This item is only available as the following downloads:
THE NATURE AND THE EXTENT OF THE RELATIONSHIPS DIVORCED
MEN MAINTAIN WITH THEIR FIRST FAMILIES
WALTER DAVID TROPF
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
One aspect of academic research that is seldom mentioned in
the textbooks is the number of people who become peripherally
involved in a research project, either because of their proximity
to the researcher or because of the variety of support services
needed to carry out the project. A very important part of this
latter group was the nine University of Central Florida students
who assisted in gathering the data. Their enthusiasm for the
project was encouraging and their tenacity in tracking down poten-
tial respondents engendered trust. They deserve to be mentioned by
name--Suzette Morris, Stacy Daughn, Joe Cravens, Lee Deese Walls,
Sandra Hatley, Beverly Tandy, Larry Murray, LaNita Knight Edourd
and Debbie Comer. Debbie deserves honorable mention for conducting
extra interviews and fielding phone calls following the newspaper
story. The assistance given by Tom Ticknor at the UCF Computer
Center was also vital in entering and extracting the data from the
computer once they were obtained.
At the University of Florida each member of my committee made
significant contributions to my education, both in the classroom
and in helping me prepare for qualifying exams. To Felix Berardo,
Gerald Leslie, Gordon Streib, Rod Webb, and earlier, Ben Gorman I
express my appreciation. In addition, the sound advice and patient
editing of my chairman, Wilbur Bock, were a necessary adjunct to
the successful completion of my dissertation.
While whatever academic credit which may accrue to this report
must be shared with those mentioned above, the necessary encourage-
ment and understanding have come from family and friends. To my
parents, who encouraged reading, fostered curiosity and never
flinched from the impact of new knowledge even when it ran contrary
to long held beliefs, I owe more than can be expressed.
A project of this magnitude would never have been started
without the encouragement and support of Annabelle and the children.
For their faith, and not inconsiderable sacrifices, I am truly
Finally, I would like to express my appreciation to our many
friends, both in academia and out, who have expressed an interest
in my studies, tolerated my preoccupation and provided encouragement
to continue. I particularly want to mention the Evans family who
served as my hosts for the five quarters I needed a second home in
Gainesville. Their warmth and hospitality will be long remembered.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS. . .
LIST OF TABLES. . .
ABSTRACT. . .
I STATEMENT OF THE PROBLEM .
II THEORETICAL FRAMEWORK. .
Role Conflict. .
Social Structure and Deviance.
III REVIEW OF THE LITERATURE .
Major Studies. .
Minor Studies. .
Summary. . .
IV RESEARCH METHOD. .
The Questionnaire .
Operational Definitions. .
Data Analysis .
V DESCRIPTION OF THE SAMPLE. .
Characteristics of the Sample.
Personal Contacts. .
Child Support. ... ..
Relationships Between Divorced
Their Ex-Wives .. ....
Attitudes of the Present Wife.
Satisfaction with Post-Divorce
Pre-Divorce Paternal Role Perfo
Cause of Divorce .
Natural Father Versus Stepfathe
Responsibility .. ....
Sub-Sample Differences .
romance as a
TABLE OF CONTENTS (Continued)
Summary Description of the Dependent and
Independent Variables . .
VI DATA ANALYSIS AND TEST OF THE HYPOTHESES ..
Test of the First Hypothesis . .
Test of the Second Hypothesis. . .
Test of the Third Hypothesis . .
VII DISCUSSION . . .
Personal Contacts and Child Support .
Visiting and Support Arrangements. .
Custody and Living Arrangements. .
VIII SUMMARY AND EVALUATION . .
Problems in the Research . .
Validity of the Findings . .
Findings in Relation to Theory . .
New Areas for Study and Investigation. .
APPENDIX A ORIGINAL LETTER TO RESPONDENTS .
APPENDIX B MODIFIED LETTER TO RESPONDENTS. .
APPENDIX C QUESTIONNAIRE . .
APPENDIX D DESCRIPTION OF RESPONDENTS BY SELECTED
VARIABLES . . .
APPENDIX E DATA CONCERNING DURATION AND FREQUENCY
OF VISITS . . .
REFERENCES . . .......
BIOGRAPHY . . .
LIST OF TABLES
1 LENGTH AND FREQUENCY OF VISITING BY
MARITAL STATUS. . ... 44
2 FREQUENCY OF MAIL AND PHONE CONTACTS BY
MARITAL STATUS. . ... 46
3 RESPONSE OF DIVORCED FATHER TO CHILD'S AND
EX-WIFE'S REQUEST FOR EXTRA MONEY BY
MARITAL STATUS. . ... 52
4 DIVORCED FATHER'S RELATIONSHIP WITH EX-WIFE AT
TIME OF DIVORCE AND AT TIME OF INTERVIEW. 54
5 EX-WIFE'S SUPPORT OF DIVORCED FATHER'S
RELATIONSHIP WITH HIS CHILDREN. ... 56
6 NATURE OF THE DISCUSSION BETWEEN DIVORCED FATHER
AND EX-WIFE REGARDING THEIR CHILDREN AT THE
TIME OF THE DIVORCE AND THE TIME OF THE
INTERVIEW. . .... 58
7 NATURE OF THE DISCUSSION BETWEEN DIVORCED FATHER
AND EX-WIFE REGARDING THEMSELVES AT THE
TIME OF THE DIVORCE AND THE TIME OF THE
INTERVIEW. . . .. 60
8 REASON GIVEN FOR CHANGE IN DISCUSSIONS BETWEEN
EX-SPOUSES CONCERNING CHILDREN ... 63
9 REASON GIVEN FOR CHANGE IN DISCUSSION BETWEEN
EX-SPOUSES CONCERNING SELVES .... .64
10 ATTITUDES OF PRESENT WIFE TOWARD THE CONTACTS
WITH FIRST CHILDREN AT TIME OF REMARRIAGE AND
AT TIME OF INTERVIEW . ... 68
11 ATTITUDES OF PRESENT WIFE TOWARD THE MONEY SPENT
ON FIRST CHILDREN AT TIME OF REMARRIAGE AND
AT TIME OF INTERVIEW . .. 69
LIST OF TABLES (Continued)
12 ATTITUDES OF PRESENT WIFE TOWARD CONTACTS
DIVORCED MAN HAS WITH EX-WIFE AT TIME OF
REMARRIAGE AND AT TIME OF INTERVIEW. .
13 EFFECT OF DIVORCE ON THE SATISFACTION OF
THE FATHER'S PERFORMANCE IN FIVE
PATERNAL ROLES. . .. 73
14 CHILD REARING PRACTICES AS A CAUSE OF DIVORCE
FOR FIVE PATERNAL ROLES . .... .76
15 AMOUNT OF RESPONSIBILITY THE NATURAL FATHER
ASSIGNS TO SELF AND STEPFATHER FOR FIVE
PATERNAL ROLES. . .... 78
16 SUB-SAMPLE DIFFERENCES IN VISITING FREQUENCY
AT DIVORCE. . . ... 88
17 FREQUENCY OF VISITS BY MARITAL STATUS
CONTROLLING FOR DISTANCE . .. 89
18 ARRANGEMENTS FOR VISITING BY RELIGION ...... .95
19 NUMBER OF FATHERS HAVING CUSTODY OF ONE OR MORE
OF THEIR CHILDREN BY SOCIAL CLASS ... .97
20 NUMBER OF FATHERS HAVING CHILDREN LIVE WITH
THEM BY SOCIAL CLASS. . .... 97
21 DESCRIPTION OF RESPONDENTS BY SELECTED
VARIABLES . .... 156
22 NUMBER OF FATHERS VISITING THEIR CHILDREN
FOLLOWING SEPARATION BY DURATION AND
FREQUENCY . .... 160
23 NUMBER OF FATHERS VISITING THEIR CHILDREN
FOLLOWING DIVORCE BY DURATION AND
FREQUENCY . .... ..
24 NUMBER OF FATHERS VISITING THEIR CHILDREN
FOLLOWING OWN REMARRIAGE BY DURATION AND
FREQUENCY . .
25 NUMBER OF FATHERS VISITING THEIR CHILDREN
FOLLOWING REMARRIAGE OF EX-SPOUSE BY
DURATION AND FREQUENCY. . .
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
THE NATURE AND THE EXTENT OF THE RELATIONSHIPS DIVORCED
MEN MAINTAIN WITH THEIR FIRST FAMILIES
Walter David Tropf
Chairman: E. Wilbur Bock
Major Department: Sociology
This study examines the relationships that continue between
father and child and between ex-spouses after divorce. A conven-
ience sample was used consisting of 101 men. The better educated,
white Protestant predominated with Protestants being in the majority
across all social classes.
Based on an assumption of post-divorce normlessness and a
review of the literature on role conflict and social structure,
three hypotheses were developed for testing. The first predicted
that personal father-child contacts and voluntary child support
would decrease with the remarriage of either ex-spouse.
The second hypothesis predicted that the way in which arrange-
ments for child support and visiting were made at the time of the
divorce would depend on the ex-wife's satisfaction with pre-divorce
paternal role performance. If paternal role performance was not an
issue in the divorce, arrangements for support and visiting would
more likely be derived by mutual consent. If performance was an
issue, arrangements would more likely be negotiated by a third party.
The third hypothesis predicted that middle class fathers would
be more likely to have custody, or have their children living with
them, than either upper class or lower class fathers.
From chi-square analysis, statistically significant decreases
in visiting frequency were found after the remarriage of either
ex-spouse (P < .05). The decrease was greater following the remar-
riage of the ex-wife. Frequency of mail and phone contacts remained
relatively stable with phone contacts increasing slightly after the
remarriage of the ex-wife. Voluntary contributions for support
increased after the remarriage of the father but decreased after the
remarriage of the ex-wife.
An examination of 11 other variables showed that the decline
in visiting after remarriage was consistent. Only geographic dis-
tance reached significance as an intervening variable. Those men
living closer to their children visited more frequently than those
living over 150 miles away (P < .01 at the remarriage of the father
and P < .001 at the remarriage of the ex-wife). Other variables
which showed indications of negative influences on visiting were
lower class standing, being Protestant, being delinquent in ordered
support payments and the ex-wife opposing visiting.
Child support arrangements were negotiated by a third party in
79% of the cases while third party negotiations for visiting
occurred in 37% of the cases. In neither case did pre-divorce
paternal role performance have an effect on how these arrangements
Two intervening variables influenced how visiting arrangements
were made. Catholics used third party intervention more frequently
than Protestants (P < .05). Middle class fathers arranged visiting
by mutual consent more frequently than lower class fathers (NS).
Middle class men were found more likely to have custody of
their children or have their children living with them, than either
upper or lower class men. The custody finding was significant
(P < .05) and the living arrangement finding approached significance
(P = .08). Although not significant, Catholics were less likely to
have custody or have a child living with them than Protestants.
From the data pertaining to the second hypothesis and an exam-
ination of the literature on role bargaining, an additional hypothe-
sis was developed predicting that heavily sanctioned roles are more
likely to be negotiated through institutionalized means than by
mutual consent. Further research is suggested to test the prediction.
STATEMENT OF THE PROBLEM
Since World War II the problem of divorce has attracted an
increasing amount of attention as both the public and those
affected began to struggle with the meaning of marital separation
for individuals and society at large. In addition to the investi-
gations of the social scientist, contributors to the literature
have included judges and attorneys, social workers, other helping
professionals and the participants themselves. This growing
interest in divorce has resulted in a spate of personal accounts,
legal and tax advice, economic studies and counseling books as well
as the formation of Parents Without Partners. This group, organ-
ized in 1958 as a self-help group to assist divorced parents in
their struggle against the isolation imposed by society on the
divorced person, has proved to be an important source of informa-
tion (Bernard, 1971, p. 9; Weiss, 1975, p. x).
Sell (1978) has compiled a comprehensive bibliography of all
divorce related literature published in the United States from 1970
to 1978. He has classified this literature into 21 different cate-
gories, indicative of the wide variety of disciplines and publics
interested in this topic.
For the family sociologist the foundation of divorce research
has been the social-demographic study. The primary thrust of these
studies is the examination of such variables as length of marriage,
age at marriage, education, occupation, residence, religion and
race as factors relating to the occurrence of divorce. Typical
of post-World War II studies of this nature are Jacobsen (1950),
Ackerman (1963), Bernard (1966) and Bumpass and Sweet (1972). The
findings have been remarkably consistent and have succeeded in
isolating significant variables of importance. Their usefulness
was considerably enhanced by the Bumpass and Sweet research which
made use of multi-variate analysis to shed light on the relative
importance of these variables.
One aspect of divorce that has received relatively little
attention is the continuing relationship between ex-spouses, and
between father and child, after marital separation. The purpose of
this paper is to examine these post-divorce relationships. The
basic assumptions on which the examination will be made are that
norms are lacking to help ex-spouses define post-divorce roles,
that this normlessness will produce role conflict and that existing
social structures make it difficult to experiment with new roles.
Based on these assumptions a theoretical framework will be
developed from which hypotheses will be constructed for testing.
This will be followed by a review of the literature on post-divorce
relationships and behaviors which might yield information pertaining
to the assumptions and the hypotheses. Succeeding chapters will
deal with the methods employed for gathering the data, a discus-
sion of the findings, the data analysis and test of the hypotheses,
a discussion of the analysis and a summary and evaluation of the
The assumption that few norms exist to assist former family
members in establishing functional relationships after divorce has
had considerable acceptance among social scientists who have been
students of family life. It has been noted that institutional
structures do not exist to order relationships between ex-spouses,
between noncustodial parents and children or between divorced per-
sons and the larger kin network and the community (Goode, 1965;
Bohannen, 1971; Mead, 1971). Goode in particular has pointed out
that the kinship structure has been ambiguous concerning the rights
of divorced persons to material or emotional support, re-admission
to the former kinship structure, the formation of new families and
the proper behavior of ex-spouses to each other (pp. 8-15). He also
suggests that this failure to institutionalize post-divorce posi-
tions functions to encourage a rapid resumption of marital status
by making divorced persons uncomfortable with positions so lacking
in role specificity (pp. 206-208).
The second assumption follows from the first. It is that this
lack of role prescriptions to guide post-divorce relationships will
produce conflict between occupants of post-divorce statuses.
The third assumption is that those few social structures that
do exist may no longer be seen as functional by some occupants of
post-divorce statuses; however, the existence of these structures
will serve to prohibit experimentation with new role definitions
which might prove to be more satisfying.
Although several post-divorce statuses will be considered in
this study the one of primary interest is the divorced male as
father. It has been suggested that the normative order for paternal
behavior varies from culture to culture and has changed over time
in contrast to maternal roles which are seen as invariant in both
time and space. Little research has been devoted to paternal roles
and what has been done is based on information collected from
mothers and children rather than from the fathers themselves.
Nevertheless, the father role has been observed to be deteriorating
in Western culture and divorce is a potential threat to that role
(Benson, 1968, pp. 3-22; Lynn, 1974, pp. 5-12). Following divorce
the obligations attendant upon this role must be fulfilled from a
distance both physical and social in nature. Thus the obligations
assumed by the individual father become more difficult to fulfill.
Two theoretical conceptions seem to offer a base from which
hypotheses can be developed to examine post-divorce behavior. One
is found in the literature pertaining to role conflict and role
bargaining and the second in the literature pertaining to social
structure and deviance. These three concepts (role conflict, role
bargaining and social structure) will be discussed and their connec-
tion to post-divorce behavior explained. From these discussions
hypotheses for testing will be developed.
Most of the research concerning role conflict has been con-
ducted within bureaucratic structures (Kahn, Wolfe, Quinn, Snoeck,
& Rosenthal, 1964, p. vii). The delineation of significant con-
cepts can be found in these studies as well as the theoretical
studies which preceded them. While role conflict can come from
several sources and be of varying types, basic to all role conflict
is the fact that various members of a role set can hold differing
expectations of the requirements of a given role.
Goode (1960) suggests that the norms of society are not a suf-
ficient guide for individual action. Continuity in social roles
can be affected by variations in social class, conflicts over values
and ideals, disagreement over the interpretation of obligations and
changes in social position. In addition to problems of continuity,
conflict can arise over the allocation of resources to a given role
or the role may have conflict inherent in it. He concludes that
the total number of role obligations for any one person will always
exceed his available resources.
This is also supported by Merton (1957) who has discussed the
conflict inherent in role-sets which he defines as the complement of
role relationships attached to a particular status. That is, an
actor in any one of his roles will have a number of people with
whom he will engage in mutual interaction and each person will make
his own particular demands upon the actor. Merton proposes, as does
Goode, that conflict is residual in such structures.
Role conflict for the divorced man can assume particular
significance over time. When first divorced the number of role
obligations lessens. He is no longer a husband. In fact he may,
in the initial post-divorce period, increase his parental role com-
mitments and spend more time with his children than before (Weiss,
1975, pp. 187-198). Differing expectations concerning his paternal
role can come from a multitude of others besides the expectations
of self, ex-wife, and children. Significant others in the situation
could consist of family on both sides as well as a whole array of
new relationships as both ex-wife and divorced father add to their
role sets through remarriage. For the divorced father the poten-
tial accumulation of new role obligations is quite high. Beyond
new role obligations the potential number of people peripherally
involved outside of his role-sets can become even higher (Bohannen,
1970, pp. 132-137). For example, if a divorced father wishes to
have his children for a summer visit he may have to take into
consideration the plans of his ex-wife and her husband, his
ex-wife's stepchildren and their mother and the new children of his
ex-wife and her husband. On his side he may have to consider the
plans of his new wife, his stepchildren, the new children of the
present marriage and his new wife's ex-husband plus the ex-husband's
new wife and their children. That these new relationships are
likely to exist is not imaginative. Studies indicate that the
median length of time between divorce and remarriage is now approx-
imately three years and that four out of five persons of middle
age will remarry after their first divorce (Current Population
The stresses imposed by conflicting expectations, the accumula-
tion of additional role obligations, inadequate resources and lack
of time are not limited to the divorced father. The ex-wife has
to cope with a variety of instrumental and expressive roles which
have been vacated by the ex-husband, some of which she may be
ill-prepared to fulfill (Bogue, 1949, p. 212; Weiss, 1975, pp. 167-
When conflict becomes sufficiently great to preclude adequate
role performance steps will be taken to resolve it. Many accommoda-
tions can be made to reduce strain and resolve conflict but one
alternative, which is consistently referred to in the literature,
is some form of role reduction or role abandonment (Buchard, 1954;
Getzels & Guba, 1954; Goode,1960; Shaw & Constanzo, 1970; Stouffer,
1949; Thibaut & Faucheux, 1964; Toby, 1964; Wardell, 1955). Goode
suggests two basic ways in which roles can be reduced. One is by
manipulating the number of roles and the other is by bargaining
between roles. One method of manipulation is the elimination of
the role through the establishment of techniques to keep others
from continuing the role relationship.
However, eliminating a role and negotiating a new role through
bargaining are quite different solutions to the problem of role
conflict. One is essentially negative while the other at least
includes the possibility of mutually acceptable role re-definitions.
A further examination of each of these potential courses of action
as they relate to divorced fathers will lead to the formulation of
two hypotheses predicting how some post-divorce role conflicts may
The increase in the family role obligations of the divorced man
is a function of remarriage and may include those of husband, step-
father and new father. With each additional obligation, the avail-
able resources for allocation as father to his first children are
reduced. Subject to external pressures to reduce his obligations
to his first children and subject to whatever internal demands to
fulfill those obligations he may feel, reduction or abandonment of
his role or the use of techniques to keep others from continuing the
relationship are possible alternatives.
At the same time his ex-wife and the children may be adding
role obligations as she remarries and furnishes the children with a
stepfather. The reduction of role relationships for them may also
be desirable. Therefore, the impetus toward role reduction or
elimination may come from either ex-spouse. The reduction or
elimination is most likely to occur in the resources allocated to
the child by the divorced father assuming he is the noncustodial
As the number of role obligations for either ex-spouse
increases, the resources allocated by the father to the children
of the previous marriage will decrease.
There is little in the literature that explains how role bar-
gaining takes place and results in the establishment of new roles.
Role bargaining is influenced by both the internal demands of the
actor as well as the relative pressures of external change (Goode,
1960). Pressure can be caused by the finite sum of resources, the
extent to which resources have already been expended and the costs
imposed by others. Others can include the ex-wife, her new husband
and the new wife of the divorced father.
Thornton and Nardi (1975) have suggested a four-stage process
delineating how new roles are acquired. While it deals with norma-
tive roles, a look at the process may help identify those points at
which role negotiations may break down because of the absence of
norms. First is an anticipatory stage based on the concept of
anticipatory socialization. Prior to assuming a new role an actor
learns the expectations of that role from generalized society and
social stereotypes. The second, formal stage occurs when the actor
becomes an incumbent of the new role and is socialized by the mem-
bers of the role set. There is high consensus both within and
between sources of information and the actor's reactions to these
expectations is postponed. He accepts the interpretation of the
role made by others in the role-set and does not attempt to modify
their interpretations by interpretations of his own. In the third
stage, which is termed the informal stage, the actor learns the
informal requirements of the role rather than the obligations of
the normative order. These informal requirements provide for some
variability in performance and to some degree the actor can shape
the role to his individual requirements. In the last stage the
actor can influence the expectations of others and "personalize"
the role to relate his own psychological needs to those of the
role. The role can now be internalized.
In applying the model to the newly divorced father several
exceptions must be noted. First is the absence of an anticipatory
stage. The strains of dissolving the personal relationship
between husband and wife may not leave much room for anticipating
the change in father-child relationships that divorce will bring.
The symbolic authority of the father is gone (Weiss, 1975, pp. 187-
198). A new role has to be negotiated without the anticipatory
socialization process that may come from reference groups (Merton,
1968, pp. 281-301).
The newly separated father, lacking the guidelines customarily
received through anticipatory socialization, must immediately move
into the second stage which is based on expectations of the role
set. Thornton and Nardi (1975) suppose high consensus at this
point based on the expectations of stage one. Since stage one is
nonexistent the father is forced to rely on the previous role
relationship with mother and child for guidance. As Getzels and
Guba (1954) point out, however, conflict in a role depends in part
on the degree of incompatability that exists. If conflict pre-
dominated in the previous role relationships, consensus is less
likely to occur. But, if the performance of the father role was
not a significant issue in the divorce, it can be assumed that the
steps in the model will take their course and the members will be
able to successfully negotiate the necessary changes in paternal
roles between themselves. If disagreement over paternal role per-
formance was so great prior to the divorce that it became a cause
for the divorce, it is unlikely that post-divorce paternal roles
will be redefined without the aid of some third party. Usually
this third party will appear within institutional structures such
as court action, legal or professional advice and friend or kin-
The use of third parties to negotiate post-divorce paternal
roles will be more likely to occur if pre-divorce paternal role
performance was seen as unsatisfactory by the ex-wife.
Social Structure and Deviance
So far role conflict has been discussed in terms of individual
interpretations of expected role behaviors in the absence of norma-
tive order. However, social structure may also fail to provide the
means by which role requirements can be achieved (Turner, 1962;
Merton, 1968; Komarovsky, 1973). Merton points out that social and
cultural structures contain two elements. The first consists of
culturally defined goals and the second of modes or institutional-
ized means of meeting those goals. While both reflect cultural
values, the values represented by the norms which regulate conduct
serve to limit the choices available for attaining cultural goals.
His main hypothesis is that nonconforming behavior may be seen as a
dissociation between cultural goals and institutionalized means.
He goes on to develop a five-stage typology of modes of indi-
vidual adaptation, the last one of which will form the basis for
the third hypothesis. He presupposes that alienation from estab-
lished goals and standards occurs when these goals and standards
have come to be seen as purely arbitrary. The result is that neither
goals or cultural means are seen as legitimate and the response is
a form of rebellion that seeks to modify social structure.
Merton concludes by suggesting that rebellion is most likely
to be engaged in by members of a rising class rather than depressed
classes. He uses as his example the 19th century bourgeoisie of
France (pp. 185-248).
Traditionally legal custody is awarded to the mother and the
child lives with the mother following divorce. Although father
custody and various forms of co-parenting have been the source of
much popular discussion of late, only one-tenth of the children
living with a divorced parent live with the father. While the
absolute number of such cases has increased, the percentage has
remained constant for the past two decades (Current Population
The human potential and liberation movements have emphasized
freedom from arbitrary sex-stereotyped roles in all areas of life.
Frequently included among their targets are the male instrumental
and the female expressive roles. The greatest participation in
these movements has been in the middle classes; therefore, it seems
most likely that they would be the first to rebel against traditional
parental roles and seek alternative ways of responding to them. One
way of doing this is for the father to take legal custody or be the
Middle class divorced parents will be more likely to reject
traditional custody and living arrangements than upper class or
lower class divorced parents.
REVIEW OF THE LITERATURE
The review of the divorce literature will consist of those
studies dealing with post-divorce relationships between ex-spouses
and between father and child. The information presented from each
study will be limited to that which sheds light on the three basic
assumptions and the variables identified in the theoretical frame-
There are three major studies that deal all, or in part, with
post-divorce relationships. The first is Goode's After Divorce pub-
lished in 1956 and re-published as Women in Divorce in 1965. In
addition to looking at both the causal and processual features of
divorce, he extensively examined post-marital activities and adjust-
ments including the continuing relationship between ex-spouses.
This latter issue was not dealt with again, at least not in such
detail, until the publication of Weiss's book Marital Separation in
1975. Weiss's work is devoted to the post-separation period and
includes information on parent-child relationships as well as those
of ex-spouses. These two works, plus the work of Heatherington,
Cox and Cox (1976), represent the major studies in the continuing
relationship between divorced men and their first families. This
latter work is in progress, but a preliminary report was published
in the Family Coordinator in 1976. These works are major in the
sense that they are comprehensive in scope and involve either large
samples or are longitudinal in nature. The Weiss (1975) and
Heatherington et al. (1976) studies involve middle class popula-
These three major studies will be discussed first, followed by
the findings of nine minor studies which also deal with either the
continuing relationships between ex-spouses or continuing father-
child relationships. These latter studies are classified as minor
in the sense that they involve either small convenience samples,
usually weighted with middle class respondents, or only deal with a
Goode's (1965) study involved 425 divorced mothers in metro-
politan Detroit who were between the ages of 20 and 38 at the time
of divorce. The length of time divorced ranged from two months to
26 months. Goode tapped a wide variety of post-divorce behaviors
and situations, a few of which are particularly pertinent to this
study. Interviews were held with the divorced woman only and com-
ments regarding the behavior of the ex-husband are based on the
Goode notes that the divorce process may continue long after
the divorce decree is granted. Ex-spouses continue to exist for
each other symbolically and, if there are children, a channel is
provided through which each can make demands on the other--she for
child support and he for visitation (pp. 287, 313).
The respondents in his study reported a low level of contin-
uous child support with only 35% reporting regular payments and 40%
reporting payments as rare or never. There were two factors mediat-
ing payment continuity. The first was ability to pay with the sig-
nificant issue not amount of earnings but steadiness of employment.
Frequent unemployment resulted in less frequent payments. The second
effect on payment continuity was the relationship with the ex-wife.
Resentment over making payments, the wife originating divorce and
her remarriage all affected payment continuity negatively (pp. 222-
Goode speculates that paternal visiting will decrease in fre-
quency as the father finds visits less pleasurable than expected,
as he takes on new role obligations and as his children move into
settled patterns of behavior for which visits may be disruptive.
Furthermore, each visit carries the burden of another separation
which may become chilling. While no frequency studies were conduc-
ted, respondents were asked to comment on the desired frequency of
visits. Expressed desires for less frequent visits were related to
steadiness of payments, her remarriage, negative post-visit
behavior on the part of the children, feelings of emotional distance
from the father expressed by the children and the ex-wife's desire
to punish her ex-husband (pp. 313-330).
The continuation of affect after divorce was examined based on
an index measuring the intensity of contacts rather than the fre-
quency. Four trends were noticed. The percentage of women reporting
a strong positive affect declined and the percentage of those
reporting indifference arose. Those reporting a friendly relation-
ship also declined. The percentage of those reporting a strong neg-
ative affect increased, but this seemed to be affected by remar-
riage. Those who were not remarried seemed directly antagonistic
toward their ex-spouse either because of what he had done or
because they perceived their situation as worsening. Those who were
remarried seemed not so antagonistic but negative in the sense that
their ex-husband suffered by comparison with the new husband (pp.
An important work in the investigation of post-separation
behavior is Weiss's study based on his association with Parents
Without Partners and information gleaned from discussions with par-
ticipants in Seminars for the Separated established by Weiss and
his colleagues. Definitely qualitative in nature the study is based
on personal reports of both pre- and post-separation experiences.
Weiss (1975, pp. 103-112, 197-198) discusses both earlier and
later phases of the post-separation behavior of spouses toward each
other. At both times a wide variety of behaviors is reported.
These range from continued sexual activity to aggressive hostility,
as well as combined behaviors indicating a marked ambivalence toward
both the other spouse and the separation. In the later phase of
separation, current experiences and new associations tend to cause
old events to subside and lose significance. However, this
long term relationship may also reflect a change from divisiveness
to cooperation or from amicability to hostility. If divisive
issues are settled the former is likely to occur, but if disagree-
ments over support or visitation continue unresolved relationships
may worsen over time.
The four issues that are most likely to engender conflict are
property division, support, custody and visitation. With the excep-
tion of visiting, conflict is almost inevitable since what one
party gets the other does without. Even in vistation the plans
made by one party may be at odds with the convenience of the other.
In discussing the continued relationship with the children Weiss
has formulated a useful set of expected paternal roles. He refers
to such traditional roles as family head, protector and provider.
The first implies such duties as inculcating values, giving advice
and maintaining discipline. The father is also, at least symbolic-
ally, the guarantor of the family's physical safety. Lastly, he
provides for the basic necessities. Because of his absence from
the daily routine his ability to provide advice, to discipline and
to convey by precept and example is limited. He is also unable to
protect his children from predatory adults, bad companions or their
own poor judgment and inexperience. While he can continue support
he becomes an income source with no control over expenditures. His
contribution may be viewed by the spouse as payment for services
rendered rather than a continued responsibility for the well-being
of the household (pp. 179-180, 188-190).
Men refer to the unnaturalness of the relationship with their
children after separation, the energy required to maintain the
relationship and their concerns over failing to meet their obliga-
tions. In the early weeks of separation a sustained effort at fre-
quent visiting may be attempted which later proves impossible to
maintain. At the other extreme, at least the idea of ending con-
tact is likely to occur. The children's ages will be a significant
factor in determining length and frequency of visits. Pre-school
children require supervision if lengthy visits are planned and
adolescents have commitments of their own which may place limita-
tions on both the frequency and the nature of the visits. Distance
may also become a significant factor if either partner moves to
another region of the country. Weiss feels that income level will
enter as a significant determinant in these cases with lower income
fathers unable to see their children at all and middle and upper
income fathers resorting to occasional yearly visits when the
children are small and visits being arranged for part or all of vaca-
tions when they are older (pp. 191-198).
Custody is traditionally granted to the ex-wife and the grant-
ing of custody to the father contrary to wishes of the ex-wife is
not likely to be supported by the courts. In the past the customary
reason for the father requesting custody has been allegations of
infidelity on the part of the ex-wife which she chooses not to con-
test. Weiss feels that this is giving way to voluntary surrenders
of custody on the part of ex-wives which she agrees to for her own
reasons (pp. 108-109, 169-171).
Heatherington et al. (1976) have presented some of their pre-
liminary findings of a longitudinal study of the impact of divorce
on the functioning of 96 families. Part of their goal was to study
the re-organization of the family in the two-year period following
divorce. A comparison was made of divorced and intact families
matched for social class, age, education and length of marriage
of the parents;and age, sex and birth order of the children. All
respondents were white and middle class. Investigative measures
included laboratory observations, interviews and structured diaries
applied at two months, one year and two years after divorce.
Conflict was a significant characteristic of the exchanges
between ex-spouses with couples reporting that 66% of exchanges
involved conflict. The most common causes of conflict were finances
and support, child rearing and intimate relations with others.
Marked ambivalence, continued dependency and continuing sexual rela-
tions were also noted. Conflict and attachment both abate with
time with new intimate relationships (including remarriage) being
the most salient factor.
Contacts between divorced father and child decreased steadily
over time. At two months, divorced fathers were having about as
much face-to-face interaction as were fathers in intact homes. One-
quarter of the divorced fathers were having more face-to-face
interaction than they had before the divorce. By the end of the
two-year period, 19 fathers were seeing their children one or more
times a week, 14 every two weeks, 7 every three weeks, and 8 once a
month or less. The nature of the relationship also changed with
divorced fathers demanding more mature behavior, communicating
better and being more consistent by the end of the research period.
They also became less nurturant and more detached. The divorced
fathers were extremely permissive and indulgent at first but became
increasingly restrictive over the two-year period. They were never
as restrictive as fathers in intact homes, however. Eight fathers
coped with the stress of separation from their children by reducing
contacts and ten reported improved relationships and more enjoyment
in their interchanges with their children.
The remaining studies deal with post-divorce norms, conflicts
in continuing relationships and satisfaction with post-divorce
fathering, visiting arrangements and frequency and father custody.
Goetting (1979) questions whether or not the 20 years that have
elapsed since the Goode study have not served to introduce at least
some degree of normative order into post-divorce relationships. To
test out this hypothesis she designed a questionnaire to tap degrees
of consensus regarding the appropriateness of behavior in specific
situations involving both the continuing relationship between
ex-spouses and shared responsibilities for child care and nurtur-
ance. Overall, she found more consensus than expected in that there
was agreement on 69% of the items in the questionnaire. In terms
of the continuing relationship between ex-spouses, consensus was
lacking on the appropriate amount of social distance to be main-
tained; however, men showed less concern over this issue than did
In the area of child support there was a lack of consensus
over how much the ex-wife should ask for extra financial assistance
and how the father should respond to such requests. There was also
disagreement over how much each parent should attempt to influence
the other in matters of child management.
Goldsmith (1979) examined the continuing relationship between
ex-spouses by interviewing both men and women who had been divorced
for approximately one year. In all cases the mother had legal cus-
tody and lived within two and one-half hours of the ex-spouse by
ground travel. More conflict was found over issues relating to
co-parenting than over issues concerning their relationship as
ex-spouses. A majority of 84% reported continued telephone or per-
sonal contacts regarding child rearing issues. The most frequent
contacts concerned medical and school problems with everyday deci-
sions being the least likely to be shared. About three-fourths char-
acterized these contacts as stressful, conflictual or involving
basic differences in child rearing. Men expressed the greatest
dissatisfaction with these co-parental contacts with 51% reporting
dissatisfaction compared to 19% of the women. Both parents
expressed dissatisfaction with visiting as a means of maintaining
a satisfactory relationship between father and child.
Wide variations were reported in the feelings of each ex-spouse
toward the other with neither positive or negative expressions
predominating. However, 59% report their feelings changing over
time with such reasons given as time, adjustment to a new life,
remarriage of either and personal growth and maturity. Interaction
with the ex-spouse itself was also given as a cause of changed feel-
ings. Interactions other than those involving co-parenting were
restricted to friendly or "kin" contacts such as talking about
family members other than the children.
Several other studies explore the themes of continued relation-
ships marked by conflict and dissatisfaction with one aspect of
the relationship or another. Newsome (1979) examined the expressed
dissatisfactions of both ex-partners. Fifty-six men and 84 women
were interviewed with only a few of them being ex-partners of the
same marriage. The length of time of the divorce was not specified.
As in the Goldsmith study, all lived near their ex-spouses. The
major dependent variable was dissatisfaction with the divorce,but
relationships with the children were one factor examined in terms of
the man's dissatisfaction. Concerning everyday involvement with
his children, 53% of the men expressed some degree of dissatisfac-
tion and 40% expressed dissatisfaction with the amount of influence
they felt they had over their child. The general theme expressed
by the fathers was one of being an intruder in the lives of their
children rather than a father. In another measure of dissatisfac-
tion it was determined that those men who expressed a strong sense
of emotional closeness to their children expressed a higher degree
of satisfaction with the divorce and that this is the single most
important determinant of satisfaction with the divorce.
Dissatisfaction with fathering after divorce was also noted
by Grief (1977) who measured perceptions of role change among a
small group of divorced men. Overall, lack of contact with their
children produced dissatisfaction, personal devaluation and a sense
of role loss for fathers.
Older fathers and fathers in lower social class groups spent
less time with their children and were less satisfied with the
degree of influence they had on their children.
Both the Goldsmith and Newsome studies involved persons who had
been divorced a relatively short length of time. In a recent study
of 500 ever-divorced men and women Albrecht (1980) examined the con-
tinuing contact between ex-spouses who appear to have been divorced
longer. The mean age at divorce was 30.9 years and the mean age at
the time of interview was 45.1 years. The time elapsed indicates
that the mean length of divorce for the sample was 14.2 years. Only
11% of the sample reported continuing contact between ex-spouses
once the divorce was obtained. More than 70% reported little or
no contact. Unfortunately, the published report does not indicate
how these contacts may have changed over time, but it is assumed
that the author is referring to contacts at the time of the study
and not at the time of the divorce.
Finally, a study by Cline and Westerman (1971) affirms the
possibility of conflict after divorce. In an examination of 102
families who appeared in a family court over a two-year period they
found that 52% had hostile interactions requiring intervention. The
two biggest areas of conflict were money and the children.
Two additional studies deal with the ways in which personal
contacts between father and child are arranged. Newsome (1977)
examined the hypotheses that the father's access to his children
will be controlled by the ex-wife and that the greater his contribu-
tion to the first family the more access he will be allowed. He
also predicted that the sentiments of the ex-wife concerning both
the divorce and paternal rights will act as intervening variables
affecting access. Generally, he found these two hypotheses sup-
ported; however, the role played by the ex-wife's sentiments was
not always clear.
Alexander (1980) reported on an investigation of the factors
which influence divorced parents in determining visiting arrange-
ments. She questioned whether visiting arrangements were decided in
the best interests of the child, the custodial parent, the visiting
parent or the continuing parental relationship. She determined
that the best interests of the child and the visiting parent were
given the highest consideration, in that order. These within group
differences were statistically significant. The interests of the
custodial parent and the continuing parental relationship were given
third and fourth consideration but these within group differences
were not statistically significant. She also found that hostility
between the two parents was not a significant factor in determining
the child's best interests.
Two other items of information on visiting emerged from the
Alexander study. The first is that visiting frequency was not
affected by the remarriage of the ex-spouse. Only one respondent
reported a decrease in visiting frequency for this reason. The
second refers to the use of third parties in making visiting
arrangements. She cites a study indicating that only 14% of par-
ents saw lawyers as influential in determining visiting arrange-
These studies all imply that a wide variety of factors may have
an impact on the satisfactions received from visiting in addition to
visiting frequency. It must be noted, however, that visiting may
not occur at all. Anspach (1976) studied kinship relations after
divorce and, while his study did not deal directly with the father-
child relationship, slightly over half of the women studied
reported that the fathers of their children did not visit.
The limited amount of data on father custody deals mostly with
the legal aspects of custody or consists of case studies reporting
the problems men have with single parenting. Only one study could
be found that discusses some of the social characteristics of men
who receive custody as well as some of the structural arrangements
that inhibit father custody. Gersick (1979) interviewed 40 men who
had been divorced between nine and twelve months. Half of the group
had had custody and the other half did not. He reports a tendency
for men in the higher social class categories to have custody com-
pared to those in the lower classes. He also found that those who
had custody tended to be older and more established than those who
did not have custody. His study bears out the supposition of Weiss
concerning the changing reasons for women relinquishing custody.
In 18 of the 20 cases of father custody the ex-wife consented to
the arrangement prior to the divorce hearing. In over half of
these it was the choice of the wife, or of the child or the result
of common consent.
Gersick found that the attitudes of the attorneys were gener-
ally not supportive of father custody and the sample as a whole
expressed dissatisfaction with their attorneys. The author sug-
gests a social class effect in the attorney-client relationship
since the professional and middle class men seemed to be able to
use their attorneys while the working class men felt that their
attorneys used them.
The continuing relationship between ex-spouses and between
father and child are the two common interests upon which this liter-
ature review is based. Consistently, the relationship between
ex-spouses is seen as ambivalent, shifting and ultimately lessen-
ing in most cases. Changes include moving toward better relation-
ships, an increase in indifference or worsening relationships.
Remarriage seems to be the most salient factor in reducing the
intensity of the relationship but there are indications that time
and maturity are also important.
Visiting between father and child seems to be characterized
largely by dissatisfaction for both parents. Goode's original
expectation that visits will decrease over time is supported by both
Weiss and Heatherington et al. Goode's findings on visiting and
Alexander's can be seen as conflicting. Goode reports that
attitudes of the ex-wife toward visiting by the father can be nega-
tively affected by a variety of conditions including both his
behavior and her attitudes toward him. Alexander reports that both
parties attempt to conduct visiting in a way which makes the inter-
ests of the child the paramount consideration.
Three other characteristics of the post-divorce relationship
are noted. One is that conflict is common, especially over matters
of child support and the father-child relationship. Secondly, sup-
port payments are most affected by the father's continued employ-
ment and his relationship with his ex-wife. Third, divorce makes
significant inroads into the father's satisfaction with his fulfill-
ment of the paternal role.
In terms of social structure, Goetting suggests that more norms
exist to guide post-divorce relationships than were previously
thought to exist while Weiss and Gersick note the presence of
strong structural constraints acting to prohibit father custody.
THE RESEARCH METHOD
The procedures for collecting and analyzing the data needed
to test the hypotheses include the selection of a sample, the
development and administration of a questionnaire, the formulation
of operational definitions and the selection of the appropriate
The best way to assure that all the desired variables would
occur in the sample would be to locate a universe consisting of
remarried men who had minor children by their first marriage and
minor stepchildren living in their home. The possibility of identi-
fying such a group seemed limited but contacts were made with two
county school districts (Orange and Volusia Counties) and one of
the large inner-city churches in Orlando. All three groups were
afraid of diverse reactions from their constituents if they
assisted in identifying members for research purposes and the
school authorities also cited federal regulations concerning the
confidentiality of school records.
Parents Without Partners seemed a likely source even though
members leave the organization after remarriage. However, organiza-
tions manned entirely by volunteers often keep minimal records and
the local chapter could not locate their record of former members.
Social service agencies comprise another community institution
that has contact with divorced men; however, the problem of confi-
dentiality again arises. A contact was made with the Epi-Center,
a social service agency in Orlando offering marital separation
seminars to men and women in the greater Orlando area. Dr. Larry
Webb, Executive Director of the Center, agreed to send a letter to
former seminar participants explaining the research project and
asking them to return an enclosed post card if they wished to par-
ticipate in the study. Twenty-one names were drawn from the Epi-
Center files. They were selected on the basis of an address in
Orange or Seminole County, only one previous marriage and the prob-
ability of having had children by that marriage. (The records of
the Center were not always complete.) Since participants in the
seminars are at various stages in the separation-divorce process it
was possible that some of those selected would have reunited and it
was unknown as to whether or not they would have remarried. To
increase the possibility of the latter event no participants were
selected who had entered the seminars after March 1978. Twenty-one
letters were sent out. Seven were returned by the post office as
undeliverable. Nine persons did not respond and five returned cards
indicating an interest. Of these five, one did not have children
and one could not arrange a convenient time for an interview. Three
interviews were completed and these three constitute the first sub-
By now it was obvious that the desired universe could not be
identified. Therefore, the definition was revised to include men,
remarried or not, who had had children by a former marriage which
had ended in divorce. There were three types of county records
which were considered as possible sources and contacts were made in
Orange, Seminole and Volusia Counties to investigate the usefulness
of these records. The use of the first type, divorce records, was
quickly abandoned when investigation indicated that two sets of
records would have to be researched and then the male's address
would be more often absent than not. Also, divorce records would
not indicate a remarriage. Access to the second type, support pay-
ment records, was refused in two counties and this avenue was not
pursued with the third county since this source would eliminate the
voluntary support arrangements which divorced couples might employ
and thus knowingly bias the sample.
The third type of county record which could be employed would
be the county marriage license applications. These records indicate
the number of previous marriages, age of the applicants and the
address at the time of application. They do not indicate if there
were children by the previous marriage nor would the applications
indicate if the marriage actually took place. In addition, there
was always the possibility that the second marriage was no longer
intact. It was felt, however, that the use of the marriage license
applications would be the best source for assuring that at least
some of the sample would have an intact remarriage. The final
decision was to identify male applicants with one previous marriage,
between the ages of 25 and 45, who had filed for a marriage license
no less than one year before it was anticipated that interviews
would begin. One previous marriage was chosen as a limit to
avoid confusing the interview schedule with too many possible sets
of children with differing custodial parents. The age limit was
selected to increase the possibility of respondents having children
who were still minors. The one year following remarriage was sel-
ected to give time for visiting arrangements which might have been
affected by the remarriage to settle into an established pattern.
Accordingly records were examined between the dates of November
1977 and June 1978. This meant that some of the applications were
filed as long as 20 months before letters were sent to potential
respondents, a factor which no doubt accounts for the great attri-
tion which took place.
Using the formula indicated above, 599 names were selected from
the records of Orange and Seminole Counties. Of these, only 187
(31.2%) could be tentatively identified and located by using phone
books and local directories. Letters on University of Central
Florida stationery and signed by the Director of Graduate Studies
and Research were sent out in June 1979, inviting the receivers to
participate in the study (see Appendix A). Of the first wave of
letters sent (122) only 23 were returned undeliverable. Of the 99
remaining, 24 could not be located by phone. Since these 24 let-
ters had obviously been forwarded the letter was modified in subse-
quent mailings to encourage receivers who had moved to contact the
researcher if they wished to participate in the study (see Appendix
B). Sixty-five more letters were sent using the revised letter but
only one person responded to this suggestion.
Following is the breakdown of responses to the 187 letters
which were sent out in four different mailings over a four-week
Breakdown of Responses
Letter returned as undeliverable 23
Letter forwarded, but could not locate 38
No answer to repeated phone calls 15
Mistaken identity 12
Refusal to participate 25
No children by first marriage 26
Remarried first wife 4
Agreed to participate, but appointment
not arranged or kept 7
Total failures to interview 148
Thirty-nine interviews were completed and one was eliminated
since answers were inconsistent and the interviewer reported that
the respondent seemed to be under the influence of drugs. The
second sub-sample thus consists of 38 cases (County Records).
Reliance on a snow ball sample had been rejected in the
planning phase of the study since the researcher found it difficult
to identify potential participants who fit all the criteria, par-
ticularly remarriage. The few that were found were used to pre-
test the interview schedule and some of these are included in the
final sample when appropriate. However, as the interviews pro-
ceeded, more and more potential respondents came to the attention of
the researcher with many being identified by the research assist-
In these cases, respondents were assigned to another inter-
viewer to avoid any element of personal bias. Twenty snow ball
respondents are included in the analysis and these comprise the
third sub-sample (Snow Ball).
When it became apparent that the sample from the marriage
application files would fall far short of the goal of 100 cases it
was decided to try to obtain media attention regarding the research
and invite men to participate in the study. Some time in early
August 1979 the Orlando Sentinel Star published a feature section
entitled "The Male in Society" which included a short article on
this research project and invited participation from divorced males
with first families. The name of the researcher was included with
his office phone number. Over the next several weeks, 48 phone
calls were received and 40 interviews were completed. Of the eight
not interviewed, two were in the first divorce process, three lived
too far away, two could not arrange a convenient time for an inter-
view and one had no children.
The initial calls from this sample were screened by a secretary
who checked for address, phone, marital status and children by the
first marriage. Comments recorded by the secretary indicated that
this sample would probably be biased since many of the respondents
obviously had an "axe to grind" and were calling to report unpleas-
ant experiences of one kind or another. It is anticipated that
this potential bias will add an extra dimension to the research
since this sub-sample can be compared with the others for differences
in significant family experiences. The fourth sub-sample thus
contains 40 cases (Newspaper).
In summary then, the total sample contains 101 cases of men who
had children by a previous marriage. This sample is composed of
four sub-samples identified as the Epi-Center sub-sample (3), the
county records sub-sample (38), the snow ball sub-sample (20) and
the newspaper sub-sample (40).
A trial questionnaire was developed and pre-tested on ten sub-
jects. From this experience a final questionnaire was printed and
then administered by the researcher and nine undergraduate sociology
and psychology majors at the University of Central Florida in
Orlando. A structured questionnaire was used with some open-ended
questions (see Appendix C). All of the interviewers had some
experience in interviewing and were recommended by their faculty
advisors. An additional four hours of instruction were provided by
the researcher along with on-going supervision throughout the data
collection process. The data were collected between June 1, and
August 31, 1979.
In order to test the hypotheses each of the dependent and
independent variables must be operationally defined. To do this
each variable necessary to the testing of the hypotheses will be
identified and the indicator used to measure it specified.
In the first hypothesis the resources allocated to the
children of the first marriage constitute the dependent variable.
These resources will be specified as the frequency of visits, mail
and phone contacts and voluntary contributions of support upon the
request of either the child or the ex-wife. The independent var-
iable is the increasing number of role obligations which are
measured according to three marital statuses. These are separation
and divorce, which give a measure of minimal role obligation, and
the remarriage of either ex-spouse as a measure of role increase.
In the second hypothesis the dependent variable is the use of
a third party to negotiate post-divorce roles. The two roles to be
negotiated are visiting and child support. Indicators of third
party involvement will be a court order issued without the know-
ledge of, or against the wishes of, either ex-spouse; or the reli-
ance upon an attorney or other third party to negotiate visiting or
support. Mutual agreement prior to the divorce proceedings, with
or without a court order, will be considered evidence of the
absence of third party involvement. The independent variable is
the ex-wife's perceptions of unsatisfactory paternal role perform-
ance. This will be measured by the extent to which paternal role
performance was a causal factor in the divorce.
In the third hypothesis the dependent variable is the rejec-
tion of traditional custody and living arrangements. Rejection of
traditional arrangements will be indicated by the father having
custody of one or more of his children or having one or more of
his children maintain their residence with him anytime after the
divorce. The independent variable is social class which will be
measured by Hollingshead's Two Factor Index. This index is based
on education and occupation and had been in use since 1957. It has
a reliability of .906. There are five classes which are identified
by number and defined by occupational position rather than by name
(Hollingshead & Redlich, 1958). For this study Class I will be
defined as upper class, Classes II and III combined as middle
class and Classes IV and V combined as lower class (Kessler &
Cleary, 1980, p. 468).
Other independent variables will be examined for the possible
effects they may have on the dependent variables. These include
the attitudes of the ex-wife toward the father and toward his con-
tinued contact with the children, the attitude of the father's
present wife, geographic distance, religion and sub-sample differ-
ences. Comparisons of religion will be limited to differences
between Catholics and Protestants.
The data were analyzed by using the Statistical Package for
the Social Sciences (Nie, Hull, Jenkins, Steinbrenner & Bent,
1975). Since the data were predominantly nominal, statistical
analysis was confined to the Chi-square test of independence and
either the Phi coefficient or Cramer's V depending on table size.
Tables were frequently reduced by combining frequencies to either
simplify analysis or to compensate for too many small or missing
cells. When cells less than five occur, SPSS automatically applies
the Yates correction for Chi-square (243). In the analysis the
corrected Chi-square will be used unless otherwise indicated. The
level of significance to be obtained in order for hypotheses to be
accepted is .05.
Since the sample is obviously of a convenience nature general-
izations cannot be made to the population of divorced fathers at
large. The use of tests of significance and measures of associa-
tion is not intended to estimate population parameters but rather
to test hypotheses and establish the location of individuals in
relationships to others (Farber, 1957; Coleman, 1961; Zelditch, 1961;
DESCRIPTION OF THE SAMPLE
This descriptive report of the questionnaire findings will
begin with a breakdown of the social and demographic characteristics
of the sample. This will be followed by frequency distributions of
the dependent and independent variables which will be used in the
next chapter to test the three hypotheses for acceptance or rejec-
Since the sample consists of 101 respondents the absolute num-
bers and the percentages are about equal. In such cases, when the
difference between the two figures is less than two, either the
absolute number or the percentage will be given, not both. In all
other instances the percentage will be given followed by the abso-
lute number in parentheses.
Characteristics of the Sample
The men in the sample ranged in age from 23 to 59 years (mean
= 37.4, median = 36.9). Ninety-six were Caucasian, four were black
and one was Spanish-American. Sixty were Protestant, 24 were
Catholic and two were Jewish. Thirteen were either nonaffiliated
or of some other religious group and two did not respond. According
to Hollingshead's Two Factor Index of Social Class, nine respond-
ents fell into Class I, 31 were Class II, 48 were Class III, ten
were Class IV and one was Class V. Two did not respond. There was
no statistically significant relationship between religion and social
class. Protestants predominated at all class levels except in Class
V which contained only one case. Only two men had less than a high
school education, 20 were high school graduates, 41 had some college,
21 were college graduates and 14 had a graduate degree. Three did
not respond. Given the time of the last census and the rapid
growth and change in the Central Florida area, the population esti-
mates would be so unreliable as to make a comparison on these demo-
graphic variables meaningless. However, it seems probable that the
sample is more likely to be better educated, Protestant and Caucasian
than the general population in Central Florida.
The length of the respondent's first marriages ranged from nine
months to 23 years (mean = 10.5 and median = 9.8). The length of
time between the divorce and the interview ranged from two months
to 15.8 years (mean = 4.8 and median = 3.9).
Sevety-six of the men had remarried and 86.8% (66) of these
second marriages were still intact. Of the remainder, two were
separated and 10.5% (8) had divorced their second wives. Twenty-
five had not remarried.
Of the ex-wives of the first marriage, 49 had remarried and 50
had not. Only two men did not know anything about the marital
status of their first wives. Of the 49 who had remarried, 35 were
still remarried, ten divorced and in the other four cases the men
were not sure.
Twenty-four of the respondents had one or more stepchildren
living with them. (This includes one man who had his stepchildren
in his care although their mother had left the family.) Fourteen
men had children born to their second marriage. Half of these
were under one year of age and the oldest was 14 years of age.
Twenty-one men had one or more of their own children living
with them and of these, 19 had custody of one or more of their
In terms of the geographic distance separating the fathers
from their children, 51.7% (46) lived 50 miles or less from their
children, 6.7% between 51 and 150 miles, 5.6% between 151 and 300
miles and 36.0% (32) lived more than 300 miles away. (See Appendix D.)
Visiting arrangements were reached by mutual consent in 62.8%
(59) of the cases. The use of a third party to negotiate visiting
arrangements occurred in 37.2% (35) of the cases.
No change in the original agreement was reported by 68.1% (64)
of the men. Changes by mutual agreement were reported by 11.7% and
9.6% reported that the request for change came at the instigation
of one party with the other party opposing it. Other reasons for
changes were reported by 10.6% (6) of the respondents without the
specific reasons being given.
Visiting is an extremely difficult behavior to define opera-
tionally since visits can vary both in length and frequency with
little relative change in the actual amount of time spent with the
child. For example, if a father has been seeing his children one
weekend a month for approximately 48 hours each visit and he changes
to one month-long visit per year he has decreased the frequency of
his visits from 12 per year to one per year. However, the actual
amount of time spent with his children has increased from 576 hours
(12 months x 48 hours) to 720 hours (30 days x 24 hours). Ignoring
the meaning of this for either father or child, it is difficult to
judge whether this should be seen as an increase or a decrease in
the amount of time allocated to the children of his first family.
In an attempt to cope with this problem respondents were asked to
state not only how often they visited their child but how long visits
lasted. The frequency of visits was scaled as follows: daily,
weekly, bi-weekly, monthly, quarterly, holidays and birthdays,
yearly, less than yearly and never. Length of visits was scaled at
two hours or less, half-day, all day, overnight or weekend and a
week or more. Respondents were asked to indicate the visiting pat-
tern at differing marital statuses. These were during separation,
after the divorce and after the remarriage of either, which indi-
cates a sequential series of events.
Appendix E indicates both the length and frequency of visits
for each marital status with both row and column totals. In order
to simplify the presentation both the frequencies and the lengths
of visits were dichotomized and are presented in Table 1. This
table includes multiple responses. That is, if the father had his
children visit once a month for the weekend and also had them visit
o- S- (n cr t
0o 01 na) 0= -C-
S- C- 0) L. 41 -C
L. r- LL 4-
0) a) C:
(U S- l) S- 0
.J h S --*
-) *r- .CM
*r- > I, C
- (0 *r- *a
_ OC c> 0
eo = r
- 0 00
r- I M
for a week or more once a year, both visiting arrangements were
recorded. Visiting lengths were divided into short visits of a day
or less and longer visits (overnight or more). Frequency of visits
was also divided into less frequent (once per month or less) and
more frequent (more than once per month).
As can be seen the longer visits are increasingly utilized as
the marital statuses proceed through separation, divorce and the
remarriage of either ex-spouse. The longer visit is utilized more,
however, after the remarriage of the ex-wife than it is after the
remarriage of the father. As the visits increase in length they
decrease in frequency. Visiting more than once a month is most com-
mon during the separation period (77.0%) but decreases to 64.9%
after divorce and drops to 47.5% and 43.7% respectively with the
father's remarriage and the remarriage of the ex-spouse.
Longer visits make up 44.8% of all visits during separation,
but increase to 58.5% after divorce and to 66.1% with the father's
remarriage and 78.1% with the remarriage of the ex-wife.
A closer examination of the appendix shows that the quarterly
and the yearly visits are the frequencies showing the greatest
increase with the remarriage of either spouse; the weekly visit
showing the greatest decrease. During separation, 14.1% report
that they never visit. The frequency drops slightly to 12.6% after
divorce and then rises to 22.6% and 30.5% respectively with the
remarriage of the father and the ex-wife (not shown).
Mail and phone contacts were both examined for frequency and
have also been dichotomized for ease of examination (see Table 2).
a) .C --
S- 4) -C
cr s- 0
Mail is obviously not a heavily utilized method for keeping in
touch. At most, only 36 men report using the mail and this was
during the period of divorce. Phone calls of more than once per
month are relied upon heavily during the separation period (83.1%)
and continue to be a major source of contact during the subsequent
marital statuses (after divorce, 68.3%; after the father's remarriage,
57.8%; and after the ex-wife's remarriage, 69.6%).
The men who never use mail or phone contacts remains fairly
stable at all statuses. The never mail group remains between 50%
and 65% and the never phone group between 20.7% and 28.9% (not shown).
The general trend has been for visits to increase in both fre-
quency and length after the divorce when compared to the separation
Following the remarriage of either ex-spouse, visits tend to
increase in length but decrease in frequency although this is more
marked for the remarriage of the ex-wife. In a separate analysis of
frequency only (which will be used later in Chapter VI) the same
pattern appears in that the number of men reporting more frequent
visits decreases and the number of men reporting less frequent
visits increases with changes in marital status. None of this
indicates whether the actual amount of time the father spends with
his children is increasing, decreasing or remaining the same. The
maximum visiting length specified in the questionnaire was "a week
or more." This could include anything from a week to the entire
summer. This visiting length increased from 8.7% during separation
to 25.4% after the remarriage of the ex-wife. Therefore, it is pos-
sible in some cases that with the remarriage of either ex-spouse
more time is spent with the children in terms of clock hours than
was spent with them during the separation or divorce periods.
The underlying logic for using frequency of visits as an indi-
cator of role involvement is strengthened by examining the number of
men who reported never visiting their children. The pattern is the
same although the progression is reversed. The number who never
visit drops slightly after divorce and then increases with the
remarriage of each ex-spouse.
While it seems clear that changing marital statuses results in
fewer direct contacts between father and child, the relative stabil-
ity of mail and phone contacts at each marital status indicates
that a reduction in personal contacts does not imply a reduction
Most of the respondents indicated that they were involved in
making support payments. Arrangements for child support were
derived by mutual consent in only 20.9% (18) of the cases. The use
of a third party to negotiate child support payments occurred in
79.1% (68) of the cases. In the majority of the cases third party
intervention was in the form of a court order contrary to the wishes
of one party or the other. Of these, 56 were against the wishes of
the man and three were against the wishes of the ex-wife.
Changes in the amount of support after the original order or
agreement occurred in 39.5% (34) of the cases. Eleven of these
were by mutual agreement leaving the court order unchanged. Three
were by court order at the man's request and seven were by court
order at the request of the ex-wife. (Thirteen failed to report the
process.) Twelve different reasons for a change in support were
given. The ex-wife requested more money in six of the cases and
the man requested a reduction due to income loss in four of the
cases. Other reasons included the remarriage of the ex-wife, change
in custody, child becoming of age and the ex-wife running away.
Fathers were also asked how many times a year they sent money
or gifts to their children on a voluntary basis. Of the 84 men
responding to this question 16.7% (14) indicated that they never did
and 42.9% (36) stated that they limited voluntary contributions to
birthdays and holidays. Those who were more involved included 17.9%
(15) who sent money or gifts five or less times in addition to hol-
idays and birthdays.
Insurance policies and savings plans were used as an indicator
of financial support for children. Thirty-two men said they kept up
life insurance policies on their children. Fifty-two had made their
children beneficiaries to their life policies, 62 covered their
children with some kind of health policy and 11 had some other kind
of insurance. Savings plans for their children were maintained by
33 of their fathers. (These data are from the total sample and
include fathers who had custody. In addition, some men are
reported more than once.)
Respondents who were making payments were asked if they had
ever stopped for three months or more or if they had ever missed or
been late with payments for a period of time up to three months.
Twenty-three (26.7%) indicated that they had stopped payments or
both stopped and missed payments one or more times. The most fre-
quently given reason was a shortage of funds. This was mentioned ten
times. Unemployment and illness or injury accounted for six occa-
sions of stopped or missed payments. Two times were due to the
ex-wife denying visitation, two for retaliation for other felt
grievances and one because the man stated his ex-wife owed him money.
Asked if the stopped, missed or late payments had ever been
made up, 55.6% (15) indicated yes and 44.4% (12) said no.
Respondents were asked to indicate if they had ever been asked
to send extra funds (beyond the support agreement) for their child
at either the request of the child or the request of the ex-wife.
The same question was asked for each of the four marital statuses
previously described. The choices were that they always sent extra
funds, they sometimes refused, they sometimes could not or they
never sent extra funds. There was a large number of nonresponses to
this inquiry and it is not certain if this was due to the fact that
requests for extra funds were never made or due to confusion over
the wording of the question.
Since the first three categories indicate that the father was at
least disposed to send money some of the time these were collapsed
into one "favorable" response. The "never" response is categorized
By looking at Table 3, it can be seen that a request from the
child receives an increased favorable response after the divorce
when compared with the separation period (from 13.6% to 22.7%).
When the request comes from the ex-wife the favorable response drops
slightly from 35.6% to 32.5% for the same change in status.
With the remarriage of the divorced father the favorable
responses rise to 27.9% at the child's request and to 48.3% at the
After the remarriage of the ex-wife the favorable responses
drop sharply to 20.0% at the child's request and 25.0% at the
Over half the cases (68.5%) reported that child support was
imposed by the court contrary to the wishes of one party or the
other but usually contrary to the wishes of the father. Later
changes in the original support arrangement occurred in 39.5% of the
cases with most of them being by mutual agreement leaving the court
order unchanged. (Note, however, that there was a large number of
nonresponses concerning the change process.)
Of the 84 fathers who reported voluntarily sending money
beyond child support requirements, 59.6% said they did not do so or
only sent money or gifts for birthdays and holidays. The other
40.5% sent money or gifts at least a few times per year in addition
to birthdays and holidays.
O .- 0
Ck r_- Cr
While the data on insurance coverage are overlapping, at least
62% of fathers carry some kind of insurance on their children with
the most common form being health insurance.
A large group of respondents (73.3%) report never missing or
being late with their payments. Of those who stopped or missed pay-
ments, 55.6% said they made up the payments at a later date.
Favorable responses to requests for extra funds from either the
child or the ex-wife vary sharply according to the marital status
of both the father and the ex-wife. The most prominent finding is
that favorable responses to a request from either child or ex-wife
increased after the father remarried and decreased after the remar-
riage of the ex-wife.
Relationships Between Divorced Fathers and Their Ex-Wives
The men were asked to respond to a six-point scale indicating
their relationship with their ex-wife both at the time of the
divorce and at the time of the interview. The six choices on the
scale were "antagonistic," "unfriendly," "neutral," "some friendly,"
"good friends," and "other."
As Table 4 indicates the number of men reporting antagonistic
or unfriendly relationships had dropped by the time of the interview.
Those reporting unfriendly relationships dropped almost two-thirds
between the time of the divorce and the time of the interview. The
group reporting neutral relationships increased--again by about
two-thirds. The group reporting generally positive relationships
showed an increase.
i-- : 0 C r--
N y ---
CM j 4-
n C0 r- I-
- U I-
0 -I-- t.
*r- -0 -4
4-> C C
*- S*- S.
o W r( L... LL
C) *r- -
4-. 4- 3 E 0
C C 0) 0 0
2C Z C
Generally, then, there was a reduction of hostilities with a
marked increase in neutral feelings and some increase in friendly
Respondents were also asked if they perceived their ex-wife as
opposing or supporting their relationship with their children at both
the time of the divorce and at the time of the interview. Very few
were neutral on this issue, and support and opposition were about
equal (see Table 5).
The apparent stability of this relationship disappears, however,
when the reasons given for change are examined on a case by case
basis. Thirty-eight men reported changes over the time period
examined. These changes were in both directions with some reporting
their ex-wives withdrawing support and some reporting an increase in
support. The father's own maturity and the influence of his
children were the two most frequently mentioned reasons for change,
but only four men indicated these reasons. All other reasons were
individual. Some examples were the ex-wife afraid of his influence
on the children, father taking custody, engaging in power plays and
the removal of a restraining order.
In an attempt to determine something of the nature and extent
of post-divorce relationships between ex-spouses, respondents were
asked to indicate if they discussed certain matters with their
ex-wives at these same two points in time. The questions concerned
both contacts they had regarding the children and contacts they may
have had which would indicate a continuing relationship between the
c -> C.n (\j
two adults. The possible subjects they might have discussed concern-
ing the children included visiting, support, health problems, school
or behavior problems and any good news concerning the children.
Possible subjects concerning their relationship included per-
sonal problems of either, household management, problems from the
past, mutual friends or relatives and mutual financial interests other
than child support.
In all of these areas respondents were asked to respond in terms
of agreement, compromise, fighting, he refusing to discuss the mat-
ter, she refusing to discuss the matter, avoiding the topic or no
need to discuss the topic. They were also asked to give the reasons
for any changes that were reported.
In order to reduce the data to a more manageable form the
nature of the discussions has been categorized as positive
(usually agree and compromise), negative (fight, he or she refuses
to discuss and avoid topic) and no discussion. Table 6 gives the
responses to discussions involving the children and Table 7 the
responses to discussion involving the two adults.
In all topic areas concerning the children, positive discus-
sions decrease over time. Negative responses decrease only slightly
for discussions concerning the children's health, school or behavior
problems and good news. The absence of discussions for all topics
increases only slightly except on the topic of support in which "no
need to discuss" increases from 28.2% to 44.0%.
The pattern of responses to discussions concerning the
ex-spouses is quite different. In all topic areas save one, both
D o O O o
SO & O O O O O
ct > CI
D "- (n-
0 0 0 0 0 0
*Im C- C\M C--
o I- 'A
Mc IC in 0-
I L- 0- L
U 0 U
0 0 4 M 0 *
; LU 'A 0' A>
u U CMJ Cj
I- 0 0
o >) m
L' -UU 2 c" e
C- F- '-' >4
o0 *o -U
00 &5S- E
I > Lnn -r- -n-D
LU) CL o
0) 4S 0000
C) V)I C, Z
c0) 0 Q > 0 C_) C) (23
O O C OC
Z C r O O
O CO CO O O
0 0 0 0 0
0 0 0 0 0 0
u in Co o d r
Q '4 r- C> O
3 *r- 0
U > -
L 0 -0 C C
S 4- U,
a) z 00O C0 0 0 0
: *r- m- 0
0 *r- CM CM i- ,-
% o to 0
u 1S- -
= 00 0
oo u E CD E
*,- CD 0 -v- SC 0
3 *i *i 3 -C U 0
C-t) 0 > CA 0 CA 0
LU m 0-
LD0 O LOn ,-
oh Ocl OS OC L0O
th mO L -
C0 r- C
U' .) mo
o U) C) Ln
CMj C i-
CO C- cr
- i- <0
v v v
0 4-) E C U
LS C a) ) ( C )n
. 0a) *r-) O 4-
E .0 S- > C
- r- 0 LL *r- *r a)
to 0 D) S- L + LL. L
C .C 00 Q E a)
0 wC (- 5 .C
AL E to a) ) C
S- >Sz (n *4-.)Q c-
a) 0 EU 3 41)
a- ~ Z" 0
0 000 0 m d-
0 0 0 0 0
r- & t 0 0 0 0 0
o0 o0 0
0 zco fn cn to
oF. C*- uo 0. UV CM
n 0 L C
*1* 0 N ..0 0
0 O M CM *
0) L+n C
+. a 3 Z LA LA sV .
oo 4 ) C m .- CM J-
a) C0 U)
C 0 C\ C%-
0C n 0 0
S -r- O L --
01 1 4 --) a
) 0- i- a) U
00 C -0 CM Ln LA c
i- a 0 L (D M
S_ C U) UJ ) CU) U)
4- C U) i- U) rI + .Q
00 0E .0 S-> CU) E
r- ) 0 LL*- *i 0)
4JU 01 1 0 0 5- 4-) LLU i. Z
U U) C .Cr 0.. I r- Ur 0 *
U) 3 0 U) C 10- S.L4-
S* U) 0 10 )-
tI) 0D 0. 3 Q0 0
positive and negative responses decrease between the time of the
divorce and the time of the interview. Accordingly, the topics
over which no discussion takes place increase, some quite substan-
tially. The one topic in which little change occurs concerns dis-
cussions that take place about mutual friends and relatives. The
response, when positive at the time of the divorce, tends to remain
positive. The decrease is only from 28.5% to 26.4%.
Respondents were also asked to give their perceptions of the
reasons these changes took place. This was posed as an open ques-
tion in the interview schedule and the responses were later cate-
gorized by the researcher. A large number of responses was so
idiosyncratic as to defy categorization; therefore, the "other" cat-
egory is quite large. The remaining primary categories in which
multiple responses occurred were the passage of time, severing con-
tacts and the remarriage of either party. While the probability is
high that the first two are related reasons, they did seem to emerge
as distinct reasons in the minds of the respondents.
Table 8 gives the reasons for change in discussions concerning
the children, and Table 9 reasons for change in discussions between
the two adults.
The basic reason for change in discussions concerning the
children are idiosyncratic reasons (36.4%). The next most important
reason is severed contacts (31.8%). Reasons for changes in discus-
sions between the ex-spouses concerning themselves are about evenly
divided between the passage of time (27.3%), idiosyncratic reasons
REASON GIVEN FOR CHANGE IN DISCUSSIONS BETWEEN
EX-SPOUSES CONCERNING CHILDREN
Men Reporting Change
Reason for Change Number of Responses Percentage of Total
Time 14 8.1
Learned to Agree 7 4.0
Children Emancipated 6 3.5
Children Did Planning 7 4.0
Severed Contact 55 31.8
Remarriage of Either 12 6.9
Financial Ties Ended 0 0.0
Custody Change 9 5.2
Other 63 36.4
TOTALS 173 100.0
REASON GIVEN FOR CHANGE IN DISCUSSIONS BETWEEN
EX-SPOUSES CONCERNING SELVES
Men Reporting Change
Reason for Change Number of Responses Percentage of
Time 53 27.3
Learned to Agree 4 2.1
Children Emancipated 0 0.0
Children Did Planning 0 0.0
Severed Contact 51 26.3
Remarriage of Either 22 11.3
Financial Ties Ended 7 3.6
Custody Change 5 2.6
Other 52 26.8
TOTALS 194 100.0
(26.8%) and severed contacts (26.3%). In both types of contacts it
would appear that time, severed contacts and idiosyncratic reasons
are more important reasons for change than the remarriage of either
party. Remarriage accounts for less than 12% of the reasons for
change in either case.
A general measure of the relationship between ex-spouses indi-
cates a reduction of hostile feelings, an increase in neutral feel-
ings and some increase in friendly feelings between the time of the
divorce and the time of the interview.
The support the ex-wife gives to the continuing relationship
between father and child shows changes in both directions with
approximately one-third of the men reporting that the ex-wife moved
from support to opposition or vice versa.
In a more detailed examination of continuing contacts broken
down by person involved and subject matter the tendency is for con-
tacts between ex-spouses to remain active over matters concerning the
children but to eventually disappear over matters concerning the
two adults. It is significant, however, that contacts concerning
the children which involve socially structured behaviors (visiting
and support) show a slight decrease in negative relationships. More
importantly, in the area of support, there is a marked move toward
nonproblematic behavior in that support moves the most noticeably
toward ceasing to be a topic of discussion. On the other hand,
those topics that are marked by a lack of normative order
(children's health, school and behavior problems and sharing of good
news) show an increase in negative relationships.
Discussions concerning the continuing relationship between
ex-spouses generally show a diminution over time which the respond-
ents largely attribute to factors related to time. These are the
passage of time itself and the severing of contacts over time.
Together these comprise 53.6% of the reasons given.
Another salient factor among the reasons for change is the
large number of idiosyncratic responses regarding the changing nature
of the relationships between ex-spouses whether the subject of dis-
cussion is the children or the adults themselves. In the absence of
normative order individuals resort to personal role negotiations
which may be conducted with a varying degree of success or failure.
Attitudes of the Present Wife
It seems reasonable to assume that the attitudes of the present
wife may influence the contacts the divorced man has with both his
children and his ex-wife. The men who had remarried were asked to
indicate how their present wife reacted to the amount of time and
money they allocated to their first children, and also how she
reacted to the contacts he had with his ex-wife. These attitudes
were examined over time with respondents asked to characterize atti-
tudes at the time of the remarriage and at the time of the interview.
The responses included in the scale were "usually resentful,"
"sometimes resentful," "not sure," "seems to accept," "encourages"
For ease of discussion the responses have been dichotomized
into "resentment" ("usually resents" and "sometimes resentful") and
"acceptance" ("seems to accept" and "encourages"). The "not sure"
and "other" categories were dropped from analysis since the number
of these responses were minimal.
Tables 10 and 11 show the attitudes of the present wife toward
the amount of time and the amount of money the divorced father allo-
cates to his first children. In both cases the reported attitudes
of the present wife are those of marked acceptance with little
change over time. Change that does occur is toward increased
Acceptance of time allocated to the children is reported by
82.5% (47) at remarriage, rising to 88.5% (46) at the time of the
interview. Acceptance of money allocated to the children is reported
at 80.0% (44) at the remarriage, increasing slightly to 81.5% (44)
at the time of the interview.
Attitudes of the present wife toward contacts with the ex-wife
show a marked drop in resentment and an increase in acceptance over
time (see Table 12). Men who report that their present wife indi-
cated resentment had dropped from 56.3% (27) at the time of remar-
riage to 32.6% (15) at the time of the interview. Those reporting
present wife acceptance had increased from 43.7% (21) at remarriage
to 67.4% (31) at the time of the interview.
Respondents were also asked to respond to an open-ended ques-
tion explaining the reasons for changes in attitude. For unknown
C 0 LLJ
reasons only 21 men responded to this question. Once again idio-
syncratic reasons predominated with 47.6% (10) giving a wide variety
of explanations which could not be classified. The other three
reasons which could be classified were the passage of time, 23.8%
(5); the ex-wife being seen as less of a threat by the present wife,
19.0% (4); and the present wife becoming more accepting of his
children, 9.5% (2).
Present wives were also evaluated by the respondents as to the
degree of interest they showed in their stepchildren. Described as
interested and involved with the stepchildren were 74.6% (44); inter-
ested but not involved, 5.1% (3); neutral, 8.5% (5); avoiding
involvement, 5.1% (3); disliking the children, 1.7% (1); and other
responses, 5.1% (3).
It seems obvious that the vast majority of present wives are
very supportive from the beginning of the resources allotted by
divorced men to their first children. In addition, a similar number
take an active interest in the lives of their stepchildren. There
is also a marked degree of tolerance for contacts the divorced man
maintains with the ex-wife which develops over time. However, this
finding must be evaluated in the light of the findings in the pre-
vious section that contacts between ex-spouses diminish substan-
tially over time. It is much easier to be tolerant or even support-
ing of occasional contacts than it is of on-going contacts.
Satisfaction with Post-Divorce Fathering
From Weiss's discussion of the father's position in the home
reported in the literature review, five paternal roles were con-
ceptualized. These are the inculcation of values, provision of
financial support, advice giving, discipline and protection (pp. 187-
198). Respondents were asked to indicate whether they were very
satisfied, usually satisfied, usually dissatisfied or very dissat-
isfied with their own performance in each of these roles.
Table 13 gives the breakdown of the responses. The combined
satisfaction scores indicate that 70.1% (68) were satisfied with the
support role, 57.5% (56) with the values role, 56.2% (54) with the
advice role, 58.4% (56) with the disciplinarian role and 51.6% (48)
with the protective role. When examining the degrees of satisfac-
tion or dissatisfaction, however, it can be seen that, with the
exception of the support role, those who were satisfied were more
often only usually satisfied but those who were dissatisfied were
more likely to be very dissatisfied.
In a related question, respondents were also asked how they felt
about the part they had been able to play in their children's lives
since the divorce. Sixteen percent (15) saw themselves as very
satisfied and 22.3% (21) as reasonably satisfied. Another 23.3%
(22) felt that there would have to be some changes made before they
would be satisfied. Of the remainder, 10.6% (10) were mostly dis-
satisfied, 23.8% (24) were very dissatisfied and 3.2% (3) gave
c z -- 0 In n
0 Cj C C
) CI C In O 'n C)
Lj 0 C On "- N C"
U S- CJ CMi C\ C 0
S N o
0 z CO CO o
LL n i- N On
C/) r L 0
*r- -- o r-- o o
L0 0 *
S M o-
SC --r 0
.. CO CO O
0 *- C ')-' C
HL U S- >LD O
co La a UC
:LLJ r- h O
z ( m :Z- CM C-
CO O n CO C
V=) a)) 0 In to C)
CYJ C m- 0C
o eM r Cn
O Nn 4- m
U.. 0 U) (n --
L- r- 1- U c o -
-4-) 4- 4-) 4-m L 0 U)
z 0 >o > Q
*_} *-) r- <
4- )0 04 -) >-
i *r S- r- -
o u) U, (U U)
S r* < >
^- S- 3 3 S
0 0 t1 < <
Ol: > ) =3
In other efforts to tap the effects of divorce on the father-
child relationship, fathers were asked about the emotional strain
of contacting their children when they knew they would have to say
"good-bye" when it was over. The question was whether or not they
had ever avoided contacting their children for this reason.
The greatest response to this question was "no" with 32.6%
(29) indicating they never had a problem with this. Another 25.8%
(23) indicated that it used to be a problem and they had become
accustomed to it. Still another 25.8% (23) indicated that although
it was difficult for them they still saw their children as much as
possible. Eight men indicated that they did avoid, or sometimes
would avoid, seeing their children for this reason.
Respondents were also asked if they had ever considered sur-
rendering their children for adoption either for the sake of the
child or for their own sake. A large majority of 85.1% (80) indi-
cated that they had never considered adoption for the sake of either
party. Twelve men said that they had given the matter some degree
of consideration for either themselves or the children. One of
these stated that he was currently giving the matter serious con-
sideration for his own sake. Three respondents had surrendered a
child for adoption by the stepfather.
In evaluating the above responses it should be noted that some
of the men who had one or more of their children living with them
also responded to those questions and it is likely that there is
some bias in the direction of satisfactions with fathering. Even
ignoring this possible bias it appears that divorce makes substan-
tial inroads into the satisfactions men derive from fathering after
divorce. At least 40.0% of the respondents expressed some degree
of dissatisfaction with four of the five paternal roles. The one
role in which more satisfaction was expressed was the support role
for which normative behaviors are delineated by law. In the other
measure of satisfaction with fathering, 57.7% expressed some degree
of dissatisfaction. In spite of these dissatisfactions the vast
majority of respondents claimed they never had considered abandon-
ing the paternal role by surrender for adoption.
Pre-Divorce Paternal Role Performance as a
Cause of Divorce
Respondents were asked to indicate to what extent they saw
disagreement over child rearing practices as a cause of their
divorce. They were to indicate if this was a major issue, a minor
issue or no issue for each of the five paternal roles discussed
above. Table 14 gives the findings. Disagreement over child rear-
ing practices does not seem to be a causal factor in 75.6% of the
cases (the mean of the "no issue" responses for the five roles).
Natural Father Versus Stepfather Responsibility
Since it is a central assumption of this study that post-divorce
roles lack institutionalization respondents were asked whom they felt
should have the most responsibility for fulfilling the five paternal
roles if the ex-wife remarries. This question was asked of all men
C0 LO r-
r- 1 0
o 0o -"
m- r- 00
m0 c,) OD
whether their ex-wife had remarried or not. This means that some
men were anticipating what their attitude would be and for the
remainder, the choice was based on experience. When this factor was
controlled few differences appeared between the two groups and these
were not statistically significant.
The choices were that the stepfather had exclusive or most
responsibility, the natural father had exclusive or most responsi-
bility or that the responsibility should be equally shared. The
results are given in Table 15.
Ignoring for the moment the equal responsibility responses, the
support role is seen as more of the natural father's responsibility
with a combined 46.7% (42) respondents selecting him as compared
with a combined 35.5% (32) selecting the stepfather. In the other
four roles there is a slight trend toward more stepfather responsi-
bility, particularly in the discipline and protection roles which
require that the parent be present in order to fulfill the roles.
However, the development of such a wide distribution of responses
indicates a lack of sanctions for either parental position and sug-
gests instead a high degree of role ambiguity.
In the previous chapter the possibility of sample bias was
noted due to the manner in which the fourth sub-sample was obtained.
This group consisted of men who had responded to a newspaper
article explaining the research and requesting volunteers to be
interviewed. It was noted by the telephone receptionists and the
L) t.0 O N
CM Cn co
r- CM %-
*r- .) U"
4.-) 4.) 0
Cra 0 .4- 0
4- U 4- r(A
Q. X m 0 to U)
L) U ( 3 L i
+- 4->) 0
0V) ) U J
U- U--- ,-
interviewers that this group expressed a marked amount of resentment
concerning their divorce experience. This impression was confirmed
when analyzing the additional comments that respondents were invited
to make at the end of the interview. Almost three times as many
of this group accused the courts of bias against the male in divorce
proceedings as the other three sub-samples. They also mentioned
more frequently the negative effects of divorce on their ability to
assume parental responsibility. On the other hand, they did not
mention as frequently the negative effects of divorce on their per-
sonal well-being as did the county records sub-sample.
In examining the data for differences in these sub-samples
which might bias the findings, the first one (Epi-Center) was
dropped from the analysis since it only contained three cases. The
analysis is therefore of the second sub-sample (county records), the
third sub-sample (snow ball) and the fourth sub-sample (newspaper).
There were no statistically significant differences in the
three sub-samples regarding social class, religion or education.
In fact, the distribution of the three sub-samples on social class
was almost identical (p > .99). Race was not analyzed since 95% of
the total sample were Caucasian.
Sub-sample differences did appear in terms of two of the
dependent variables. As might be expected, fewer of the newspaper
and snow ball sub-samples had custody of one or more of their
children or had one or more of their children living with them.
Custody was held by 38.9% (7) of the newspaper group and 5.6% (1) of
the snow ball group compared with 55.6% (10) of the county records
groups. Having a child live with them involved only 35.0% (7) of
the newspaper sub-sample and 10.0% (2) of the snow ball sub-sample
as compared with 55.0% (11) of the county records group. Not only
was none of the cases large in number but none of the differences
was statistically significant. Since custody or living arrangements
are likely to be the cause of differences in attitudes toward the
court rather than an effect, these differences will not be con-
sidered further as an independent variable.
The newspaper sub-sample did show a statistically significant
effect on one of the categories of the dependent variable "frequency
of visits." The impact of this on the findings concerning the first
hypothesis will be discussed at the appropriate time.
All of the independent variables which will be used in the
analysis of the data were examined for sub-sample differences. None
demonstrated any statistically significant differences. In spite
of the complaints of the newspaper respondents about court biases,
differences between them and the other two samples on support and
visiting arrangements did not exceed 11 percentage points.
Summary Description of the Dependent
and Independent Variables
The following summarizes the significant findings of each of
the dependent and independent variables presented above.
1. Arrangements for visiting were more likely to be made by
mutual consent than through third party intervention. The frequency
of personal contacts between divorced fathers and their children are
associated with changes in marital status. Visiting frequency
reduces following the remarriage of either ex-spouse with the
remarriage of the ex-wife having a more noticeable effect than the
remarriage of the father. Arrangements for visiting were more likely
to be made by mutual consent. Changes in the frequency of mail and
phone contacts were less marked when examined by marital status.
They tend to increase after the remarriage of the ex-wife when com-
pared with their frequency at the time of the divorce.
2. Arrangements for child support were more likely to be made
through third parties than by mutual consent. The great majority of
respondents reported that they never missed a payment and more than
half maintained some type of insurance on their children. Requests
for extra funds were affected by the marital status of either
ex-spouse with the remarriage of the father resulting in an increase
in favorable response and the remarriage of the ex-wife resulting
in a decrease.
3. Relationships with the ex-wife generally moved toward a
reduction in hostile feelings and an increase in neutral or friendly
feelings. Attitudes of the ex-wife toward the continuing relation-
ship between fathers and children showed shifts both from support
to opposition and vice-versa between the time of the divorce and the
time of the interview.
4. Contacts with the ex-wife tended to remain active over
matters concerning the children but to eventually disappear over
matters concerning the adults. Contacts concerning the children
were less problematic if they involved institutionalized roles.
5. Attitudes of the present wife toward resources allocated to
the children of the first marriage and toward the ex-wife were
remarkably supportive. Any change was in the direction of improve-
6. Displeasure was expressed over the satisfaction to be found
in fathering after divorce. However, this did not seem to result
in reducing or abandoning the father role to any significant degree.
7. Dissatisfaction with child rearing practices does not seem
to be a significant factor as a cause of divorce.
DATA ANALYSIS AND TEST OF THE HYPOTHESES
In this chapter the data relative to the three hypotheses will
be analyzed to determine whether the hypotheses will be accepted or
rejected. Each hypothesis will be restated, the pertinent data
reviewed and a finding of acceptance or rejection determined. In
each case this will be followed by an examination of other inde-
pendent variables which might be expected to have some affect on
the results which were achieved.
Test of the First Hypothesis
The first hypothesis proposed that both personal contacts with
the children and financial support will decrease with changes in
the marital status of either ex-spouse. By referring bac' to the
bottom half of Table 1 on page 44 it can be seen that there is an
increase in less frequent visiting and a decrease in more frequent
visiting with the assumption of a new marital status by either
ex-spouse. These differences are significant at the p < .05 level
(X2 = 14.31, d.f. = 3). The association is weak, however (Cramer's
V = .23). Mail contacts are used by only a third of the sample and
remain relatively stable. Phone contacts also remain relatively
stable but show some change opposite to the predicted direction by
increasing after the remarriage of the ex-wife (see Table 2 on page
Of the several measures of financial support included in the
questionnaire only one measured support at the four marital statuses
employed in the study. This measured voluntary giving at the
request of either the child or the ex-wife. It was used since it
would provide a test of involvement free of legal coercion. By
re-examining Table 3 on page 52 it can be seen that the remarriage
of the father actually increases his favorable response to a request
for extra funds whether it comes from the child or the ex-wife.
After the remarriage of the ex-wife the favorable responses decrease
at the request of either the child or the ex-wife. However, none of
these differences reaches statistical significance.
The null hypothesis of no difference is rejected when using
frequency of visits as an indicator of personal contacts since visit-
ing does reduce at a statistically significant level with the remar-
riage of either ex-spouse. The null hypothesis is accepted, how-
ever, when using mail or phone contacts as a measure of personal
contact. The null hypothesis is also accepted when measuring vol-
untary support. The response is in the predicted direction for
visiting but shows no direction, or is in the opposite direction,
for mail and phone contacts and for support.
Examination of Additional Variables
In order to construct a model of visiting frequency that could
be entered into the computer and used to examine visiting frequency
by other variables, it was necessary to eliminate visiting length
from the data presented in the findings in Chapter V. This required
summarizing frequency of visits so that just the visiting mode
reflecting the greatest number of visits was used. In other words,
if a father had his children visit every month for a day and quar-
terly for a week, only the monthly visit was counted since it repre-
sented the greatest frequency. Therefore, the total n reported in
the following discussion will be less than that of the table in
Chapter V. The maximum reduction in n for any one marital status is
21. Since the trend of reducing visits for the remarriage statuses
is maintained any distortion will not substantially affect the valid-
ity of the findings.
Both the weakness of the association between visiting frequency
and common sense indicate that other factors may be operating in
the reduction of the number of visits between father and child.
Eleven independent variables were examined for their effect on
visiting frequency. Nine of these proved to be statistically not
significant although four showed directions of interest. The two
that achieved statistical significance were geographic distance and
the sub-sample differences.
The nine that failed to achieve significance were religion,
social class, continuity of support, degree of emotional difficulty
experienced by the father when visiting, the attitudes of the cur-
rent wife toward the ex-wife and toward the amount of time and money
allocated to the children, the relationship between the father and
the ex-wife and the extent to which the ex-wife supported or opposed
visiting. The four of these that showed directions of interest were
religion, social class, continuity of support and the ex-wife's
attitudes toward visiting. In the discussion which follows these
four variables will be treated first followed by the evidence on
the two significant variables.
Religion, Class, Support and Attitude Variables
Catholics visited more frequently than Protestants at all mari-
tal statuses. Differences were small at the separation and divorce
statuses but at the father's remarriage 76.9% (10) of the Catholics
visited more frequently compared to 51.6% (16) of the Protestants.
After the remarriage of the ex-wife 80.0% (4) of the Catholics
visited more frequently compared to 40.9% (9) of the Protestants.
After the remarriage of the ex-wife, Class I and II respondents
visited more frequently than Class III and IV respondents. Although
the total number of respondents in these four categories was only
28, the trend is unmistakable. In Classes I and II, 75.0% (3) and
70.0% (7) of the fathers visited more frequently compared with 30.0%
(3) of Class III and 25.05% (1) of Class IV.
While continuity of support did not have much affect on visit-
ing frequency at the separation and divorce status, after the remar-
riage of the father 62.9% (22) of those who never stopped payment
visited more frequently compared with 50.0% (7) of the fathers who
had stopped payment. After the remarriage of the ex-wife the con-
trast is even greater. Of those who had never stopped payment 52.4%
(11) visited more frequently compared with 37.5% (3) of the fathers
who had stopped payment.
The relationship between the father and the ex-wife also showed
no effect on visiting frequency at the time of the divorce. How-
ever, in terms of supporting the father's relationship with his
children, 84.2% (32) of fathers whose ex-wife supported the rela-
tionship visited more frequently compared with 65.7% (23) of those
whose ex-wife opposed the relationship. While this finding did not
attain statistical significance it approached significance at the .07
level. Unfortunately, it is not possible to compare this relation-
ship at the two remarried statuses since the questionnaire design
did not yield the necessary data.
Sub-sample and Geographic Variables
Sub-sample differences in visiting frequency did appear with
the newspaper group engaging in more frequent visits at all marital
statuses. However, these were only statistically significant for
the divorced status. The snow ball sub-sample, which only contained
ten cases, increased visiting after divorce and after the father's
remarriage. This group did not visit at all after the remarriage
of the ex-wife. However, both the newspaper and the county records
groups followed the expected trend. The newspaper group dropped in
frequency from 90.0% (29) at separation to 73.3% (14) at the remar-
riage of the father and to 62.5% (10) at the remarriage of the
ex-wife. The county records group went from 82.6% (19) to 44.0%
(11) and 33.3% (4) for the same statuses. Table 16 shows the dif-
ferences in the sub-samples for the divorced status only.
SUB-SAMPLE DIFFERENCES IN VISITING FREQUENCY AT DIVORCE
County Records Snow Ball Newspaper
of Visits % N* % N % N
Less Frequent (40.7) 11 (29.4) 5 (12.1) 4
More Frequent (59.3) 16 (70.6) 12 (87.9) 29
TOTALS (100.0) 27 (100.0) 17 (100.0) 33
*Number of Fathers
X2 = 6.46 with d.f. = 2, p < .05
Cramer's V = 2.9
Geographic distance was also examined as an independent var-
iable certain to have an effect on visiting frequency. Of those
fathers who did not have custody of their children 51.7% (46) lived
less than 50 miles away. Of the remainder 6.7% (6) lived between 51
and 150 miles, 5.6% (5) between 151 and 300 miles and 36.0% (32)
over 300 miles. Since the two intermediate distances contained only
11 cases combined, distance was dichotomized as "less than 150
miles" and "more than 151 miles."
The data in Table 17 indicate that when controlling for dis-
tance the trend toward visits in the more frequent category decrease
and visits in the less frequent category increase after the remar-
riage of either spouse. However, the change for those fathers liv-
ing further away is much greater than for those living closer. For
those living the greater distance more frequent visiting decreases
N LO CO
- r r-
- m I
CL 0 S-
U 4J 4-)
C C (0
0 Q L-
to $- 0
L o U
I M CO I 0
S) -3 CD)
CO C)C- 0
CD (u Z S- z 0
( E 4-
+P 0 0 0
C 0 C LL.
.-- -C --
-) I- *r-
O a) e
u L ) Z CO -
0 a r- CN
S- S- -
0 r-- C O NC
LU > z;-"
-I *r- C CC O
0CM CM Z 4-)
+ 1 o vi
a) E a)
0- n m o S.- ( O
40. 0 L OC-
a, LL- a)
U I L u w V -0
a i c <( O l<- C V U E
uZ i L L- O O LLQ
S LL L C -
C 4- 0 0
L.- 0M -i
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EPH932I03_A3FPRW INGEST_TIME 2012-03-08T18:32:20Z PACKAGE AA00003884_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC