Variable: Singleness Orientation
23. Regressions on the "Singleness Orientation" by Sex
24. Sample Differences by Age
Variable: Assertiveness
25. Sample Differences by Age
Variable: Dependency
26. Sample Differences by Age
Variable: Social Influence Index
27. Sample Differences by Age
Variable: Singleness Orientation
28. Regressions by Age
29. Sample Differences by Religiosity
Dependent Variable Singleness Orientation
30. Regressions by Religiosity
Dependent Variable Singleness Orientation
Correlated with Assertiveness, Dependency and
the Social Influence Index
112
SAMPLE DIFFERENCES BY SUBFILES
TABLE 10
VARIABLE ASSERTIVENESS
Subfile: Mean: N: Std. Dev:
CLIENTS
9.78 18 3.56
SECULARS 10.51 157 4.33
RELIGIOUS 9.88 69 4.06
OTHERS
9.78 37 4.78
Subfile Differences Difference
Between the
Means:
t: df: Sig. *
OTHERS-CLIENTS: 0.0 0.
-
n.s.
OTHERS-SECULARS: 0.73 0.85 192 n. s.
OTHERS-RELIGIOUS: 0.10 0.93 104 n.s.
CLIENTS-SECULARS: 0.73 1.24 173 U.S.
CLIENTS-RELIGIOUS: 0.10 0.10 85 n.s.
SECULARS-RELIGIOUS: 0.63 1.05 224 n.s.
♦Differences between the means were considered "not significant" where the h
value of tau-alpha exceeded .05. * * '
TABLE 11
VARIABLE DEPENDENCY
Subfile: Mean: N: Std. Dev:
CLIENTS: 7.33 18 3.90
SECULARS 8.70 157 4.10
RELIGIOUS 7.78 69 4.46
OTHERS 8.11 37 4.01
Subfile Differences: Difference
Between the
Means:
t: df: Sig. *
OTHERS-CLIENTS 0.78 0.69 53 n.s.
OTHERS-SECULARS 0.59 0.80 192 n.s.
OTHERS-RELIGIOUS 0.33 0.38 134 n.s.
CLIENTS-SECULARS 1.37 1.41 173 n.s.
CLIENTS-RELIGIOUS 0.45 0.42 85 n.s.
SECULARS-RELIGIOUS 0.92 1.46 224 n.s.
♦Differences between the means were considered "not significant" where the value ofn
tau-alpha exceeded .05. 01
TABLE 12
VARIABLE SOCIAL INFLUENCE INDEX
Subfile: Mean: N: Std. Dev:
CLIENTS 12.69 18 2.53
SECULARS 11.94 157 2.72
RELIGIOUS 12.60 69 2.29
OTHERS 12.84 37 2.66
Subfile Differences: Difference
Between the
Means:
t: df: Sig.*
OTHERS-CLIENTS 0.15 0.30 53 n.s.
OTHERS-SECULARS 0.90 2.71 192 .01
OTHERS-RELIGIOUS 0.24 0.69 104 n.s.
CLIENTS-SECULARS 0.75 1.75 173 n.s.
CLIENTS RELIGIOUS 0.09 0.20 85 n.s.
SECULARS-RELIGIOUS 0.66 2.76 224 .01
..........
♦Differences between the means were considered "not significant" where the value h
of tau-alpha exceeded..05. £
TABLE 13
VARIABLE SINGLENESS ORIENTATION
I ............ 1 * ' ■ .........-................ ■ 1 ■ 1.
Subfile: Mean: N: Std. Dev:
CLIENTS 4.56 18 1.72
SECULARS 4.79 150 1.95
RELIGIOUS 5.34 67 2.23
OTHERS 6.22 37 2.47
Subfile Differences
•
Difference Between t: df: Sig.*
the Means
OTHERS-CLIENTS 1.66 2.90 53 .01
OTHERS-SECULARS 1.43 3.29 185 .001
OTHERS-RELIGIOUS 0.88 1.80 102 n.s.
CLIENTS-SECULARS 0.23 0.52 166 n.s.
CLIENTS"RELIGIOUS 0.78 1.60 83 n.s.
SECULARS-RELIGIOUS 0.55 1.74 215 n.s.
♦Differences between the means were
tau-alpha exceeded .05.
considered "not significant" where the value ofH
TABLE 14
VARIABLE NEGATIVE AFFECT
Subfile: Mean: N: Std. Dev:
CLIENTS 2.94 18 1.63
SECULARS 2.17 155 1.69
RELIGIOUS 2.21 68 1.77
OTHERS 2.95 37 1.89
Subfile Differences: Difference
Between the Means: t: df: Sig.*
OTHERS-CLIENTS 0.01 0.02 53 n.s.
OTHERS-SECULARS 0.78 2.30 190 .05
OTHERS-RELIGIOUS 0.74 1.74 103 n.s.
CLIENTS-SECULARS 0.77 1.90 171 n.s.
CLIENTS-RELIGIOUS 0.73 1.52 84 n.s.
SECULARS-RELIGIOUS 0.04 0.13 221 n.s.
♦Differences between
tau-alpha exceeded .05.
the means were considered "not significant" where the value of £
00
TABLE 15
VARIABLE POSITIVE AFFECT
Subfile: Mean: N: Std. Dev:
CLIENTS 4.17 18 1.15
SECULARS 3.38 155 1.53
RELIGIOUS 3.54 68 1.51
OTHERS 3.30 37 1.58
Subfile Differences: Difference
Between the
Means:
t: df: Sig.*
OTHERS-CLIENTS 0.87 2.33 53 .05
OTHERS-SECULARS 0.08 0.28 190 his.
OTHERS-RELIGIOUS 0.24 0.76 103 n.s.
CLIENTS-SECULARS 0.79 2.67 171 .01
CLIENTS-RELIGIOUS 0.63 1.93 84 n.s.
SECULARS-RELIGIOUS 0.16 0.56 190 n.s.
♦Differences between the means were considered "not significant" where the value of £
tau-alpha exceeded .05.
II. REGRESSIONS BY SUBFILES
120
121
TABLE 16
SINGLENESS ORIENTATION AS DEPENDENT VARIABLE
Dependent Variable: Singleness Orientation,
Correlated with Assertiveness, Dependency,
and the Social Influence Index.
Subfile 1; CLIENTS, SECULARS, and RELIGIOUS:
Multiple R: .40 Fisher's Rz: .424 N:
Subfile 2; OTHERS:
Multiple R: .50 Fisher's Rz: .549 N:
Difference between Rz's: .125
Std. Error of Difference: .239
Confidence that difference is significant: <95%
207
22
122
TABLE 17
NEGATIVE AFFECT AS DEPENDENT VARIABLE
Dependent Variable; Negative Affect,
Correlated with Assertiveness-SII Dif-
ference, Dependency-SII Interaction,
and Assertiveness-Dependency Interaction.
Subfile 1: CLIENTS, SECULARS, and RELIGIOUS;
Multiple R; .13 Fisher's Rz; .131 N:
Subfile 2; OTHERS;
Multiple R; .11 Fisher's Rz; .110 N:
Difference between Rz's: .021
Std. Error of Difference; .2 39
Confidence that difference is significant: <95%
207
22
123
TABLE 18
POSITIVE AFFECT AS DEPENDENT VARIABLE
Dependent Variable; Positive Affect/
Correlated with Involvement with Others,
Assertiveness, and Dependency.
Subfile 1; CLIENTS, SECULARS, and RELIGIOUS:
Multiple R: .33 Fisher's Rz: .343 N:
Subfile 2; OTHERS:
Multiple R: .35 Fisher's Rz: .366 N:
Difference between Rz's: .02 3
Std. Error of Difference: .2 39
Confidence that difference is significant: <95%
207
22
III. SAMPLE DIFFERENCES BY
DEMOGRAPHIC CATEGORIES
124
125
TABLE 19
SAMPLE DIFFERENCES BY SEX
Variable: Assertiveness
Mean Std. Dev. N
Male 11.47 4.15 109
Female 9.29 4.12 161
Difference between
the means:
t: d.f. : Sig: *
2.18 4.25 268 . 0 0 1
*Differences between the means were considered "not sig
nificant" where the value of tau-alpha exceeded .05.
TABLE 20
SAMPLE DIFFERENCES BY SEX
Variable: Dependency
Mean Std. Dev. N
Male 7.47 4.02 109
Female 8.96 4.18 161
Difference between
the means:
t: d.f. : Sig:
1.49 2.94 268 . 0 1
126
TABLE 21
SAMPLE DIFFERENCES BY SEX
Variable: Social Influence Index
Mean Std. Dev. N
Male 12.72 2.59 109
Female 11.99 2.54 161
Difference between t:
the means:_______ ____
0.73 3.38
TABLE 22
SAMPLE DIFFERENCES BY SEX
Variable: Singleness Orientation
Mean Std. Dev. N
Male 5.15 2 . 1 2 103
Female 5.05 2.13 159
Difference between
the means:
t: d.f. : Sig.:*
0 . 1 0.37 260 n.s.
*Differences between the means were considered "not sig
nificant" where the value of tau-alpha exceeded .05.
d.f.: Sig.:
268 . 0 0 1
TABLE 23
REGRESSIONS ON THE "SINGLENESS ORIENTATION" BY SEX
Independent Variable
Males
N = 109
Females
N = 161
Multiple "R": .64* Multiple "R": .61*
Beta** Simple "r" Beta** Simple "r"
Assertiveness -0.44 .07 -0 . 1 2 . 0 0
Dependency 1.48 -.24 0.95 -.24
Social Influence Indicator 0.54 .59 0.72 .50
Assertiveness-Dependency Interaction -1 . 2 1 .18 -1.35 - . 1 0
Assertiveness-SII Interaction 0.65 .34 -0.31 .23
Dependency-SII Interaction -1.33 - . 0 1 -1.47 - . 0 1
Assertiveness-Dependency-S11 Interaction 0.94 . 0 1 1.96 .09
♦Difference between "R's": not significant, (p > .05).
**At the final step of the regression.
128
TABLE 24
SAMPLE DIFFERENCES BY AGE
Variable: Assertiveness
Class Mean Std. Dev. N
1. 30 & under 8.75 3.72 55
2. 31 - 40 9.78 4.22 82
3. 41 - 50 11.15 4. 06 104
4. 51 - 60 10. 50 5.09 34
5. Over 60 7. 67 3.79 3
Classes Difference between
the means
t: d.f. : Sig:*
1 - 2 1.03 1. 50 135 n.s.
1 - 3 2.40 3.75 157 . 0 0 1
1 - 4 1.75 1.75 87 n.s.
1 - 5 1.08 0.48 56 n.s.
2 - 3 1. 37 2.24 184 ; 05
2 - 4 0.72 0.73 114 n.s.
2 - 5 2 . 1 1 0.94 83 n.s.
3 - 4 0.65 0.67 136 n.s.
3 - 5 3.48 1.57 105 n.s.
4 - 5 2.83 1 . 2 0 35 n.s.
*Differences between the means were considered "not
significant" where the value of tau-alpha exceeded .05.
129
TABLE 2 5
SAMPLE DIFFERENCES BY AGE
Variable: Dependency
Class Mean Std. Dev. N
1. 30 & under 9.22 3.97 55
2. 31 - 40 8 . 24 4.43 82
3. 41 - 50 8 . 36 4.15 104
4. 51 - 60 7.32 3.92 34
5. Over 60 4. 6 6 3.05 3
Classes Difference between
the means
t: d.f. : Sig: *
1 - 2 0.98 1. 35 135 n.s.
1 - 3 0 . 8 6 1.28 157 n.s.
1 - 4 1.90 2 . 2 1 87 .05
1 - 5 4.56 2.48 56 . 0 2
2 - 3 0 . 1 2 0.19 184 n.s.
2 - 4 0.92 1 . 1 1 114 n.s.
2 - 5 3.58 1. 96 83 n.s.
3 - 4 1.04 1. 32 136 n.s.
3 - 5 3.70 2.04 105 .05
4 - 5 2 . 6 6 1.41 35 n.s.
♦Differences between the means were considered
"not significant" where the value of tau-alpha
exceeded .05.
130
TABLE 2 6
SAMPLE DIFFERENCES BY AGE
Variable: Social Influence Index
Class Mean Std. Dev. N
1. 30 & under 12.60 2. 38 55
2. 31
-
40 12.29 2. 37 82
3. 41
-
50 12.29 2.87 104
4. 51 -
60 11.57 2.72 34
5. Over 60 1 1 . 2 1 3.28 3
Classes: Difference
the means
between
:
t: d.f. : Sig: *
1 - 2 0. 31 1 . 1 2 135 n.s.
1
-
3 0. 31 1 . 1 0 157 n.s.
1
-
4 1.03 2.70 87 . 0 1
1
-
5 1. 39 1.06 56 n.s.
2 - 3 0 . 0 0 . 0 1 184 n.s.
2
-
4 0.72 1.99 114 .05**
2 - 5 1 . 08 0.83 83 n.s.
3 - 4 0.72 1.96 136 .05
3
-
5 1.08 0.82 105 n.s.
4
-
5 0.36 0.26 35 n.s.
♦Differences between the means were considered "not
significant" where the value of tau-alpha exceeded .05.
**Borderline
131
TABLE 27
SAMPLE DIFFERENCES BY AGE
Variable: Singleness Orientation
Class Mean Std. Dev. N
1. 30 and under 5.20 2.17 55
2. 31 - 40 5.04 1.90 81
3. 41 - 50 5.12 2.27 97
4. 51 - 60 4.91 2.24 33
5. Over 60 6 . 0 0 3.61 3
Classes: Difference
the means
between
•
t: d.f. : Sig:*
1 - 2 0.16 0.45 134 n.s.
1-3 0.08 0 . 2 2 150 n.s.
1-4 0.29 0.59 8 6 n.s.
1-5 0.80 0.38 56 n.s.
2-3 0.08 0.26 176 n.s.
2-4 0.13 0.29 1 1 2 n.s.
2-5 0.96 0.46 82 n.s.
3-4 0 . 2 1 0.46 128 n.s.
3-5 0 . 8 8 0.42 98 n.s.
4-5 1.09 0.51 34 n.s.
♦Differences between the means were considered "not
significant" where the value of tau-alpha exceeded .05.
TABLE 28
REGRESSIONS BY AGE
(Independent Variables: Assertiveness,- Dependency, Social Influence Index).
1. Dependent Variable: Singleness Orientation
Multiple R: N: Diff. Between R's:
Age 50 and under .44 193
Age 51 and over .47 33 .03 (n.s.)*
2. Dependent Variable: Negative Affect
Age 50 and under .17 193
Age 51 and over .32 33 .15 (n.s.)*
3. Dependent Variable: Positive Affect
Age 50 and under .30 193
Age 51 and over .50 33 . 2 0 (n.s.)*
♦Confidence interval of significance: <95%
TABLE 29
SAMPLE DIFFERENCES BY RELIGIOSITY
Dependent Variable Singleness Orientation
Religiosity Class : Mean Singleness Orientation: Std. Dev: N:
1. Very Religious 5.84 1.95 19
2. Moderately Religious 4.87 2.09 129
3. Not Religious 5.12 2.18 119
Class Differences: Difference Between the Means: t: df: Sig: *
1 - 2 0.97 2.01 146 .05
1-3 0.72 1.47 136 n.s.
2-3 0.25 0.93 246 n.s.
♦Differences between the means were considered "not significant" where the value of
tau-alpha exceeded .05.
********
Religiosity Class 1 ("Very Religious") Broken Down by Age
Age Class: Mean Singleness Factor: N:
1. 30 and under 5.60 5
2. 31-40 4.75 8
3. 41-50 7.60 5
4. 51-60 7.00 1
5. Over 6 0 0
u>
( j j
Class Total 5.10 19
TABLE 30
REGRESSIONS BY RELIGIOSITY
DEPENDENT VARIABLE SINGLENESS ORIENTATION CORRELATED WITH ASSERTIVENESS
DEPENDENCY AND THE SOCIAL INFLUENCE INDEX
Religiosity Class 1 ("Very Religious")
Multiple R:
Religiosity Class 2
.87 Fisher's
("Moderately Religious")
Rz: 1.333 N: 17
Multiple R:
Religiosity Class 3
.56 Fisher's
("Not Religious")
RZ: .633 N: 119
Multiple R: .63 Fisher's
R2;
.741 N: 106
Class Pairs: Difference Between Rz's: Std. Error of Diff: Sig.:
1 - 2 0.7 .283 .05
1-3 0.592 .285 .05
2-3 0.108 .138 n.s.*
♦Confidence interval of significance <95%.
u>
A P P E N D I X C
DESCRIPTIVE STATISTICS OF THE VARIABLES
135
APPENDIX C
Contents
Table
31. Independent, Directly Scored Variables
32. Dependent, Directly Scored Varaibles
33. Pearson's Correlation Coefficients for Individual
Social-Interpersonal Items, Correlated with the
Singleness Orientation
34. The Computed Variables
136
TABLE 31
INDEPENDENT, DIRECTLY SCORED VARIABLES
(Total Sample. N: 281)
Variable: Assertiveness
Mean 10.21
Variance: 18.26
Range: 20.00
Standard Error: 0.26
Kurtosis: -0.79
Minimum: 0.0
Std Dev: 4.27
Skew: -0.05
Maximum: 20.00
************** ** ** ** ** *** *** ** ** ** ** *** ** ** **
Variable: Dependency
Mean: 8.31
Variance: 17.42
Range: 20.00
Standard Error: 0.25
Kurtosis: -0.01
Minimum: 0.0
Std Dev: 4.17
Skew: 0.43
Maximum: 20.00
u>
TABLE 32
DEPENDENT, DIRECTLY SCORED VARIABLES
(Total Sample. N: 281)
Variable: Negative Affect
Mean: 2.34
Std. Error: 0.25 Std. Dev: 1.75
Variance:
3.06
Kurtosis: -1.24 Skewness: 0.43
Range:
5.0
Minimum: 0 . 0 Maximum: 5.0
***** ************ ****** * * * * * ****** ******* * * *
Variable: Positive Affect
Mean:
3.46 Std. Error: 0.09 Std. Dev: 1.52
Variance:
2.30
Kurtosis: -1.24 Skewness: -0.78
Range:
5.0
Minimum: 0 . 0 Maximum: 5.0
***** ************ ****** * * * * * ****** ******* * * *
Variable: Singleness Preference
Mean: 2.54 Std. Error: 0.07 Std. Dev: 1.17
Variance: 1.39
Kurtosis: -0.47 Skewness: 0.45
Range: 5.0 Minimum: 0 . 0 Maximum: 5.0
***** ************ ****** * * * * * ****** ******* * * *
Variable: Singleness Expectation
Mean:
2.67 Std. Error: 0.07 Std. Dev: 1.20
Variance:
1.45
Kurtosis: -0.96 Skewness: 0.19
Range: 5.0 Minimum: 0.0 Maximum: 5.0
< jU
00
TABLE 33
PEARSON'S CORRELATION COEFFICIENTS FOR INDIVIDUAL SOCIAL-INTERPERSONAL ITEMS
CORRELATED WITH THE SINGLENESS ORIENTATION
(Based Upon 281 Cases)
Items used in computing the "Social Influence Index!' are
indicated by an asterisk (*). Items used in computing
"Involvement with Others" are indicated by two asterisks (**).
Item
i i £ * *
Significance
*
1. A person can have a normal...life without
being married. .43 . 0 0 1
*
2. Being single has... advantages over...marriage. .29 . 0 0 1
*
3. Sometimes I'm not sure what is expected of me. .23 . 0 0 1
4. There is a place and function for single per
sons. . 1 1 .031
*
5. Members of my family would like me married. . 2 1 . 0 0 1
6 . My religion teaches it is better to be mar
ried. .04 .256
7. Sex outside of marriage is morally wrong. . 1 0 .047
8 . I now have a companion who would like to get
married . 1 2 .027
i _ j
*
9. I find it hard to be a parent and a single
person at the same time. .13 .014
u>
VO
TABLE 33
PEARSON'S CORRELATION COEFFICIENTS FOR INDIVIDUAL SOCIAL-INTERPERSONAL ITEMS
CORRELATED WITH THE SINGLENESS ORIENTATION
(Based Upon 281 Cases) (Continued)
Items used in computing the "Social Influence Index" are
indicated by an asterisk (*). Items used in computing
"Involvement with Others" are indicated by two asterisks (**).
Item
H I I
Significance
*10. Most of my friends are married .14 . 0 1 1
11. The children are with me most of the time. . 1 2 . 0 2 0
1 2 . ...It is far better to be married than single
on my job. .04 .276
*13. Time spent at work each week. .16 .004
**14. Time spent in same-sex group activities. -.13 .016
**15. Time spent in mixed-sex singles' activities. -.13 .017
*16. Time spent on recreation and hobbies. .19 . 0 0 1
*17. I prefer to live as part of a warm and
caring family. .30 . 0 0 1
**18. My sex life is fully satisfying to me. •
o
o
.476
**19. I have friends with whom I feel perfectly
close and trusting. - . 0 2 .382
*20. Being married is essential to my goals. . 6 6 . 0 0 1
L40
TABLE 34
THE COMPUTED VARIABLES
(Total Sample. N: 281)
Variable: Social Influence Index
Mean: 12.27 Std. Error: 0.16 Std. Dev: 2.64
Variance: 6.98 Kurtosis: -0 . 6 6 Skewness: 0.01
Range: 1 2 . 0 Minimum: 6 . 0 Maximum: 18.0
***** ********* *********** * * * ** * * * * * * * * * * * * * * *
Variable: Involvement With Others
Mean: 10. 67 Std. Error: 0.14 Std. Dev: 2.26
Variance: 5.10 Kurtosis: 0.63 Skewness: -0.76
Range: 1 2 . 0 Minimum: 3.0 Maximum: 15.0
***** ********* *********** * * * * *** * * * * * * * * * * * * *
Variable: Singleness Orientation
Mean: 5.11 Std. Error: 0.13 Std. Dev: 2.14
Variance: 4.56 Kurtosis: -0 . 2 1 Skewness: 0.46
Range: 9.0 Minimum: 1 . 0
■ ' ■ H*
Maximum: 1O.)0*»
L _ !
A P P E N D I X D
STATISTICS SUPPORTING THE FINDINGS
142
APPENDIX D
Contents
Table
35. Zero-Order Correlation Coefficients
36. Correlation of Age with Singleness Preference
37. Correlation of Age with Singleness Expectation
38. Multiple Regression Showing the Effects of
Positive Affect and Negative Affect as
Possible Predictors of Singleness Orientation
39. Multiple Regression: Negative Affect as
Dependent Variable (Hypothesis 7)
40. Multiple Regression: Positive Affect as
Dependent Variable (Hypothesis 8 )
Figure
11. Aid Program; Singleness Preference as
Dependent Variable
12. Aid Program: Singleness Expectation as
Dependent Variable
143
TABLE 35
ZERO-ORDER CORRELATION COEFFICIENTS
Assertiveness
Dependency
Negative
Affect
Social
Influence
Indicator
Singleness
Orientation
Assertiveness
Dependency
Interaction
As sertivenes s
SII
Interaction
Dependency
SII
Interaction
Combined
Interaction
Assertiveness-
SII
Assertiveness
**
- . 0 1 -.14 .32 .04 .61 .91 .13 .64 .54
Dependency
**
. 1 2 -.23 -.26 .71 -.09 .89 .60 -.09
Negative Affect
**
-.06 -.07 - . 0 0 -.14 .09 - . 0 2 .07
Social Influence Indicator
**
.52 .04 .64 . 2 0 .31 .15
Singleness Orientation
**
-.14 .28
i
•
o
• e *
.05 . 1 2
As s er tivenes s-Dependency
Interaction
**
.49 .75 .94 -.37
Assertiveness-SII Interaction
**
.19 .65 -.43
Dependency-SII Interaction
**
•
00
-.03
Combined Interaction
**
-.33
Assertiveness-SII
144
*
*
TABLE 36
CORRELATION OF AGE WITH SINGLENESS PREFERENCE
Count I Age
I
I
I I I 2 1 3 1 4 1 5 1
Row
Total
Singleness _________ ^_________^________ x________ ^________^
Preference ^ x 9 I 1 5 j 24 I 9 1 1 1 5 8
I I I I I I 21.0
-I--------- 1---------- 1----------1----------1--------- 1
2 I 20 I 26 I 29 I 11 I 0 1 8 6
I I I I I I 31.2
3 1 13 I 26 I 30 I 10 I I I 80
I I I I I I 29.0
4 I 8 1 11 I 9 1 I I 0 1 29
I I I I I I 10.5
5 1 3 1 4 I 12 I 3 1 I I 23
I I I I I I 8.3
Column -I-------- 1-------- 1-------- 1-------- 1--------1
Total 53 82 104 34 3 276
19.2 29.7 37.7 12.4 1.1 100.0
ETA = 0.00519 with Singleness Preference Dependent.
Number of missing observations = 5
145
TABLE 37
CORRELATION OF AGE WITH SINGLENESS EXPECTATION
Count I Age
I
I
I -.1 o l I A I ^ 1
Row
Total
Singleness _________ -___ -___ -__
Expectation ± x i 0 i 14 I 20 I 7 I 0 I 51
I I I I I I 18.4
2 1 16 I 33 I 24 I 10 I 2 1 85
I I I I I I 30.7
3 1 17 I 14 I 23 I 6 1 0 1 60
I I I I I I 21.7
4 I 11 I 18 I 25 I 9 I 0 1 63
I I I I I I 22.7
5 1 I I 3 1 11 I 2 1 I I 18
I I I I I I 6.5
-I-------- 1-------- 1 -------- 1-------- 1 -------- 1
Column 55 103 34 3 277
Total 19.9 37.2 12.3 1.1 100.0
Somers' D (Asymmetric) = 0.05720 with Singleness Expectation Dependent.
Number of missing observations = 4
147
TABLE 38
MULTIPLE REGRESSION SHOWING THE EFFECTS OF
POSITIVE AFFECT AND NEGATIVE AFFECT AS
POSSIBLE PREDICTORS OF SINGLENESS ORIENTATION
Dependent Variable: Singleness Orientation
Multiple R: .54 N: 281
Variable Beta* Simple R
Dependency -0 . 1 2 -.23
Assertiveness -0 . 1 2 - . 0 0
Social Influence Index 0.51 .49
Negative Affect
in
o
•
0
1
-.06
Positive Affect -0 . 1 2 -.09
*At final step of the regression.
148
TABLE 39
MULTIPLE REGRESSION: NEGATIVE AFFECT AS
DEPENDENT VARIABLE (Hypothesis 7)
Multiple R: .13 N: 281
Variable Beta* Simple R
Assertiveness-SII
(Absolute difference) 0.03 .07
Assertiveness-Dependency
Interaction -0 . 1 2 - . 0 0
Dependency-SII
Interaction 0.18 .09
*At final step of regression.
149
TABLE 40
MULTIPLE REGRESSION: POSITIVE AFFECT AS
DEPENDENT VARIABLE (Hypothesis 8 )
Multiple R: .38 N: 281
Variable Beta* Simple R
Assertiveness 1.67 .27
Involvement with Others 0 . 6 8 .18
Dependency 1.30 - . 0 1
Interaction:
Dependency,
Assertiveness,
Involvement 1.89 .18
Interaction:
Dependency
Assertiveness,
-1.98 . 1 1
Interaction:
Involvement
Dependency,
-1.31 .06
Interaction:
Involvement
Assertiveness,
-1.53 .30
*At final step of the regression.
FIGURE 11
AID PROGRAM: SINGLENESS PREFERENCE AS DEPENDENT VARIABLE
Singleness Preference, Dependent Var.
Range: 0 - 5
11. Assertive
Value.: 1* 4
Mean Sing.Prf: 2.24
9. Age
31-40, 51-60
Mean Sing.Prf: 2.60
Independent Vars.
Assertiveness
Dependency
Social Influence
Index (All re
coded to a range
Age
Sex
Religiosity
cn
o }Rejected by program
1. Sample
N: 281
Mean Sing.Prf.
2.53
8 . Age N: 44
Under 30, 41-50 over 60
Mean Sing.Prf: 2.16
Mean Sing.Prf:
3.69
7. Age
41-50, 61 over
Mean Sing.Prf.
3. 33
3. SII
Value: 3, 4
Mean Sing.Prf:
3.04
6 . Age
Under 40, 51-60
Mean Sing.Prf:
1.94
4. Dependency
Value: 2, 3,
5. Dependency
Value: 0, 1
Mean Sing.Prf:
2.39
N: 193
Mean Sing.Prf:
2.16
SII
Mean Sing.Prf: 1.76
10. Assertive
AID PROGRAM:
FIGURE 12
SINGLENESS EXPECTATION AS DEPENDENT VARIABLE
Singleness Expectation: Dependent Var.
Range: 0 -
Independent Vars: N
Assertiveness
Dependency
Social Influence Index
(All recoded to a
range of 0-4)
Age
Sex
Religiosity
} Rejected by program
<_n
1. Sample
N: 281
Mean Sing.Exp:
2.66
11. Age
41-50, 61 over
Mean Sing.Exp:
3.74
Mean Sing.Exp:
3. SII
Value: 3
N: 192
Mean Sing.Exp:
2.33
SII
Mean Sing.Exp:
2.90
5. Dependency
Value: 0
7. Assertive
Value: 0, 1, 2
N: 110
Mean Sing.Exp: 2,35
10. Age
Under 40, 51-60
Mean Sing.Exp:
3.12
£. Assertive
Value: 3, 4
Mean Sing.Exp: 1.94
Mean Sing.Exp: 2.57
9. Dependency
Value: 1, 4
8 . Dependency
Value: 2, 3
Mean Sing.Exp: 2.20
4. Dependency
Value: 1, 2, 3
N: 163
Mean Sing.Exp:
2.22

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Coalition Formation In Conjoint Marriage Counseling

PDF

The Attitude Of Resistance To The Adoption Of An Older Child

PDF

An Analysis Of Factors In The Family'S Withdrawal Of A Patient From A Hospital For The Mentally Retarded

PDF

Personality And Interpersonal Factors Associated With The Duration Of Marriage Counseling

PDF

The Effect Of Counselor Gender And Sex-Role Attitudes On Change Of Female Clients' Sex-Role Attitudes

PDF

Familial Role Typology, Accuracy Of Perception, And Mutual Needs Among Pre-Nuptial Partners

PDF

Complementarity, Homogeneity, Heterogeneity And Marital Stability

PDF

An Investigation Of Complementary Needs Between Marital Partners

PDF

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

PDF

An Analysis Of The Positive And Negative Functions In The Contemporary Usage Of An Ancient Rite Of Passage

PDF

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

PDF

On Becoming A Parent: Attitude And Feeling Changes

PDF

Power Relationships In Marital Discord

PDF

An Analysis Of The Changing Focus Of Marriage Counseling

PDF

Personality Characteristics: Ideal And Perceived In Relation To Mate Selection

PDF

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

PDF

Strategies Of Marital Communication

PDF

Atomism and social control: Causes and consequences of alcohol use.

PDF

Time Duration In Marathon Groups: Effects Upon Self-Concept And Self-Actualization

PDF

Design And Testing Of A Group Cohesiveness Scale To Measure The Salience Of Ethnic Identity Among Polish-Americans In The Los Angeles Metropolitanarea

Asset Metadata

Creator Olander, Edward Bernard (author)

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

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 Broderick, Carlfred Bartholomew (committee chair), Feldman, Albert G. (committee member), Peterson, James A. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-842594

Unique identifier UC11356558

Identifier 7421494.pdf (filename),usctheses-c18-842594 (legacy record id)

Legacy Identifier 7421494.pdf

Dmrecord 842594

Document Type Dissertation

Rights Olander, Edward Bernard

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