Divorce

Crisis, Challenge, Or Relief?

David Chiriboga, William A. Galston

• Publisher: NYU Press
• Year: 1991
• ISBN:
  9780814772164
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

Not since William Goode's Women in Divorce in the 1950's have we had such a comprehensive study of adjustment to divorce. This longitudinal work views divorce as a transition process which may have positive or negative outcomes over time. In addition to statistical analysis, the book includes very interesting case studies to demonstrate the dynamic events occurring as individuals refashion their lives after the breakup of their marriages. Researchers on divorce and the interested public will find this book very valuable for years to come."
—Colleen L. Johnson, Ph.D.Professor
Medical Anthropology, University of California, San Francisco

We are witnessing a steady increase in the overall number of older adults who are divorced, yet the majority of divorce research has concerned itself with persons in the younger adult years. This unique, groundbreaking book addresses the critical need for information on the impact of divorce on individuals in all age groups, and pays special attention to age as a factor in the effects of divorce on both men and women.

Written by an interdisciplinary team of social and behavioral scientists, Divorce: Crisis, Challenge or Relief? provides the invaluable results gained from their life span study of divorced adults. Divorce is the product of hundreds of interviews containing a host of very specific questions conducted with divorced adults between the ages of 20 and 79, both just after their divorce and again several years later.

1
INTRODUCTION

David A. Chiriboga

The social phenomenon we call divorce has gone from the unusual and suspect to the common and socially accepted in the United States in a space of less than thirty years. After a temporary surge in divorces immediately following World War II, when the soldiers returned home, there was a period of stability until the mid-1950s. From 1957 to 1977, the incidence of divorce in the United States among persons 44 years or younger more than doubled, while marriage rates began a slow decline (Glick 1984). For some social scientists the converging rates of marriage and divorce heralded the impending demise of the American family. Doom and gloom prophesying was enhanced by the fact that divorce was seen as being most prevalent among those potentially least protected from its impact: the young, minorities and the impoverished.

More recent U.S. figures reveal inconsistencies in demographic patterns. Marriage rates began to climb once again by the late 1970s but then resumed a gradual decline. The corresponding figures for divorce have vacillated back and forth. Most recently, from 1983 through 1990, the trend has been in the downward direction. During this period, the divorce rate per thousand population dropped from approximately 5.1 to 4.7 (National Center for Health Statistics 1990). Despite these fluctuations, the incidence of divorce remains high. Estimates are that 1,163,000 couples, or well over two million persons, divorced during 1989 (National Center for Health Statistics 1990). If one takes into consideration the children and the parents of divorcing persons, it is reasonable to assume that over six million persons were directly affected by divorce during that single year.

Today the future of marriage and the family looks modestly optimistic. Using 1985 data from the U.S. Bureau of the Census, Arthur Norton and Jeanne Moorman (1987) report evidence that lifelong probabilities of divorce will reach their peak with the cohort currently in their thirties, and may have already begun to wane. Looking just at the figures for women in the last five years of each decade (e.g., 25–29, 35–39), Norton and Moorman report that of those in their thirties, approximately 56% will divorce at some point in their lives. In contrast, women in their twenties have a slightly—but encouragingly—lower 54% probability of divorce. Rates are also lower for those in middle and later ages: it is 36% for women in their forties and 24% for women in their fifties.

Regardless of whether there is reason to be more or less optimistic about the fate of the American family, the fact remains that many millions of lives are touched each year by divorce. Current estimates are that one out of every five families with children under age 18 is headed by a single parent (Taeuber and Valdisera 1986). Moreover, the fluctuations in the rate of marriage and divorce inform us that the social fabric of adult life is gradually changing, and that a variety of patterns may be emerging within American society. For example, notwithstanding the peaking of divorce rates, we are witnessing a steady increase in the overall number of adults who are divorced. In the long run, the prevalence of divorced persons may have social fallout such as greater numbers of older adults without social supports, or greater numbers of people who find it harder to play a care-giving role to both younger and older generations. There is, for example, evidence that divorce may interfere with the support provided by adult children to dependent parents (Cicirelli 1983).

ORIGINS OF OUR STUDY

A Social Condition of Need

In the late 1970s the social phenomenon of divorce was relatively understudied, although the research of Waller (1930), Goode (1956) and Hetherington, Cox and Cox (1976) had emphasized the significance of the area and the more clinically oriented work of Krantzler (1973) and Weiss (1975) had demonstrated the need for clinical intervention. It was rapidly becoming clear that divorce affected the lives of many people. In 1976 over three million persons, or roughly 1.5% of the population, were directly involved in dissolution of a marriage (Bloom, Asher and White 1978); 14% of the adult population had at one time been divorced (Norton and Glick 1979). Divorce emerged from these statistical descriptions as an increasingly accepted alternative to an unhappy marriage. The total number of Americans contending with the effects of marital dissolution, including spouses, children and those relatives and friends indirectly affected (Hetherington, Cox and Cox 1976), was projected to run into the millions.

THE LONGITUDINAL STUDY OF DIVORCE

In 1977, a multidisciplinary team of researchers began the first wave of interviews for what was to become known as the Longitudinal Study of Divorce. Data were collected on 333 persons who had recently filed for divorce in San Francisco and Alameda counties, California. The main thesis outlined by our study was that during the period immediately following separation, an individual could undergo either or both psychological growth and psychological dysfunction. Of particular interest were the personal and social characteristics associated with adjustment and maladjustment to separation.

One very important decision we had to make at the very beginning was at what points in the divorce process respondents should be interviewed. Prior research (Chiriboga 1972; Lazarus 1966) had suggested that psychological disruption during the months immediately following a stressful event such as marital separation, while important, might not correspond directly to how individuals function in the long run. Therefore, the study design called for interviews to be conducted during the period immediately after separation (up to eight months) and in the long run (defined for convenience as between three and four years following separation).

By 1980 the research team was ready to begin a follow-up of the original participants. In this follow-up, conducted approximately three and a half years after the initial interviews, two lines of inquiry were emphasized: (1) identification of persons standing at long-term risk and (2) clarification of the stressors, mediators and responses to the divorce process. We were interested in determining where people ended up after they had gone through an experience like divorce.

Sample Selection Procedures

We drew a random sample of potential respondents from records of persons who had filed petitions for divorce at the county clerk offices in San Francisco and Alameda counties, California. An individual’s eligibility for participation in the study was based on the following criteria:

1. Separated from spouse and engaged in the legal process of divorce.

2. Married for more than one year and over the age of 20.

3. Separated for a period no greater than eight months.

4. Spouse not a participant. Only one person of a divorcing couple could participate in the study; when both spouses were randomly drawn, one was selected by random assignment.

Although these inclusion criteria were followed in subject screening, due to inaccuracies in court records, 23 persons with separations longer than eight months were contacted and interviewed; these subjects were eliminated from analyses focused on the initial stages of divorce.

Of those potentially available, 18% were unreachable by mail or phone, 18% either refused outright or were unable to participate when requested, and 64% agreed to participate. In actual numbers, this amounted to 333 men and women.

Follow-up. Tracking people over three and a half years can be hard, especially when they are in the throes of a divorce. In spite of this, we were able to find and interview approximately 85% of the original sample at follow-up. Thus, 283 of the original 333 individuals comprised the sample at follow-up. Of the remaining 15%, 9% refused and 6% could not be located.

Description of the Sample

The 199 women and 134 men we interviewed ranged in age from 20 to 79, with 75% being under 40 years of age. It was the first marriage for three-quarters of the sample; about half had been married between 5 and 19 years. Approximately 40% had no children and 40% had one or two children. The average length of separation was six months at the time of contact.

The sample was reasonably well-educated, with over half having attended at least one year of college. Examination of sex differences showed that a greater percentage of the men were highly educated. More men reported incomes of $10,000 and over compared with the females in the sample. These differences are similar to what one would expect in the U.S. population as a whole.

Follow-up. Three and a half years after the initial interviews, a 66% majority (188 people) had divorced but not remarried. Twenty-one percent were remarried and 4% (10 people) had reconciled. Some differences were found when this distribution was broken down into different age-groups. For example, a greater percent of men aged 50 and over had remarried compared to women of the same age: 20% versus 15%. In general, however, sex differences in marital status at follow-up were not marked.

Interviewing Procedure

Baseline data collection occurred between April 1, 1976 and May 31, 1977. Addresses and phone numbers of the respondents were obtained from records at the county clerk offices. Respondents were sent a letter about the Divorce Study, informing them that a staff member would contact them within a week. If the interviewer was unable to reach the respondent by telephone, a second letter was sent requesting that they fill out a card and return it to the Divorce staff. The interviews were held at a location chosen by the respondent; the majority were held at the respondent’s home, place of work, or an office at the University of California. All interviewers were trained by the staff and evaluated prior to starting their job. The interview contact proceeded in two stages. During the first stage, the intent of the interview, and the areas covered, were explained. The respondents were then asked to read the Experimental Subject’s Bill of Rights and Consent Form which they had to sign before the interview began. The second stage was the actual interviewing: each interview took approximately three hours and consisted of structured and unstructured questions. A few of the interviews were done by mail when there were no other options.

Follow-up. The follow-up procedure was very similar to that at baseline. The interview took approximately three hours and also consisted of structured and unstructured questions, many of which duplicated those from the first interview. Procedural standards were the same in every other respect as those at baseline.

Instruments

The items for the baseline interview were developed through consultations with the staff of the California Divorce Law Research Project (Dr. Weitzman), the Divorce Project (Drs. Wallerstein and Kelly), the Life Events and Adaptation in Adulthood Project (Drs. Pearlin and Lieberman), and the Divorce, Role, Health Status and Service Systems Project (Drs. Sussman and Kitson). Reviews of existing studies on divorce and stress also influenced the selection of items. Decisions on the content of the interview were based on a desire to enhance the potential for generalizing from the findings as well as to facilitate collaboration with other ongoing projects in the Human Development and Aging Program at the University of California, San Francisco.

The main topics covered in the interview were: demographic, life before separation, the divorce process, relationships, physical health, mental health, goals, activities, time perspectives, stress and coping, and emotional well-being. The instruments used in the interview included several that also were used in the companion study of normative transitions (Lowenthal et al. 1975; Fiske and Chiriboga 1990): the Adjective Rating Scale (adapted from Block 1961), the Goal Sort, the Life Events Questionnaire (developed in collaboration with Drs. Mardi Horowitz and Richard Rahe), the California Symptoms Checklist (developed in collaboration with Drs. Robert Butler and Alexander Simon), the Life Evaluation Chart (adapted from an instrument originally developed by Jean MacFarlane), the Leisure Query, Activities Checklist and Bradburn (1969) Morale Scales.

In addition to the instruments that overlapped with those in the companion study, we concluded a number of instruments drawn from other studies of divorce or similar life crises, such as the Social Supports Schedule (adapted from the work of Pearlin and Lieberman), and Kitson and Sussman’s modification of Goode’s (1956) Trauma Index.

Follow-up Interview. The follow-up interview was very similar to the first interview. Many of the additional questions and instruments were drawn from other studies in order to facilitate cross-study comparisons in the future. Materials were drawn from the following projects: the Cleveland Divorce Study (Case Western Reserve: Kitson and Sussman), the Separation and Divorce Study (Pennsylvania State University: Spanier), Stress and Coping Study (University of California, Berkeley: Lazarus), Divorced Family Systems (Northwestern University: Goldsmith and Ahrons), Divorced Mothers Project (Donahue and Colletta), and the Family Mediation Research Project (University of Georgia: Weber). In addition to specific questions, the following structured instruments were added to the follow-up interview: the Ways of Coping Inventory (Lazarus and Folkman 1984), Hassles Index (based on the work of Lazarus), and Self-Other Semantic Differential (based on the work of Osgood).

A COMPARISON GROUP: THE LONGITUDINAL STUDY OF TRANSITIONS

The Divorce Study originally began as a counterpart to another longitudinal study that focused on the impact of normative transitions. The Longitudinal Study of Transitions, led by Professor Marjorie Fiske, examined how people adapt to the normative crises and challenges that are associated with different stages of life. It included a sample of men and women aged 16 to 65 at their first interviews in 1969 and 1970.

The longitudinal study consisted of 216 subjects drawn from the San Francisco Bay Area, the same geographic locale as subjects in the Divorce Study. This was essentially a purposive sample, although it was grounded in a probability sampling approach (Fiske and Chiriboga 1990). Four groups were selected for study, each of which could be expected to undergo one or more normative transitions within three to four years: high school seniors facing challenges of entering the adult world, newlyweds facing issues revolving around parenthood, middle-aged parents whose youngest child was a high school senior who presumably would be leaving the home, and late middle-agers facing retirement. Gender and ages of participants were as follows:

• 25 high school males aged 16–18

• 27 high school females aged 17–18

• 25 newlywed males aged 21–38

• 25 newlywed females aged 20–34

• 27 empty nest males aged 44–61

• 27 empty nest females aged 39–57

• 30 pre-retirement males aged 53–65

• 30 pre-retirement females aged 45–67

Interviewing began in late 1968 and continued until late 1980, at which point the 168 remaining participants had each been interviewed at five points in time. The retention rate of about 78% over 12 years of study reflects a high level of commitment to the project on the part of the respondents. If we discount the 18 who died during the course of study, the retention rate was actually about 85%. There was little or no evidence of any bias in who did not complete the study; dropouts did not differ in mental health, sociodemographic characteristics, self-image or stress exposure (Fiske and Chiriboga 1990).

Transitions Study as a Comparison Group

From its very beginning, the Divorce Study was designed to build on the findings of the Transitions Study. This was because in many ways the transition of divorce presents itself as an appealing contrast to more normative transitions. Unlike normative transitions, divorce can occur at any age and is much less of an expected event. Perhaps more importantly, like normative transitions, divorce can also be viewed as a stress condition of major significance.

Over several years of study, the research team of the Transitions Study began to realize that no matter how major the changes involved in a normative transition, the more or less expectable nature of these changes acts as a buffering or protective factor (Lowenthal et al. 1975; Fiske and Chiriboga 1990). This realization was the inspiration for the second study, and the reason that a substantial overlap in instrumentation was built into the Divorce Study.

Whenever we compare Divorce and Transitions Study participants in this book, we use a subset of Transitions participants who were not themselves in the throes of marital separation or divorced. Comparison between the two samples on a variety of sociodemographic factors indicated little difference, save that the Divorce Study included more persons from minority groups, and women in the Divorce Study reported significantly lower family incomes than did women in the Transitions Study. However, income levels for life prior to the marital separation were comparable to those reported by Transitions Study women.

Since the Divorce Study began when the Transitions Study had already been in the data collection phase for about seven years, we had to decide at what two contact points the groups should be compared. After some deliberation, we selected the second- and five-year contact points from the Transitions Study, since the time interval approximated that for the Divorce Study, and because during this time interval most Transitions Study participants were themselves experiencing normative transitions.

IN ANTICIPATION

In the chapters that follow we shall address a series of issues and questions related to stress and adaptation that our research suggests may underlie much of what happens in marital separation and divorce. In each chapter we also try to identify risk factors and what might be called benefit factors: things that help and things that hinder working through the problems created by divorce. One common theme is that the risk and benefit factors associated with adaptation during the impact phase of divorce, immediately following marital separation, are often different from those associated with adaptation in the long run.

Most of our findings are based on statistical procedures such as regression ...