Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Divorce after long-term marriage : psychological well-being and parent-adult child relationships Jacoby, Carole Diane 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1996-0586.pdf [ 6.26MB ]
Metadata
JSON: 831-1.0087213.json
JSON-LD: 831-1.0087213-ld.json
RDF/XML (Pretty): 831-1.0087213-rdf.xml
RDF/JSON: 831-1.0087213-rdf.json
Turtle: 831-1.0087213-turtle.txt
N-Triples: 831-1.0087213-rdf-ntriples.txt
Original Record: 831-1.0087213-source.json
Full Text
831-1.0087213-fulltext.txt
Citation
831-1.0087213.ris

Full Text

DIVORCE AFTER LONG-TERM MARRIAGE, PSYCHOLOGICAL WELL-BEING AND PARENT-ADULT CHILD RELATIONSHIPS by CAROLE DIANE JACOBY B.A., The Un i v e r s i t y of B r i t i s h Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n THE FACULTY OF GRADUATE STUDIES School of Family and N u t r i t i o n a l Sciences Family Studies We^-ScceprT\this tmesis as ^conforming tg2~requ:p:ed standard THE UNIVERSITY OF BRITISH COLUMBIA September 1996 ® Carole Diane Jacoby, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Pepartmcnt of T ^ m " , ^ "Wu,-yv,^o^Jl SoteA*ce< The University of British Columbia Vancouver, Canada Date Q c H W r 6 ; DE-6 (2/88) Abstract The purpose of t h i s study was to answer the following questions: what are the e f f e c t s of recent divorce a f t e r long-term marriage on the psychological well-being of parents and on parent-adult c h i l d r e l a t i o n s h i p s , and what are the e f f e c t s of parent-adult c h i l d r e l a t i o n s h i p s on the psychological well-being of recently divorced parents? Analyses used prospective l o n g i t u d i n a l data from two waves of the National Survey of Families and Households. The sample f o r the study consists of 1584 respondents who were married and l i v i n g with t h e i r spouse at Time One; married at least nineteen years at time of f i n a l separation or mean time of f i n a l separation; were ei t h e r continuously married at Times One and Two (the comparison group) or married at Time One and separated or divorced at Time Two; and who had at least one c h i l d nineteen years or older at the time of f i n a l separation or mean time of separation. Study r e s u l t s show that divorce a f t e r long-term marriage had negative e f f e c t s on s e l f - r e p o r t s of happiness, depression and parent-adult c h i l d contact; but reported l e v e l s of contact, support given and support received di d not moderate the negative e f f e c t s of divorce on parent's psychological well-being. A major f i n d i n g of the study i s that the e f f e c t s of divorce on psychological well-being and parent-adult c h i l d r e l a t i o n s h i p s , and the ef f e c t s of these r e l a t i o n s h i p s on postdivorce psychological well-being, were not s i g n i f i c a n t l y d i f f e r e n t f o r divorced mothers or divorced fathers. The r e s u l t s of the study were interpreted within the framework of i d e n t i t y theory, and implications of the research and suggestions f o r future research are also discussed. iii Table of Contents Abstract i i Table of Contents i i i L i s t of Tables v i i Acknowledgements. . v i i i Dedication ix Chapter One Introduction 1 I. Purpose of the Study 2 II . Some D e f i n i t i o n s 4 A. Divorce 4 B. Divorce A f t e r a Long-Term Marriage 5 C. Psychological Adjusment 6 Chapter Two L i t e r a t u r e Review 7 I. The E f f e c t s of Divorce on Psychological Weil-Being 7 A. Dimensions of Psychological Weil-Being 8 1. Happiness 8 2. Psychological Distress 9 3. Self-Esteem 11 B. Gender Differences 12 1. Happiness 13 2. Psychological Distress 14 3 . Self-Esteem 16 C. Summary 16 I I . The E f f e c t s of Divorce A f t e r Long-Term Marriage on Psychological Well-Being 17 A. Dimensions of Psychological Weil-Being 23 i v 1. Happiness 23 2. Psychological Distress 25 3. Self-Esteem 26 B. Summary 27 I I I . Parent-Adult C h i l d Relationships and Psychological Weil-Being 27 A. Parent-Adult C h i l d Relationships 27 B. Parent-Adult C h i l d Relationships and Psychological Weil-Being 30 IV. Divorce, Parent-Adult C h i l d Relationships and Psychological Weil-Being 32 A. Long-Term E f f e c t s of E a r l i e r Divorce on Parent-Adult C h i l d Relationships and Parent's Psychological Weil-Being 32 B. Recent Divorce, Parent-Adult C h i l d Relationships and Psychological Weil-Being 35 C. Summary 43 Chapter Three Research Problem 45 I. Theoretical Perspective 45 A. Theoretical Concepts 46 B. Linking S o c i a l Structure, Role Relationships, and Role-Identities 47 C. Linking Role Relationships, Role-Identities and Psychological Weil-Being 49 II . Hypotheses 52 I I I . Control Variables 60 V IV. Methods 65 A. The Data Set and Sample 65 B. Dependent Variables 68 1. Psychological Weil-Being 68 a. Happiness 6 8 b. Depression 70 c. Self-Esteem 71 2. Parent-Adult C h i l d Relationships 71 a. Contact 73 b. Support 74 C. Independent and Conditional Variables 76 D. Control Variables 76 1. Income/Assets 77 2. Race 78 3. Age of Respondent 78 4. M a r i t a l Cohort 78 5. Degree of S o c i a l Integration 78 6. R e l i g i o s i t y 80 7. Geographic Distance 80 E. Tests for M u l t i c o l l i n e a r i t y 81 vi Chapter Four Results ' 82 Chapter Five Discussion 95 I. Divorce A f t e r Long-Term Marriage and Psychological Well-Being 96 A. Happiness 96 B. Depression 97 C. Self-Esteem 98 II . Divorce A f t e r Long-Term Marriage and Parent-Adult C h i l d Relationships 100 A. Contact 100 B. Support Given 103 C. Support Received 105 I I I . Divorce A f t e r Long-Term Marriage, Parent-Adult C h i l d Relationships and Psychological Well-Being 107 IV. Strengths and Limitations I l l A. Limitations I l l B. Strengths 113 C. Implications 115 a. Family Studies 115 b. Theoretical Development 116 c. Family P r a c t i t i o n e r s 117 d. Family-Related P o l i c y 118 V. Conclusion 119 References 122 Appendix A Research Questions 133 Appendix B Tables 151 v i i L i s t of Tables Table 1. Means/Proportions and Standard Deviations for Study Variables 152 Table 2. H i e r a r c h i c a l OLS Estimates for the E f f e c t s of Divorce on Unhappiness 153 Table 3. H i e r a r c h i c a l OLS Estimates for the E f f e c t s of Divorce on Depression 154 Table 4. H i e r a r c h i c a l OLS Estimates for the E f f e c t s of Divorce on Self-Esteem 155 Table 5. H i e r a r c h i c a l OLS Estimates for the E f f e c t s of Divorce on Contact 156 Table 6. H i e r a r c h i c a l OLS Estimates for the E f f e c t s of Divorce on Support Given 157 Table 7. H i e r a r c h i c a l OLS Estimates for the E f f e c t s of Divorce on Support Received 158 Table 8. H i e r a r c h i c a l OLS Estimates for the E f f e c t s of Divorce and Parent-Adult C h i l d Relationships on Depression 159 \ vi i i Acknowledgements This thesis represents not only the end product of months of concentrated focus, but also the culmination of my years of study i n the School of Family and N u t r i t i o n a l Sciences. Therefore I would l i k e to take t h i s opportunity to thank the f a c u l t y of family studies for the s o l i d academic foundation they have given me and for the r o l e models they have been and continue to be. In p a r t i c u l a r I would l i k e to thank the members of my thesis committee: Dr. Rick Bulcroft, f or the structure he provided me, his patient teaching of research s k i l l s , and h i s unwavering b e l i e f i n my competencies; and Dr. Dan Perlman, for the wisdom inherent i n his c r i t i q u e s . I have also been fortunate i n having Dr. Edward Kruk of So c i a l Work as the external member of my committee, and I thank him for his support of my career goals and for broadening my perspectives on divorce. I look forward to continuing my studies i n divorce and family mediation with him. And l a s t , but not least, I would l i k e to say "thank you" to Margaret, whose support throughout the process w i l l always be remembered and appreciated. ix Dedication This thesis i s dedicated to my granddaughter C a i t l i n , born July 26, 1996. 1 Chapter One Introduction Divorce i n middle or l a t e r l i f e i s no longer the rare occurrence i t once was. While the divorce rates for younger adults are s t a b i l i z i n g or d e c l i n i n g s l i g h t l y , divorce rates for older adults are increasing and are expected to continue to increase, including rates for couples ending long-term marriages (Berardo, 1982; Cooney, 1994; Goodman, 1993; Smyer & Hofland, 1982; Uhlenberg & Myers, 1981; Uhlenberg, Cooney, & Boyd, 1990; Weingarten, 1988). Cooney (1994, c i t i n g National Center for Health S t a t i s t i c s , 1991) noted that i n 1988 approximately 20% of a l l divorces i n the United States involved couples married f i f t e e n years or longer. The rates are high i n Canada also: S t a t i s t i c s Canada (1996) recently reported that i n 1994 12.8% of a l l Canadian divorces involved couples married between f i f t e e n and nineteen years, and 21.3% involved couples married twenty years or longer (Canada, 1996). When couples divorce a f t e r a long-term marriage, reverberations are f e l t throughout the e n t i r e family system. However, despite repeated c a l l s f o r more research i n t h i s area, the extant body of research l i t e r a t u r e remains small (Hennon, 1983; Hagestad, Smyer, & Stierman, 1984; Uhlenberg & Myers, 1981; f o r reviews see Brubaker, 1990; Kitson & Morgan, 1990) . In t h e i r seminal three-generational study of the impact of divorce i n middle age on family r e l a t i o n s h i p s , Hagestad, Smyer, and Stierman (1984) r e f e r to the "young adult bias" i n divorce research and t h i s bias appears to be continuing, as the increased divorce rates noted above are being accompanied by growing research i n t e r e s t i n the e f f e c t s of divorce a f t e r long-term marriage on adult children's well-being and development. Although researchers are beginning to investigate how being divorced a f f e c t s family 2 r e l a t i o n s h i p s when ch i l d r e n are grown, much less attention i s being given to how becoming divorced a f t e r a long-term marriage a f f e c t s family-r e l a t i o n s h i p s and how changes i n family r e l a t i o n s h i p s may a f f e c t the parent's psychological adjustment to the divorce. Parent-child r e l a t i o n s h i p s are, f o r most people, a primary kinship t i e that remains of c e n t r a l s i g n i f i c a n c e throughout the l i f e course (Barnett, Marshall, & Pleck, 1992; Eggebeen, 1992; Moss, Moss, & Moles, 1985; Rosenthal, 1987; Rossi & Rossi, 1990; T r o l l , 1982) and, o v e r a l l , parent-adult c h i l d r e l a t i o n s h i p s i n middle and l a t e r l i f e are characterized by high l e v e l s of emotional closeness, contact and s o c i a l support ( C i c i r e l l i , 1983 ; Mancini & Blieszner, 1989; Rossi & Rossi, 1990; T r o l l , 1982) . Fortunately so, because the q u a l i t y of these r e l a t i o n s h i p s has been found to have an ef f e c t on the psychological well-being of both generations (Le v i t t , Guacci, & Weber, 1992; Umberson, 1992). When thinking about a divorce by parents a f t e r a long-term marriage the question arises as to what the e f f e c t s of the divorce on parent-adult c h i l d r e l a t i o n s h i p s might be, and how the parent-adult c h i l d r e l a t i o n s h i p might a f f e c t the parent's psychological adjustment to the divorce. The idea behind t h i s study i s that the r e l a t i o n s h i p s between divorced parents and t h e i r adult c h i l d r e n have the p o t e n t i a l to s i g n i f i c a n t l y influence the divorce adjustment of the parent. I. Purpose of the Study The purpose of t h i s study i s to answer the following three questions: f i r s t , what are the e f f e c t s of recent divorce a f t e r long-term marriage on the psychological well-being of older parents; second, what are the e f f e c t s of recent parental divorce on parent-adult c h i l d r e l a t i o n s h i p s ; and t h i r d , what are the e f f e c t s of parent-adult c h i l d r e l a t i o n s h i p s on the psychological well-being of recently divorced parents? These questions are 3 examined using prospective l o n g i t u d i n a l data from two waves of the National Survey on Families and Households (NSFH), a n a t i o n a l l y representative data set developed by prominent family scholars i n the United States s p e c i f i c a l l y to address family issues. It i s frequently noted that there has been l i t t l e empirical research on middle and l a t e r l i f e divorce, and l i t t l e t h e o r e t i c a l development (Ahrons & Rodgers, 1987; Bengtson, Rosenthal, & Burton, 1990; Brubaker, 1990; Coleman & Ganong, 1993; Hennon, 1983; Lloyd & Zick, 1986). In t h i s study, i d e n t i t y theory was used to provide a t h e o r e t i c a l framework to guide the development of hypotheses and the i n t e r p r e t a t i o n of r e s u l t s , i n conjunction with the e x i s t i n g l i t e r a t u r e . / Identity theory was chosen for three main reasons. One, i n recent years i d e n t i t y theory has been used i n studies to explain differences i n parent-adult c h i l d r e l a t i o n s h i p s and the association between these r e l a t i o n s h i p s and the psychological well-being of both generations (Atkinson, 1989; Marks, 1995; Mutran & Reitzes, 1984; Roberts & Bengtson, 1993; Umberson, 1992). In these studies, the researchers have not attempted to measure the t h e o r e t i c a l concepts d i r e c t l y , however the conceptual ideas have been used to provide a framework for the research and the findings. Two, the nature of parent-child r e l a t i o n s h i p s over the l i f e course suggests that the parent r o l e - i d e n t i t y i s important to most parents and that the continuing presence of active, supportive parent-child r e l a t i o n s h i p s w i l l have an impact on the parent's postdivorce psychological well-being. And three, i d e n t i t y theory i s p a r t i c u l a r l y relevant when examining gender differences i n parent-adult c h i l d r e l a t i o n s h i p s because i t provides t h e o r e t i c a l l i n k s between the i n s t i t u t i o n a l , group (family) and i n d i v i d u a l le v e l s of analysis; for example, i t l i n k s s o c i a l s t r u c t u r a l factors such as 4 marital status and normative expectations for behavior with the family-in t e r a c t i o n s of i n d i v i d u a l s . I I . Some D e f i n i t i o n s A. Divorce The study of "divorce" i s a c t u a l l y the study of a process which extends from the time of i r r e s o l v a b l e c o n f l i c t and decision, through the physical separation, the le g a l process and f i n a l divorce decree, to the psychological and pragmatic adjustments of the postdivorce period. According to Chiriboga (1982), i t i s during the t r a n s i t i o n a l period following separation that the major impact of divorce occurs for most people. Consequently many prominent researchers engaged i n conducting large-scale l o n g i t u d i n a l studies of divorce measure the timing of t h e i r studies from the date of the f i n a l separation before respondents f i l e d for divorce, rather than from the time of the f i n a l divorce decree (e.g. Bloom and his colleagues - Bloom & Caldwell, 1981; Bloom & Clement, 1984; Bloom, Hodges, Kern, & McFaddin, 1985; Chiriboga and his colleagues - Chiriboga, 1982; Chiriboga, Catron and Associates, 1991; Chiriboga, Roberts, & Stein, 1989; Kitson and Holmes, 1992; Spanier & Thompson, 1984). The point at which the actual divorce decree was received i s not generally noted (for an exception see Kitson & Holmes, 1992). As a r e s u l t of t h i s predominant perspective, researchers often use the term "divorced" to r e f e r to persons engaged i n the divorce process, or use the term "separated/divorced." In t h i s study, the term "divorced" w i l l be used to denote those respondents who separated or divorced between Wave I and Wave II of the data c o l l e c t i o n (a period of approximately f i v e years). 5 B. Divorce A f t e r a Long-term Marriage There has yet to develop a c l e a r l y - d e f i n e d l i t e r a t u r e on mid- and l a t e r - l i f e divorce. At present, information on the divorces of adults divorcing i n mid- or l a t e r - l i f e a f t e r a long-term marriage can be found i n both studies of divorce i n general and i n studies of l a t e r l i f e divorce i n p a r t i c u l a r . For example, i n some studies of divorce, the upper age range may include a few respondents i n t h e i r f o r t i e s or older (e.g. Kitson & Holmes, 1992; Spanier & Thompson, 1984; Wallerstein & Blakeslee, 1989), while i n other studies, the lower age for adults designated as "older" w i l l be 45 or 50 (e.g. Hammond & Muller, 1992: Kitson & Roach, 1989). Most of the information on divorce a f t e r a long-term marriage i s subsumed by the generic t i t l e of " l a t e r l i f e divorce." There i s , however, no consensus i n the l i t e r a t u r e as to the d e f i n i t i o n of " l a t e r l i f e divorce," with some researchers using chronological age and some using duration of marriage. There i s l i t t l e consistency i n the choice of a chronological age marker f or l a t e r l i f e divorce; some researchers have used age 50 (e.g. Gander, 1991; Gander & Jorgensen, 1990; Hammond & Muller, 1992), some 55 (e.g. Barresi & Hunt, 1990), some 60 (e.g. Weingarten, 1988), and some 65 (e.g. Uhlenberg & Myers, 1981). When duration of marriage i s s p e c i f i e d , i t i s frequently used i n combination with other c r i t e r i a ; f o r example, Gander and Jorgensen (1990) r e s t r i c t e d t h e i r sample to respondents f i f t y years and older who had been married at le a s t f i f t e e n years, and Wright and Maxwell (1991) r e s t r i c t e d t h e i r sample to respondents who had been divorced a f t e r nineteen or more years of marriage and had at least one c h i l d over the age of eighteen. In t h i s study, the c r i t e r i a f o r sample s e l e c t i o n was that respondents had to have been married at least nineteen years at time of separation and have at least one adult c h i l d nineteen years of age or older at time of separation (or mean time of separation for the continuously married). The choice of nineteen years as a marker for long-term marriage i s congruent with previous l i t e r a t u r e (e.g. Wright & Maxwell, 1991) and r e f l e c t s the focus of the study on long-term r e l a t i o n s h i p s . The age of nineteen was chosen for adult c h i l d r e n because of the l e g a l connotations: by nineteen the "children" are l e g a l l y "adults," parents no longer have l e g a l r e s p o n s i b i l i t i e s , and the parent-child r e l a t i o n s h i p becomes more voluntary. C. Psychological Adjustment "Psychological adjustment" i s defined i n t h i s study as "psychological well-being" (also r e f e r r e d to as "subjective well-being"; see Weingarten & Bryant, 1987), which can be simply defined as "contentment with oneself and l i f e " (Spanier & Thompson, 1984) . This concept i s widely used as an outcome in studies of divorce, marriage and parenthood (Doherty, Su, & Needle, 1989; Gove & Shin, 1989; Gove, Style, & Hughes, 1990; McLanahan & Adams, 1987; Marks, 1995; Menaghan, 1989), although i t i s not unidimensional. Factor analyses by Bryant and Veroff (1984) have d i f f e r e n t i a t e d three basic dimensions: p o s i t i v e evaluations (e.g. happiness), negative evaluations (e.g. s t r a i n or psychological d i s t r e s s ) , and personal adequacy or competence (e.g. self-esteem). Therefore i n t h i s study, psychological adjustment was operationalized i n terms of happiness, depression and self-esteem. 7 Chapter Two Li t e r a t u r e Review The extant body of l i t e r a t u r e on parent-adult c h i l d r e l a t i o n s h i p s and l a t e r l i f e divorce i n general and divorce a f t e r a long-term marriage i n p a r t i c u l a r i s very small. The plan f o r t h i s l i t e r a t u r e review, therefore, i s to e s t a b l i s h a context within which divorce a f t e r a long-term marriage and parent-adult c h i l d r e l a t i o n s h i p s can be better understood. In order to do t h i s , the l i t e r a t u r e review has been divided into two major sections. The f i r s t section reviews the research l i t e r a t u r e on the e f f e c t s of divorce on psychological well-being i n general, gender differences, and how these e f f e c t s may d i f f e r by age. The second section reviews the l i t e r a t u r e on the nature of parent-adult c h i l d r e l a t i o n s h i p s i n general and how they can influence a parent's psychological well-being, how e a r l i e r l i f e divorce a f f e c t s parent-adult c h i l d r e l a t i o n s h i p s and psychological well-being, and f i n a l l y , what we know about how a recent divorce a f f e c t s parent-adult c h i l d r e l a t i o n s h i p s and psychological well-being. Although the l i t e r a t u r e review includes research conducted with younger respondents, the intent i s not to compare younger and older respondents, but rather to lay out a pattern of e x i s t i n g research into which the present study w i l l l o g i c a l l y f i t . I. The E f f e c t s of Divorce on Psychological Well-Being The e f f e c t s of divorce on psychological well-being have been found to vary according to the dimension being measured (see f o r example, Gove, Hughes, & Style, 1983; Marks, 1995; Mastekaasa, 1994b). Consequently, although some researchers have used scales which measure o v e r a l l psychological well-being (e.g. Gander, 1991; Gander & Jorgensen, 1990), 8 most researchers report findings f or p a r t i c u l a r dimensions. The l i t e r a t u r e review which follows i s organized around three commonly-measured dimensions of psychological well-being which r e f l e c t the three basic dimensions discussed i n Chapter One. These three dimensions are happiness, psychological d i s t r e s s and self-esteem. A. Dimensions of Psychological Well-Being 1. Happiness Cross-sectional studies comparing divorced respondents and respondents i n i n t a c t marriages have shown that divorce has a negative e f f e c t on s e l f - r a t i n g s of happiness (Connidis & McMullin, 1993; Gove & Shin, 1989; Gove et a l . , 1983; Hatch & S t u l l , 1987; Mastekaasa, 1994b; Marks, 1995). Longitudinal studies have shown a changing pattern of psychological well-being over the course of the divorce process. Hetherington, Cox, and Cox (1978) interviewed respondents at two months postdivorce, one year l a t e r and two years l a t e r . They found divorced respondents to be s i g n i f i c a n t l y less happy at a l l three time points, but the difference to be smaller at Time 3. Chiriboga, Catron and associates (1991) report s i m i l a r f i n d i n g s : at Time 1 (separated less than eight months) separated respondents were less happy than married respondents; at Time 2 (approximately three years later) separated/divorced respondents and married respondents were s i m i l a r i n happiness. S i m i l a r l y , Booth and Amato (1991) conducted a prospective l o n g i t u d i n a l study i n which a l l respondents were married at Time 1 (1980) and there were two groups of divorced persons: Group 1 consisted of persons who divorced between Time 1 (1980) and Time 2 (1983), and Group 2 consisted of persons who divorced between Time 2 and Time 3 (1988). The 9 r e s u l t s show the changing e f f e c t s of marital d i s t r e s s and d i s r u p t i o n over time. At Time 1, when both divorced groups are compared to the continuously-married group, respondents i n Group 1 (who divorced between Time 1 and Time 2) were experiencing greater unhappiness than controls, while respondents i n Group 2 (who divorced between Time 2 and Time 3) were experiencing s i m i l a r l e v e l s of unhappiness. At Time 2, respondents i n both divorce groups (recently divorced and soon-to-be divorced) were experiencing greater unhappiness than the continuously-married group. At Time 3, respondents i n Group 1 were once again experiencing s i m i l a r l e v e l s of unhappiness to the continuously-married group, while respondents i n Group 2 were s t i l l experiencing higher l e v e l s of unhappiness. 2. P s y c h o l o g i c a l D i s t r e s s S i m i l a r to the r e s u l t s f o r happiness, c r o s s - s e c t i o n a l studies have shown that divorced respondents rate themselves higher on ind i c a t o r s of depression ( i . e . higher l e v e l s of depression) than do married respondents (Kurdek, 1991; Marks, 1995; Mastekaasa, 1994b; Riessman & Gerstel, 1985) . Longitudinal studies also show divorced persons r a t i n g themselves as s i g n i f i c a n t l y more depressed than married persons. For example, Kitson and Holmes (1992) report r e s u l t s of a study conducted between 1974 and 1979, i n which respondents were interviewed three times. Time 1 was at the time of f i l i n g f o r divorce (an average of one year postseparation); Time 2 was an average of eleven months postdivorce (an average of two years postseparation); Time 3 was an average of three years postdivorce (an average of four years postseparation). Results show that the separated/divorced respondents were experiencing higher 10 l e v e l s of depression and subjective d i s t r e s s than the married comparison group at a l l three time points. It i s not cle a r , however, whether persons who subsequently divorce are more depressed p r i o r to the divorce than persons who are continuously married. Menaghan and Lieberman (1986) analyzed a l o n g i t u d i n a l data set and confined t h e i r sample to persons married at Time 1 (1972) and s t i l l married to and l i v i n g with the same spouse at Time 2 (1976; n=758); and those married at Time 1 but divorced at Time 2 (n=32). Results from t h i s analysis showed those who stayed married and those who subsequently divorced to be s i m i l a r i n l e v e l s of depression at Time 1, but those who subsequently divorced to be higher i n l e v e l s of depression at Time 2. Menaghan and Lieberman found, however, that when they added indicators of current economic problems, perceived d e t e r i o r a t i o n i n economic s i t u a t i o n and perceived a v a i l a b i l i t y of personal support to t h e i r regression equation, the marital group differences i n depression change became s t a t i s t i c a l l y n o n s ignificant. In contrast to Menaghan and Lieberman (1986), Booth and Amato (1991) found that respondents who divorced between 1980 and 1983 reported higher l e v e l s of psychological d i s t r e s s i n 1980 (predivorce) and i n 1983 (postdivorce), but s i m i l a r l e v e l s to the continuously married group i n 1988. Respondents who divorced between 1983 and 1988, when compared with those who remained married, reported s i m i l a r l e v e l s of psychological d i s t r e s s i n 1980, higher l e v e l s i n 1983 and s i m i l a r l e v e l s i n 1988. Thus Booth and Amato's study, i n contrast to Menaghan and Lieberman 1s study, shows l e v e l s of psychological d i s t r e s s to be higher both before and a f t e r a divorce, but to d i s s i p a t e with time. The d i f f e r e n c e between these studies may be p a r t i a l l y explained by the small s i z e of Menaghan and 11 Lieberman's sample, n=32 with only twelve men. Also, Doherty and colleagues (1989) note that the a t t r i t i o n rate i n Menaghan and Lieberman's study was 58%. Booth and Amato (1991) also analyzed the e f f e c t of time since divorce on happiness and psychological d i s t r e s s while c o n t r o l l i n g f or predivorce l e v e l s of s t r e s s . Results of t h i s analysis show no differences between the divorced group and nondivorced group a f t e r 25 months; t h i s pattern held for both divorced respondents who remarried and those who d i d not. 3 . S e l f - E s t e e m The e f f e c t s of divorce on self-esteem are less c l e a r than those for happiness and psychological d i s t r e s s . While some cr o s s - s e c t i o n a l studies have shown divorce to have l i t t l e s i g n i f i c a n t e f f e c t on self-esteem (Marks, 1995; Weingarten, 1985), some have shown negative e f f e c t s . Umberson and Gove (1989) f o r example, found that everdivorced respondents rated themselves lower on self-esteem than married respondents. Si m i l a r to cross-sectional studies, r e s u l t s from l o n g i t u d i n a l studies are also mixed, with some showing negative e f f e c t s of divorce on self-esteem and others i n d i c a t i n g that self-esteem may follow a s i m i l a r pattern to that of happiness and psychological d i s t r e s s , i . e . i n i t i a l decline and subsequent recovery. For example, Doherty, Su, and Needle (1989), i n t h e i r small prospective l o n g i t u d i n a l study, found no differences i n self-esteem between those who divorced and those who were continuously married, at e i t h e r predivorce or postdivorce interviews (average twelve months and f i v e months r e s p e c t i v e l y ) . Chiriboga, Catron and associates (1991) also found that self-evaluations on most dimensions of the self-concept d i d not change s i g n i f i c a n t l y over the 3.5 years of 12 t h e i r study. They d i d f i n d , however, that the divorce sample scored s i g n i f i c a n t l y higher than the married comparison group on negative s e l f , vulnerable s e l f , and mastery at Times 1 and 2 (both postdivorce), and interpreted t h i s to mean that the divorce process may involve p o s i t i v e and negative changes i n o v e r a l l self-image. In contrast, Kitson and Holmes (1992) found that newly separated respondents (average f i v e months) rated themselves lower i n self-esteem than married respondents, but that these differences disappeared by Times 2 and 3 (one and three years postdivorce r e s p e c t i v e l y ) . The r e s u l t s from a l l these studies suggest that the r e l a t i v e s t a b i l i t y of self-esteem throughout the divorce process may be an a r t i f a c t of measurement ( i . e . gains and losses balance out), or may be an i n d i c a t i o n that self-esteem measures such as the widely-used Rosenberg Self-Esteem Scale (Rosenberg, 1989) tap a more stable p e r s o n a l i t y a t t r i b u t e rather than a dimension of psychological well-being. B. Gender Differences One of the most predominant themes i n divorce research has been the exploration and explanation of gender differences i n divorce adjustment and psychological well-being. In 1972 Jesse Bernard concluded that men and women experience "his" and "her" marriages. Since then, many divorce researchers have s i m i l a r l y concluded that when marriages break down, men and women also experience "his" and "her" divorces (Connidis & McMullin, 1993; Doherty et a l . , 1989; Kitson & Morgan, 1990; Price & McKenry, 1988; Sel t z e r & Brandreth, 1994; Wallerstein, 1986). Recent reviews of the research, however, have c a l l e d the findings on gender differences "curiously mixed" (Kitson & Morgan, 1990, p. 916), "contradictory" (Price & McKenry, 1988, p. 61), and "inconclusive" (Raschke, 1987, p. 612). 13 While some studies have shown no gender differences i n o v e r a l l psychological well-being (Gander, 1991; Gove & Shin, 1989; Spanier & Thompson, 1984), the more general pattern i s for women and men to d i f f e r across various dimensions of well-being. 1. Happiness Chiriboga and his colleagues began conducting research on separation and divorce i n the 1970s, and t h e i r conclusion i s that women and men have d i f f e r e n t areas of v u l n e r a b i l i t y (Chiriboga, 1982; Chiriboga et a l . , 1978). In several e a r l y studies, they found separated/divorced men to be less happy than separated/divorced women; women were about average f o r the general population while men were about twice as unhappy as the general population (Chiriboga et a l , 1978). Separated/divorced women, however, reported more psychological d i s t r e s s and emotional tension (Chiriboga, 1982) . The researchers' o f t e n - c i t e d conclusion i s that separated/divorced men experience a lower o v e r a l l sense of well-being than women, while women experience greater emotional turmoil than men (Chiriboga, 1982; Chiriboga et a l . , 1978). Results from cross-sectional studies, such as those c i t e d above by Chiriboga and colleagues (see also Hammond & Muller, 1992) suggest that separated/divorced men are less happy than separated/divorced women. When juxtaposed with research findings that married men are happier than married women (see Coombs, 1991 for a review), one imp l i c a t i o n i s that divorce can have a negative e f f e c t on men's happiness. Cause-and-effect cannot be determined from cross-sectional studies, however, as the same men who are unhappy following a divorce may also have been unhappy before i t . 14 Other studies have found no gender differences i n se l f - r e p o r t e d happiness f o r separated/divorced men and women. Gove and Shin (1989) found no s i g n i f i c a n t gender differences on any of the dimensions they measured: happiness, l i f e s a t i s f a c t i o n , home l i f e s a t i s f a c t i o n , s e l f -esteem, psychological d i s t r e s s , f e e l i n g trapped, and meaninglessness. Booth and Amato (1991), i n a lon g i t u d i n a l study, also found no differences i n e f f e c t s of separation/divorce by gender. Other l o n g i t u d i n a l studies have found that gender differences i n happiness disappear over time. Chiriboga, Catron and associates (1991) found l e v e l s of reported happiness to change over time, with men s i g n i f i c a n t l y less happy than women at Time 1 (less than eight months postseparation), but no gender differences at Time 2 (approximately 3 years l a t e r ) . Bloom and his colleagues (1985) found that at Time 1 (an average of eight weeks postseparation) women reported increased happiness compared to a year previously, but that at Times 2, 3, 4 and 5 (six months l a t e r , 18 months l a t e r , 30 months l a t e r and 4 years l a t e r , respectively) there were no s i g n i f i c a n t gender d i f f e r e n c e s . 2. P s y c h o l o g i c a l D i s t r e s s In s o c i a l surveys, psychological d i s t r e s s i s generally operationalized using s e l f - r e p o r t measures of depression and symptomatic psychological d i s t r e s s ; measures on which women t y p i c a l l y rate themselves higher ( i . e . more distressed) than do men (Aneshensel et a l . , 1981; Connidis & McMullin, 1993; Kurdek, 1991; Riessman & Gerstel, 1985) . Men, in contrast, have been found to have higher rates of p s y c h i a t r i c treatment, suicide, morbidity and mortality (see Riessman & Gerstel, 1985) . Women, therefore, appear to experience more symptoms of mild to 15 moderate depression, while men may experience more symptoms of severe depression (Chiriboga et a l . , 1991; Riessman & Gerstel, 1985). Kitson and Morgan (1990) have suggested that, because the responses of women to divorce have been studied more than the responses of men, e x i s t i n g measures of psychological d i s t r e s s may be more appropriate f o r women than f o r men and they recommend that new measures be developed that are more s e n s i t i v e to the reactions of men. Umberson and Williams (1992), f o r example, have noted that v i o l e n t behavior and alcohol and drug abuse can be "functional equivalents" of depression and anxiety (Horwitz & White, 1991) and that men are more l i k e l y to engage i n these behaviors than are women (Wingard, 1984) . Although some recent studies have included measures of adjustment thought to r e f l e c t more adequately men's responses to separation and divorce (e.g. Cooney, Hutchinson, & Leather, 1995; Thoits, 1992; Umberson & Williams, 1993), the standard p r a c t i c e i n divorce research i s to r e l y on established s e l f - r e p o r t measures of psychological d i s t r e s s . Consequently, i t i s not s u r p r i s i n g when studies using these s e l f -evaluative measures f i n d separated/divorced women reporting higher l e v e l s of depression than separated/divorced men (e.g. Chiriboga et a l . , 1978; Wallerstein & Kell y , 1980) . What i s i n t e r e s t i n g i s when the r e s u l t s d i f f e r from the expected pattern. Aneshensel, Frerichs, and Clark (1981) for example, i n t h e i r c r oss-sectional study, found the e f f e c t of divorce on depression to be stronger f o r men than f o r women. They found married women to be s i g n i f i c a n t l y more depressed than married men, but found no s i g n i f i c a n t differences between the depression scores of divorced women and divorced men. This f i n d i n g i s accounted f o r by the considerably higher depression scores of divorced men as compared to married men. 16 In t h e i r prospective l o n g i t u d i n a l study, however, Menaghan and Lieberman (1986; see also Menaghan, 1985) found recently-divorced women and men to have higher l e v e l s of depression than continuously-married women and men, but found no gender differences i n the divorced group ei t h e r predivorce or postdivorce ( c o n t r o l l i n g f o r current economic circumstances, perceived d e t e r i o r a t i o n i n standard of l i v i n g , and cur r e n t l y a v a i l a b l e confidants). 3. Self-Esteem There do not appear to be any s i g n i f i c a n t gender differences i n self-esteem e i t h e r predivorce or postdivorce (Doherty et a l , 1989). In a lon g i t u d i n a l study, Kitson and Holmes (1992) found no gender differences i n self-esteem at any of the three postseparation time points. Gove and Shin (1989) and Weingarten (1985) also found no s i g n i f i c a n t e f f e c t s of divorce on self-esteem ratings i n t h e i r c r o s s - s e c t i o n a l studies. C. Summary Ove r a l l , the research findings on the e f f e c t s of divorce on happiness, depression and self-esteem are mixed. Cross-sectional research shows negative e f f e c t s f or happiness and depression, but mixed r e s u l t s f o r self-esteem (negative or no differences between married and divorced). Longitudinal research, however, indicates that the negative e f f e c t s of divorce on ratings of happiness and depression generally attenuate with time. Research findings on gender differences i n divorce adjustment are also mixed. Cross-sectional research shows eit h e r men to report being less happy or no gender differences, women to report being more depressed or no gender differences, and no gender differences between men and women 17 i n self-esteem. Longitudinal studies show that i n i t i a l gender differences i n happiness d i s s i p a t e with time, while Menaghan & Lieberman (1986) found no predivorce or postdivorce gender differences i n ratings of depression. There also do not appear to be any gender differences over time i n self-esteem. How, then, might we characterize the e f f e c t of divorce on psychological well-being? It seems f a i r l y c l e a r that, for most people, divorce has at least short-term negative e f f e c t s on s e l f - r a t i n g s of happiness and depression and p o s s i b l y on self-esteem, but that these negative e f f e c t s may attenuate with time. Regarding gender differences i n divorce adjustment, again i t seems f a i r l y c l e a r that men report lower l e v e l s of happiness and women report higher l e v e l s of depression, but that these gender differences l i k e l y also attenuate over time. There appear to be no gender differences i n self-esteem. I I . The E f f e c t s of Divorce A f t e r Long-Term Marriage on Psychological Well-Being The p r e v a i l i n g assumption i n the divorce l i t e r a t u r e has been that adjustment to divorce at older ages must be more d i f f i c u l t than adjustment to divorce at younger ages (Berardo, 1982; Price & McKenry, 1988; Raschke, 1987), but t h i s assumption i s beginning to be challenged (see Kitson & Morgan, 1990) . The question becomes, then, what was the o r i g i n a l assumption based on, and what resources might persons who have been married longer have that may have been overlooked thus far? One way to look at these issues i s to use Bohannan's (1970) six-p a r t conceptualization of the divorce process as an organizing framework. Bohannan (1970) i d e n t i f i e d s i x overlapping experiences of divorce (or s i x "divorces" as they are commonly known): the emotional, l e g a l , economic. 18 community, psychic, and coparental. By looking at each of these "divorces," we can begin to d i f f e r e n t i a t e areas of p o t e n t i a l d i f f i c u l t y from areas of p o t e n t i a l support. According to Bohannan (1970), the emotional divorce i s u s u a l l y the f i r s t stage of the process and i s centered around the problem of the d e t e r i o r a t i n g marriage. In Bohannan's perspective, a natural process i n marriage i s for the partners to grow i n new d i r e c t i o n s but also to develop strong bonds of interdependence. If partners do not "grow together as they grow apart" (p. 30), they may begin to f e e l imprisoned by the marriage bonds rather than secure i n them. As they grow apart, a f f e c t i o n and t r u s t can disappear. It has often been noted that the emotional q u a l i t y of marriages tends to change over time: i n the e a r l y years there i s more l i k e l y to be an intense, passionate q u a l i t y to the r e l a t i o n s h i p ; as time goes on, i t i s l i k e l y that a more companionate q u a l i t y begins to develop (see Brehm, 1985). For adults i n short-term marriages, the emotional divorce may be quite d i f f i c u l t and tempestuous, while for adults i n long-term marriages, the emotional divorce may be tempered both by the stage i n the r e l a t i o n s h i p and by the l i k e l i h o o d that emotions have mellowed over a longer period of time. Consequently the emotional divorce may be less d i f f i c u l t f o r adults a f t e r a long-term marriage than f o r adults a f t e r a short-term marriage. The l e g a l divorce r e f e r s to the l e g a l aspects of the divorce. In North America, the l e g a l system and the courts have the r e s p o n s i b i l i t y for the formal d i s s o l u t i o n of marriages. Bohannan proposed h i s s i x "divorces" before the advent of "no-fault divorce," therefore the l e g a l divorce as he discussed i t i s concerned p r i m a r i l y with l e g a l grounds, 19 e.g. adultery, drunkenness, i n c o m p a t i b i l i t y . Although revised divorce laws have removed or diminished some of the emphasis on f i n d i n g l e g a l grounds, the l e g a l divorce s t i l l e n t a i l s some involvement with the l e g a l system and i t s adversarial perspective for a l l divorcing persons. The l e g a l divorce i s c l o s e l y connected to the economic divorce, as the l e g a l divorce includes such procedures as d i v i s i o n of property and other assets (e.g. pension plans, health insurance), perhaps alimony, and i n the case of minor children, custody, access and c h i l d support. One advantage f o r parents of adult c h i l d r e n i s that the l e g a l divorce w i l l not l i k e l y be complicated with issues of custody, access and c h i l d support (exceptions could be f o r parents of c h r o n i c a l l y - i l l or otherwise dependent adult c h i l d r e n ) . The economic divorce deals with the reassignment of property and the d i v i s i o n of finances. In community property j u r i s d i c t i o n s , assets acquired a f t e r the marriage (except perhaps inheritances) are divided equally between the former marital partners. Two of the major issues i n the economic divorce are c h i l d support and alimony (generally now c a l l e d "maintenance") which i s set by the court based on one partner's need (usually the wife's) and on the other partner's a b i l i t y to pay. The economic aspects of a divorce a f t e r a long-term marriage can be p a r t i c u l a r l y d i f f i c u l t f o r older women (the "displaced homemakers") who have devoted t h e i r l i v e s to homemaking with only b r i e f , i f any, entries into the workforce (Morgan, 1991) . Although older men also face the d i v i s i o n of family assets, they are more l i k e l y to be well-established i n careers and able to recoup t h e i r f i n a n c i a l losses. Younger divorced mothers with custody of young c h i l d r e n also face economic d i f f i c u l t i e s as many of them struggle to provide f o r t h e i r 20 f a m i l i e s with l i t t l e or no c h i l d support from the nonresidential fathers. Fathers who do f u l f i l l t h e i r f i n a n c i a l r e s p o n s i b i l i t i e s to the divorced f a m i l i e s may f i n d themselves supporting two fa m i l i e s i f they remarry. Overall, the economic divorce can be the source of much c o n f l i c t and stress i n both short-term and long-term divorced f a m i l i e s . The community divorce r e f e r s to the adjustments required by divorced persons as friendships, extended family r e l a t i o n s h i p s , and other s o c i a l network r e l a t i o n s h i p s change and perhaps end. Included i n t h i s divorce would be r e l a t i o n s h i p s with former in-laws and with grandchildren (Weingarten, 1988) . The reshaping of these r e l a t i o n s h i p s can mean enormous upheaval f o r persons who are embedded i n extensive s o c i a l and family t i e s which have developed over many years. Bohannan (1970) notes that just as people change communities when they marry, they also change communities when they divorce. Because divorce i s not as common i n mid- and l a t e r - l i f e , these d i v o r c i n g adults may lack the peer support that other adults may have when going through t h i s t r a n s i t i o n . However, these older parents may turn to t h e i r adult chi l d r e n , children-in-law and grandchildren f o r support and continued l i f e meaning (Ahrons & Rddgers, 1987; Weingarten, 1988) . The psychic divorce r e f e r s to the issues d i v o r c i n g persons face as they r e - e s t a b l i s h (or perhaps, i n the case of older women e s p e c i a l l y , establish) t h e i r i n d i v i d u a l autonomy. Aft e r separation and divorce, the psychic task i s to become a whole, complete, autonomous i n d i v i d u a l a f t e r being part of a couple f o r a shorter or longer period of time. Bohannan (1970) viewed the psychic divorce as being the most d i f f i c u l t of the divorces, but also as having the p o t e n t i a l to be the most constructive. 21 A c r u c i a l issue i n the psychic divorce i s to come to terms with the loss of the marital r o l e i d e n t i t y of "spouse." The longer a couple has been married, the more deeply the i d e n t i t y of "spouse" has become ingrained i n a person's self-concept. At the same time, when a couple have c h i l d r e n , the i d e n t i t y of parent i s also a s i g n i f i c a n t part of the parent's self-concept, and the i d e n t i t y of parent also becomes more i n t e g r a l to the self-concept over time. Weingarten (1988) found i n her study of Divorce-after-Sixty that one source of a f f i r m a t i o n of i d e n t i t y for her respondents was t h e i r adult c h i l d r e n and grandchildren. The s i x t h divorce, and the one that Bohannan (1970) considered to be the most l i k e l y source of continuing pain, i s the coparental divorce. While c h i l d l e s s couples i n c o n f l i c t have the option of discontinuing a l l contact, couples with c h i l d r e n have continuing parental r e s p o n s i b i l i t i e s , and thus are faced with the task of renegotiating t h e i r parental roles while detaching from t h e i r marital r o l e s . This can be a huge task for couples who were not able to resolve t h e i r c o n f l i c t s while married (Peck, 1989; Wallerstein & Kelly, 1980) . A complicating f a c t o r i s that i n the coparental divorce many of the unresolved issues of the other "divorces" get played out (Ahrons & Rodgers, 1987), and the contact required for a continuing parental r e l a t i o n s h i p can be an ongoing reminder of the grievances associated with the divorce. It i s i n the coparental divorce that a c l e a r d i f f e r e n c e emerges between divorces involving minor c h i l d r e n and divorces i n v o l v i n g adult c h i l d r e n . Parents of minor c h i l d r e n may f i n d t h e i r c h i l d r e n to be a source of l i f e meaning and support, but they are s t i l l faced with issues of custody, access, c h i l d support and coparenting (see Kitson, Babri, Roach, & P l a c i d i , 1989; Pledge, 1992). Parents of adult c h i l d r e n , 22 however, have had many years to b u i l d up strong and meaningful t i e s with t h e i r adult c h i l d r e n and the strength of these t i e s may form a major part of t h e i r i d e n t i t y . When parents with grown c h i l d r e n divorce, t h e i r adult c h i l d r e n represent a p o t e n t i a l resource unavailable to younger parents. Weingarten (1988) found that her respondents (primarily women) perceived the r e l a t i o n s h i p with t h e i r c h i l d r e n as c r i t i c a l to personal well-being. They i d e n t i f i e d the supportiveness of t h e i r c h i l d r e n as being one of the most important factors i n f l u e n c i n g t h e i r adjustment. Bohannan's conceptualization of divorce i s us e f u l because i t captures the complexity of the divorce process, and also subdivides i t into s i x dimensions that can be studied and analyzed independently (see Weingarten, 1988) . By looking at each of the "divorces" separately, i t can be seen that they can present d i f f e r e n t challenges f o r older and younger adults and for men and women. Congruent with the assumption that adjustment to divorce i n l a t e r l i f e would be extremely d i f f i c u l t (Berardo, 1982; Price & McKenry, 1988), the psychic and community divorces would l i k e l y be more d i f f i c u l t f o r adults ending a long-term marriage. However, the l e g a l and economic divorces are l i k e l y d i f f i c u l t for d i v o r c i n g persons of any age or l i f e stage, and the emotional and coparental divorces would l i k e l y be less d i f f i c u l t f o r older adults and for parents of adult c h i l d r e n . Bohannan's s i x "divorces," therefore, suggest that divorce at any age w i l l i n i t i a t e a serie s of psychological adjustments, and that these adjustments may be d i f f e r e n t f o r adults divorcing a f t e r a long-term marriage, but not n e c e s s a r i l y more d i f f i c u l t . Some of our knowledge about the e f f e c t s of mid- or l a t e r - l i f e divorce on psychological well-being can be gleaned from studies that compare younger and older respondents within one sample, or across 23 samples. The mixed r e s u l t s from these studies lend credence to the suggestion that many adults have the resources to e f f e c t i v e l y cope with a divorce a f t e r a long-term marriage (see Ahrons & Rodgers, 1987) . The purpose of t h i s study i s not to compare the postdivorce psychological well-being of older and younger adults. However, because the research l i t e r a t u r e on m i d l i f e and l a t e r l i f e divorce i s extremely lim i t e d , one goal of t h i s l i t e r a t u r e review i s to place the phenomenon of psychological adjustment to divorce a f t e r a long-term marriage within the context of psychological adjustment to divorce i n general. Within t h i s context, some studies have reported differences i n postdivorce psychological well-being between younger and older respondents. A. Dimensions of Psychological Well-Being An attempt has been made to present the findings of the studies i n t h i s s ection i n the same format as those i n the previous section. However, because t h i s body of l i t e r a t u r e i s very small, r e s u l t s from studies reporting age differences and those reporting gender differences w i l l be integrated for each of the three dimensions reviewed previously: happiness, psychological d i s t r e s s , and self-esteem. 1. Happiness Research on the e f f e c t s of l a t e r divorce on s e l f - r e p o r t s of happiness has y i e l d e d mixed r e s u l t s . Chiriboga (1982), Chiriboga, Roberts, and Stein (1978) and Bloom, White, and Asher (1979) found recently divorced older adults to be less happy than recently divorced younger adults; and Chiriboga and colleagues (1978) note that, although national surveys have indicated a small increase by age of people f e e l i n g not too happy (e.g. Bradburn, 1969,- Bradburn & Caplovitz, 1965), older 24 divorced persons i n t h e i r study reported l e v e l s of unhappiness considerably exceeding the national averages. In contrast, Hammond and Muller (1992), using NSFH data, compared younger respondents (under 50 and curr e n t l y separated or divorced) and older respondents (over 50, separated or divorced past the age of 50 and cur r e n t l y separated or divorced). Results show no s i g n i f i c a n t differences between younger and older respondents for o v e r a l l happiness. S i m i l a r l y , Gander (1991) compared two samples of recently divorced persons: one sample being composed of persons under 50 and the second of persons 50 and over. Using the General Weil-Being Scale (Fazio, 1977, c i t e d i n Gander, 1991), she found that the two groups d i d not d i f f e r i n t h e i r o v e r a l l scores of emotional well-being. On the ca t e g o r i c a l scores, however, where there were differences, the older persons scored higher. For example, the older group scored higher on good s p i r i t s , happiness, s a t i s f a c t i o n and firm c o n t r o l . F i n a l l y , Gove and Shin (1989), i n a study of everdivorced adults, also used a composite measure of psychological well-being and found that older divorced adults reported higher l e v e l s of psychological well-being than d i d younger divorced adults, while on the dimension of happiness there were no s i g n i f i c a n t differences by age. With respect to gender differences, Connidis and McMullin (1992), i n t h e i r sample of everdivorced respondents 55 and over, found the ef f e c t s of divorce on avowed happiness to be negative and s i g n i f i c a n t f o r men and nonsignificant f or women. In addition, they found divorced men to be less happy than married men, but no corresponding differences f or women. The findings from t h i s study, therefore, show the e f f e c t s of 25 divorce on avowed happiness to be stronger f o r older men than for older women. Consistent with t h i s f i n d i n g , Hammond and Muller (1992) found older divorced women (50 and older) to report higher l e v e l s of happiness than older divorced men, and Chiriboga (1982) found i n his study that 35% of the separated/divorced men i n t h e i r 40s and 60% i n t h e i r 50s+ reported being "not too happy," while the comparable figures for women were 25% and 50%. 2. P s y c h o l o g i c a l D i s t r e s s S i m i l a r to the r e s u l t s for happiness, research on the e f f e c t s of l a t e r divorce on s e l f - r e p o r t s of depression has also y i e l d e d mixed r e s u l t s . While two recent studies found that younger and older divorced respondents d i d not d i f f e r on s e l f - r a t i n g s of depression (Gander, 1991; Kitson & Holmes, 1992), Roach and Kitson (1989) i n t h e i r study of recently divorced and recently widowed women found that older divorced women reported greater psychological well-being ( i . e . less depressive symptoms) than younger divorced women. Simi l a r to these r e s u l t s found by Roach and Kitson (1989), Gove and Shin (1989), i n a study of everdivorced adults, found an inverse r e l a t i o n s h i p between s e l f - r e p o r t s of psychological d i s t r e s s and age; older divorced adults i n t h e i r study rated themselves as less d i s t r e s s e d p s y c h o l o g i c a l l y than younger divorced adults. Longitudinal studies that report r e s u l t s f o r younger and older women suggest that older women may experience more depressive symptoms i n i t i a l l y than younger women, but that they may also recover more quickly. Wallerstein and K e l l y (1980), for example, found older women to be s i g n i f i c a n t l y more depressed than younger women at Time 1 (about s i x 26 months postseparation). However, by Time 2 (about 18 months postseparation), there were no s i g n i f i c a n t differences i n s e l f - r a t i n g s of depression between younger women and older women. Wallerstein and Kelly's (1980) f i n d i n g i s echoed by Chiriboga and colleagues (1991) who comment that the middle-aged women i n t h e i r study were "doing better" (p. 281) than younger women by Time 2 (around three years postseparation; no s t a t i s t i c s are given; measurement included happiness and depressive symptoms). The researchers suggest that t h i s i s " i n part because the l a t t e r were often faced with the sometimes c o n f l i c t i n g demands of r a i s i n g c h i l d r e n and reentering the world as a singl e person" (p. 281). With respect to gender differences, Connidis and McMullin (1993) found the e f f e c t s of divorce on s e l f - r e p o r t s of depression to be s i m i l a r to the e f f e c t s of divorce on avowed happiness; p o s i t i v e and s i g n i f i c a n t for men (i.e higher s e l f - r a t i n g s of depression) and nonsignificant for women. While divorced men rated themselves as s i g n i f i c a n t l y more depressed than married men, divorced women did not rate themselves as more depressed than married women. 3 . S e l f - E s t e e m In general, there appear to be no differences i n self-esteem between younger divorced adults and older divorced adults. Kitson and Holmes (1992) and Gove and Shin (1989) found no s i g n i f i c a n t age differences i n self-esteem ratings by t h e i r respondents. Chiriboga and colleagues (1991) found older separated/divorced persons to be higher i n self-esteem than younger separated/divorced persons. 27 With respect to gender differences, Gove and Shin (1989) also found no s i g n i f i c a n t gender differences i n the e f f e c t s of divorce on the s e l f -esteem of younger and older respondents. B. Summary As noted i n the introduction to t h i s section, research findings on age differences i n divorce adjustment are mixed. E a r l i e r c r o s s - s e c t i o n a l research shows more negative e f f e c t s of divorce on older respondent's reports of happiness than on younger respondent's, while more recent studies show no differences or higher ratings f o r older respondents, and cros s - s e c t i o n a l r e s u l t s f o r depression show e i t h e r no age differences or older respondent's reporting lower l e v e l s of depression. Longitudinal studies indicate that older women may be i n i t i a l l y more di s t r e s s e d than younger women, but that they recover more quickly. There do not appear to be any age differences i n s e l f - r a t i n g s of self-esteem. Regarding gender differences, the e f f e c t s of divorce on happiness have been found to be stronger f o r older men than f o r older women, and may also be stronger f o r depression. Again there appear to be no gender differences i n self-esteem. I I I . Parent-Adult C h i l d Relationships and Psychological Well-Being A. Parent-Adult C h i l d Relationships One of the most profound changes that increased longevity i s bringing to North American fa m i l i e s i s the extended period of time i n which men and women are parents of adult c h i l d r e n (Rossi & Rossi, 1990) . Hagestad (1987) has noted that i n most of the research on parents and children, " c h i l d " has been defined c h r o n o l o g i c a l l y (hence "adult child") rather than as a family r o l e that continues as long as there i s a 28 surviving parent. The use of the rather awkward term, "adult c h i l d , " i s i t s e l f an i n d i c a t i o n of the newness of t h i s evolving period i n the l i v e s of familes (Hagestad, 1981). Apart from the marital bond, parent-child r e l a t i o n s h i p s for most people are t h e i r primary kinship t i e s (Rossi & Rossi, 1990), and throughout the l i f e course the par e n t - c h i l d r e l a t i o n s h i p remains of central s i g n i f i c a n c e to both parents and chil d r e n (Barnett et a l . , 1992; Eggebeen, 1992; Rossi & Rossi, 1990; T r o l l , 1982). For most people the parent-adult c h i l d r e l a t i o n s h i p i s characterized by frequent i n t e r a c t i o n , a f f e c t i v e closeness, s a t i s f a c t i o n and low le v e l s of c o n f l i c t , plus high l e v e l s of s o c i a l support and help exchange (Mancini & Blieszner, 1989; Rossi & Rossi, 1990,- T r o l l , 1982) . Gender of both parent and c h i l d have been found to be key variables i n p a r ent-child r e l a t i o n s h i p s (Bengtson et a l . , 1990; Rossi & Rossi, 1990; T r o l l , 1989). Women have been c a l l e d the family "kinkeepers," (Rosenthal, 1985) because of t h e i r r o l e i n maintaining family t i e s , and family "monitors" (Hagestad et a l . , 1984) because of t h e i r r o l e i n monitoring family r e l a t i o n s h i p s . Women are the connecting l i n k s i n the extended family (Rossi & Rossi, 1991). Rossi and Rossi (1990) argue that s o c i o l o g i s t s and psychologists have underestimated the strengths and a b i l i t i e s of women ( p a r t i c u l a r l y older women) by fusing the wife r o l e and the mother r o l e . They suggest that This fusion of parent and spouse roles encourages the associ a t i o n of strength, reserve, and dominance with the family r o l e s of men, and subordination, emotionality, and dependence with the family roles of women, y i e l d i n g a f u n c t i o n a l l y complementary balance between the expressive wife-mother and the instrumental husband-father. (p. 14). 29 However, as Rossi and Rossi continue, once the wife r o l e i s d i f f e r e n t i a t e d from the mother r o l e , i t seems c l e a r that i t i s not as mothers but as wives that many women are dependent, timid, and submissive, while . . . as mothers women have to be and are active, productive, and strong, (p. 14). F i n a l l y , Hagestad and Smyer (1982) concluded that "women dis p l a y impressive strength i n an area which seems to be the men's A c h i l l e s ' heels: interpersonal bonds to family members, across generations" (p. 9). Within f a m i l i e s , the strongest intergenerational bonds have been found between mothers and t h e i r daughters (Rossi & Rossi, 1990; Rossi & Rossi, 1991; T r o l l , 1986). The salience of gender i n family r e l a t i o n s i s shown c l e a r l y i n the study by Rossi and Rossi (1990) . Throughout t h e i r analyses they found t i e s among women to be stronger, more frequent, more r e c i p r o c a l and less contingent on circumstances than those of men. Overall , mothers have been found to have strong and stable r e l a t i o n s h i p s with both adult daughters and adult sons (Rossi & Rossi, 1990). The Rossis theorize that across the l i f e course, women have a greater developmental stake i n maintaining close r e l a t i o n s with t h e i r parents, s i b l i n g s , and adult c h i l d r e n than men do, because of the much higher p r o b a b i l i t y that they w i l l need family help and support at some point and that others w i l l need help from them, i . e . there i s anticipated interdependency. From t h i s perspective, the more intensive investment women make i n parenting may pave the way for greater r e c i p r o c i t y i n exchanges with c h i l d r e n i n m i d l i f e and a "payoff" (p. 458) i n l a t e r l i f e when women are widowed or otherwise f i n d themselves i n need. In contrast to the strength of mother-child r e l a t i o n s h i p s , Rossi and Rossi (1990) note that one of t h e i r most s t r i k i n g findings i s the v u l n e r a b i l i t y of father's r e l a t i o n s h i p s with t h e i r adult c h i l d r e n . In 30 t h e i r r o l e s as fathers, sons, and grandfathers, the men i n Rossis' study were found to hold a more precarious p o s i t i o n within the family than d i d the women i n t h e i r roles as mothers, daughters and grandmothers. The Rossis comment that t h i s conditional q u a l i t y to men's family r e l a t i o n s h i p s can be seen i n almost a l l the regression equations, with higher R square values i n the equations dealing with men's re l a t i o n s h i p s with parents and c h i l d r e n than those dealing with women's re l a t i o n s h i p s with parents and children, thus "emphasizing i n a dramatic way the more contingent q u a l i t y of men's t i e s to the family than women's" (p. 499). B. Parent-Adult C h i l d Relationships and Psychological Weil-Being Close, supportive r e l a t i o n s h i p s between parents and t h e i r adult c h i l d r e n have been shown to have a p o s i t i v e e f f e c t on the psychological well-being of both generations. In an o f t e n - c i t e d 1983 study of older parents, Quinn found health to be the strongest p r e d i c t o r of psychological well-being and q u a l i t y of the p a r e n t - c h i l d r e l a t i o n s h i p to be the next strongest pr e d i c t o r and i n a 1992 three-generational study of Anglo/European and L a t i n American women, L e v i t t , Guacci, and Weber found r e l a t i o n s h i p q u a l i t y to be p o s i t i v e l y associated with well-being for each generation. The l i n k between parent-adult c h i l d r e l a t i o n s h i p s and parental well-being i s perhaps most f o r c e f u l l y i l l u s t r a t e d by a recent study conducted by S i l v e r s t e i n and Bengtson (1991) . These researchers found that close parent-child r e l a t i o n s increased s u r v i v a l time among parents who had experienced a s o c i a l loss (death of a c h i l d , parent or spouse,-divorce from a spouse: the measure was dichotomized into loss or no l o s s ) . Results showed t h i s b u f f e r i n g e f f e c t to be stronger f o r the type of loss that was less l i k e l y to occur (e.g. death of a spouse compared to 31 death of a parent; t h e i r sample of divorced respondents was too small to analyze separately). S i l v e r s t e i n and Bengtson concluded that the mortal health r i s k s associated with the stress of experiencing a s i g n i f i c a n t s o c i a l loss can be p a r t i a l l y o f f s e t by close r e l a t i o n s h i p s with adult c h i l d r e n . They suggest that the heightened f e e l i n g s of s e c u r i t y engendered by these r e l a t i o n s h i p s and the symbolic or instrumental interventions by adult c h i l d r e n promote the well-being of parents. These r e s u l t s are also consistent with research which i s showing a l i n k between close r e l a t i o n s h i p s and immunological functioning (Kiecolt-Glaser, Fisher, Ogrocki, Stout, Speicher & Glaser, 1987; see also Kitson & Morgan, 1990) . S i l v e r s t e i n and Bengtson concluded that i t i s the parents' perceptions of t h e i r r e l a t i o n s h i p s which are consequential f o r t h e i r well-being, regardless of whether or not the parents' perceptions are congruent with t h e i r children's perceptions. In coming to t h i s conclusion they r e f e r to the symbolic i n t e r a c t i o n i s t premise that s i t u a t i o n s experienced as r e a l by ind i v i d u a l s have r e a l consequences (Thomas & Znaniecki, 1927). In summary, research evidence shows that p a r e n t - c h i l d r e l a t i o n s h i p s are important to parents across the l i f e course, and that these r e l a t i o n s h i p s can have a s i g n i f i c a n t impact on psychological well-being. It also appears that parent-adult c h i l d r e l a t i o n s h i p s may become p a r t i c u l a r l y s a l i e n t i n times of family c r i s i s , such as when parent or c h i l d loses a spouse/partner through death or divorce. 32 IV. Divorce, Parent-Adult C h i l d Relationships and Psychological Well-Being A. Long-term E f f e c t s of E a r l i e r Divorce on Parent-Adult C h i l d Relationships and Parent's Psychological Weil-Being Patterns of postdivorce parenting established when ch i l d r e n are young appear to have long-term e f f e c t s on parent-child r e l a t i o n s : the few studies which have examined long-term e f f e c t s of divorce on parent-adult c h i l d r e l a t i o n s h i p s have been showing negative e f f e c t s f o r fathers and adult children, and nonsignificant (but generally s l i g h t l y negative) e f f e c t s f o r mothers and adult c h i l d r e n . For example, fathers who were divorced report s i g n i f i c a n t l y less contact and less support exchange with t h e i r adult c h i l d r e n than continuously-married fathers, but the reports of divorced mothers f or contact and support exchange do not d i f f e r s t a t i s t i c a l l y from those of continuously-married mothers (Bulcroft & Bulc r o f t , 1991; Bumpass & Sweet, 1991; Cooney & Uhlenberg, 1990) . A recent study by Aquilino (1994a) from the adult c h i l d ' s perspective, which investigated the e f f e c t of custody arrangements on the re l a t i o n s h i p s of young adults (19-34) with t h e i r parents, h i g h l i g h t s the key r o l e that custody arrangements have on parent-adult c h i l d r e l a t i o n s h i p s . Aquilino's study i s noteworthy because i t includes both custodial and noncustodial mothers and fathers. His r e s u l t s show, for example, that adult c h i l d r e n whose mothers had custody reported l i t t l e d i f f e r e n c e i n r e l a t i o n s h i p q u a l i t y , geographic distance and contact with t h e i r mothers as compared to adult c h i l d r e n i n in t a c t f a m i l i e s , but reported lower r e l a t i o n s h i p q u a l i t y , less contact and greater geographic distance from t h e i r fathers. In contrast, adult c h i l d r e n whose fathers 33 had custody reported l i t t l e difference i n distance and contact and higher r e l a t i o n s h i p q u a l i t y with t h e i r fathers as compared to adult c h i l d r e n i n int a c t f a m i l i e s , but reported greater geographic distance, a trend to less contact, and no differences i n r e l a t i o n s h i p q u a l i t y with t h e i r mothers. The differences i n father-adult c h i l d q u a l i t y and contact between father-custody f a m i l i e s and mother-custody f a m i l i e s were very large, and Aquilino notes that one of the most s t r i k i n g findings i n t h i s study i s the enormous impact that custody has on fathers' long-term r e l a t i o n s with t h e i r c h i l d r e n following divorce. Recent research from both the adult c h i l d ' s and the parent's perspectives, however, indicates that the e f f e c t s of divorce on parent-c h i l d r e l a t i o n s h i p s vary by the timing of the divorce. Aquilino (1994a), for example i n his study from the adult c h i l d ' s perspective, found the age of the c h i l d at separation from noncustodial parents to be p o s i t i v e l y r e l a t e d to r e l a t i o n s h i p q u a l i t y and contact and negatively r e l a t e d to geographic distance, i . e . the older the c h i l d was at the time of separation/divorce, the higher the r e l a t i o n s h i p q u a l i t y , the greater the contact, and the less the geographic distance between adult c h i l d r e n and noncustodial parents. Research conducted from the perspective of the divorced parent i s showing a s i m i l a r pattern of r e s u l t s . In several recent studies, e a r l i e r marital d i s r u p t i o n has been found to have a negative e f f e c t on i n t e r a c t i o n with adult c h i l d r e n (Bulcroft & Bu l c r o f t , 1991; Bumpass & Sweet, 1991; Cooney & Uhlenberg, 1990; Crimmins & Ingegneri, 1990). Thus from the parent's perspective as well as from the adult c h i l d ' s , timing of the divorce appears to have a s i g n i f i c a n t e f f e c t on parent-adult c h i l d r e l a t i o n s h i p s . This appears to be e s p e c i a l l y true f o r fathers, i . e . the 34 e a r l i e r i n the family l i f e cycle the separation and divorce occur, the more negative the e f f e c t s on subsequent f a t h e r - c h i l d r e l a t i o n s (Bulcroft & B u l c r o f t , 1991; Bumpass & Sweet, 1991; Cooney & Uhlenberg, 1990). Aq u i l i n o (1994b) recently summed up the e f f e c t s of e a r l i e r marital d i s r u p t i o n on parent-adult c h i l d r e l a t i o n s h i p s as follows: parental separation and divorce r e s u l t i n less instrumental and economic support exchange, less emotional support exchange, lower parent-child contact, greater geographic distance between parent and c h i l d r e n , and lower perceived r e l a t i o n s h i p q u a l i t y with adult c h i l d r e n . The family d i s r u p t i o n e f f e c t s are stronger for fathers than f o r mothers . . . (pp. 909, 910) . There has also been l i t t l e research attention paid to the e f f e c t that parent-adult c h i l d r e l a t i o n s h i p s may have on the psychological well-being of parents who divorced e a r l i e r i n l i f e . In a recent study using NSFH data, Marks (1995) examined the e f f e c t s of marital status and support exchange with adult c h i l d r e n on psychological well-being (happiness, depression, self-esteem). Regarding support exchanges, she found that separated/divorced m i d l i f e fathers were s i g n i f i c a n t l y more l i k e l y than fathers i n f i r s t marriages to report neither g i v i n g nor g e t t i n g instrumental support during the past month (72.3% compared to 40.1%) and more l i k e l y to report neither g i v i n g nor g e t t i n g emotional support (61.5% to 48.3%). Separated/divorced mothers were also more l i k e l y than mothers i n f i r s t marriages to report neither g i v i n g nor g e t t i n g instrumental support (45.7% compared to 42.4%) and s i g n i f i c a n t l y more l i k e l y to report neither g i v i n g nor getting emotional support (48.4% to 38.8%). Over a l l , Marks found marital status to be a stronger predictor of psychological well-being than support exchanges. However, for mothers ( o v e r a l l ) , support exchanges with adult c h i l d r e n had a greater e f f e c t on self-esteem than d i d marital status. 35 Umberson (1992) also examined marital status, gender, contact and support. She found that divorced parents reported less frequent contact with c h i l d r e n aged 16 and older, less s o c i a l support from children, and more r e l a t i o n s h i p s t r a i n and parental d i s s a t i s f a c t i o n , while mothers (overall) reported more contact, more s o c i a l support from c h i l d r e n and less parental d i s s a t i s f a c t i o n than fathers. O v e r a l l , she found that r e l a t i o n s h i p s t r a i n and parental d i s s a t i s f a c t i o n were more strongly-r e l a t e d to parent's psychological d i s t r e s s than were contact and s o c i a l support. Contact with children, however, was more strongly r e l a t e d to well-being f o r divorced parents than for married parents ( s t a t i s t i c s are not given). B. Recent Divorce, Parent-Adult C h i l d Relationships and Psychological Well-Being When parents of adult c h i l d r e n divorce i n m i d l i f e , both mothers and fathers have had many years to develop close r e l a t i o n s h i p s with t h e i r sons and daughters. Custody and i t s constraints are no longer an issue, and adult c h i l d r e n have greater autonomy i n determining the shape the parent-adult c h i l d r e l a t i o n s h i p w i l l take. Although recent research from the adult c h i l d ' s perspective i s showing that divorce l a t e r i n l i f e may have s i m i l a r e f f e c t s on parent-child r e l a t i o n s h i p s to divorce e a r l i e r i n l i f e , research findings are mixed. In Aquilino's 1994 cross-sectional study (1994b) using NSFH data, respondents were young adults 19-34 whose parents had divorced a f t e r the c h i l d r e n were 18. Regression r e s u l t s show that the parents' divorce had a negative e f f e c t on parent-adult c h i l d r e l a t i o n s h i p s . Respondents with divorced parents reported s i g n i f i c a n t l y lower p a r e n t - c h i l d r e l a t i o n s h i p q u a l i t y than d i d respondents with nondivorced parents, s i g n i f i c a n t l y 36 lower contact, and greater geographic distance from divorced fathers but not from divorced mothers. These e f f e c t s were stronger f o r f a t h e r - c h i l d r e l a t i o n s h i p s than for mother-child r e l a t i o n s h i p s , and stronger for father-daughter than for father-son. With regard to support exchanges, Aquilino found that adult c h i l d r e n reported no differences i n the support they gave to t h e i r divorced parents, but sons reported r e c e i v i n g less support than sons i n i n t a c t f a m i l i e s (there were no s i g n i f i c a n t d ifferences for daughters). In Cooney's 1994 cross-sectional study, respondents were young adults 18-23 whose parents had divorced within the previous f i f t e e n months. Regression r e s u l t s show negative e f f e c t s on f a t h e r - c h i l d r e l a t i o n s but not on mother-child r e l a t i o n s . Respondents whose parents had divorced reported s i g n i f i c a n t l y less contact with t h e i r fathers than did respondents from in t a c t f a m i l i e s ; and daughters, but not sons, reported s i g n i f i c a n t l y less intimacy with t h e i r fathers. There were no differences i n contact and intimacy with mothers. Cooney also found that intimacy and i n t e r a c t i o n were more strongly c o r r e l a t e d i n divorced f a m i l i e s than i n i n t a c t f a m i l i e s . She suggests t h i s may indicate the family r e l a t i o n s h i p s become more voluntary a f t e r divorce. The seminal work i n the research area of m i d l i f e divorce and family r e l a t i o n s h i p s from the divorcing parent's perspective was conducted by Hagestad and her colleagues i n the l a t e 1970s and e a r l y 1980s. Respondents i n the studies had been l e g a l l y divorced for 12-18 months at the time of the interview. Gender differences are of p a r t i c u l a r i n t e r e s t i n these studies. Hagestad and Smyer (1982) discuss patterns of divorcing i n middle age i n terms of "orderly" and "disorderly" divorces (p. 164). In o r d e r l y divorces, the marital d i s s o l u t i o n follows the sequence found 37 i n the process of gett i n g married, i . e . the person experiences a psychological and interpersonal divorce p r i o r to the l e g a l one; i n di s o r d e r l y divorces, some or a l l of the r e l a t i o n s h i p "ceasings" are l e f t undone at the time of the divorce. Hagestad and Smyer c l a s s i f i e d 61/93 of the divorces as orderly, and found that between 40% and 50% of both men and women perceived that they had had t o t a l control over the divorce process; women were more l i k e l y to indicate p a r t i a l control ( i . e . they did not i n i t i a t e the divorce, but they were i n control of when and how the divorce would proceed), and a greater number of the men indicated no co n t r o l . Women indicated being aware of marital problems before men, and were more l i k e l y to discuss these problems with t h e i r c h i l d r e n (two-th i r d s of the women compared to one-quarter of the men). According to Hagestad and Smyer, divorce i n "our" culture (presumably the dominant North American culture) involves the loss of a s o c i a l l y - v a l u e d r o l e and the entry into a status which has few cl e a r r o l e d e f i n i t i o n s . Therefore, i f there i s order i n the t r a n s i t i o n , i t i s because the person divorcing has created i t , and i f there i s s o c i a l support, i t i s because the person has sought i t out. They conclude that middle-aged divorced men may be more f i n a n c i a l l y secure than middle-aged divorced women, but that the women may be more able to take control of the divorce process, create lee-time f o r themselves, and seek out t r a n s i t i o n a l support, p a r t i c u l a r l y i n s i t u a t i o n s where the divorce was not o r i g i n a l l y sought by them. In 1984, Hagestad, Smyer & Stierman (1984) published a study based on the same data c o l l e c t i o n as Hagestad & Smyer (1982) . Respondents i n t h i s study are described as being 43 men and 50 women who had been divorced between twelve and eighteen months previously when they were between the ages of 40 and 59, had been married at le a s t sixteen years 38 (the average was 25), and had at least one c h i l d sixteen years or older. Major findings center around gender dif f e r e n c e s . In addit i o n to the re s u l t s reported i n Hagestad and Smyer (1982), t h i s study reports the following f i n d i n g s : (1) women were more l i k e l y to inform the ch i l d r e n of the impending divorce, and were more l i k e l y to turn to c h i l d r e n f o r support before and a f t e r the divorce; (2) nearly 25% of the mothers named a c h i l d as the person most h e l p f u l during the worst part of the divorce process, compared to 5% of the fathers; (3) 90% of the mothers s a i d that r e l a t i o n s h i p s with t h e i r c h i l d r e n remained unchanged or improved since the divorce, compared to 58% of the fathers; (4) more than one-third of the fathers reported that t h e i r r e l a t i o n s h i p with at least one of t h e i r c h i l d r e n had deteriorated (no figures are given for mothers); (5) one-t h i r d of the fathers reported that divorce had made them less e f f e c t i v e as a parent, compared to 4% of mothers; and (6) 92% of mothers reported that t h e i r effectiveness as a parent was the same or improved (no figures are given f o r f a t h e r s ) . Hagestad, Smyer and Stierman conclude that f o r t h i s cohort of women, time and energy invested i n family r e l a t i o n s h i p s appears to pay o f f i n times of c r i s i s . These mothers seem to have an "unshakable f a i t h i n the strength of t h e i r t i e s with c h i l d r e n " (p. 256); they expect to be able to count on t h e i r children, and i n most cases t h e i r expectations are met. Weingarten (1988) conducted a q u a l i t a t i v e study of f i f t y - f i v e women and eight men, most of whom ( a l l but three men and three women) attended meetings of the support group Divorce-After-Sixty i n Ann Arbor, Michigan during a four-year period. It i s therefore l i k e l y that f o r most of the respondents the divorce was f a i r l y recent (however, at le a s t one of the respondents was involved i n the ongoing coordination of the group). The 39 majority of the respondents perceived the supportiveness of t h e i r c h i l d r e n to be one of the most important factors i n t h e i r postdivorce adjustment; and some said that t h e i r r e l a t i o n s h i p s with t h e i r c h i l d r e n were more p o s i t i v e a f t e r the breakup of t h e i r marriage. However, changes i n r e l a t i o n s h i p s with c h i l d r e n could s t i l l be perceived as problematic. When r e l a t i o n s h i p s t r a i n s were present, communication and the children's divided l o y a l t i e s were mentioned as areas of d i f f i c u l t y . The general perception was that r e l a t i o n s h i p s with t h e i r c h i l d r e n were c r i t i c a l to t h e i r personal well-being. Also, Weingarten found that respondents who focused on parental r o l e s a t i s f a c t i o n rather than loss of the spousal r o l e generally perceived themselves to be less d i s t r e s s e d than respondents who were s t i l l invested i n the spousal r o l e . Several recent studies have examined the e f f e c t s of divorce on support exchange between parents and t h e i r adult c h i l d r e n . These studies show that older divorced mothers depend more on support from t h e i r adult c h i l d r e n than do older divorced fathers. Hammond and Muller (1992), using NSFH data, compared younger respondents (under 50 and cu r r e n t l y separated or divorced) and older respondents (over 50, separated or divorced past the age of 50 and cu r r e n t l y separated or divorced). Looking at the older age group only, r e s u l t s show that the older women r e l i e d s i g n i f i c a n t l y more on c h i l d r e n for emotional support during the divorce process than did the older men. In addition, older divorced women rated themselves s i g n i f i c a n t l y higher than older divorced men on ind i c a t o r s of being a parent, caring for children, and o v e r a l l happiness. These r e s u l t s from Hammond and Muller's study suggest that older divorced women are more s a t i s f i e d with t h e i r r e l a t i o n s h i p s with t h e i r c h i l d r e n than are older 40 divorced men, and are happier and more s a t i s f i e d with t h e i r postdivorce l i v e s . Wright and Maxwell (1991) examined the adult c h i l d ' s r o l e as a provider of s o c i a l support to recently divorced parents. Respondents were 78 men and 152 women who had been married 19 or more years and had at least one c h i l d over the age of 18; questions were asked about the geographically c l o s e s t c h i l d . Results show that mothers were more l i k e l y than fathers to rank c h i l d r e n as the most h e l p f u l source of support (46.2% compared to 18.3%). Fathers were less l i k e l y than mothers to seek support from t h e i r c h i l d r e n and to receive i t . They were also more l i k e l y to mention that at least one c h i l d had severed t i e s with them and consequently of f e r e d no support at a l l (no figures are given). Overall, mothers received more support than fathers i n a l l four categories: socioemotional aid, services, advice, and f i n a n c i a l a i d . The adult c h i l d ' s divided l o y a l t i e s i s a rec u r r i n g theme i n the research on l a t e r divorce, and can be seen from both the c h i l d ' s perspective and from the parent's perspective. Cooney and colleagues (1986), for example, found that 67% of her young adult respondents worried about l o y a l t y c o n f l i c t s , and that women more frequently i d e n t i f i e d fathers as the sole target of t h e i r anger. Booth and Amato (1994) found that c o a l i t i o n s were us u a l l y between daughters and mothers, and i n a recent l o n g i t u d i n a l study, Cooney and colleagues (Cooney, Hutchinson & Leather, 1995) found that daughters are more l i k e l y than sons to report being c l o s e r to one parent than to the other. From the parent's perspective, Wright and Maxwell (1991) found "sidetaking" to be a s i g n i f i c a n t predictor of t o t a l support given to parents, and also found that more fathers than mothers reported that 41 c h i l d r e n had severed t i e s and therefore provided no support. Rossi and Rossi (1990) suggest that marital unhappiness causes a realignment of the bonds i n the family, with sons and daughters more l i k e l y to remain close to the mother than to the father. In a c l i n i c a l study which included both parents and adult c h i l d r e n as respondents, Jones and Jones (1993) found that when an adult c h i l d i d e n t i f i e s a parent at f a u l t f o r the marital c o n f l i c t , i t i s t y p i c a l l y the father and i n such cases, the parents u s u a l l y divorce (these r e s u l t s are not reported by gender of c h i l d r e n ) . Following a divorce, there i s evidence to suggest that r e l a t i o n s h i p s between parents and t h e i r adult c h i l d r e n can influence the parent's psychological adjustment to the divorce. Two recent studies by Gander (1991) and Gander and Jorgensen (1990) indicate that the current "closeness" between older divorced parents and t h e i r a d u l t - c h i l d r e n may be the most s i g n i f i c a n t p r e d i c t o r of the parent's psychological well-being and postdivorce adjustment. (Neither study provides information about how "closeness" was measured). In her 1991 study, Gander compared two samples of recently divorced persons: one sample being composed of persons under the age of 50, and the second of persons 50 and over who had been married at least f i f t e e n years. Looking at the older respondents only, 40% of the older respondents reported that r e l a t i o n s h i p s with t h e i r c h i l d r e n were "better" since the divorce, 45% that they were the "same," and 15% that they were "worse." Gander does not provide a breakdown of these figures by gender of parent or gender of c h i l d . Regression analyses show that degree of c o n f l i c t during the divorce process and current family closeness predicted general well-being for the older respondents; only c o n f l i c t a f t e r the divorce predicted general well-being for the younger respondents ( a l l of whom were 4 2 c u s t o d i a l parents: 185 mothers, 21 fa t h e r s ) . The older respondents reported experiencing s i g n i f i c a n t l y less c o n f l i c t a f t e r the divorce than the younger respondents. Gander also found that the older women and older men d i d not d i f f e r i n t h e i r General Well-Being scores. Gander and Jorgensen (1990) examined s o c i a l support and postdivorce adjustment i n a study which appears to have been based on the same data c o l l e c t i o n as Gander (1991) . In t h i s study, respondents (67 women and 44 men) are described as being divorced persons 50 years and older, separated up to 48 months, divorced within the past two years, married at least 15 years, with no more than two divorces. Approximately f o u r - f i f t h s of the respondents had been married once; the average length of marriage was 29.7 years. Information on whether or not a l l respondents are parents i s not given. Gander and Jorgensen used both the General Weil-Being Scale (GWB: Fazio, 1977, c i t e d i n Gander & Jorgensen, 1990), which assesses a respondent's subjective fe e l i n g s of well-being and d i s t r e s s , and the S o c i a l Adjustment Scale of Self-Report (SAS-SR: Weissman & Paykel, 1974), which assesses r o l e performance i n s i x areas over a two-week period, to determine which s o c i a l support variables are the best predictors of general well-being and postdivorce adjustment. Regression r e s u l t s show that, i n order of si g n i f i c a n c e , the four s o c i a l support variables that best predict general well-being are: (1) frequency of contact with a f r i e n d f o r help with personal problems during the past month (negative association; i . e . more contacts for help, less well-being) ,• (2) present closeness with c h i l d r e n (positive association) ; (3) gender (nine women were depressed compared to four men); and (4) number of people i n s o c i a l support network i n an emergency (positive a s s o c i a t i o n ) . The two s o c i a l support variables that were found to be the 43 best p r e d i c t o r s of postdivorce adjustment on the SAS-SR are: (1) present closeness with c h i l d r e n ; and (2) s a t i s f a c t i o n with number of friends (both associations are p o s i t i v e ) . Based on these congruent findings. Gander and Jorgensen conclude that the "closeness of older divorced persons with t h e i r c h i l d r e n i s the factor which appears to be most s a l i e n t i n i t s p r e d i c t i v e power for both p o s i t i v e emotional well-being and postdivorce adjustment" (p. 50). C. Summary The l i t e r a t u r e reviewed i n t h i s chapter has shown that, o v e r a l l , divorce has negative e f f e c t s on psychological well-being, that these e f f e c t s may attenuate over time, and that they vary by gender and by age. For those i n d i v i d u a l s who are parents, the patterns of postdivorce pa r e n t - c h i l d i n t e r a c t i o n which are established when c h i l d r e n are young have been found to have long-term e f f e c t s on parent-child r e l a t i o n s h i p s . Recent studies of parent-adult c h i l d r e l a t i o n s h i p s from the perspective of young adult c h i l d r e n who experienced marital d i s r u p t i o n during t h e i r childhood are showing patterns s i m i l a r to those found for younger ch i l d r e n , i . e . custody arrangements i n childhood are found to have s i g n i f i c a n t e f f e c t s on the r e l a t i o n s h i p s between parents and c h i l d r e n l a t e r i n l i f e , with the r e l a t i o n s h i p with the nonresidential father being most at r i s k . We know very l i t t l e about what happens to parent-adult c h i l d r e l a t i o n s h i p s when parents divorce a f t e r a long-term marriage and custody arrangements are no longer an issue. However, the few studies from the perspective of young adult c h i l d r e n whose parents had recently divorced are showing a s i m i l a r pattern to that found when parents divorced e a r l i e r 44 i n l i f e : f a t h e r - c h i l d r e l a t i o n s h i p s appear to be more negatively affected than mother-child r e l a t i o n s h i p s . Across the l i f e course, parent-child r e l a t i o n s h i p s are important to both parents and children, and the q u a l i t y and strength of these r e l a t i o n s h i p s i s consequential to the psychological well-being of both generations. When parents divorce a f t e r a long-term marriage, both parents have had the c h i l d ' s l i f e t i m e to develop close and supportive r e l a t i o n s h i p s . In times of c r i s i s , therefore, adult c h i l d r e n represent an important p o t e n t i a l resource f o r both parents. Rossi and Rossi (1990), however, have found much greater s t a b i l i t y i n mother-child r e l a t i o n s across time and circumstance than i n f a t h e r - c h i l d r e l a t i o n s . They suggest that women's greater investment i n t h e i r r e l a t i o n s h i p s with t h e i r c h i l d r e n pays o f f i n l a t e r l i f e when women are i n need. There has been very l i t t l e empirical research from the parents' perspective which examines the e f f e c t s of divorce a f t e r long-term marriage on parent-adult c h i l d r e l a t i o n s h i p s , and the subsequent e f f e c t of these r e l a t i o n s h i p s on the divorce adjustment of the parents. However, the i n d i c a t i o n i s that the parent-adult c h i l d r e l a t i o n s h i p influences the postdivorce psychological well-being of the parent, and that the q u a l i t y and strength of these r e l a t i o n s h i p s varies by gender of parent. Research from the young adults' perspective indicates that they also vary by gender of c h i l d . 45 .Chapter Three Research Problem I. T h e o r e t i c a l Perspective In recent years there has been a renewed i n t e r e s t i n using i d e n t i t y theory to explain differences i n parent-adult c h i l d r e l a t i o n s h i p s and the as s o c i a t i o n between these r e l a t i o n s h i p s and the psychological well-being of both generations (Atkinson, 1989; Marks, 1995; Mutran & Reitzes, 1984; Roberts & Bengtson, 1993; Umberson, 1992). The studies c i t e d do not measure the concepts of the theory d i r e c t l y , but do use i t s conceptual ideas to provide a framework for t h e i r research and findings. For example, Umberson (1992; c i t i n g Mutran & Reitzes, 1984) refers to the influence of the s o c i a l - s t r u c t u r a l context i n which the family i s embedded on parent-adult c h i l d r e l a t i o n s h i p s and t h e i r e f f e c t on the psychological well-being of both generations, and Marks (1995) suggests that the successful enactment of roles and i d e n t i t i e s w i l l enhance psychological well-being through s o c i a l esteem and i n t r i n s i c g r a t i f i c a t i o n . The NSFH also does not allow for the d i r e c t measurement of i d e n t i t y theory concepts such as r o l e - i d e n t i t i e s and t h e i r r e l a t i v e salience. The conceptual o r i e n t a t i o n provided by i d e n t i t y theory does, however, appear promising as a guiding perspective for work i n t h i s area. Consequently, although t h i s study i s not proposed as a test of i d e n t i t y theory, the t h e o r e t i c a l perspective presented by the theory w i l l be used i n combination with the extant research l i t e r a t u r e to develop hypotheses and to explain study f i n d i n g s . 46 A. T h e o r e t i c a l Concepts The main proponent of i d e n t i t y theory has been Sheldon Stryker (see Stryker, 1980; Stryker, 1981; Stryker, 1987; Stryker & Serpe, 1982) who adheres to a s o c i a l s t r u c t u r a l version of symbolic interactionism. According to t h i s perspective the " s e l f " , which i s shaped through s o c i a l i n t e r a c t i o n , guides and organizes behavior; s o c i a l i n t e r a c t i o n i n turn, however, i s shaped and constrained by s o c i a l structures which set the context for the i n t e r a c t i o n and hence influence who w i l l i n t e r a c t with whom i n what settings and with what resources. Two s o c i a l structures which are central i n i d e n t i t y theory are "positions" and " r o l e s . " Stryker (1980, p. 54) defines "positions" as " r e l a t i v e l y stable, morphological components of s o c i a l structure," and "roles" as the "shared behavioral expectations [carried by] these po s i t i o n s . " To c l a r i f y t h i s with an example: there are u n i v e r s a l s o c i a l p o s i t i o n s of wife and husband, but the roles of wife/husband are d i f f e r e n t i n d i f f e r e n t contexts (e.g. c u l t u r e s ) , i . e . the behavioral expectations are d i f f e r e n t . Identity theory seeks to explain only r o l e - r e l a t e d behavior, not a l l s o c i a l behavior (Stryker & Serpe, 1982). According to Stryker (1981), the " s e l f , " being a r e f l e c t i o n of a complex, d i f f e r e n t i a t e d yet organized society, must also be both d i f f e r e n t i a t e d and organized. He defines the " s e l f " as a "structure of i d e n t i t i e s r e f l e c t i n g roles played i n d i f f e r e n t i a t e d networks of i n t e r a c t i o n s " (Stryker, 1987), and " i d e n t i t i e s " as " i n t e r n a l i z e d p o s i t i o n a l designations that e x i s t insofar as the person p a r t i c i p a t e s i n structured role r e l a t i o n s h i p s " (Stryker, 1981, p. 23). Thus a person's " s e l f " might 47 include such i d e n t i t i e s (or r o l e - i d e n t i t i e s ) as spouse, parent, paid worker, f r i e n d , neighbor. Stryker postulates that i d e n t i t i e s are organized, or ranked, into salience h i e r a r c h i e s , with "salience" being defined as the " p r o b a b i l i t y of the various i d e n t i t i e s being invoked i n a given s i t u a t i o n or over many si t u a t i o n s " (1981, p. 24). The r e l a t i v e salience of various i d e n t i t i e s i n a salience hierarchy i s determined by the commitment a person has to s p e c i f i c i d e n t i t i e s , with commitment r e f l e c t i n g " i n t e r a c t i o n a l costs i n the form of r e l a t i o n s h i p s foregone were the person no longer to hold the positions and play the roles underlying the i d e n t i t i e s " (Stryker, 1987, p. 91). The basic premise of Stryker's i d e n t i t y theory i s that "commitment af f e c t s i d e n t i t y salience which, i n turn, a f f e c t s behavioral choices" (Stryker, 1981, p. 24). The idea i s that i f a person i s highly committed to a r o l e - i d e n t i t y , that r o l e - i d e n t i t y i s highly s a l i e n t , and the person w i l l be motivated to perform those behaviors which are congruent with personal and normative expectations f o r the performance of that p a r t i c u l a r r o l e - i d e n t i t y . B. Linking S o c i a l Structure, Role Relationships and Role-Identities According to Stryker (1987), i t i s through commitment that the s o c i a l structure enters into r o l e r e l a t i o n s h i p s , as the constraints on r o l e behavior and the choices a v a i l a b l e to i n d i v i d u a l s are influenced or shaped by s o c i o c u l t u r a l and socioeconomic variables such as age, gender, s o c i a l class and power r e l a t i o n s . These larger s o c i a l s t r u c t u a l systems, therefore, are an i n t e g r a l part of symbolic i n t e r a c t i o n i s t - b a s e d t h e o r i z i n g . 48 One of the strengths of i d e n t i t y theory i s t h i s l i n k that i t provides between the i n s t i t u t i o n a l , group and i n d i v i d u a l l e v e l s of analysis, i . e . the recognition that i n d i v i d u a l r o l e - i d e n t i t i e s and family i n t e r a c t i o n patterns are shaped and constrained by the ideologies and opportunity structures which e x i s t i n a given society. In North America, for example, the "motherhood mandate" (mothers have primary r e s p o n s i b i l i t y for family work) and "fatherhood mandate" (fathers have primary r e s p o n s i b i l i t y for the f i n a n c i a l support of t h e i r families) are reinforced by s t r u c t u r a l b a r r i e r s to employment equity for women and s t r u c t u r a l constraints to family involvement f o r men (Cohen, 1987; Lewis & O'Brien, 1987; Kruk, 1992) . C u l t u r a l l y dominant ideologies of "masculinity" and "femininity" also continue to constrain the ro l e behaviors of fathers and mothers (Lewis & O'Brien, 1987). Consequently, while research on work and the famiy continues to show that mothers spend more time doing family and household work than fathers, current fatherhood research shows that most men value t h e i r r o l e s as fathers highly and that t h e i r family roles impact t h e i r psychological well-being more than t h e i r work roles (e.g. Cohen, 1987; Pleck, 1985). When discussing gender roles and gender r o l e - i d e n t i t i e s within the context of family, researchers often r e f e r to G i l l i g a n ' s (1982) s e l f - i n -r e l a t i o n theory of morality (Cooney, 1994; Cooney et a l ; , 1986; Scott & Alwin, 1989; Thoits, 1992) . G i l l i g a n (1982) contends that through s o c i a l i z a t i o n and experience men tend to develop a sense of i d e n t i t y and s e l f as separate or independent i n r e l a t i o n to others (leading to a morality based on equality and justice) and that women tend to develop a sense of i d e n t i t y and s e l f as connected and interdependent (leading to a morality based on i n c l u s i o n and equity). Consequently she concluded that 49 "Women not only define themselves i n the context of r e l a t i o n s h i p but also judge themselves i n terms of t h e i r a b i l i t y to care" (1982, p. 17). To the extent that men or women define themselves as interdependent i n t h e i r r e l a t i o n s h i p s , t h e i r self-evaluations could be expected to be influenced by sa l i e n t family r e l a t i o n s h i p s . C. Linking Role Relationships, Role-Identities and Psychological Well-Being Identity theory provides us with t h e o r e t i c a l l i n k s between r o l e -i d e n t i t i e s , r o l e r e l a t i o n s h i p s and psychological well-being. Stryker o r i g i n a l l y hypothesized that "the greater the commitment, the higher the i d e n t i t y salience, the greater the impact of ro l e performance on s e l f -esteem w i l l be" (1981, p 24). While Stryker himself has not developed his ideas of an ass o c i a t i o n between i d e n t i t y salience and self-esteem or other dimensions of psychological well-being, other t h e o r i s t s have. Roberts and Bengston (1993), for example, state that i d e n t i t y theory "views i d e n t i t y structures as interfaces between s o c i a l experiences - such as p a r e n t - c h i l d r e l a t i o n s h i p s - and an in d i v i d u a l ' s psychological well-being" (p. 263). They suggest that the self-concept incorporates both a s e l f - d e f i n i t i o n a l dimension (I am a father/mother) and a se l f - e v a l u a t i v e dimension (I am a good father/mother). They assert that t h i s s e l f -evaluative process i s informed by a person's own evaluative b e l i e f s (I believe I am a good mother/father) and by the appraisals received from others as r o l e - i d e n t i t i e s are enacted (You are a good mom/dad). Pos i t i v e self-evaluations develop when ind i v i d u a l s believe that t h e i r r o l e performance i s appropriate and perceive that the r e l a t i o n s h i p s which sustain the r o l e - i d e n t i t y r e f l e c t p o s i t i v e appraisals of t h e i r r o l e performance (see Roberts & Bengtson, 1993). 50 The most thorough explanation of the l i n k between r o l e behaviors and psychological well-being from the perspective of i d e n t i t y theory has been developed by Peggy Thoits (1983, 1985, 1991, 1992) . According to Thoits (see esp. 1983, 1991), the key to the r e l a t i o n s h i p between the multiple s e l f (the s e l f being composed of multiple r o l e - i d e n t i t i e s ) and psychological well-being l i e s i n i d e n t i t y enactment (1983) . Thoits (1985) proposes that r o l e r e l a t i o n s h i p s and the associated r o l e enactments can be p s y c h o l o g i c a l l y b e n e f i c i a l i n three main ways. F i r s t , r o l e - i d e n t i t i e s provide i n d i v i d u a l s with a sense of e x i s t e n t i a l s e c u r i t y based i n ongoing i d e n t i t y enactment: i n d i v i d u a l s know who they are i n r e l a t i o n to others, know how to behave appropriately and, by being embedded i n a system of r e c i p r o c a l r e l a t i o n s h i p s , experience a sense of belonging which w i l l have a p o s i t i v e e f f e c t on the i n d i v i d u a l ' s psychological well-being (or perhaps even more importantly, d e f i c i t s i n these areas w i l l have a negative e f f e c t ) . Disruptions i n a valued r e l a t i o n s h i p , therefore, could be expected to have negative e f f e c t s on an i n d i v i d u a l ' s psychological well-being. Second, i m p l i c i t i n i d e n t i t y theory i s the idea that evaluations of r o l e performance (by s e l f and by others) w i l l be based on normative expectations for behavior i n a given r o l e . For example, f o r most people the parent-adult c h i l d r e l a t i o n s h i p i s characterized by frequent i n t e r a c t i o n and high l e v e l s of support exchange (Mancini & Blieszner, 1989; Rossi & Rossi, 1990) . Parents and adult c h i l d r e n who are involved i n p o s i t i v e , supportive r e l a t i o n s h i p s therefore would be l i k e l y to evaluate t h e i r r o l e enactments p o s i t i v e l y (and to perceive that others are evaluating them p o s i t i v e l y ) which, according to i d e n t i t y theory, would have 51 a p o s i t i v e e f f e c t on t h e i r psychological well-being (and conversely, negative evaluations would have a negative e f f e c t on well-being). Third, successful role performances can bring not only s o c i a l approval as r o l e expectations are met, but also i n t r i n s i c g r a t i f i c a t i o n and a sense of competence and self-esteem. If, for example, a parent i s motivated to perform supportive a c t i v i t i e s (e.g. instrumental or emotional support) and has the resources to carry out these a c t i v i t i e s , the successful enactment of these a c t i v i t i e s ( i . e . congruence between expectations and behavior) would have a p o s i t i v e e f f e c t on psychological well-being. Thus i d e n t i t y theory suggests avenues through which the parent-adult c h i l d r e l a t i o n s h i p could influence the postdivorce psychological well-being of parents. When couples divorce a f t e r a long-term marriage, the parent r o l e - i d e n t i t y and the parent-adult c h i l d r e l a t i o n s h i p are l i k e l y highly s a l i e n t components of the parent's i d e n t i t y . Following a divorce, the continuing parent-adult c h i l d r e l a t i o n s h i p could function as an affirmation of the parent's established sense of i d e n t i t y and also as a source of s e c u r i t y when the parent seeks to develop new r o l e - i d e n t i t i e s . P o s i t i v e r e f l e c t e d appraisals from the adult c h i l d r e n could also be expected to enhance the psychological well-being of the parent. 52 I I . Hypotheses Based on i d e n t i t y theory and the l i t e r a t u r e , three major hypotheses were developed. Hypothesis 1: Divorce a f t e r a long-term marriage w i l l have a negative e f f e c t on parent's psychological well-being. Previous research has shown that divorce has at least a short-term negative e f f e c t on psychological well-being as measured by happiness, depression and self-esteem. Longitudinal studies show that f o r most people, these negative e f f e c t s attenuate with time (Booth & Amato, 1991; Chiriboga et a l . , 1991; Kitson & Holmes, 1992). The author has found no l o n g i t u d i n a l studies of the e f f e c t s of recent divorce a f t e r long-term marriage on psychological well-being. There are t h e o r e t i c a l reasons to expect negative e f f e c t s of divorce on psychological well-being. When in d i v i d u a l s divorce a f t e r a long-term marriage, they disrupt a r o l e - i d e n t i t y (spouse) which has l i k e l y been a s a l i e n t part of t h e i r self-concept and which i s highly valued by society. Identity theory suggests that the loss of a highly s a l i e n t r o l e - i d e n t i t y and the i n t e r r u p t i o n of a long-standing source of i d e n t i t y a f f i r m a t i o n would have negative e f f e c t s on an i n d i v i d u a l ' s psychological well-being. This expectation i s supported by Bohannan's idea of the psychic divorce, which r e f e r s to the i d e n t i t y issues divorcing persons face as they re-e s t a b l i s h (or establish) t h e i r i n d i v i d u a l autonomy. The longer people have been married, the more d i f f i c u l t i t may be for them to s h i f t t h e i r i d e n t i t i e s away from the r o l e - i d e n t i t y of spouse. 53 Hypothesis 2: Divorce a f t e r a long-term marriage w i l l have d i f f e r e n t i a l e f f e c t s on mother-adult c h i l d r e l a t i o n s h i p s and father-adult c h i l d r e l a t i o n s h i p s . Hypothesis 2a: Divorce a f t e r a long-term marriage w i l l have a p o s i t i v e e f f e c t on mother-adult c h i l d r e l a t i o n s h i p s . For women now i n mid- or l a t e r - l i f e , e s p e c i a l l y those who have devoted themselves p r i m a r i l y to homemaking, the mother r o l e - i d e n t i t y and the mother-child r e l a t i o n s h i p have l i k e l y been cen t r a l to her sense of s e l f . The primacy of these mother-child bonds i s r e f l e c t e d i n the strong and stable r e l a t i o n s h i p s most mothers have with t h e i r c h i l d r e n throughout the l i f e course (Rossi & Rossi, 1990). Identity theory suggests that following a divorce a f t e r a long-term marriage, a mother would be motivated to perform those r o l e behaviors which would maintain and enhance her rel a t i o n s h i p s with her ch i l d r e n . Research by Hagestad, Smyer and Stierman (1984) supports t h i s expectation: 90% of the middle-aged mothers i n t h e i r study reported unchanged or improved r e l a t i o n s h i p s with t h e i r c h i l d r e n following a middle l i f e divorce (compared to 58% of the fa t h e r s ) . Middle-aged mothers may also invest i n the parent r o l e - i d e n t i t y i n compensation for the loss of the spouse r o l e - i d e n t i t y (Gecas & Seff, 1990; Mutran & Reitzes, 1984). Again research supports t h i s expectation: mothers who divorce l a t e r i n l i f e have been found to turn to t h e i r c h i l d r e n f or support (Hagestad et a l . , 1984; Hammond & Muller, 1992). Adult c h i l d r e n have also been found to respond to the mother's need for help and support (Wright & Maxwell, 1991). The research l i t e r a t u r e suggests that adult c h i l d r e n f e e l a r e s p o n s i b i l i t y towards t h e i r parents 54 and, i n the event of a divorce, f e e l a p a r t i c u l a r r e s p o n s i b i l i t y towards t h e i r mother (Bonkowski, 1989; Cooney, 1989; Cooney, Smyer, Hagestad & Klock, 1986; Wright & Maxwell, 1991). When a mother has invested heavily i n her r e l a t i o n s h i p s with her ch i l d r e n over the l i f e course, t h e i r support and p o s i t i v e r e f l e c t e d appraisals could be expected to have a p o s i t i v e e f f e c t on her evaluation of the mother-adult c h i l d r e l a t i o n s h i p s . Subtest 2a: The p o s i t i v e e f f e c t of divorce on mother-adult c h i l d r e l a t i o n s h i p s w i l l be greater for mother-daughter r e l a t i o n s h i p s than for mother-son r e l a t i o n s h i p s . There i s some research evidence showing that the e f f e c t s of divorce on parent-adult c h i l d r e l a t i o n s h i p s may be d i f f e r e n t f or daughters and for sons (Aquilino, 1994b; Cooney, 1994; Rossi & Rossi, 1990). However, there has not been enough empirical research conducted from the parent's perspective on which to base hypotheses. In view of t h i s , two subtests w i l l be conducted to examine whether postdivorce parent-daughter and parent-son r e l a t i o n s h i p s d i f f e r . Rossi and Rossi (1990), i n p a r t i c u l a r , have stressed that the mother-daughter bond i s the strongest of the four parent-child dyadic bonds. Cooney and her colleagues (1986), i n a study of recent l a t e r l i f e divorce from the young adult's perspective, found that 62% of the young women reported improved r e l a t i o n s h i p s with t h e i r mothers. The strength of c h i l d -parent bonds i n general i s shown by the figures for sons: 50% of the young men also reported improved re l a t i o n s h i p s with t h e i r mother. There i s also research evidence showing that adult daughters are more l i k e l y to take sides i n a divorce than are adult sons, and that when they do, they are more l i k e l y to take the mother's side (Cooney et a l . , 1986; Cooney et a l , 1995) . Adult c h i l d r e n may also perceive that the mother needs more care during and a f t e r the divorce than the father and, although both sons and daughters provide support for the mother, daughters provide more socioemotional support (Wright & Maxwell, 1991). Hypothesis 2b: Divorce a f t e r a long-term marriage w i l l have a negative e f f e c t on father-adult c h i l d r e l a t i o n s h i p s . In general, men's family r e l a t i o n s h i p s have been found to be more vulnerable across time and circumstance than women's (see Rossi & Rossi, 1990). As noted above, when parents divorce a f t e r a long-term marriage, adult c h i l d r e n may f e e l pressured to take sides and the research indicates that when t h i s occurs the c h i l d r e n are more l i k e l y to take the mother's side and that t h i s realignment of family bonds may occur before the divorce (Jones & Jones, 1993; Rossi & Rossi, 1990). In addition, fathers are f a r more l i k e l y to report l o s i n g touch with at least one of t h e i r c h i l d r e n than women are (Hagestad et a l . , 1984; Wright & Maxwell, 1991). These findings suggest that the mother and the adult c h i l d r e n may form c o a l i t i o n s during a marriage break-up, and the father may f i n d himself occupying a peripheral r o l e i n the family before a divorce occurs (Booth & Amato, 1991; Jones & Jones, 1993; Rossi & Rossi, 1990) . There are other ways i n which the father may f i n d himself at a disadvantage a f t e r a divorce. The geographical distance between divorced fathers and t h e i r adult c h i l d r e n i s greater than that between divorced mothers and t h e i r adult c h i l d r e n (Aquilino, 1994b), and geographical distance has been found to be the most s i g n i f i c a n t p r e d i c t o r of parent-adult c h i l d i n t e r a c t i o n and support (Rossi & Rossi, 1990) . Also, during 56 the marriage the mothers may have f a c i l i t a t e d the f a t h e r - c h i l d i n t e r a c t i o n , and consequently following a divorce the father and c h i l d r e n may have to develop new patterns of i n t e r a c t i o n . For example, mothers generally organize family r i t u a l s (e.g. birthdays, Christmas) and may continue to do so postdivorce. In a recent study of changes i n family r i t u a l s a f t e r a " l a t e - l i f e " divorce from the adult c h i l d ' s perspective, Pett and her colleagues (Pett, Lang & Gander, 1992) found that when the respondents reported the loss of parent's involvement i n family r i t u a l s , i t was e s p e c i a l l y f o r fathers (225 fathers compared to 54 mothers). Research from the adult c h i l d ' s perspective i s showing divorce to have negative e f f e c t s on adult children's evaluation of t h e i r r e l a t i o n s h i p s with t h e i r fathers (Aquilino, 1994b; Cooney, 1994). This research may be r e f l e c t i n g the d i f f i c u l t i e s adult c h i l d r e n have i n coping with c o n f l i c t i n g allegiances to both parents (see Cooney et a l . , 1995). From the perspective of i d e n t i t y theory, these negative r e f l e c t e d appraisals from the adult c h i l d r e n , whether temporary or long-term, could be expected to have a negative e f f e c t on the father's perception of the father-adult c h i l d r e l a t i o n s h i p . Subtest 2b: The negative e f f e c t of divorce a f t e r long-term marriage on father-adult c h i l d r e l a t i o n s h i p s w i l l be greater for father-daughter r e l a t i o n s h i p s than f o r father-son r e l a t i o n s h i p s . The extant l i t e r a t u r e indicates that the father-daughter r e l a t i o n s h i p i s most at r i s k following a divorce (Aquilino, 1994b; Booth & Amato, 1994; Cooney, 1994; Cooney et a l . , 1986; Rossi & Rossi, 1990). Studies by Aquilino (1994b) and Cooney (1994) from the adult c h i l d ' s perspective, for 57 example, show the negative e f f e c t s of m i d l i f e divorce on parent-adult c h i l d r e l a t i o n s h i p s to be greater for the father-daughter r e l a t i o n s h i p than for the father-son r e l a t i o n s h i p . In addition, although both adult sons and adult daughters may experience c o n f l i c t i n g l o y a l t i e s to t h e i r parents during a m i d l i f e divorce, there i s some i n d i c a t i o n that daughters may have greater d i f f i c u l t y remaining close to both parents (Cooney et a l . , 1986; Cooney et a l . , 1995). These negative r e f l e c t e d appraisals from daughters could be expected to negatively influence the father's evaluation of the father-daughter r e l a t i o n s h i p . From the perspective of i d e n t i t y theory, i t could also be argued that fathers with a t r a d i t i o n a l sex role o r i e n t a t i o n would be more committed to the r o l e - i d e n t i t y "father-of-a-son." It i s possible that the p o t e n t i a l l y greater number of shared meanings i n the same-sex parent-child r e l a t i o n s h i p would f a c i l i t a t e greater father-son bonding (and conversely, mother-daughter bonding). Hypothesis 3: Strong parent-adult c h i l d r e l a t i o n s h i p s w i l l moderate the negative e f f e c t s of divorce a f t e r long-term marriage on the psychological well-being of the parent. "Strong" parent-adult c h i l d r e l a t i o n s h i p s w i l l be operationalized i n t h i s study as high ratings of contact and support. Several recent studies of m i d l i f e and l a t e r l i f e divorce have indicated that r e l a t i o n s h i p s with adult c h i l d r e n may play a key r o l e i n the divorce adjustment of older parents. Weingarten (1988), for example, found that the majority of her respondents perceived the supportiveness of t h e i r c h i l d r e n to be one of the most important factors i n t h e i r postdivorce 58 adjustment. Gander (1991) found current family closeness (undefined) to predict the postdivorce psychological well-being of her older respondents (age 50 and o l d e r ) , while Gander and Jorgensen (1990), i n a r e l a t e d study based on the same sample (and also not defining closeness), found present closeness with c h i l d r e n to predict both psychological well-being and s o c i a l adjustment. Identity theory suggests that strong parent-adult c h i l d r e l a t i o n s h i p s can be an alternate source of i d e n t i t y a f f i r m a t i o n when the marital r e l a t i o n s h i p i s disrupted. Thoits, who i n her 1992 study measured r o l e -i d e n t i t y salience d i r e c t l y , found that her respondents rated the r o l e -i d e n t i t i e s of parent and spouse the highest of seventeen possible roles, with parent being s l i g h t l y higher than spouse (and much higher than l o v e r ) . Referring back to her 1985 explanation of the l i n k between r o l e behaviors and psychological well-being, we would expect strong parent-adult c h i l d r e l a t i o n s h i p s to provide the parent with a continuing f e e l i n g of e x i s t e n t i a l s e c u r i t y ( i . e . purpose and meaning i n l i f e ) , p o s i t i v e evaluations of r o l e performance (from s e l f and others), s o c i a l approval, i n t r i n s i c g r a t i f i c a t i o n , and a sense of competence and self-esteem. Hypothesis 3a: The moderating e f f e c t s of strong p a r e n t - a d u l t c h i l d r e l a t i o n s h i p s on p s y c h o l o g i c a l w e l l - b e i n g w i l l be greater f o r women than f o r men. For the cohorts of women now i n mid- or l a t e r - l i f e p a r t i c u l a r l y , the r o l e - i d e n t i t y of mother and the mother-child r e l a t i o n s h i p have l i k e l y been central components of s e l f - i d e n t i t y . The commitment of mothers to these r e l a t i o n s h i p s i s r e f l e c t e d i n the strong, stable bonds most mothers have with t h e i r c h i l d r e n across time and circumstance (see Rossi & Rossi, 1990) . 59 Research shows that most women turn to t h e i r c h i l d r e n f o r support when a long-term marriage breaks down, and that c h i l d r e n respond to t h e i r mother's needs (Hagestad et a l , 1984; Wright & Maxwell, 1991) . Because of these expectations and normative patterns of behavior, the continuing presence of suportive mother-adult c h i l d r e l a t i o n s h i p s could be expected to have a substantial impact on the mother's postdivorce psychological well-being. Strong mother-adult c h i l d r e l a t i o n s h i p s could o f f s e t the negative e f f e c t s of divorce or, conversely, poor re l a t i o n s h i p s i n addition to a marriage breakdown could be doubly d i s t r e s s i n g . G i l l i g a n ' s (1982) s e l f - i n - r e l a t i o n theory of morality also suggests that r e l a t i o n s h i p s may be more c l o s e l y linked to psychological well-being for women than f o r men. G i l l i g a n theorizes that some people (primarily men) come to define themselves and morality i n terms of independence and separateness, and others (primarily women) come to define themselves and morality ( i . e . r i g h t , wrong) i n terms of interdependence and connection i n rel a t i o n s h i p s and consequently evaluate themselves by t h e i r a b i l i t y to care. I f t h i s i s the case, i t could be argued that the parent-adult c h i l d r e l a t i o n s h i p i s l i k e l y to have a greater impact on the self-evaluations of women than of men. The p i c t u r e f o r fathers i s d i f f e r e n t . Men's r e l a t i o n s h i p s with t h e i r f a m i l i e s have been shown to be more precarious than women's, i . e . more contingent on circumstances such as marital discord or divorce (see Rossi & Rossi, 1990) . In addition, studies are showing that older fathers are u n l i k e l y to confide i n t h e i r c h i l d r e n when a marriage breaks down or to turn to t h e i r c h i l d r e n f o r support (Hagestad et a l . , 1984,- Wright & Maxwell, 1991). In contrast, older fathers are more l i k e l y than older mothers to turn to parents f o r support (Hagestad et a l . , 1984). It may also be that for fathers seeking support from grown c h i l d r e n i s contrary to t h e i r perceptions of the father r o l e (see Hagestad et a l , 1984; Wright & Maxwell, 1991) . Nevertheless, research from the adult c h i l d ' s perspective i s showing that l a t e r divorce has a negative e f f e c t on the adult c h i l d ' s perception of the f a t h e r - c h i l d r e l a t i o n s h i p , and from the perspective of i d e n t i t y theory, these negative r e f l e c t e d appraisals could be expected to have a negative e f f e c t on the father's evaluations of the r e l a t i o n s h i p s and consequently on his psychological well-being. The hypothesis, therefore, i s not that the parent-adult c h i l d r e l a t i o n s h i p a f f e c t s mothers' postdivorce psychological well-being and not fathers', but that the impact of the parent-adult c h i l d r e l a t i o n s h i p on psychological well-being w i l l be greater for mothers than for father. To date no research has addressed t h i s issue. I I I . Control Variables When t e s t i n g the hypotheses i n the preceding section, a number of control v a r i a b l e s were added to the regression equations to i d e n t i f y t h e i r p o t e n t i a l as competing hypotheses and to reduce the chance of spurious associations between the independent and dependent v a r i a b l e s . The control variables chosen for t h i s study have been shown i n previous research to influence both divorce adjustment and parent-child r e l a t i o n s h i p s . These control v a r i a b l e s are income, assets, race, age of respondent, marital cohort, degree of s o c i a l integration, and r e l i g i o s i t y . 1. Income/Assets Divorce rates, parent-child r e l a t i o n s h i p s and psychological well-being have a l l been found to vary with socioeconomic status (see Price & McKenry, 1988; Raschke, 1987). The variables chosen to control for SES i n 61 t h i s study are income and assets. Raschke (1987) noted that of the three components of SES (income, education, occupation) family income has the most s i g n i f i c a n t influence on marital s t a b i l i t y . Assets are included i n the proposed study because i n m i d l i f e and l a t e r l i f e , perceived f i n a n c i a l s e c u r i t y i s l i k e l y to be s i g n i f i c a n t l y r e l a t e d to a person's psychological well-being (Berardo, 1982; see Kitson et a l . , 1989; Weingarten, 1988). For example, the ownership of a family residence i s a considerable source of f i n a n c i a l security, and i t i s more l i k e l y that the mother w i l l remain i n the house l i v i n g with c h i l d r e n than w i l l the father. This p o t e n t i a l f i n a n c i a l resource would be missed i f only income i s evaluated (for home ownership see Uhlenberg et a l , 1990). Previous research has found a c l e a r inverse r e l a t i o n s h i p between income and other measures of socioeconomic status and divorce (see Booth et a l . , 1986; Price & McKenry, 1988,- Raschke, 1987; White, 1990). The f i n a n c i a l well-being of the parents has also been found to have an e f f e c t on parent-adult c h i l d r e l a t i o n s h i p s . Cooney and colleagues (1995), for example, found that at Time One (15 months postdivorce), the adult c h i l d ' s perception of the parent (mother or father) as a p o t e n t i a l source of f i n a n c i a l support showed a strong p o s i t i v e c o r r e l a t i o n with the adult c h i l d ' s intimacy r a t i n g for that parent. Income has also been shown to have an inverse r e l a t i o n s h i p with psychological well-being; lower le v e l s of depression are associated with higher l e v e l s of income (Gore & Mangione, 1983; Reskin & Coverman, 1985). 2. Race The NSFH i s an American data set, and the controls for r a c e / e t h n i c i t y consequently had to be chosen within that context. The categories for race 62 included i n the NSFH are as follows: White, Black, Hispanic (Mexican-American, Puerto Rican, Cuban, Other Hispanics) and Other (Native and Asian Americans). Most of the American research which includes race as a var i a b l e compares Black Americans and White Americans. This research shows that the divorce rate i s higher f o r Black Americans than f o r White Americans; among the reasons given f o r t h i s difference i s the greater representation of Black Americans i n the lower SES groups, a factor also c o r r e l a t e d with divorce (see Price & McKenry, 1988; Raschke, 1987) . Differences i n parent-adult c h i l d r e l a t i o n s h i p s have also been found between Black Americans and White Americans (Scott & Alwin, 1989,- Lawton, S i l v e r s t e i n , & Bengtson, 1994,- Umberson, 1992), and some of these differences are also associated with the context within which many Black Americans parent, e.g. lone-parent fa m i l i e s and lack of employment opportunities for men. Psychological adjustment to divorce has been found to vary between Black Americans and White Americans, with White Americans reporting higher l e v e l s of depression (Menaghan & Lieberman, 1986). 3. Age of Respondent Chronological age has been found to be corr e l a t e d with divorce rates, parent-child r e l a t i o n s h i p s and psychological well-being. F i r s t , divorce s t a t i s t i c s show that older people are much less l i k e l y to divorce than are younger persons (Booth et a l . , 1986; White, 1990). Second, the rel a t i o n s h i p s that parents have with t h e i r c h i l d r e n change throughout the l i f e span. Bengtson and Kuypers (1971), f o r example, advanced the idea that parents have a "developmental [or generational] stake" i n the parent-adult c h i l d r e l a t i o n s h i p , i . e . they have a greater investment or emotional stake i n the c o n t i n u i t y of the r e l a t i o n s h i p and w i l l therefore be motivated to 63 minimize intergenerational tensions. This idea i s supported by research which shows parents to rate the q u a l i t y of parent-adult c h i l d r e l a t i o n s h i p s higher than the c h i l d r e n do (e.g. Rossi & Rossi, 1990) . Third, age i s associated with psychological well-being. For example, i n general age has been found to be negatively associated with s e l f - r a t i n g s of happiness (Bradburn, 1969). 4. M a r i t a l Cohort Individuals who marry during d i f f e r e n t h i s t o r i c a l periods are exposed to d i f f e r e n t c u l t u r a l expectations and patterns of s o c i a l i z a t i o n . For example, i n d i v i d u a l s marrying when t r a d i t i o n a l gender roles were normative would l i k e l y have d i f f e r e n t expectations for marriage and the roles of wives/husbands and mothers/fathers than would i n d i v i d u a l s marrying when more contemporary gender roles are normative. The general expectation for people marrying e a r l i e r i n t h i s century was for the marriage to l a s t u n t i l the death of one spouse, regardless of marital q u a l i t y or personal s a t i s f a c t i o n . Respondents who married with t h i s expectation would be less l i k e l y to divorce, and i f they did, the divorce would l i k e l y represent a greater adjustment for them than i t would for respondents who married when t h i s s o c i e t a l expectation was weaker. 5. Degree of S o c i a l Integration The i n t e g r a t i o n of i n d i v i d u a l s into s o c i a l networks i s correlated with the propensity to divorce, i n d i v i d u a l psychological well-being, and the nature of parent-child r e l a t i o n s h i p s . In t h i s study, s o c i a l integration, which can be defined as the "existence of s o c i a l r e l a t i o n s h i p s " (Umberson & Gove, 1989, p. 443), includes the presence of active kinship and non-kinship t i e s . O verall, research has shown that 64 s o c i a l i n t e g r a t i o n retards divorce (Glenn & Shelton, 1985; see White, 1990). Also, supportive s o c i a l networks can lessen the impact of s t r e s s f u l l i f e events (see George, 1990), and s o c i a l i n t e g r a t i o n has been found to have a p o s i t i v e e f f e c t on postdivorce psychological well-being (Gander & Jorgensen, 1990; Kitson & Roach, 1989; see Raschke, 1987). In addition, the s o c i a l - s t r u c t u r a l context within which r e l a t i o n s h i p s are embedded influences the nature of those r e l a t i o n s h i p s (Umberson, 1992); thus the degree of the family's i n t e g r a t i o n into the community can be expected to influence the nature of the parent-child r e l a t i o n s h i p (see Peterson & R o l l i n s , 1987). Another form of s o c i a l integration which can be expected to influence the p r o b a b i l i t y of divorce, through mechanisms such as s o c i a l control, i s the presence of other adults i n the household. Also, following a divorce, the presence of another adult i n the household has been found to be p o s i t i v e l y r e l a t e d to happiness and depression, i . e . greater happiness and less depression (Kurdek, 1991; Mastekaasa, 1994a). 6. R e l i g i o s i t y R e l i g i o s i t y was measured by frequency of attendance at r e l i g i o u s services. This v a r i a b l e has been found to be negatively r e l a t e d to divorce (Booth et a l . , 1986; see Raschke, 1987; White & Booth, 1991). There are also reasons to expect that r e l i g i o s i t y would have an influence on parent-adult c h i l d r e l a t i o n s h i p s and psychological well-being (for a discussion of t h i s topic, see Marciano, 1987) . For example, most r e l i g i o n s place a high value on family l i f e and family r e l a t i o n s h i p s , and a high value on interpersonal r e l a t i o n s h i p s and i n d i v i d u a l peace of mind. Frequent church attendance also connotes an integration into a s p i r i t u a l community which can enhance well-being. In general, r e l i g i o s i t y has been found to be 65 negatively r e l a t e d to divorce, and when there i s a divorce, to be p o s i t i v e l y r e l a t e d to divorce adjustment (see Price & McKenry, 1988) . R e l i g i o s i t y was measured at Time One because i n t h i s study i t was p r i m a r i l y considered to be a c u l t u r a l v a r i a b l e which influences the context of the parent-child r e l a t i o n s h i p . IV. Methods A. The Data Set and Sample Data used i n t h i s study come from the National Survey of Families and Households, which was designed s p e c i f i c a l l y to document the nature and v a r i a b i l i t y of family l i f e . The NSFH was developed i n response to a request from the Center for Population Research of the National I n s t i t u t e of C h i l d Health and Human Development (NICHD) i n the United States and has been funded by the same agency. From i t s outset the NSFH was designed to provide a data resource for the research community at large, and was developed i n consultation with leading family scholars; the research team led by Larry Bumpass and James Sweet included representatives from the areas of family sociology, s o c i a l demography, s o c i a l psychology, and family economics, and consultants included such leading scholars as William Aquilino, Frank Furstenberg, Greer L i t t o n Fox and Andrew Ch e r l i n . The NSFH was designed to be l o n g i t u d i n a l and the second wave of data became ava i l a b l e i n 1995. No comparable l o n g i t u d i n a l Canadian data set e x i s t s . The NSFH c o l l e c t e d data by personal interview and self-administered questionnaire from a national, multistage p r o b a b i l i t y sample of 13,017 respondents. One adult per household (age 19 or older) was randomly selected as the primary respondent (age 18 i f married, or i f there were no persons 19 and older i n household; the l a t t e r f or main sample only). 66 Individuals i n the following subgroups were double sampled: one-parent and reconstituted f a m i l i e s , cohabitors, minority f a m i l i e s (Black Americans, Mexican Americans, Puerto Ricans), and recently married persons. In Wave I, there were 9,663 primary respondents i n the main sample and 3,374 i n the oversample. Analyses i n the proposed study are based on l o n g i t u d i n a l data from Wave I, c o l l e c t e d between March 1987 and May 1988, and Wave II, c o l l e c t e d between la t e 1992 and 1994. Detailed information about the NSFH can be found i n Sweet, Bumpass and C a l l (1988). The sample f o r t h i s study consists of respondents chosen on the basis of four c r i t e r i a . One, respondents must be cu r r e n t l y married and l i v i n g with t h e i r spouse at Time One. Two, they must have been married nineteen years or more at the time of f i n a l separation or, for the continuously-married respondents, at the mean time of separation (3.3 years from Time One). Three, they must have at least one b i o l o g i c a l or adopted c h i l d age nineteen or older at the time of f i n a l separation or, for the continuously-married, at the mean time of separation. Four, at Time Two respondents categorized as "married" must have been continuously married to the same partner; those categorized as "divorced" must be cu r r e n t l y separated or divorced from t h e i r spouse at Time Two ( i . e . without an intervening remarriage and divorce). Based on the c r i t e r i a noted above, the selected sample i s composed of 1542 continuously married respondents (873 women; 669 men) and 42 separated/divorced respondents (23 women; 19 men). At Time One, the mean age of respondents i n the f u l l sample was 54.3 years (52.9 for women; 56.1 for men), with a range of 31-85. The mean age for respondents who remained married was 54.45 (53.03 for women; 56.3 for men) and for respondents who separated or divorced between Time One and Time Two, the mean age was 48.4 67 years (46.3 for women; 51.1 for men) with a range of 34-70. Also at Time One, the mean length of marriage was 34.9 years for the f u l l sample, 35.1 for the continuously married and 29.2 for those who separated or divorced. Regarding race, the number of respondents coded as "nonwhite" i n the divorced group was considerably higher than i n the f u l l sample. 1327 respondents i n the f u l l sample were coded as "white" (83.8%) and 257 as "nonwhite," (16.2%), while i n the divorced sample, 25 were coded as "white" (59.5%) and 17 as "nonwhite" (40.5%). This difference l i k e l y r e f l e c t s , i n part, the higher divorce rates i n the American Black population (White=25; Black=ll; Others=6). Respondents i n t h i s study were selected from the f u l l sample (main plus oversample), and standard weighting procedures were followed. In general, proportional weighting procedures assign values between zero and one for categories over-represented i n the f u l l sample, and greater than one for cases under-represented i n the f u l l sample. When the sample weights are summed together, they are equal to the number of cases i n the f u l l sample. Because a subset of the f u l l sample was selected i n t h i s study, an add i t i o n a l step was necessary. F i r s t , the average case weight i n the subset was calculated and then used to standardize case weights before using the weighting procedure. The var i a b l e "weight2," which was used to weight the sample, was computed by d i v i d i n g "mufinw93," the f i n a l weight for the f u l l sample at Time Two, by the average case weight i n the subset (1.493). By using t h i s procedure, the sample i n the study i s representative of the United States population because "mufinw93" corrects for Time One and Time Two i n terms of representativeness of the sample. 68 B. Dependent Variables There are three sets of hypotheses i n t h i s study. For the f i r s t and t h i r d sets of hypotheses the dependent va r i a b l e i s psychological well-being; for the second set, parent-adult c h i l d r e l a t i o n s h i p s . (Appendix A contains a complete l i s t i n g of questions used f o r the study v a r i a b l e s ) . 1. Psychological Well-Being The three dimensions of psychological well-being chosen for measurement i n t h i s study are happiness (positive a f f e c t ) , depression (negative a f f e c t or s t r a i n ) , and self-esteem (personal adequacy). Previous research has shown these dimensions to be f a c t o r i a l l y d i s t i n c t (e.g. Bryant & Veroff, 1982) . a. Happiness The concept of "happiness" can be measured i n a v a r i e t y of ways: two of the most widely-used i n the divorce l i t e r a t u r e are (i) Bradburn's (1969) happiness scale, which measures p o s i t i v e a f f e c t , negative a f f e c t , and provides an a f f e c t balance calculated by subtracting negatives from p o s i t i v e s ; and ( i i ) the one-item "how are things" measure, which i s considered to represent a respondent's o v e r a l l , global evaluation of general happiness. One advantage of Bradburn's scale i s that i t i s able to capture some of the multidimensionality of "happiness," but one disadvantage i s that some of the items for negative a f f e c t overlap with items on depression scales such as the CES-D. There i s also a continuing debate i n the psychological l i t e r a t u r e about whether "happiness" r e f l e c t s the degree to which pleasant emotions outweigh unpleasant emotions, or whether "happiness" (and i t s two major components, p o s i t i v e and negative affect) are determined more by r 69 p e r s o n a l i t y than by s i t u a t i o n (e.g. Eysenck, 1990). Eysenck (1990), for example, argues that happiness stems mainly from p e r s o n a l i t y and that one's pe r s o n a l i t y influences one's emotional reaction to l i f e ' s events ( i . e . p e r s o n a l i t y influences emotional reactions which are then evaluated as happiness/unhappiness). The idea of happiness being determined (at least i n part) by p e r s o n a l i t y i s supported by research showing that respondent's s e l f - r a t i n g s of happiness are f a i r l y stable over periods of 2-3 years (Eysenck, 1990) . And although women are generally found to experience (or to report experiencing) more intense emotional states than men, women's and men's s e l f - r e p o r t s of global happiness have been found to be very s i m i l a r (Bryant & Veroff, 1982; Bryant & Veroff, 1984; Eyesenck, 1990). In addition, Eysenck points out that, although there l i k e l y i s a s o c i a l d e s i r a b i l i t y bias i n responses to happiness questions, repondent 1s s e l f -ratings of happiness are u s u a l l y reasonably close to the ratings of psychologists. The NSFH uses the global one-item happiness measure, and the question which assesses happiness i s as follows: " F i r s t , taking things a l l together, how would you say things are these days. C i r c l e the number that best describes how you f e e l . " Ratings are on a seven point scale ranging from 1 (very unhappy) to 7 (very happy). Respondents i n t h i s study generally reported high l e v e l s of happiness with mean scores of 5.652 at Time One and 5.561 at Time Two (out of a possible range of 1-7) . Because the d i s t r i b u t i o n f o r t h i s v a r i a b l e was skewed i n a negative d i r e c t i o n , the v a r i a b l e scores were reversed (1=7, 2=6, 3=5, 4=4, 5=3, 6=2, 7=1) and logged (skew of transformed scores at Tl=.078; T2=.058). To f a c i l i t a t e i n t e r p r e t a t i o n of r e s u l t s , the v a r i a b l e 70 was r e l a b e l l e d unhappiness so that higher scores ind i c a t e higher l e v e l s of unhappiness. Although logging scores transforms variables to meet the normality assumption and i s c o r r e c t l y used to c a l c u l a t e s i g n i f i c a n c e scores, logged scores cannot be accurately compared to other scores that have not been logged. Therefore i n t h i s study, reported means are i n the o r i g i n a l metric for a p a r t i c u l a r v a r i a b l e , and the int e n t i o n was that when regression equations were ca l c u l a t e d f or the purpose of comparing v a r i a b l e s (e.g. i n t e r a c t i o n e f f e c t s ) , logged scores would f i r s t be converted back to the o r i g i n a l metric by c a l c u l a t i n g the a n t i - l o g ( a n t i - l o g minus one i f scores were c a l c u l a t e d using the natural log plus one to correct f o r zero scores i n the d i s t r i b u t i o n ) . However, when analyses were run, none of the i n t e r a c t i o n terms reached s i g n i f i c a n c e . b. Depression In the NSFH, depression i s measured by a twelve-item version of the Center f o r Epidemiological Studies Depression Scale (CES-D). Respondents were asked to indicate the number of days i n the past week they might have f e l t depressed, f e a r f u l , lonely, sad and so on. Items are coded so that higher scores ind i c a t e higher depression. Cronbach's alphas f o r the depression scale were .93 at both Times One and Two. Congruent with reported high l e v e l s of happiness, most respondents i n the study reported low l e v e l s of depression. Mean scores f o r depression were 11.541 at Time One and 12.219 at Time Two (out of a possible range of 0-84) . Because the d i s t r i b u t i o n f or t h i s v a r i a b l e was skewed i n a p o s i t i v e d i r e c t i o n , scores were logged plus one, to adjust f o r the skew to 0 (skew of transformed scores; Tl=.094; T2=.131). 71 c. Self-Esteem The NSFH measures self-esteem with a modified (three-item) version of the Rosenberg Self-Esteem Scale (Rosenberg, 1989) . On a f i v e - p o i n t L i k e r t -type scale where "1" represented "strongly agree" and "5" represented "strongly disagree", respondents were asked to indicate i f they agreed or disagreed with statements such as "I am able to do things as well as other people." Respondents reported high l e v e l s of self-esteem with means of 12.195 at Time One and 11.993 at Time Two (out of a possible range of 3-15) . The d i s t r i b u t i o n f o r t h i s v a r i a b l e was also skewed i n a p o s i t i v e d i r e c t i o n (skew at Tl=1.199; T2=.851). Therefore the o r i g i n a l v a r i a b l e was reconstructed by reversing and trichotomizing each of the items and then re-adding these items into a scale; the d i s t r i b u t i o n f o r the r e s u l t i n g scale was approximately normal (skew at Tl=.158; T2=.064). Mean scores f o r the reconstructed scale were 6.4 02 at Time One and 6.265 at Time Two (out of a possible range of 3-9). Cronbach's alphas for the o r i g i n a l items i n the scale were .66 at Time One and .67 at Time Two, and f o r the recoded items, .70 at Time One and .69 at Time Two. 2. Parent-Adult C h i l d Relationships In the family l i t e r a t u r e , parent-adult c h i l d r e l a t i o n s h i p s are commonly examined from the perspective of Bengtson and colleagues' model of intergenerational s o l i d a r i t y (see Bengtson & Roberts, 1991; Bengtson et a l . , 1990; Rosenthal, 1987). The model proposes that there are s i x e s s e n t i a l dimensions of parent-child r e l a t i o n s : a s s o c i a t i o n a l s o l i d a r i t y (frequency of i n t e r a c t i o n ) , a f f e c t i o n a l s o l i d a r i t y (subjective judgments concerning the q u a l i t y of family i n t e r a c t i o n s ) , consensual s o l i d a r i t y (agreement on values, b e l i e f s ) , functional s o l i d a r i t y (helping and exchange 72 of resources), normative s o l i d a r i t y (commitment to norms of familism), and s t r u c t u r a l s o l i d a r i t y (opportunity structure r e f l e c t e d i n number of family members, type, and geographic proximity). A l l s i x dimensions are r a r e l y measured i n one research project (for an exception see Rossi & Rossi, 1990) . The dimensions chosen to measure parent-adult c h i l d r e l a t i o n s h i p s i n t h i s study are a s s o c i a t i o n a l and functional s o l i d a r i t y . These dimensions were chosen because of t h e i r relevance to i d e n t i t y theory and because of t h e i r use by other researchers i n t h i s area (e.g. Aquilino, 1994b; Booth & Amato, 1994; B u l c r o f t & B u l c r o f t , 1991; Cooney, 1994; White, 1992) . Identity theory suggests that the continuing presence of active, p o s i t i v e parent-adult c h i l d r e l a t i o n s h i p s w i l l have p o s i t i v e e f f e c t s on the psychological well-being of parents. The c r i t e r i a f o r the s e l e c t i o n of the adult c h i l d r e n i n t h i s study are congruent with those for the parents which were out l i n e d above. S p e c i f i c a l l y , the adult c h i l d r e n must be the respondent's b i o l o g i c a l or adopted chil d r e n , nineteen years of age or older at the time of f i n a l separation (or mean time of separation for the continuously-married respondents), and may be l i v i n g either i n or out of the respondent's household. There are a number of methods that researchers can use to c a l c u l a t e scores f o r parent-adult c h i l d r e l a t i o n s h i p s : some of these are using maximum scores, average, total/aggregate, s e l e c t i o n of the geographically closest c h i l d , random s e l e c t i o n , oldest or youngest c h i l d , and designation of a s p e c i f i c c h i l d by the parent. The choice of method va r i e s according to the research purpose of a study and any conceptual framework or theory used. For t h i s study, the d e c i s i o n was made to c a l c u l a t e average scores. 73 and t h i s d e c i s i o n was made p r i m a r i l y on t h e o r e t i c a l grounds. As noted above, i d e n t i t y theory suggests that the parent-adult c h i l d r e l a t i o n s h i p s i n which the parent r o l e - i d e n t i t y i s enacted w i l l influence the parent's psychological well-being. Although i t i s possible that parent r o l e - i d e n t i t y and i t s salience and influence on well-being are s p e c i f i c to "a" c h i l d rather than generally across children, the theory has not been developed enough thus f a r to make t h i s assumption. Therefore, the choice was made to cal c u l a t e average scores, which w i l l give an i n d i c a t i o n of the parent's o v e r a l l experience i n the parental r o l e (see Umberson, 1989a). The o r i g i n a l i n t e n t i o n i n t h i s study was to use predivorce and postdivorce measures of parent-adult c h i l d r e l a t i o n s h i p s i n a doubly multivariate repeated measures MANOVA, but the s i z e of the sample precluded t h i s method of an a l y s i s . Because the sample siz e at Time One was small, i t was decided to sel e c t the sample based on ch i l d r e n who were nineteen at the time of separation (or at the mean time of separation f o r the nondivorced) rather than nineteen at Time One. This decision meant that support could not be measured at Time One because these questions were asked only f o r chi l d r e n nineteen and older at that time (and, i n addition, the support measures were changed between Time One and Time Two). Consequently, r e l a t i o n s h i p measures are for Time Two only. a. Contact Wave Two of the NSFH includes two questions on contact f o r nonresidential c h i l d r e n age nineteen or over. The f i r s t question i s , "During the past 12 months, how often did you see ( c h i l d ) ? " ; and the second question i s , "During the past 12 months, how often d i d you t a l k on the telephone or receive a l e t t e r from (c h i l d ) ? " Response categories range from "1 - not at a l l " to "6 - more than once a week." 74 For the purposes of t h i s study, the v a r i a b l e of i n t e r e s t i s whether or not there i s a continuing r e l a t i o n s h i p between the parent and the c h i l d ; therefore, the mode of contact ( v i s i t s , l e t t e r s , phone c a l l s ) i s less important than the existence of and frequency of such contact. Consequently, these two measures were combined into one contact measure, averaged over a l l sons and a l l daughters as noted previously. Because the r a t i n g categories are o r d i n a l measures and therefore cannot be added, the mode of contact with the higher r a t i n g i s used. Contact with c o r e s i d e n t i a l c h i l d r e n i s coded at the maximum (6). Previous researchers have also combined contact i n person, by phone and by mail into one measure (see Aquilino, 1994a, 1994b; Booth & Amato, 1994; Umberson, 1992). Aquilino uses NSFH data, but he does not describe the method he used to combine these measures. Parents i n t h i s study were generally i n frequent contact with t h e i r adult c h i l d r e n . Only four respondents reported having no contact at a l l with any of t h e i r adult children, and with (6) being coded as maximum contact, the mean score f o r average contact i s 5.2. Scores f o r response category (7), "some other frequency," were deleted from analyses because there i s no way of t e l l i n g whether these were meant to be some other higher or some other lower. The o r i g i n a l d i s t r i b u t i o n of scores showed a negative skew (1.273), so scores were squared to transform the shape of the d i s t r i b u t i o n to more c l o s e l y approximate a normal d i s t r i b u t i o n (skew T2: .720) . b . Support The interview schedule f o r Wave Two includes several questions about support given to and received from adult c h i l d r e n . There are four questions about instrumental support that the respondent may have given to 75 a c h i l d i n the l a s t month: help with shopping, errands, or transportation; help with housework, yard work, car repairs, or other work around the house; help with c h i l d care while c h i l d r e n were working; or help with c h i l d care at other times. In addition there i s one question asking i f the respondent had given a c h i l d advice, encouragement, moral or emotional support. If the respondent d i d provide these forms of support, a "yes" i s recorded by household member number of the c h i l d . The questions f o r r e c e i v i n g support from c h i l d r e n are i d e n t i c a l except that no c h i l d care questions were asked. Although parents i n the study reported being i n frequent contact with t h e i r adult c h i l d r e n , they were generally not highly involved i n g i v i n g them support i n the response categories provided by the NSFH2. With a possible range of 0 to 5 (5 categories of support per c h i l d ) , the mean score f o r average support given i s 1.099. The d i s t r i b u t i o n of scores f o r t h i s v a r i a b l e was also skewed i n a p o s i t i v e d i r e c t i o n (1.055), so scores were logged plus one to adjust for the skew to 0 (skew T2: .130). Scores f o r average support received were lower than f o r average support given, with a mean scores of .676 out of a possible range of 0 to 3 (3 categories of support per c h i l d ; the NSFH2 includes c h i l d care the parent gives but not c h i l d care the parent may r e c e i v e ) . The d i s t r i b u t i o n of scores f o r t h i s v a r i a b l e was also skewed i n a p o s i t i v e d i r e c t i o n , and scores were also logged plus one (skew T2: .611). The subtests i n Hypothesis 2 predict that the e f f e c t of divorce on mother-child r e l a t i o n s h i p s w i l l be greater for mother-daughters than for mother-sons and f o r father-daughters than for father-sons. As noted previously the scores for contact, support given and support received are average scores. Because the study uses average scores rather than the 76 score from one selected c h i l d , the subtests were analyzed by c a l c u l a t i n g the proportion of adult c h i l d r e n who are female. The d i s t r i b u t i o n for the "femcomp" measure was approximately normal (skew=.063). C. Independent and Conditional Variables The independent v a r i a b l e for the three sets of hypotheses i s divorce, which i s operationalized as a change i n marital status between Time One and Time Two from cu r r e n t l y married and l i v i n g with spouse (coded 0) to c u r r e n t l y separated or divorced from spouse at Time Two ( i . e . no intervening marriage or divorce; coded l ) . Conditional variables for Hypothesis Two are gender of parent and gender of c h i l d (coded female=l; male=0). For Hypothesis Three, the three measures of the parent-adult c h i l d r e l a t i o n s h i p were entered as conditional variables to test f o r the hypothesized moderating e f f e c t of the parent-adult c h i l d r e l a t i o n s h i p on the parent's post-divorce psychological adjustment. In Hypothesis 3a, gender of parent i s also a c o n d i t i o n a l v a r i a b l e because the e f f e c t s were predicted to be greater f o r women than for men. D. Control Variables This i s a study of the e f f e c t s of a s p e c i f i c change i n marital status (divorce a f t e r a long-term marriage) on parent-adult c h i l d r e l a t i o n s h i p s and consequently on parent's psychological well-being. The primary focus of the study i s on comparing parent-adult c h i l d r e l a t i o n s h i p s and parent's psychological well-being for those parents who divorce and f o r those who remain married rather than on analyzing the process variables which contribute to changes i n r e l a t i o n s h i p s and well-being over the divorce process. The control variables chosen, therefore, are those which are 77 applicable to both divorced and married respondents. As noted i n the previous section, the variables used as controls i n t h i s study are income, assets, race, age of respondent, marital cohort, degree of s o c i a l integration, and r e l i g i o s i t y . 1. Income/Assets Income i s measured using an NSFH constructed measure of household income at Time One ( e s s e n t i a l l y a measure of s o c i a l class) and Time Two (capturing change). As with income, an NSFH constructed measure of assets (including material assets, savings, and investments, minus debts) was used. Assets were also measured at Time One and Time Two. The d i s t r i b u t i o n s f o r assets and household income were highly skewed (assets Tl=16.523, T2=4.219; household income Tl=6.46, T2=2.91), with mean scores f o r assets of $120,530.02 at Time One and $128,571.37 at Time Two, and mean scores f o r household income of $46,567.67 at Time One and $53,285.37 at Time Two. Logging d i d not bring the skew f o r these d i s t r i b u t i o n s down to a workable number, so these v a r i a b l e s were transformed by trichotomizing. Assets were recoded as follows: Time One -(lo thru 42953 = 1) (43000 thru 94363=2) (95000 thru hi=3) and Time Two - (lo thru 42999 = 1) (43000 thru 94999=2) (95000 thru hi=3), with r e s u l t i n g skews of -.007 at Time One and -.194 at Time Two. Household income was recoded as follows: Time One - (lo thru 28349 = 1) (28350 thru 49450=2) (49451 thru hi=3) ,-and Time Two (lo thru 30804 = 1) (30805 thru 58000=2) (58001 thru hi=3) , with r e s u l t i n g skews of .00 at both times. 78 2. Race Based on the sample numbers as reported above, race i s treated as a dummy v a r i a b l e ; nonwhite (0), white (having more respondents - 1). 3. Age of Respondent Age was dichotomized into younger (31-49) and older (50-85) age groups. The age of 50 was chosen as the d i v i d i n g l i n e because respondents who were 50 and older i n 1987-1988 belong to the pre-Baby Boom generations. 37.5% of the respondents were i n the younger age group, and 62.5% were i n the older age group. 4. M a r i t a l Cohort The measure used for marital cohort i n t h i s study was "years married." When i n t e r c o r r e l a t i o n s were calculated, i n t e r c o r r e l a t i o n s between age of respondent and years married were high (.88794). Although t h i s was not unexpected, i t indicated that a choice had to be made between c o n t r o l l i n g f o r age and c o n t r o l l i n g f o r marital cohort. Because age i s the more consistently-used measure f o r l a t e r divorce i n the l i t e r a t u r e , the decision was made to use age of respondent as the c o n t r o l . However, i n an attempt to capture some of the cohort e f f e c t s , age was dichotomized (31-49, 50-85) and included as an i n t e r a c t i o n term i n analyses. 5. Degree of S o c i a l Integration S o c i a l i n t e g r a t i o n was operationalized using the items i n Question 16 of the self-administered questionnaire (SE-2). Question 16 asks how often respondents spend a s o c i a l evening with r e l a t i v e s , neighbors, co-workers and/or fr i e n d s , attend a s o c i a l event at t h e i r church or synagogue [note: t h i s question does not include mosques or other places of fellowship], go 79 to a bar or tavern, or p a r t i c i p a t e i n a group r e c r e a t i o n a l a c t i v i t y such as bowling, g o l f , or square dancing. Response categories are "never (0), several times a year, about once a month, about once a week, several times a week (4)." Although o r i g i n a l l y included as a control v a r i a b l e , p a r t i c i p a t i o n i n voluntary organizations was not included i n the f i n a l analyses because the relevant questions were not equivalent at Time One and Time Two. At Time One respondents were asked to indicate which of seventeen organizations they p a r t i c i p a t e d i n , and at Time Two they were asked to indi c a t e which of three groupings of organizations they p a r t i c i p a t e d i n . The combination of organizations i n these questions precluded measurement of exactly which organizations the respondent was i n d i c a t i n g . O v e r a l l , respondents were not p a r t i c u l a r l y involved i n the s o c i a l a c t i v i t i e s l i s t e d , with mean scores of 7.531 at Time One and 10.602 at Time Two (out of a possible range of 0-28). The d i s t r i b u t i o n s f o r t h i s v a r i a b l e approached normality at .525 and .331. The scores f o r other adults i n the household were dichotomized into (0=no other adults) and (l=other adults excluding adult c h i l d r e n ) . At Time One 1575 respondents (99.4%) reported another adult i n the household, and 9 (.6%) reported no other adult i n the household (respondents who were l i v i n g apart from t h e i r spouse for reasons other than marital problems, e.g. m i l i t a r y service, were included i n the sample). At Time Two 1547 respondents (97.7%) reported at least one other adult i n the household , and 37 reported none (2.3%). 80 6. R e l i g i o s i t y The scores for r e l i g i o s i t y (measured as church attendance) were widely d i s t r i b u t e d , ranging from 0 to 2555 times a year. To transform the d i s t r i b u t i o n of scores for t h i s v a r i a b l e into an approximately normal d i s t r i b u t i o n , three o u t l i e r s (520, 1144, 2555) were coded missing and the remaining scores were logged plus one (skew before transformations 15.514; a f t e r transformations -.432). 7. Geographic Distance Geographic distance was added as a control because i t has been found to be a s i g n i f i c a n t p r e d i c t o r of support exchanges between parents and adult c h i l d r e n (e.g. Rossi & Rossi, 1990). The relevant question i n the Time Two interview asks parents how many miles away each nonresidential adult c h i l d l i v e s , and these distances were summed and averaged across c h i l d r e n . The geographic distance f o r adult c h i l d r e n l i v i n g i n the household at Time Two was coded as 0. The mean score for distance at Time Two was 363.0 miles, with a possible range of 0 to 9000 miles, and the d i s t r i b u t i o n of scores was skewed i n a p o s i t i v e d i r e c t i o n ( i . e . most parents and adult c h i l d r e n l i v e r e l a t i v e l y close together). To approximate a normal d i s t r i b u t i o n , the scores f o r t h i s v a r i a b l e were logged (skew T2 before transformations 4.393; a f t e r transformations -.163). 81 E. Tests f o r M u l t i c o l l i n e a r i t y A c o r r e l a t i o n matrix showed only age of respondent and years married to be highly c o r r e l a t e d (.894) and, with years married dropped, the variables included i n analyses met the assumption of independence. Tolerance s t a t i s t i c s also indicated no problems with m u l t i c o l l i n e a r i t y or s i n g u l a r i t y . 82 Chapter Four Results There were three sets of hypotheses i n t h i s study, each with three dependent v a r i a b l e s . The f i r s t hypothesis predicted a negative e f f e c t of divorce a f t e r a long-term marriage on psychological well-being, measured as happiness (unhappiness), depression, and self-esteem. The second set of hypotheses predicted p o s i t i v e e f f e c t s of divorce on mother-adult c h i l d r e l a t i o n s h i p s , measured as contact, support given and support received, and negative e f f e c t s on father-adult c h i l d r e l a t i o n s h i p s . And the t h i r d set predicted that parent-adult c h i l d r e l a t i o n s h i p s would have a p o s i t i v e e f f e c t on postdivorce psychological well-being, and that t h i s e f f e c t would be greater f o r mothers than for fathers. To t e s t these hypotheses, a series of h i e r a r c h i c a l regression procedures were used. The exact nature of these procedures i s discussed i n respect to each hypothesis; however, the general pattern f o r the f i r s t and t h i r d set of hypotheses was as follows (in the second set divorce was entered f i r s t ) : the Time One score on the dependent v a r i a b l e was entered f i r s t i n order to control f or the e f f e c t s of Time One on Time Two (thereby r e s u l t i n g i n an analysis of change) ,- then Divorce was added to calculate zero-order effects,- next the complete set of controls were added to determine which of the controls were s i g n i f i c a n t . In succeeding regressions only s i g n i f i c a n t controls were entered, and the two- and three- way i n t e r a c t i o n terms were entered l a s t to test for s i g n i f i c a n t differences by gender, age and divorce. In h i e r a r c h i c a l regression, e f f e c t s are re-estimated at each step. Therefore i n a l l steps a f t e r step 2 the divorce e f f e c t i s net of a l l other variables i n the model. When t h i s e f f e c t ( i . e . of 83 divorce) diminishes i n subsequent steps i t i s an i n d i c a t i o n that the i n i t i a l e f f e c t was spurious or explained by one or more of the co n t r o l s . Hypothesis 1: Divorce a f t e r a long-term marriage w i l l have a negative e f f e c t on p a r e n t ' s p s y c h o l o g i c a l w e l l - b e i n g . To test the f i r s t hypothesis, separate analyses were performed f or unhappiness, depression and self-esteem. For each dependent va r i a b l e , two series of regressions were run. In the f i r s t s e r i e s . Time One psychological well-being and divorce were entered i n the f i r s t step, and the complete set of controls i n the second step. In the second se r i e s , Time One psychological well-being and divorce were entered i n the f i r s t step, s i g n i f i c a n t controls i n the second, a l l two-way in t e r a c t i o n s with respondent's gender, divorce and age on the t h i r d , and the three-way i n t e r a c t i o n of respondent's gender, divorce and age i n the fourth. For t e s t i n g a l l hypotheses i n the study, the R2 change and i t s s i g n i f i c a n c e l e v e l were examined at each step to determine which regression model best f i t the data. Interactions were only considered i f the ent i r e block added s i g n i f i c a n t l y to the variance explained. The r e s u l t s of the f i n a l regression models t e s t i n g the e f f e c t of divorce on psychological well-being are shown i n Tables 2, 3 and 4 (Appendix B). Because the scores f o r happiness were reversed, r e s u l t s w i l l be present i n terms of "unhappiness" i n order to s i m p l i f y i n t e r p r e t a t i o n . Regression r e s u l t s show Time One unhappiness to be the strongest predictor of Time Two unhappiness (B=.305, p<=.001), and that divorce had a p o s i t i v e 84 e f f e c t on the change i n unhappiness between Time One and Time Two (B=.054, p<.05) with the divorced respondents reporting higher l e v e l s of unhappiness at Time Two than the continuously-married respondents. There are also main e f f e c t s f o r gender of parent (B=.080, p<.05) and Time Two s o c i a l a c t i v i t i e s (B=-.055, p<.05), with gender of parent being p o s i t i v e l y r e l a t e d to change i n unhappiness and s o c i a l a c t i v i t i e s negatively r e l a t e d to change i n unhappiness. Overall, women reported higher l e v e l s of unhappiness at Time Two than did men, while respondents who were more s o c i a l l y active at Time Two reported lower l e v e l s of unhappiness than those who were less s o c i a l l y a c t i v e . None of the interactions were s i g n i f i c a n t , i n d i c a t i n g that the e f f e c t of divorce on parent-adult c h i l d contact did not s i g n i f i c a n t l y d i f f e r for mothers or fathers, or for younger and older respondents. The adjustecd R2 s t a t i s t i c f o r the f i n a l model (R2 =.118, p<.001) shows that around 12% of the variance i s accounted f o r i n the model. Becoming separated or divorced between Time One and Time Two also had a s i g n i f i c a n t l y p o s i t i v e e f f e c t on the change i n depression between Time One and Time Two (B=.052, p<.05), with separated/divorced respondents reporting s i g n i f i c a n t l y higher l e v e l s of depression at Time Two than continuously-married respondents. As with unhappiness. Time One depression i s the strongest p r e d i c t o r of Time Two depression (B=.414, p<.001). There are also main e f f e c t s f o r gender of respondent (B=.090, p<.001) which was p o s i t i v e l y r e l a t e d to change i n depression between Time One and Time Two, and household income at Time Two which was negatively r e l a t e d (B=-.118, p<.001) . Over a l l , at Time Two women i n the study reported higher l e v e l s of depression than d i d men and respondents with higher incomes reported lower l e v e l s . 85 Again, none of the interactions were s i g n i f i c a n t , i n d i c a t i n g that there were no s i g n i f i c a n t differences i n the e f f e c t divorce had on depression f o r fathers and mothers, or younger and older respondents. The adjusted R2 s t a t i s t i c f o r the f i n a l model (R2=.216, p<.001) shows that around 22% of the variance i s accounted for i n the model. Divorce showed no s i g n i f i c a n t e f f e c t on the change i n self-esteem between Time One and Time Two. As with the preceding analyses, Time One self-esteem i s the strongest predictor of Time Two self-esteem (B=.390, p<.001), although there were main e f f e c t s for both Time Two household income and for gender of respondent, with Time Two household income being p o s i t i v e l y r e l a t e d to change i n self-esteem between Time One and Time Two (B=.105, p<.001) and gender of respondent being negatively r e l a t e d (B=-.090, p<.001). Respondents with higher income at Time Two reported higher l e v e l s of self-esteem and o v e r a l l , women reported lower l e v e l s of self-esteem than men. None of the int e r a c t i o n s were s i g n i f i c a n t , i n d i c a t i n g again that the ef f e c t s of divorce on self-esteem d i d not d i f f e r s i g n i f i c a n t l y for mothers and fathers, or for younger and older respondents. The adjusted R2 s t a t i s t i c f o r the f i n a l model (R2 =.180, p<.001) shows that around 18% of the variance i s accounted f or i n the model. In summary, the r e s u l t s support the hypothesis f o r the measures of happiness and depression, but not for self-esteem. 86 Hypothesis 2: Divorce a f t e r a long-term marriage w i l l have d i f f e r e n t i a l e f f e c t s on mother-adult c h i l d r e l a t i o n s h i p s and father-adult c h i l d r e l a t i o n s h i p s . Hypothesis 2a: Divorce a f t e r a long-term marriage w i l l have a p o s i t i v e e f f e c t on mother-adult c h i l d r e l a t i o n s h i p s . Hypothesis 2b: Divorce a f t e r a long-term marriage w i l l have a negative e f f e c t on father-adult c h i l d r e l a t i o n s h i p s . Separate analyses were performed for contact, average support given and average support received. For each dependent var i a b l e , three series of regressions were run. In the f i r s t s e ries, divorce was entered i n the f i r s t step and the complete set of controls i n the second step. In the second s e r i e s . Time Two depression was added to the set of s i g n i f i c a n t controls, and the e f f e c t s i n the f i r s t series were re-estimated. Time Two depression was added to the regression a f t e r subsequent tests on Hypothesis 3 revealed a s i g n i f i c a n t p o s i t i v e r e l a t i o n s h i p between support received and Time Two depression which, as t h i s i s c o r r e l a t i o n a l data, could be i n d i c a t i n g that higher l e v e l s of parental depression may stimulate greater involvement of c h i l d r e n i n the parent-adult c h i l d r e l a t i o n s h i p . Because mothers reported higher l e v e l s of depression than fathers, the e f f e c t s for depression could confound any gender differences i n the e f f e c t s of divorce on parent-adult c h i l d r e l a t i o n s h i p s , and so Time Two depression was added as a control for the second set of hypotheses. F i n a l l y , i n the t h i r d s e r i e s divorce was entered i n the f i r s t step, s i g n i f i c a n t controls plus the s i g n i f i c a n t Time 87 Two depression measure i n the second step, a l l two-way inte r a c t i o n s between divorce, gender and age i n the t h i r d , and the three-way i n t e r a c t i o n (divorce by gender of parent by age) on the fourth step. The r e s u l t s of the regression models t e s t i n g the e f f e c t s of divorce on parent-adult c h i l d r e l a t i o n s h i p s are shown i n Tables 5, 6 and 7 (Appendix B). These r e s u l t s show divorce to have a negative e f f e c t on Time Two average contact (B=-.056, p<.05). There are also main e f f e c t s f o r Time One age (B=-.169, p<.001), geographic distance (B=-.488, p<.001), gender of parent (B=.075, p<.01). Time One household income (B=.091, p<.001), Time One assets (B=.072, p<.01), and Time Two depression (B=.069, p<.001). Overall, respondents who were older and those who l i v e d farther away reported less contact; women and those respondents who had higher household income at Time One, higher assets at Time One and higher l e v e l s of depression at Time Two a l l had higher contact with t h e i r adult c h i l d r e n than t h e i r counterparts i n each category. None of the interactions were s i g n i f i c a n t , including divorce by gender of parent and divorce by age, i n d i c a t i n g that the e f f e c t s of divorce on change i n contact did not d i f f e r f o r mothers or fathers, or for those respondents over or under f i f t y . In the f i n a l regression model, the adjusted R2 s t a t i s t i c (R2 =.322, p<.001) shows that around 32% of the variance i s accounted f or i n t h i s model. Becoming separated or divorced had no s i g n i f i c a n t e f f e c t on parent's i reports of average support given to adult c h i l d r e n . There were s i g n i f i c a n t main e f f e c t s f o r Time One age (B=-.113, p<.001), geographic distance (B=-.136, p<.001), gender of parent (B=.108, p<.001), Time One household income (B=.154, p<.001), Time Two assets (B=.102, p<.001), Time One s o c i a l a c t i v i t i e s (B=.091, p<.01), and Time Two depression (B=.083, p<.01). 88 Overall, respondents who were older and those who l i v e d farther away-reported less support giving, and women and those respondents who had higher household income at Time One, higher assets at Time Two, more s o c i a l a c t i v i t i e s at Time One, and higher le v e l s of depression at Time Two reported gi v i n g more support to t h e i r adult c h i l d r e n . Again none of the interactions were s i g n i f i c a n t , i n d i c a t i n g that the ef f e c t s of divorce on average support given did not d i f f e r f o r mothers and fathers, or f o r older and younger respondents. In the f i n a l regression model, the adjusted R2 s t a t i s t i c (R2 =.114, p<.001) shows that around 11% of the variance i s accounted f or i n t h i s model. The r e s u l t s also show divorce to have no s i g n i f i c a n t e f f e c t on the average support that parents reported receiving from t h e i r adult children. There were, however, main e f f e c t s f o r age (B=-.121, p<.001), geographic distance (B=-.108, p<.001), gender of parent (B=.123, p<.001), and Time Two depression (B=.122, p<.001). Overall, older respondents and those who l i v e d farther away received less support, and women and respondents who had higher l e v e l s of depression received more support. Again, none of the i n t e r a c t i o n terms were s i g n i f i c a n t , i n d i c a t i n g that there were no gender (of parent) differences i n the e f f e c t s of divorce on average support parents received from t h e i r adult c h i l d r e n . The adjusted R2 i n the f i n a l model (R2 =.068, p<.001) shows that the model accounts f or around 7% of the variance. In summary, the r e s u l t s showed only p a r t i a l support f or the second set of hypotheses. As predicted, divorce had a negative e f f e c t on father-adult c h i l d contact, but also on mother-adult c h i l d contact. There were no s i g n i f i c a n t e f f e c t s on eit h e r support given or support received. 89 Subtest 2a: The p o s i t i v e e f f e c t of divorce a f t e r a long-term marriage on mother-adult c h i l d r e l a t i o n s h i p s w i l l be greater for mother-daughter r e l a t i o n s h i p s than for mother-son r e l a t i o n s h i p s . Subtest 2b: The negative e f f e c t of divorce a f t e r a long-term marriage on father-adult c h i l d r e l a t i o n s h i p s w i l l be greater for father-daughter r e l a t i o n s h i p s than fo r father-son r e l a t i o n s h i p s . In these subtests, the v a r i a b l e "femcomp" was added to the regression equations to test whether a higher proportion of daughters among the divorced respondent's adult c h i l d r e n made a s i g n i f i c a n t d i f f e r e n c e i n the reported l e v e l s of average contact, average support given, and average support received. Separate analyses were performed for the three dependent v a r i a b l e s . For each dependent variable, two series of regressions were run. In the f i r s t s e r i e s , divorce was entered i n the f i r s t step, gender of parent and "femcomp" i n the second, and s i g n i f i c a n t controls (including Time Two depression) as a block on the t h i r d . In the second se r i e s , divorce was entered i n the f i r s t step, gender of parent and "femcomp" on the second, s i g n i f i c a n t controls on the t h i r d , a l l two-way inte r a c t i o n s between divorce, gender of parent and "femcomp" on the fourth, and the three-way i n t e r a c t i o n (divorce by gender of parent by femcomp) on the f i f t h . When the regression models for Hypotheses 2a and 2b were rerun with the addition of "femcomp," there was a main e f f e c t of "femcomp" for contact (B=.103, p<.001) and f o r support giving (B=.064, p=.05), but none for support received. Overall, respondents with a higher proportion of daughters i n the composition of t h e i r families reported higher l e v e l s of 90 contact and of support g i v i n g . Again none of the inte r a c t i o n s reached s i g n i f i c a n c e , i n d i c a t i n g that the e f f e c t s of divorce on parent-daughter and on parent-son r e l a t i o n s h i p s did not d i f f e r by gender of parent. The adjusted R2 for the f i n a l contact model (R2 =.332, p<.001) showed that the model accounted f o r around 33% of the variance; the adjusted R2 f o r the f i n a l support given model (R2 =.117, p<.001) showed that the model accounted for around 12% of the variance; and the adjusted R2 for the f i n a l support received model (R2 =.068, p<.001) showed that the model accounted f o r around 7% of the variance. The nonsignificant i n t e r a c t i o n e f f e c t s f o r divorce by gender by "femcomp" d i d not support the gender differences predicted i n subtests 2a and 2b. Hypothesis 3: Strong parent-adult c h i l d r e l a t i o n s h i p s w i l l moderate the negative e f f e c t s of divorce a f t e r a long-term marriage on the psychological well-being of the parent. Separate analyses were performed f o r unhappiness, depression and s e l f -esteem. For each dependent variable, three series of regressions were run. In the f i r s t s e r i e s , Time One psychological well-being and divorce were entered i n the f i r s t step, and the s i g n i f i c a n t controls from Hypothesis One as a block i n the second. In the second series, the three r e l a t i o n s h i p measures of contact, support given, and support received were entered as a block i n a t h i r d step. In the t h i r d series, Time One psychological well-being and divorce were entered i n the f i r s t step, s i g n i f i c a n t controls on the second, the three r e l a t i o n s h i p measures on the t h i r d , and a l l two-way inter a c t i o n s between divorce and the r e l a t i o n s h i p measures on the fourth. 91 In the analyses f or unhappiness, the R change was nonsignificant when the three r e l a t i o n s h i p measures were added i n a block, i n d i c a t i n g that these measures of average contact, average support given and average support received d i d not have s i g n i f i c a n t e f f e c t s on the change i n parent's l e v e l s of unhappiness between Time One and Time Two. There were main e f f e c t s f o r gender of parent (B=.076, p<.01), Time One unhappiness (B=.305, p<.001), and Time Two s o c i a l a c t i v i t i e s (B=-.059, p<.05), i n d i c a t i n g that o v e r a l l , women and those respondents who reported more unhappiness at Time One reported more unhappiness at Time Two, and those respondents who reported more s o c i a l a c t i v i t i e s at Time Two reported less unhappiness. When the i n t e r a c t i o n terms were added i n a block, the R2 change s t a t i s t i c was nonsignificant, i n d i c a t i n g that the f i t of the model was not improved by adding i n the i n t e r a c t i o n terms. That i s , parent-adult c h i l d r e l a t i o n s h i p s d i d not s i g n i f i c a n t l y moderate the negative e f f e c t s of divorce on happiness. In the f i n a l model, the e f f e c t of divorce on unhappiness (p=.07) i s reduced only s l i g h t l y from that found i n the analyses f or Hypothesis 1 (p<.05). The adjusted R2 (R2 =.117, p<.001) indicates that the f i n a l model accounts f or around 12% of the variance (similar to Hypothesis 1) -In the analyses f or depression, the addition of the r e l a t i o n s h i p measures i n a block produced a s i g n i f i c a n t R2 change (R2 =.018, p<.001), with support received showing a s i g n i f i c a n t p o s i t i v e r e l a t i o n s h i p with Time Two depression (B=.120, p<.001). Time One depression s t i l l showed a s i g n i f i c a n t main e f f e c t (B=.409, p<.001), and the e f f e c t of divorce was reduced s l i g h t l y (from p<.05 to p=.051). These r e s u l t s indicate that respondents who reported r e c e i v i n g more support from t h e i r children, those who were more depressed at Time One, and those who were divorced had higher l e v e l s of 92 depression at Time Two. There were also s i g n i f i c a n t main e f f e c t s f o r gender of parent (B=.070, p<.01) and Time Two household income (B=-.135, p<.001): women reported higher l e v e l s of depression and those respondents who reported higher household income at Time Two reported lower l e v e l s of depression. Overa l l , divorced and continuously married parents who reported re c e i v i n g more support reported higher l e v e l s of depression rather than lower l e v e l s . In addition, the measures of contact and support given did not moderate the negative e f f e c t s of divorce on depression. In the f i n a l model for t h i s t e s t , the adjusted R2 (R2 =.232, p<.001) indicates that the model accounts f o r 23% of the variance (similar to Hypothesis 1). The r e s u l t s of the regression models t e s t i n g f o r the e f f e c t s of parent-adult c h i l d r e l a t i o n s h i p s on the parent's postdivorce psychological well-being can be found i n Table 8 (Appendix B). In the analyses f o r self-esteem, the three r e l a t i o n s h i p measures and t h e i r i n t e r a c t i o n terms with divorce also f a i l e d to produce a s i g n i f i c a n t R2 change when entered i n subsequent blocks to the regression, showing that these measures d i d not a s i g n i f i c a n t e f f e c t on change i n self-esteem between Time One and Time Two. As i n Hypothesis 1 there were s i g n i f i c a n t p o s i t i v e main e f f e c t s f o r Time One self-esteem (B=.390, p<.001) and Time Two household income (B=.105, p<.001), and s i g n i f i c a n t negative main e f f e c t s for gender of parent (B=-.096, p<.01). Overall, respondents with higher l e v e l s of self-esteem at Time One and those with high household income at Time Two reported higher self-esteem at Time Two, and women reported lower s e l f -esteem. Although the interactions terms were not s i g n i f i c a n t , the magnitude of the nonsignificant r e s u l t s f o r the e f f e c t of divorce was, however, reduced 93 s l i g h t l y (from p=.92 to p=.78). In the f i n a l model, the adjusted R s t a t i s t i c (R2 =.180, p<.001) shows that the model accounts f o r around 18% of the variance (similar to Hypothesis 1). Hypothesis 3 a : The moderating e f f e c t s of strong parent-adult c h i l d r e l a t i o n s h i p s on psychological well-being w i l l be greater for women than for men. In the analyses for t h i s hypothesis, the two-way inte r a c t i o n s with gender of parent ( i . e . gender of parent by r e l a t i o n s h i p variables) and the other two-way inte r a c t i o n s with divorce were entered as a block i n the fourth step, and i n the f i n a l step the three-way i n t e r a c t i o n s (divorce by gender of parent by r e l a t i o n s h i p variables) were entered. None of the int e r a c t i o n s produced a s i g n i f i c a n t change i n the value of R2 for any of the three measures of psychological well-being, i n d i c a t i n g that the e f f e c t of parent-adult c h i l d r e l a t i o n s h i p s on the psychological-well-being of divorced parents d i d not d i f f e r f o r mothers or for fathers. The r e s u l t s are therefore s i m i l a r to the r e s u l t s f or Hypotheses 1 and 3. In the analyses for unhappiness there were p o s i t i v e main e f f e c t s f or gender of parent (B=.076, p<.01) and negative main e f f e c t s f or s o c i a l a c t i v i t i e s at Time Two (B=-.059, p<.05), and the e f f e c t of divorce was reduced (between Hypothesis 1 and Hypothesis 3) to a trend (p=.07). The strongest p r e d i c t o r of Time Two unhappiness was Time One unhappiness (B=.305, p<.001). In the f i n a l model, the adjusted R2 i s .117 (p<.001), i n d i c a t i n g that t h i s model accounts for around 12% of the variance. In the analyses for depression, the three r e l a t i o n s h i p measures added as a block produced a s i g n i f i c a n t R2 change (R2 =.018, p<.001), with average support received having a s i g n i f i c a n t p o s i t i v e e f f e c t on l e v e l s of 94 depression (B=.120, p<.001). The r e s u l t s also show s i g n i f i c a n t negative main e f f e c t s f o r Time Two household income (B=-.135, p<.001), and p o s i t i v e main e f f e c t s f o r gender of parent (B=-.070, p<.01), depression at Time One (B=.396, p<.001), and the e f f e c t of divorce i s reduced s l i g h t l y (to p=.051). In the f i n a l model for t h i s analysis, the adjusted R2 i s .232 (p<.001), i n d i c a t i n g that t h i s model accounts f o r around 23% of the variance. In the analyses f o r self-esteem there were negative main e f f e c t s f o r gender of parent (B=-.296, p<.001), p o s i t i v e main e f f e c t s f o r Time One s e l f -esteem (B=.374, p<.001) and household income at Time Two (B=.198, p<.001), and divorce was not s i g n i f i c a n t . In the f i n a l model for t h i s analysis, the adjusted R2 i s .181 (p<.001), i n d i c a t i n g that t h i s model accounts for around 18% of the variance. Consistent with the lack of gender inte r a c t i o n s i n any of the study r e s u l t s thus f a r , the i n t e r a c t i o n terms i n the analyses f o r Hypothesis 3a also did not reach s i g n i f i c a n c e . The r e s u l t s of t h i s study d i d not, therefore, support Hypothesis 3 or 3a,- reported l e v e l s of contact, support given and support received d i d not moderate the negative e f f e c t s of divorce on happiness, depression or self-esteem for mothers or fathers. Chapter Five Discussion The purpose of t h i s study has been to answer the following three questions about divorce: f i r s t , what are the e f f e c t s of recent divorce a f t e r long-term marriage on the psychological well-being of parents,-second, what are the e f f e c t s of recent parental divorce on parent-adult c h i l d r e l a t i o n s h i p s ; and t h i r d , what are the e f f e c t s of parent-adult c h i l d r e l a t i o n s h i p s on the psychological well-being of recently divorced parents? Data f o r the study were taken from two waves of the National Survey on Families and Households, a n a t i o n a l l y representative data set developed by prominent family scholars i n the United States. The present study i s , to the best knowledge of the author, the f i r s t prospective l o n g i t u d i n a l study of divorce a f t e r long-term marriage. Reviewers of the divorce l i t e r a t u r e repeatedly note that there has been l i t t l e empirical research on middle and l a t e r l i f e divorce, and l i t t l e t h e o r e t i c a l development (e.g. Coleman & Ganong, 1993; Lloyd & Zick, 1986). There has been even less empirical research that has s p e c i f i c a l l y looked at divorces occurring a f t e r long-term marriages, or at the e f f e c t s of these divorces on family r e l a t i o n s h i p s and on the parent's psychological well-being. Because of the dearth of research i n t h i s area, there are not c l e a r t h e o r e t i c a l frameworks to follow. For t h i s study, i d e n t i t y theory was used to provide a t h e o r e t i c a l framework to guide the development of hypotheses and the i n t e r p r e t a t i o n of r e s u l t s i n conjunction with the e x i s t i n g l i t e r a t u r e . O v e r a l l , the analyses i n t h i s study produced mixed r e s u l t s . The f i r s t hypothesis, that divorce would have negative e f f e c t s on psychological well-being, was p a r t i a l l y supported with divorce having the expected e f f e c t s on change i n happiness and depression between Time One 96 and Time Two, but no s i g n i f i c a n t e f f e c t s on change i n self-esteem. The second set of hypotheses, that divorce would have p o s i t i v e e f f e c t s on mother-child r e l a t i o n s h i p s and negative e f f e c t s on f a t h e r - c h i l d r e l a t i o n s h i p s , had some support with divorce having negative e f f e c t s for f a t h e r - c h i l d contact, but also for mother-child contact. There were, however, no s i g n i f i c a n t e f f e c t s of divorce on e i t h e r support given or support received. And f i n a l l y the t h i r d set of hypotheses, that strong parent-adult c h i l d r e l a t i o n s h i p s would moderate the negative e f f e c t s of divorce on psychological well-being and that t h i s moderating e f f e c t would be greater for mothers than for fathers, was not supported. The major findings of the study are that for these respondents divorce had negative e f f e c t s on change i n s e l f - r a t i n g s of happiness and depression between Time One and Time Two, and negative e f f e c t s on parent-adult c h i l d contact for both fathers and mothers. Perhaps one of the most s t r i k i n g findings of t h i s study i s the absence of gender interactions f or any of the hypotheses. I . Divorce A f t e r Long-Term Marriage and Psychological Well-Being Hypothesis One predicted that divorce a f t e r a long-term marriage would have a negative e f f e c t on psychological well-being and t h i s hypothesis was p a r t i a l l y supported, with r e s u l t s showing negative e f f e c t s for happiness and depression but not for self-esteem. A. Happiness In general, the r e s u l t s f or happiness are congruent with previous cr o s s - s e c t i o n a l research (Gove & Shin, 1989; Marks, 1995; Mastekaasa, 1994b) and previous l o n g i t u d i n a l research (Booth & Amato, 1991; Chiriboga et a l , 1991; Hetherington et a l , 1978) showing divorce to have a negative e f f e c t on s e l f - r a t i n g s of happiness. Previous l o n g i t u d i n a l research (e.g. Booth & Amato, 1991) has shown negative e f f e c t s of divorce on s e l f - r e p o r t s of happiness to attenuate with time. For the majority of divorced respondents i n t h i s study, however, s e l f - r a t i n g s of happiness are s t i l l lower an average (mean) of 2.6 years postseparation than the s e l f - r a t i n g s of continuously-married respondents. Results from previous studies have also been mixed regarding gender and age differences i n the e f f e c t s of divorce on happiness. Regarding gender, the r e s u l t s of t h i s study are congruent with the cross-sectional research of Gove and Shin (1989) and the l o n g i t u d i n a l research of Booth and Amato (199) showing no gender differences. Regarding age, the r e s u l t s of t h i s study are congruent with the cross-sectional research of Hammond and Muller (1992) and Gove and Shin (1989), but contrary to that of Chiriboga (1982) who found that recently divorced older respondents reported less happiness than recently divorced younger respondent. B. Depression In general the r e s u l t s f or depression are also congruent with previous c r o s s - s e c t i o n a l (Kurdek, 1991; Marks, 1995, Mastekaasa, 1994b, Riessman & Gerstel, 1985) and l o n g i t u d i n a l research (Booth & Amato, 1991; Kitson & Holmes, 1992; Menaghan & Lieberman, 1986) showing divorce to have a negative e f f e c t on s e l f - r a t i n g s of depression. Similar to happiness, previous l o n g i t u d i n a l studies (e.g. Booth & Amato, 1991) have shown negative e f f e c t s of divorce on s e l f - r e p o r t s of depression attenuate with time. The r e s u l t s of t h i s study, however, show that for the majority of divorced respondents, negative e f f e c t s are s t i l l being experienced an average of 2.6 years postseparation. 98 Results of previous research have been mixed regarding gender and age d i f f e r e n c e s . Regarding gender, the r e s u l t s of t h i s study are congruent with the cross-sectional research of Aneshenshel, Frericks and Clark (1981) and the l o n g i t u d i n a l research of Booth and Amato (1991) and Menaghan and Lieberman (1986) which show no gender differences i n the e f f e c t s of divorce on depression. And regarding age, the r e s u l t s are congruent with the cross-sectional research of Gander (1991), but contrary to c r o s s - s e c t i o n a l research that shows divorced older respondents reporting less depression than divorced younger respondents (Gove & Shin, 1989; Roach & Kitson, 1989). The f i n d i n g of no age differences i s also congruent with the l o n g i t u d i n a l research of Kitson and Holmes (1992). C. Self-Esteem Unlike the r e s u l t s f or happiness and depression, the r e s u l t s f or self-esteem d i d not support the hypothesis. In previous empirical research the r e s u l t s f or the e f f e c t s of divorce on self-esteem have been mixed, and the findings of no divorce differences i n t h i s study are congruent with the cross-sectional studies of Marks (1995) and Weingarten (1985) and with the l o n g i t u d i n a l studies of Doherty, Su and Needle (1989) and Chiriboga, Catron and associates (1991). The findings of no s i g n i f i c a n t gender differences are also congruent with the c r o s s - s e c t i o n a l research of Gove and Shin (1989) and Weingarten (1985) and with the l o n g i t u d i n a l research of Doherty, Su and Needle (1989) and Kitson and Holmes (1992) . S i m i l a r l y , the present findings regarding no age differences are congruent with the cross-sectional research of Gove and Shin (1989) and the lon g i t u d i n a l research of Kitson and Holmes (1992) . The nonsignificant r e s u l t s f or the e f f e c t s of divorce on self-esteem are contrary to the hypothesized p r e d i c t i o n which was based i n part on 99 Stryker's i d e n t i t y theory; the expectation was that the loss of the spouse ro l e i d e n t i t y would have a negative e f f e c t on self-esteem. Because the spouse r o l e i d e n t i t y i s highly s a l i e n t for most North Americans (e.g. Thoits, 1992), i t was reasoned that the termination of t h i s r o l e i d e n t i t y a f t e r a long-term marriage would have negative e f f e c t s on self-esteem. An a l t e r n a t i v e explanation for t h i s unexpected r e s u l t , however, may be found i n research l i t e r a t u r e which i s suggesting that people who divorce a f t e r a long-term marriage may take a long time reaching the decision to separate (Decker & Langelier, 1978; Hagestad et a l . , 1984; Melichar & Chiriboga, 1985) . If t h i s i s the case, the spouse r o l e i d e n t i t y f or at least one partner may have become less s a l i e n t . Consequently, negative e f f e c t s f or those more invested i n the spouse r o l e i d e n t i t y may be washed out by neutral (or positive) e f f e c t s f or those less invested i n that r o l e i d e n t i t y . Perhaps, however, even for respondents highly invested i n the spouse ro l e i d e n t i t y , divorce can have both p o s i t i v e and negative e f f e c t s on self-esteem. For example, Chiriboga and associates (1991), i n t h e i r postdivorce l o n g i t u d i n a l study, found that self-evaluations on most dimensions of self-concept did not change s i g n i f i c a n t l y over the 3.5 years of t h e i r study, but did f i n d that the divorced group scored s i g n i f i c a n t l y higher than the married comparison group on negative s e l f , vulnerable s e l f , and mastery, which they interpreted as suggesting that the divorce process may involve both p o s i t i v e and negative changes i n o v e r a l l s e l f -image . 100 I I . Divorce A f t e r Long-Term Marriage and Parent-Adult C h i l d Relationships The second set of hypotheses i n t h i s study predicted that divorce a f t e r a long-term marriage would have d i f f e r e n t i a l e f f e c t s on mother-adult c h i l d and father-adult c h i l d r e l a t i o n s h i p s , with e f f e c t s being p o s i t i v e for mothers and negative for fathers. There was l i t t l e support f o r these pr e d i c t i o n s ; r e s u l t s showed negative e f f e c t s f o r both f a t h e r - c h i l d and mother-child contact and no s i g n i f i c a n t e f f e c t s f o r eit h e r support given or support received. A. Contact The measure of contact i n t h i s study was chosen i n part to allow comparison with current research from the adult c h i l d ' s perspective, and the r e s u l t s of the study are congruent with the findings of Aquilino (1994b) that divorce has a s i g n i f i c a n t l y negative e f f e c t on contact f o r both mothers and fathers. No d i r e c t comparisons with previous research of the e f f e c t s of divorce a f t e r long-term marriage on parent-adult c h i l d contact from the parent's perspective are possible because I know of no such studies to date that have included a quantitative measure of contact. Although the r e s u l t s of t h i s study show divorce to have a negative e f f e c t on contact with adult children, there i s no divorce by gender i n t e r a c t i o n , that i s , the negative e f f e c t s on contact are not s i g n i f i c a n t l y d i f f e r e n t f o r fathers than f o r mothers. Consequently these r e s u l t s do not r e f l e c t the strong mother-child bond which has been found in previous research on middle and l a t e r l i f e divorce (Hagestad et a l , 1984; Weingarten, 1988), but do support r e s u l t s from studies showing a decreasingly negative e f f e c t of divorce on f a t h e r - c h i l d contact as the ch i l d ' s age at time of divorce increases (Bulcroft & B u l c r o f t , 1991; 101 Bumpass & Sweet, 1991; Cooney & Uhlenberg, 1990) . Because the r e s u l t s of no s i g n i f i c a n t differences between the e f f e c t s of divorce on contact for mothers and fathers i s supported by research from the adult c h i l d ' s perspective but i s contrary to the t h e o r e t i c a l expectations of t h i s study, these r e s u l t s bear closer examination. F i r s t i t should be noted that previous research (e.g. Rossi & Rossi, 1990) has found high congruence between parents and c h i l d r e n on objective indi c a t o r s such as contact and f a i r l y high congruence on parent-adult c h i l d r e l a t i o n s h i p s i n general. It seems l i k e l y then, that i f parents were biased i n t h e i r reporting, they would report more contact than they were a c t u a l l y experiencing, just as parents tend to report higher l e v e l s of intergenerational r e l a t i o n s h i p q u a l i t y than c h i l d r e n do (Rossi & Rossi, 1990). If t h i s i s the case, these analyses represent a conservative test of the hypothesis. Unfortunately i n the NSFH data set, there i s no way to v e r i f y the parents' and children's reports for the sample used i n t h i s study. From the author's perspective, the most s t r i k i n g f i n d i n g i n t h i s study i s the absence of any s i g n i f i c a n t divorce by gender of parent int e r a c t i o n s f o r parent-adult c h i l d r e l a t i o n s h i p s or f o r the e f f e c t s of these r e l a t i o n s h i p s on the parent's divorce adjustment. In previous studies of divorce a f t e r long-term marriage, fathers have been found to be much more l i k e l y than mothers to report d e t e r i o r a t i n g r e l a t i o n s h i p s or l o s t contact with at least one c h i l d (Hagestad et a l , 1984; Wright & Maxwell, 1991). The use of an average contact measure i n t h i s study, rather than the more commonly used maximum contact with any c h i l d , was an attempt to capture some of the troubled r e l a t i o n s h i p s reported by some fathers and also by some mothers (Weingarten, 1988) . Consequently, the 102 f i n d i n g of no s i g n i f i c a n t differences i n the e f f e c t s of divorce on father-c h i l d and mother-child postdivorce contact may be suggesting that the sex d i f f e r e n t i a l i n problematic postdivorce parent-child r e l a t i o n s h i p s i s not as great as the two studies c i t e d above appear to i n d i c a t e . The r e s u l t s showing no s i g n i f i c a n t differences i n the e f f e c t s of divorce upon l e v e l s of contact for mothers and fathers are also congruent with the growing body of fatherhood research which i s showing that the father r o l e i s highly s a l i e n t to the majority of fathers, but that t h e i r involvement i n that r o l e i s l i m i t e d by s t r u c t u r a l , i n s t i t u t i o n a l and c u l t u r a l constraints (e.g. Cohen, 1987; Kruk, 1993; Lewis & O'Brien, 1987; Pleck, 1985). It i s i n t e r e s t i n g to speculate upon the possible changes i n these constraints that may be occurring when parents divorce at l a t e r stages i n the family l i f e cycle. For example, fathers may be more secure i n t h e i r jobs and careers and facing less occupational demands, while mothers may be returning to school or paid employment or increasing the number of hours worked, thus decreasing time a v a i l a b l e for mother-child contact. The study of women's career paths, although germane to t h i s topic, i s very complex and was, a l b e i t r e l u c t a n t l y , not included i n the present study. The important point here i s that the negative e f f e c t s of divorce on contact d i d not d i f f e r s i g n i f i c a n t l y f or fathers and mothers, thus suggesting that the r o l e i d e n t i t y of parent and the parent-adult c h i l d r e l a t i o n s h i p was equally s a l i e n t f or the mothers and fathers i n t h i s sample. The subtests for Hypothesis 2 predicted that the e f f e c t s of divorce would be greater for the mother-daughter r e l a t i o n s h i p (positive) and for the father-daughter r e l a t i o n s h i p (negative). Although the v a r i a b l e "femcomp," proportion of daughters to sons, has a p o s i t i v e main e f f e c t on 103 contact, i . e . o v e r a l l having more daughters was associated with having more contact, there was no divorce by gender of parent by "femcomp" in t e r a c t i o n , i n d i c a t i n g that having more daughters d i d not s i g n i f i c a n t l y influence the negative e f f e c t s of divorce on parent-child contact for mothers or f o r fathers. Again, these r e s u l t s suggest that from the parent's perspective, the parent-child r e l a t i o n s h i p i s equally s a l i e n t for divorced mothers and for divorced fathers, and for divorced parents of sons and divorced parents of daughters. As contact provides the opportunity f o r support exchanges and perhaps the b u i l d i n g of close parent-child r e l a t i o n s h i p s (see Moss & Moss, 1992), the nonsignificant gender composition e f f e c t s found i n t h i s study warrant further i n v e s t i g a t i o n with measures s p e c i f i c to i n d i v i d u a l daughters and sons rather than with an o v e r a l l proportionate measure. B. Support Given Being separated or divorced at Time Two d i d not have a s i g n i f i c a n t e f f e c t on the l e v e l s of average support given to adult c h i l d r e n reported by fathers and mothers i n t h i s sample. Contrary to expectations, mothers did not appear to increase t h e i r l e v e l s of support g i v i n g i n response to the termination of the spousal r o l e r e l a t i o n s h i p nor d i d father's involvement i n support g i v i n g decrease a f t e r separation or divorce. Although o v e r a l l mothers reported g i v i n g more support than fathers, the e f f e c t s of divorce on support g i v i n g were not s i g n i f i c a n t l y d i f f e r e n t f o r mothers and fathers ( i . e . there was no divorce by gender of parent i n t e r a c t i o n ) . As with contact, there was also no divorce by age i n t e r a c t i o n , i n d i c a t i n g that e f f e c t s of divorce on support given did not d i f f e r s i g n i f i c a n t l y f o r younger and older respondents. 104 Although previous studies have examined support g i v i n g from the adult c h i l d ' s perspective and from the everdivorced parent's perspective ( i . e . parents who have ever been divorced), there i s l i t t l e information on support g i v i n g from parents who have recently separated or divorced a f t e r a long-term marriage. Aquilino, i n his 1994 study from the adult c h i l d ' s perspective, found that his respondents reported r e c e i v i n g s i g n i f i c a n t l y less support from divorced parents than from continuously-married parents. In his study, however, divorce i s not n e c e s s a r i l y recent (0-15 years) and support measures (NSFH) do not d i s t i n g u i s h support from mothers from support from fathers. The study by Hagestad, Smyer and Stierman (1984), however, d i d f i n d that 90% of mothers and 58% of fathers said that r e l a t i o n s h i p s with t h e i r c h i l d r e n had remained the same or improved. The r e s u l t s from the present study suggest that on average the support-giving dimension of the parent-adult c h i l d r e l a t i o n s h i p appears to have remained the same. The absence of gender differences i n postdivorce support giving suggests that when parents divorce a f t e r a long-term marriage, fathers (as well as mothers) may have been able to b u i l d up r e l a t i o n s h i p s with t h e i r c h i l d r e n that are strong enough to make the t r a n s i t i o n from i n t a c t family to divorced family. Perhaps, as has been suggested by Kruk (1993) for divorced fathers of younger children, some of the fathers who divorced a f t e r a long-term marriage were able to e s t a b l i s h new patterns of i n t e r a c t i o n and develop closer r e l a t i o n s h i p s with t h e i r c h i l d r e n than they were able to do during what may have been a long-term unhappy marriage. Again, as was found for contact, although "femcomp" had a s i g n i f i c a n t main e f f e c t on support giving, the gender composition of the adult c h i l d r e n had no s i g n i f i c a n t influence on the e f f e c t s of divorce on 105 the l e v e l s of support g i v i n g reported by parents. So, although o v e r a l l mothers report g i v i n g more support and parents with daughters or a higher proportion of daughters report giving more support, divorced parents with daughters or a higher proportion of daughters do not report g i v i n g s i g n i f i c a n t l y more support than divorced parents of sons or those with a higher proportion of sons. This suggests that i t i s the parent r o l e i d e n t i t y that i s s a l i e n t for married and divorced i n d i v i d u a l s , and that parents w i l l perform those behaviors that are normatively expected of them whether they are parents of daughters or parents of sons. C. Support Received The r e s u l t s f or average support received are s i m i l a r to those for average support given: o v e r a l l mothers reported r e c e i v i n g more support than fathers; divorce had no s i g n i f i c a n t influence on the e f f e c t of divorce on l e v e l s of support received reported by fathers and mothers; and there was no divorce by gender i n t e r a c t i o n or divorce by age i n t e r a c t i o n . This i s congruent with Aquilino's (1994b) study from the adult c h i l d ' s perspective that his respondents reported no s i g n i f i c a n t differences i n l e v e l s of support given to divorced or continuously married parents. These findings, however, are contrary to the study by Wright and Maxwell (1991) which examined the support that parents who were recently divorced a f t e r a long-term marriage reported r e c e i v i n g from t h e i r adult c h i l d r e n . In t h e i r study, divorced mothers reported r e c e i v i n g more help than divorced fathers i n a l l four categories: socioemotional aid, services, advice, and f i n a n c i a l a i d . The discrepancy i n these r e s u l t s may be explained by the difference i n the s e l e c t i o n of the adult c h i l d . While the present study used an average support measure encompassing a l l children, the Wright and Maxwell study used a support measure for the 106 geographically c l o s e s t c h i l d to keep the questionnaire "reasonably short" and on the "assumption that c h i l d r e n who l i v e d nearby would have the greatest opportunity for providing d a i l y support" (Wright & Maxwell, 1991, pp. 28, 29). It may be that t h i s decision by Wright and Maxwell underestimated the amount of socioemotional a i d (respect, encouragement, affection) and advice exchange between father and adult c h i l d r e n who may l i v e f a r t h e r apart. The explanations for the findings of no s i g n i f i c a n t differences i n the e f f e c t s of divorce a f t e r long-term marriage on support received are s i m i l a r to those for the other two parent-child r e l a t i o n s h i p measures, contact and support given. However, the explanations d i f f e r i n that while the measure of support given r e f l e c t s the parent's current investment i n the r e l a t i o n s h i p , the measure of support received r e f l e c t s the adult c h i l d ' s response and perhaps the parent's past investment. The f i n d i n g of nonsignificant differences i n the e f f e c t s of divorce on reported l e v e l s of support received by divorced mothers and divorced fathers may be a r e f l e c t i o n of the strong parent-child r e l a t i o n s h i p s that both fathers and mothers were able to develop over the years with t h e i r c h i l d r e n . When i n t e r p r e t i n g these r e s u l t s , however, i t i s well to keep i n mind that o v e r a l l the parents i n t h i s sample reported r e c e i v i n g low l e v e l s of support and 46% reported receiving none. If there are no s i g n i f i c a n t differences i n support received by married or divorced parents, i t may be an i n d i c a t i o n that adult c h i l d r e n have d i f f i c u l t y responding to the p o s s i b l y increased need for support of t h e i r divorced parents, perhaps because of l o y a l t y c o n f l i c t s . In the analyses for support received, there was once again no divorce by gender composition of c h i l d r e n i n t e r a c t i o n , i n d i c a t i n g that the 107 proportion of daughters to sons does not have a s i g n i f i c a n t influence on the e f f e c t s of divorce on support received from adult children, although the main e f f e c t shows that o v e r a l l parents with more daughters received more support than parents with more sons. I I I . Divorce A f t e r Long-Term Marriage, Parent-Adult C h i l d Relationships and Psychological Well-Being The p r e d i c t i o n that strong parent-adult c h i l d r e l a t i o n s h i p s would moderate the negative e f f e c t s of divorce a f t e r long-term marriage on parent's psychological well-being was not supported. Parent-adult c h i l d r e l a t i o n s h i p s , as measured by l e v e l s of contact, support given and support received, had no s i g n i f i c a n t e f f e c t on Time Two happiness or self-esteem, and support received had a p o s i t i v e (rather than negative) e f f e c t on s e l f -reports of depression ( i . e . greater support received, greater depression). Contrary to expectations, there were no divorce by gender of parent by r e l a t i o n s h i p i n t e r a c t i o n s , i n d i c a t i n g that the influence of parent-adult c h i l d r e l a t i o n s h i p s on postdivorce psychological well-being d i d not d i f f e r s i g n i f i c a n t l y f o r mothers and fathers. Although i t was expected that investment i n parent-adult c h i l d r e l a t i o n s h i p s would be b e n e f i c i a l for divorced parents, these r e s u l t s suggest that the s i t u a t i o n may be more complex. Previous research from the perspective of i d e n t i t y theory has found that the parent r o l e i d e n t i t y i s s a l i e n t to both married and divorced i n d i v i d u a l s (Thoits, 1992), and i t was suggested e a r l i e r i n t h i s study that continued contact with and support from adult c h i l d r e n could function as an af f i r m a t i o n of the parent's established sense of i d e n t i t y and therefore have a p o s i t i v e influence on postdivorce psychological well-being. 108 Other research with younger divorced mothers, however, has found that c l o s e - k n i t networks were h e l p f u l f or divorced mothers who wished to maintain t h e i r e x i s t i n g i d e n t i t i e s , but loose-knit networks h e l p f u l f or those wished to e s t a b l i s h new i d e n t i t i e s (McLanahan, Wedemeyer & Adelberg, 1981). It may be therefore that high l e v e l s of investment i n the parent role i d e n t i t y could have mixed r e s u l t s f or parents who have divorced a f t e r a long-term marriage. On the one hand, as i d e n t i t y theory suggests, the successful enactment of the parent r o l e i d e n t i t y may have p o s i t i v e e f f e c t s on psychological well-being; on the other hand, continued investment i n the parent r o l e i d e n t i t y may also be a continued reminder of d i f f i c u l t i e s i n the marital r o l e r e l a t i o n s h i p and the (perhaps continuing) attachment to the former spouse, and therefore have negative e f f e c t s on psychological well-being. When I was contemplating what some of the alternate explanations for the nonsignificant r e s u l t s i n the analyses of Hypotheses 3 and 3a might be, I r e a l i z e d that an underlying assumption of the study was that s o c i a l support from adult c h i l d r e n would be b e n e f i c i a l f o r divorced parent's psychological well-being. Once t h i s assumption was uncovered, flaws i n the reasoning immediately became apparent. For one thing, the ambiguity i n the measure of emotional support, "advice, encouragement, moral or emotional support," took on greater importance. For i n addition to p o t e n t i a l l y confusing the respondent with an array of not-necessarily equivalent options, the word "advice" i t s e l f i s open to i n t e r p r e t a t i o n ; advice can be welcome or not, p o s i t i v e or not, perceived as supportive or not. And t h i s led me to the consideration of the nature of postdivorce negotiations i n the parent-adult c h i l d r e l a t i o n s h i p . 109 Family research has shown that over the l i f e course, the normative s i t u a t i o n i s for support to flow down the generational l i n e s u n t i l parents are very old, and even then r e c i p r o c i t y i n some form i s preferred (e.g., see Rossi & Rossi, 1990). Although we have l i t t l e knowledge as yet of support exchanges following divorce a f t e r a long-term marriage, evidence from other sources (e.g. Mancini & Blieszner, 1989) suggests that r e c i p r o c i t y i s a key element to consider when examining the e f f e c t s of support exchanges on psychological well-being. In t h i s study, support given and support received were kept separate, following Mutran & Reitzes (1984). It may be, however, that a r a t i o measure such as that suggested by Mangen (1987) of help given to help received would be a better measure of the strength of the parent-adult c h i l d r e l a t i o n s h i p . Another issue l i e s i n the fact that parent-adult c h i l d r e l a t i o n s h i p s are not always p o s i t i v e , and that r e l a t i o n s h i p s t r a i n s can have a strongly negative e f f e c t on psychological well-being (Rook, 1984; Umberson, 1992). So i t may be that support exchanges are occurring, but that they are somehow un s a t i s f a c t o r y i n t h e i r psychological consequences. Take, f o r example, a divorced parent who i s motivated to invest i n the parent r o l e i d e n t i t y and the parent-adult c h i l d r e l a t i o n s h i p out of a sense of o b l i g a t i o n and a need to preserve i d e n t i t y , and who finds his or her e f f o r t s not appreciated by the chil d r e n . This could happen i f a c h i l d i s angry at the parent for divorcing and needs some distance to sort things out. This response from the c h i l d could negate any psychological benefits that the increases i n support giving may otherwise have produced. Without some form of open communication, neither parent nor c h i l d would be able to express t h e i r reasons f o r t h e i r actions and reactions. 110 Support received could also create d i f f i c u l t i e s . A parent, for example, may perceive help as inappropriate, may not want to receive support from adult children, or may f i n d i t d i f f i c u l t to be i n a p o s i t i o n of having to accept support from ch i l d r e n and therefore f i n d the experience depressing rather than supporting (as the r e s u l t s suggest may be happening). The adult children, on the other hand, may not know how to support a parent who divorces i n mid- or l a t e r - l i f e , may not want to (for whatever reason), or may f i n d the s i t u a t i o n so s t r e s s f u l ( e s p e c i a l l y i f caught i n the middle) that they are unable to give t h e i r parents much support at a l l . Parents who perceive that t h e i r c h i l d r e n are not responding appropriately or as expected to t h e i r needs may become even more depressed. A divorce at any stage of the family l i f e cycle requires a renegotiation of roles and r o l e r e l a t i o n s h i p s for parents and for children, and t h i s negotiation process i s perhaps even more c r i t i c a l when parents and t h e i r adult c h i l d r e n renegotiate t h e i r r e l a t i o n s h i p s . Younger children, f o r example, cannot choose to "opt out" of a family, but older c h i l d r e n can; and while there are l e g a l and s o c i e t a l sanctions holding parents of younger c h i l d r e n to t h e i r parental r e s p o n s i b i l i t i e s , the expectations for f a m i l i e s with adult c h i l d r e n are f a r more amorphous. Results from c l i n i c a l research p a r t i c u l a r l y i s showing, however, that parents and adult c h i l d r e n adjust better i f f a m i l i e s can be restructured i n a way that f a c i l i t a t e s the c h i l d ' s interactions with both parents and allows s p e c i a l occasions, such as weddings, to be celebrated with a minimum of c o n f l i c t (Campbell, 1995; Goodman, 1993; Pett et a l , 1992). As empirical research on divorce a f t e r long-term marriage i s only beginning i n family studies, our knowledge about the forms postdivorce f a m i l i e s with I l l adult c h i l d r e n take or how f a m i l i e s s u c c e s s f u l l y negotiate r o l e and r o l e r e l a t i o n s h i p changes i s l a r g e l y speculative. Such research i s needed, however, both to further our t h e o r e t i c a l and empirical knowledge about the divorcing process and postdivorce family r e l a t i o n s h i p s and to ensure that our attempts to help divorcing f a m i l i e s i s based on a secure foundation of s c i e n t i f i c knowledge. IV. Strengths and Limitations This study was designed to advance our knowledge of divorce a f t e r a long-term marriage, parent-adult c h i l d r e l a t i o n s h i p s , and the influence of the parent-adult c h i l d r e l a t i o n s h i p on parents' divorce adjustment. Like a l l empirical research, the study has both l i m i t a t i o n s and strengths. A. Limitations The d e c i s i o n of whether or not to use secondary data requires a researcher to c a r e f u l l y balance advantages and disadvantages. For t h i s study the d e c i s i o n was that the advantage of having access to a well-respected survey with a large, n a t i o n a l l y representative sample of respondents outweighed the p o t e n t i a l l i m i t a t i o n s . According to Menaghan (1985), the s c a r c i t y of prospective lon g i t u d i n a l data i s the most important obstacle to advances i n research about divorce and i t s e f f e c t s on psychological well-being. When using secondary data, however, researchers are always faced with constructing v a l i d measures from other researchers' questions, and t h i s study was no exception. For example, the NSFH dropped the Time Two measure of parent-adult c h i l d r e l a t i o n s h i p q u a l i t y and the questions for support exchanges with adult c h i l d r e n were changed from Time One to Time Two. Consequently, there are no measures of r e l a t i o n s h i p q u a l i t y i n the 112 study nor i s there l o n g i t u d i n a l data for the parent-adult c h i l d r e l a t i o n s h i p . The number of respondents who separated or divorced between Time One and Time Two and who met the c r i t e r i a for sample s e l e c t i o n i n t h i s study also turned out to be quite small: 42 (23 women and 19 men). Therefore, although the o r i g i n a l NSFH sample was large and representative, and the sample s i z e i n t h i s study i s comparable to those i n previous studies by Menaghan and Lieberman (1986; N=32) and Doherty and colleagues (1989; N=42), caution i s s t i l l warranted i n the i n t e r p r e t a t i o n of r e s u l t s . There are also no controls i n the study for employment status, and i t i s acknowledged that the worker r o l e i d e n t i t y could be highly s a l i e n t for these respondents, p a r t i c u l a r l y following a divorce, and could represent an a l t e r n a t i v e r o l e for i d e n t i t y investment. A measure of hours worked was constructed for Time One, but constructing a comparable measure for Time Two turned out to be problematic because the data for Time Two has not been adequately cleaned at t h i s time. The i n c l u s i o n of a comparison group of continuously-married respondents has been noted as a strength of the study,- however, t h i s focus on comparisons between married and divorced respondents meant that the controls included i n the study had to be relevant to both groups, and that variables relevant only to the divorcing group could not be part of the study. Some of the p o t e n t i a l l y important variables which therefore had to be omitted include who i n i t a t e d the separation and who i n i t i a t e d divorce proceedings (not n e c e s s a r i l y the same person), length of time between separation and divorce and between divorce and the interview, the mode of dispute r e s o l u t i o n the couple has chosen (e.g. divorce counselling, l i t i g a t i o n , mediation), and changes i n t h e i r s o c i a l networks (kinship and 113 f r i e n d s h i p ) . Also, because the scores for the parent-adult c h i l d r e l a t i o n s h i p were averaged across children, i t was not possible to include c h a r a c t e r i s t i c s of the adult children, such as parental status ( i . e . whether or not respondents had grandchildren). And f i n a l l y , the r e s u l t s of t h i s study may be l i m i t e d i n t h e i r g e n e r a l i z a b i l i t y from a representative sample of the United States population to the Canadian population. One example of a d i f f e r e n c e that could create p o t e n t i a l l i m i t a t i o n s can be found i n the c u l t u r a l / r a c i a l composition of each population. In the United States the primary "nonwhite" r a c i a l group i s composed of Afro-Americans, whose divorce patterns have been found to vary from those of the "white" r a c i a l group and also from "nonwhite" r a c i a l and c u l t u r a l groups ("white" and "nonwhite" being terminologies used i n the research l i t e r a t u r e ) . B. Strengths The data for t h i s study come from the National Survey of Families and Households, a national survey which c o l l e c t e d data from a multistage, randomly-selected p r o b a b i l i t y sample of 13,017 respondents. One benefit of the well-planned nature of the study plus the c a r e f u l l y selected sample i s the increased g e n e r a l i z a b i l i t y of study r e s u l t s . Another i s that the NSFH was planned and c a r r i e d out by foremost scholars and researchers s p e c i f i c a l l y to examine family issues which increases the v a l i d i t y of the r e s u l t s . Also, the NSFH was designed to be a l o n g i t u d i n a l survey, and data f o r t h i s study come from the f i r s t two waves c o l l e c t e d i n 1987-1988 and 1992-1994. The study i s , to the best of my knowledge, the f i r s t study of divorce a f t e r long-term marriage to use prospective l o n g i t u d i n a l data. L i t e r a t u r e reviews c o n s i s t e n t l y note the lack of t h e o r e t i c a l development i n research on mid- and l a t e r - l i f e divorce (see Lloyd & Zick, 114 1986). In response to t h i s l i m i t a t i o n i n the research, t h i s study has used i d e n t i t y theory to provide a t h e o r e t i c a l framework to guide the development of hypotheses and the i n t e r p r e t a t i o n of r e s u l t s i n conjunction with the e x i s t i n g l i t e r a t u r e . The study has also extended our knowledge of divorce a f t e r long-term marriage and parent-adult c h i l d r e l a t i o n s h i p s by using multiple indicators of both parent-adult c h i l d r e l a t i o n s h i p s and psychological well-being, and by i n c l u d i n g a comparison group of parents who have been continuously married. The study was also designed to compare the e f f e c t s of divorce on parent-child r e l a t i o n s h i p s and psychological well-being for both mothers and fathers, as information from fathers who divorce i n mid- and l a t e r -l i f e i s p a r t i c u l a r l y lacking i n the research l i t e r a t u r e (e.g. Cooney & Uhlenberg, 1990; Price & McKenry, 1988). In addition, the study has examined the e f f e c t s of recent divorce a f t e r long-term marriage on parent-adult c h i l d r e l a t i o n s h i p s from the parent's perspective. The "young adult bias" mentioned by Hagestad and her colleagues (1984) appears to be continuing, with divorce i n mid- and l a t e r - l i f e being studied from the adult c h i l d ' s perspective with the emphasis on the adult c h i l d ' s developmental and adjustment d i f f i c u l t i e s . This study was designed, i n part, to complement recent studies of divorce and parent-adult c h i l d r e l a t i o n s h i p s from the adult c h i l d ' s perspective by using s i m i l a r measures (contact, support given, support received) from the parent's perspective , thus allowing comparisons between the adult c h i l d ' s and the parent's points of view (see for example, Aquilino, 1994b). The focus of the study, however, has been on the e f f e c t s of divorce and parent-adult c h i l d r e l a t i o n s h i p s on parent's psychological well-being. 115 C. Implications The r e s u l t s of t h i s study have implications for the f i e l d of family studies, for t h e o r e t i c a l development i n the area of divorce a f t e r long-term marriage, and for family p r a c t i t i o n e r s and f a m i l y - r e l a t e d s o c i a l p o l i c y . a. Family Studies The implications of t h i s study for the f i e l d of family studies are two-fold. F i r s t , research i s greatly needed on the phenomenon of divorce a f t e r a long-term marriage and i t s e f f e c t s on family r e l a t i o n s h i p s . While t h i s study has focused on parent-adult c h i l d r e l a t i o n s h i p s , other family r e l a t i o n s h i p s are also affected. For example, aside from the research of Hagestad and her colleagues (1984), we know v i r t u a l l y nothing about the e f f e c t s of divorce a f t e r long-term marriage on the parents of the divorcing i n d i v i d u a l s , nor how the parent's response may hinder or f a c i l i t a t e the divorce adjustment of parents and t h e i r adult c h i l d r e n . In addition, the assumption that mothers of grown c h i l d r e n are more invested i n p a r e n t - c h i l d r e l a t i o n s h i p s than fathers needs to be re-examined, as does the assumption that being a parent i s as central to i d e n t i t y when the c h i l d r e n are grown as i t i s during the c h i l d - r a i s i n g years when the c h i l d r e n are young (see Eisenhandler, 1992). Second, family studies textbooks and courses need to include information on divorce a f t e r long-term marriage and a recognition that some students w i l l be experiencing t h i s t r a n s i t i o n . Also, more information from the growing body of fatherhood research could be integrated into family studies textbooks and courses. For example, i n the area of work and family, i n t e g r a t i o n of current fatherhood research would draw att e n t i o n to s t r u c t u r a l and c u l t u r a l constraints that shape the role 116 involvement of both mothers and fathers, p a r t i c u l a r l y when c h i l d r e n are very young. b. T h e o r e t i c a l Development Ident i t y theory was used i n t h i s study p a r t l y because of the l i n k s i t provides between the i n s t i t u t i o n a l , family and i n d i v i d u a l l e v e l s of analysis, which allows researchers to include considerations of s t r u c t u r a l and c u l t u r a l constraints upon fathers and mothers i n t h e i r i n t e r p r e t a t i o n s . For example, a major f i n d i n g i n t h i s study was the absence of any gender differences i n the r e s u l t s f o r any of the analyses, and t h i s was interpreted as suggesting that the constraints on and opportunities a v a i l a b l e to women and men may be d i f f e r e n t f o r i n d i v i d u a l s who divorce a f t e r long-term marriage than they are for those who divorce e a r l i e r . One implication of t h i s i s that generalizations of r e s u l t s from studies of divorce e a r l i e r i n the family l i f e cycle to divorce l a t e r i n the family l i f e cycle warrant caution. A second implication i s that gender or age differences which may be found i n future studies w i l l l i k e l y need to be explained by reference to factors exogenous to parent-adult c h i l d r e l a t i o n s h i p s , as well as to r e l a t i o n s h i p dynamics endogenous to these r e l a t i o n s h i p s . It may be that i d e n t i t y theory w i l l prove most useful i n future research i f i t i s combined with another theory, as Mutran and Reitzes (1984) have done with exchange theory. Identity theory could, f o r example, be combined with s o c i a l support, s o c i a l network, family mediation or economic theories to expand our knowledge of how parent-adult c h i l d r e l a t i o n s h i p s and parent's psychological well-being are influenced by divorce a f t e r long-term marriage. 117 c. Family P r a c t i t i o n e r s This study also has implications f o r family p r a c t i t i o n e r s . Family l i f e educators, f o r example, can provide information f o r parents who are divorcing a f t e r long-term marriages and for adult c h i l d r e n whose parents are d i v o r c i n g by developing workshops i n community settings, such as those offered by Family Services. These workshops could include suggestions on how to renegotiate parent-adult c h i l d r e l a t i o n s h i p s during and a f t e r a divorce. S o c i a l workers can also a s s i s t families who are i n t r a n s i t i o n during divorce a f t e r long-term marriage. For example, they can organize support groups both f o r divorcing parents and for adult c h i l d r e n who are experiencing the divorce of t h e i r parents. In t h e i r c l i n i c a l practive, s o c i a l workers can also help f a c i l i t a t e communication between parents and adult c h i l d r e n when d i f f i c u l t i e s i n postdivorce family functioning occur (see Bonkowski, 1989) . Family therapists, as Campbell (1995) has suggested, may need to take a less structured view of family development (perhaps by working within a developmental family systems framework). Therapists working with c l i e n t s who are divorcing a f t e r long-term marriages also need to attend to the postdivorce parenting of t h e i r c l i e n t s , and to help parents understand and cope with changes i n parenting i n constructive ways that recognize both t h e i r own needs and those of t h e i r adult c h i l d r e n . And f i n a l l y , family mediators as well need to be s e n s i t i v e to po t e n t i a l d i f f i c u l t i e s i n the parent-adult c h i l d r e l a t i o n s h i p when working with c l i e n t s who are divorcing a f t e r a long-term marriage. One implication of t h i s research i s that mediators may f i n d a 118 multigenerational model of divorce mediation to be useful when adult c h i l d r e n are involved. d. Family-related s o c i a l p o l i c y There are also implications f o r s o c i a l p o l i c y . The focus of governmental and s o c i e t a l i n t e r e s t has long been on the e f f e c t s of divorce on minor c h i l d r e n and on the non-involvement of many noncustodial fathers (the l a t t e r p r i m a r i l y for economic reasons). However, there i s also a need for a t t e n t i o n to be paid to the needs of divorcing f a m i l i e s with grown ch i l d r e n . For example, s o c i a l workers and counsellors who work i n agencies that are funded by the government need to be train e d i n dealing with the issues unique to t h i s group, and to recognize the developmental issues of adult c h i l d r e n and t h e i r parents. Outreach programs may also be necessary to reach men, who i n general, do not seek out counselling for family d i f f i c u l t i e s (e.g. Kruk, 1993). There may also be implications for government p o l i c y on family caregiving. Although t h i s study found no s i g n i f i c a n t differences i n the lev e l s of support that continuously-married and divorced respondents reported, i n one way t h i s f i n d i n g i s disquieting, as the expectation was that adult c h i l d r e n would increase t h e i r l e v e l s of support to divorced parents. It may be, of course, that the parents did not need or did not want support from t h e i r c h i l d r e n by the time of the postdivorce interview. However, i t may also be that the adult c h i l d r e n were not responding to increased needs of t h e i r parents, f o r a v a r i e t y of reasons. Because of f i s c a l r e s t r a i n t s , government p o l i c y i s strongly emphasizing the ro l e of chi l d r e n as caregivers to aging parents, and one question that arises from t h i s study i s how divorce a f t e r long-term marriage w i l l influence the 119 response of adult c h i l d r e n to the increased needs of t h e i r parents for support l a t e r i n l i f e . V. Conclusion Although the rates of divorce a f t e r long-term marriage have been increasing, research into t h i s phenomenon i s only beginning. This study has contributed to our knowledge of divorce a f t e r long-term marriage by examining divorce e f f e c t s on parent's psychological well-being and parent-adult c h i l d r e l a t i o n s h i p s , and the subsequent e f f e c t of the parent-child r e l a t i o n s h i p s on the parent's divorce adjustment. Results show that these e f f e c t s vary according to the dimension being measured, with divorce having negative e f f e c t s on s e l f - r a t i n g s of happiness, depression and parent-child contact, but no s i g n i f i c a n t e f f e c t s on self-esteem, support given or support received. Also, contrary to expectations, the dimensions of parent-adult c h i l d r e l a t i o n s h i p s measured i n the study d i d not moderate the negative e f f e c t s of divorce on either happiness or depression for e i t h e r mothers or fathers. The r e s u l t s of t h i s study suggest several avenues of i n v e s t i g a t i o n that future research could follow. One of these i s to broaden the measurement of parent-adult c h i l d r e l a t i o n s h i p s to include measures of a f f e c t i o n a l s o l i d a r i t y . Researchers basing t h e i r studies on the family s o l i d a r i t y model frequently use a ten-item scale for a f f e c t i o n a l s o l i d a r i t y that asks parents to rate the understanding, t r u s t , fairness, respect and a f f e c t i o n they perceive a p a r t i c u l a r c h i l d has for them and that they have for the c h i l d . For example, "How well do you f e e l t h i s c h i l d understands you?" followed l a t e r i n the interview by "How well do you understand him or her?" ( S i l v e r s t e i n & Bengtson, 1991; see also Roberts & Bengtson, 1990). The q u a l i t i e s of the parent-adult c h i l d 120 r e l a t i o n s h i p measured i n the intergenerational a f f e c t i o n a l solidarity-scale may prove to have more of an e f f e c t on the parent's postdivorce psychological well-being than the s o c i a l support measures used i n the present study. Perhaps one of the most i n t e r e s t i n g findings i n t h i s study was the complete absence of gender differences for any of the hypotheses. For these respondents, divorce d i d not have a more negative e f f e c t on parent-adult c h i l d r e l a t i o n s h i p s for fathers than for mothers, and the e f f e c t s of those r e l a t i o n s h i p s on divorce adjustment was not s i g n i f i c a n t l y d i f f e r e n t for mothers or fathers. One of the ideas underlying t h i s study was that adult c h i l d r e n are a p o t e n t i a l resource for parents divorcing a f t e r a long-term marriage and, although the measures used di d not confirm t h i s , the r e s u l t s suggest that the father-adult c h i l d r e l a t i o n s h i p may be stronger than previous research has indicated. Another avenue of research suggested by t h i s study, therefore, i s a c l o s e r examination of the dynamics of parent-adult c h i l d r e l a t i o n s h i p s during the parent's divorce, with p a r t i c u l a r attention paid to father-adult c h i l d r e l a t i o n s h i p s . Contrary to p r e v a i l i n g assumptions i n the l i t e r a t u r e that adjustment to divorce at older ages must be more d i f f i c u l t than adjustment to divorce at younger ages (Berardo, 1982; Price & McKenry, 1988; Raschke, 1987), t h i s study showed no s i g n i f i c a n t age differences between respondents younger than f i f t y and those older than f i f t y for any of the hypotheses tested f o r divorce by age i n t e r a c t i o n s . These r e s u l t s , therefore, support Kitson's (e.g. Kitson & Morgan, 1990) contention that the negative assumptions i n the divorce l i t e r a t u r e need to be re-examined. As was done in t h i s study, researchers could use Bohannan's (1970) six-part conceptualization of divorce as a framework when considering how divorces 121 a f t e r long-term marriage when chil d r e n are grown may d i f f e r from divorces e a r l i e r i n l i f e when ch i l d r e n are young. While research, media and p o l i c y attention remains f i x e d on the causes and e f f e c t s of divorce i n younger f a m i l i e s , the number of couples who are divorcing a f t e r long-term marriages i s q u i e t l y increasing. As Ahrons and Rodgers (1987) speculate t h e o r e t i c a l l y and Hagestad and her colleagues have shown i n a three-generational study, the r i p p l e e f f e c t s (Hagestad, 1981) of divorce a f t e r long-term marriage spread up and down the family lineage, with r e l a t i o n s h i p changes i n a l l family dyads. From grandchildren to great-grandparents, family members must cope with the counter-transitions (Riley & Waring, 1976) brought about by the (perhaps unexpected and unwanted) divorce of the parents i n the middle of the family chain. Just at the stage i n the family l i f e cycle when family t r a d i t i o n s are passed on to adult c h i l d r e n who are e s t a b l i s h i n g f a m i l i e s of t h e i r own, the i d e n t i t y of a family i s irrevocably altered, leaving grandchildren, adult children, parents, grandparents, and perhaps great-grandparents with the tasks of creating a new family i d e n t i t y while at the same time r e t a i n i n g the h i s t o r y and legacy of the o l d to pass on to future generations. When considered i n these terms, the increasing rates of divorce a f t e r long-term marriage may change the shape of North American f a m i l i e s as much as, or even more than, divorce i n e a r l i e r l i f e has done. This i s the perspective from which the present study was planned, and i t i s hoped that the r e s u l t s of the study both advance our s c i e n t i f i c knowledge of divorce a f t e r long-term marriage and, i n doing so, also contribute to the development of e f f e c t i v e interventions to help parents and t h e i r adult c h i l d r e n adjust to the divorce. 122 R e f e r e n c e s Ahrons, C. R. (1986, June). Divorce when the c h i l d r e n are older. Paper presented at a meeting of the American Family Therapy Association Conference, Washington, D. C. Ahrons, C. R., & M i l l e r , R. B. (1993). The e f f e c t of the postdivorce r e l a t i o n s h i p on paternal involvement: A l o n g i t u d i n a l a n a l y s i s . American Journal of Orthopsychiatry, 63(3), 441-450. Ahrons, C. F., & Rodgers, R. H. (1987). Divorced f a m i l i e s : Meeting the challenge of divorce and remarriage. New York: Norton. Aneshensel, C. S., Frerichs, R. R., & Clark, V. A. (1981). Family roles and sex differences i n depression. Journal of Health and S o c i a l Behavior, 22, 379-393. Aquilino, W. S. (1994a). Impact of childhood family d i s r u p t i o n on young adults' r e l a t i o n s h i p s with parents. Journal of Marriage and the Family, 56, 295-313. Aquilino, W. S. (1994b). Later l i f e parental divorce and widowhood: Impact on young adults' assessment of parent-child r e l a t i o n s . Journal of Marriage and the Family, 56, 906-922. Atkinson, M. P. (1989). Conceptualizations of the parent-child r e l a t i o n s h i p : S o l i d a r i t y , attachment, crescive bonds, and i d e n t i t y salience. In J. A. Mancini (Ed.), Aging parents and adult c h i l d r e n (pp. 81-97). Lexington, MA: Lexington Books. Barnett, R. C , Marshall, N. L., & Pleck, J. H. (1992). Adult son-parent r e l a t i o n s h i p s and t h e i r associations with sons' psychological d i s t r e s s . Journal of Family Issues, 13(4), 505-525. Bengtson, V. L., & Dannefer, D. (1987). Families, work and aging: Implications of disordered cohort flow for the 21st century. In R. A. Ward & S. S. Tobin (Eds.), Health i n aging: S o c i o l o g i c a l issues and p o l i c y d i r e c t i o n s (pp. 256-289). New York: Springer. Bengtson, V. L., & Kuypers, J. A. (1971). Generational differences and the "developmental stake." Aging and Human Development, 2(1), 249-260. Bengtson, V. L., Rosenthal, C , & Burton, L. (1990). Families and aging: D i v e r s i t y and heterogeneity. In R. H. Binstock & L. K. George (Eds.), Handbook of aging and the s o c i a l sciences (3rd ed., pp. 263-287). San Diego, CA: Academic Press. Berardo, D. H. (1982). Divorce and remarriage at middle age and beyond. The Annals of the American Academy of P o l i t i c a l and S o c i a l Science, 464, 132-139. Bernard, J. (1972) . The future of marriage. New York: World Publications. Bloom, B. L., & Caldwell, R. A. (1981). Sex differences i n adjustment during the process of marital separation. Journal of Marriage and the Family, 43, 693-701. Bloom, B. L., & Clement, C. (1984). M a r i t a l sex r o l e o r i e n t a t i o n and adjustment to separation and divorce. Journal of Divorce, 7(3). 87-98. 123 Bloom, B. L., Hodges, W. F., Kern, M. B., & McFaddin, S. C. (1985). A preventive intervention program for the newly separated: F i n a l evaluations. American Journal of Orthopsychiatry, 55(1), 9-26. Bloom, B. L., White, S. W., & Asher, S. J. (1979). M a r i t a l d i s r u p t i o n as a s t r e s s f u l l i f e event. In G. Levinger & 0. C. Moles (Eds.). Divorce and separation: Context, causes, and consequences (pp. 184-200) . New York: Basic Books. Bohannan, P. (1970). The s i x stations of divorce. In P. Bohannan, Divorce and a f t e r (pp. 29-55) . Garden City, NY: Doubleday. Bonkowski, S. E. (1989). Lingering sadness: Young adults' response to parental divorce. S o c i a l Casework. 70(4), 219-223. Booth, A., & Amato, P. (1991). Divorce and psychological s t r e s s . Journal of Health and S o c i a l Behavior, 32, 396-407. Booth, A., & Amato, P. R. (1994). Parental marital q u a l i t y , parental divorce, and r e l a t i o n s with parents. Journal of Marriage and the Family, 56, 21-34 . Booth, A., Johnson, D. R., White, L. K., & Edwards, J. N. (1986). Divorce and marital i n s t a b i l i t y over the l i f e course. Journal of Family Issues, 7(4), 421-442. Bradburn, N. M. (1969) . The structure of psychological well-being. Chicago: Aldine. Bradburn, N. M., & Caplovitz, D. (1965). Reports on happiness. Chicago: Aldine. Braver, S. L., Wolchik, S. A., Sandler, I. N., Sheets, V. L., Fogas, B., & Bay, R. C. (1993). A l o n g i t u d i n a l study of noncustodial parents: Parents without c h i l d r e n . Journal of Family Psychology, 7(1), 9-23. Brehm, S. S. (1985) . Intimate r e l a t i o n s h i p s . New York: Random House. Bray, J . H., & Hetherington, E. M. (1993). Families i n t r a n s i t i o n : Introduction and overview. Journal of Family Psychology, 7(1), 3-8. Brubaker, T. H. (1990). Families i n l a t e r l i f e : A burgeoning research area. Journal of Marriage and the Family, 52, 959-981. Bryant, F. B., & Veroff, J. (1984). Dimensions of subjective mental health i n American men and women. Journal of Health and S o c i a l Behavior, 25, 116-135. Bul c r o f t , K. A., & Bulcroft, R. A. (1991). The timing of divorce: E f f e c t s on parent-child r e l a t i o n s h i p s i n l a t e r l i f e . Research on Aging, 13(2), 226-243. Burke, P. J. (1980). The s e l f : Measurement requirements from an i n t e r a c t i o n i s t perspective. S o c i a l Psychology Quarterly, 43(1), 18-29. Burke, P. J . (1991). Identity processes and s o c i a l s t r e s s . American S o c i o l o g i c a l Review, 56, 836-849. Bumpass, L., & Sweet, J. A. (1991). The e f f e c t of marital disruption on intergenerational r e l a t i o n s h i p s . (NSFH Working Paper No. 40). Madison: U n i v e r s i t y of Wisconsin, Center for Demography and Ecology. 124 Campbell, M. (1995). Divorce at m i d - l i f e : Intergenerational issues. Journal of Divorce and Remarriage, 23(1/2), 185-202. Canada. S t a t i s t i c s Canada. Divorces, May, 1996, (Catalogue No. 84-213-XMB). [mirofiche] Chiriboga, D. A. (1982). Adaptation to marital separation i n l a t e r and e a r l i e r l i f e . Journal of Gerontology, 37(1), 109-119. Chiriboga, D. A., Catron, L. S. and Associates. (1991). Divorce: C r i s i s , challenge or r e l i e f ? New York: New York U n i v e r s i t y Press. Chiriboga, D. A., Roberts, J., & Stein, J. A. (1978). Psychological well-being during marital separation. Journal of Divorce, 2(1), 21-36. C i c i r e l l i , V. G. (1983). Adult c h i l d r e n and t h e i r elder parents. In T. H. Brubaker (Ed.), Family r e l a t i o n s h i p s i n l a t e r l i f e (pp. 31-46). Beverly H i l l s , CA: Sage. Cohen, T. F. (1987). Remaking men: Men's experiences becoming and being husbands and fathers and t h e i r implications f o r reconceptualizing men's l i v e s . Journal of family issues, 8(1), 57-77. Coleman, M., & Ganong, L. H. (1993). Families and marital d i s r u p t i o n . In T. H. Brubaker (Ed.), Family r e l a t i o n s : Challenges for the future (pp. 112-128) . Newbury Park, CA: Sage. Connidis, I. A. (1989). Family t i e s and aging. Toronto: Butterworths. Connidis, I. A., & McMullin, J. A. (1993). To have or have not: Parent status and the subjective well-being of older men and women. The Gerontologist, 33(5), 630-636. Coombs, R. H. (1991). M a r i t a l status and personal well-being: A l i t e r a t u r e review. Family Relations, 40, 97-102. Cooney, T. M. (1989). Co-residence with adult c h i l d r e n : A comparison of divorced and widowed women. The Gerontologist, 6, 77 9-784. Cooney, T. M. (1994). Young adults' r e l a t i o n s with parents: The influence of recent parental divorce. Journal of Marriage and the Family, 56, 45-56. Cooney, T. M., & Uhlenberg, P. (1990). The r o l e of divorce i n men's re l a t i o n s with t h e i r adult c h i l d r e n a f t e r m i d - l i f e . Journal of Marriage and the Family, 52, 677-688. Cooney, T. M., Hutchinson, M. K., & Leather, D. M. (1995). Surviving the breakup? Predictors of parent-adult c h i l d r e l a t i o n s a f t e r parental divorce. Family Relations, 44, 153-161. Cooney, T. M., Smyer, M. A., Hagestad, G. 0., & Klock, R. (1986). Parental divorce i n young adulthood: Some preliminary f i n d i n g s . American Journal of Orthopsychiatry, 56(3), 470-477. Crimmins, E. M., & Ingegneri, D. G. (1990). Interaction and l i v i n g arrangements of older parents and t h e i r c h i l d r e n : Past trends, present determinants, future implications. Research on Aging, 12(1) , 3-35. Doherty, W. J., Su, S., & Needle, R. (1989). M a r i t a l d i s r u p t i o n and psychological well-being: A panel study. Journal of Family Issues, 10(1), 72-85. 125 Eggebeen, D. J. (1992) . Family structure and intergenerational exchanges. Research on Aging, 14(4), 427-447. Fazio, A. F. (1977) . A concurrent v a l i d a t i o n a l study of the NCHS General Well-Being Scale (DHEW Publication No. [HRA] 78-1347) . H y a t t s v i l l e , MD: National Center f o r Health S t a t i s t i c s . Furstenberg, F. F., Nord, C. W., Peterson, J. L., & Z i l l , N. (1983). The l i f e course of ch i l d r e n of divorce: M a r i t a l d i s r u p t i o n and parental contact. American S o c i o l o g i c a l Review, 48, 656-668. Gander, A. M. (1991). A f t e r the divorce: F a m i l i a l factors that predict well-being f o r older and younger persons. Journal of Divorce and Remarriage. 15(2), 175-192. Gander, A. M., & Jorgensen, L. A. B. (1990). Postdivorce adjustment: S o c i a l supports among older divorced persons. Journal of Divorce, 13(4), 37-52. Gecas, V., & Seff, M. A. (1990). S o c i a l class and self-esteem: Psychological c e n t r a l i t y , compensation, and the r e l a t i v e e f f e c t s of work and home. S o c i a l Psychology Quarterly, 53(2), 165-173. George, L. K. (1990) . S o c i a l structure, s o c i a l processes, and soci a l - p s y c h o l o g i c a l states. In R. H. Binstock & L. K. George (Eds.), Handbook of aging and the s o c i a l sciences (3rd ed., pp. 186-204). San Diego, CA: Academic Press. Gerstel, N. (1987). Divorce and stigma. S o c i a l Problems, 34(2), 172-186 . G i l l i g a n , C. (1982). In a d i f f e r e n t voice. Cambridge, MA: Harvard U n i v e r s i t y Press. Glenn, N. D., & McLanahan, S. (1981). The e f f e c t s of o f f s p r i n g on the psychological well-being of older adults. Journal of Marriage and Family, 43, 409-421. Glenn, N. D., & Shelton, B. A. (1985). Regional differences i n divorce i n the United States. Journal of Marriage and the Family, 47, 641-652. Goodman, C. C. (1993). Divorce a f t e r long term marriages: Former spouse r e l a t i o n s h i p s . Journal of Divorce and Remarriage, 20(1/2), 43-61. Gore, S., & Mangione, T. W. (1983). S o c i a l r o l e s , sex roles and psychological d i s t r e s s : Additive and i n t e r a c t i v e models of sex differences. Journal of Health and So c i a l Behavior, 24, 300-312. Gove, W. R., & Shin, H. (1989). The psychological well-being of divorced and widowed men and women: An empirical a n a l y s i s . Journal of Family Issues, 10(1), 122-144. Gove, W. R., Hughes, M., & Style, C. B. (1983). Does marriage have p o s i t i v e e f f e c t s on the psychological well-being of the indiv i d u a l ? Journal of Health and So c i a l Behavior, 24, 122-131. Gove, W. R., Style, C. B., & Hughes, M. (1990). The e f f e c t of marriage on the well-being of adults: A t h e o r e t i c a l a n a l y s i s . Journal of Family Issues, 11(1), 4-35. Hagestad, G. 0. (1981) . Problems and promises i n the s o c i a l psychology of intergenerational r e l a t i o n s . In R. W. Fogel, E. H a t f i e l d , S. 126 B. K i e s l e r & E. Shanas (Eds.), Aging: S t a b i l i t y and change i n the family (pp. 11-46). New York: Academic Press. Hagestad, G. 0. (1987) . Parent-child r e l a t i o n s i n l a t e r l i f e : Trends and gaps i n past research. In J. B. Lancaster, J. Altmann, A. S. Rossi, & L. R. Sherrod (Eds.), Parenting across the l i f e span: B i o s o c i a l dimensions (pp. 405-433) . New York: Aldine de Gruyter. Hagestad, G. O. (1988). Demographic change and the l i f e course: Some emerging trends i n the family realm. Family Relations. 37, 405-410. Hagestad, G. O., & Smyer, M. A. (1982). Dissolving long-term r e l a t i o n s h i p s : Patterns of divorcing i n middle-age. In S. Duck (Ed.), Personal r e l a t i o n s h i p s . 4: Dissolving personal r e l a t i o n s h i p s (pp. 155-188) . London: Academic Press. Hagestad, G. 0., Smyer, M. A., & Stierman, K. (1984). The impact of divorce i n middle age. In R. S. Cohen, B. J. Cohler, & S. H. Weissman, Parenthood: A psvchodynamic perspective (pp. 247-262). New York: G u i l f o r d . Hammond, R. J., & Muller, G. 0. (1992). The l a t e - l i f e divorced: Another look. Journal of Divorce and Remarriage, 17(3/4), 135-150. Hatch, L. R., & S t u l l , D. E. (1987). Unraveling the e f f e c t s of multiple l i f e changes: A reanalysis. Research on Aging. 8 (4) , 505-514. Hennon, C. B. (1983) . Divorce and the e l d e r l y : A neglected area of research. In T. H. Brubaker (Ed.), Family r e l a t i o n s h i p s i n l a t e r l i f e (pp. 149-172) . Beverly H i l l s , CA: Sage. Hetherington, E. M. (1993). An overview of the V i r g i n i a Longitudinal Study of Divorce and Remarriage with a focus on ea r l y adolescence. Journal of Family Psychology, 7(1), 39-56. Hetherington, E. M., Cox, M., & Cox, R. (1978). The aftermath of divorce. In J. H. Stevens & M. Matthews, Mother/child f a t h e r / c h i l d r e l a t i o n s h i p s (pp. 149-176). Washington, DC: National Association f o r the Education of Young Children. Hetherington, E. M., Cox, M., & Cox, R. (1982). E f f e c t s of divorce on parents and ch i l d r e n . In M. E. Lamb (Ed.), Nontraditional f a m i l i e s : Parenting and c h i l d development (pp. 233-288) . H i l l s d a l e , NJ: Lawrence Erlbaum. Horwitz, A. V., & White, H. R. (1991). Becoming married, depression, and alcohol problems among young adults. Journal of Health and So c i a l Behavior, 32, 221-237. Hughes, M. (1989) . Parenthood and psychological well-being among the formerly married: Are chi l d r e n the primary source of psychological d i s t r e s s ? Journal of Family Issues, 10(4), 463-481. Johnson, C. L. (1988) . Ex f a m i l i a : Grandparents, parents, and chi l d r e n adjust to divorce. New Brunswick: Rutgers U n i v e r s i t y Press. Jones, S. L., & Jones, P. K. (1993). The ro l e of adult c h i l d r e n i n t h e i r parent's divorce. Journal of Divorce and Remarriage, 21(1/2), 55-71. Kitson, G. C. with W. M. Holmes. (1992). P o r t r a i t of divorce: Adjustment to marital breakdown. New York: G u i l f o r d Press. Kitson, G. C , & Morgan, L. A. (1990). The multiple consequences of divorce: A decade review. Journal of Marriage and the Family, 52, 913-924. 127 Kitson, G. C., & Roach, M. J. (1989). Independence and s o c i a l and psychological adjustment i n widowhood and divorce. In D. A. Lund (Ed.), Older bereaved spouses: Research with p r a c t i c a l a p plications (pp. 167-183) . New York: Hemisphere. Kruk, E. (1991) . Discontinuity between pre- and post-divorce father-c h i l d r e l a t i o n s h i p s : New evidence regarding paternal disengagement. Journal of Divorce and Remarriage, 16(3-4), 195-227. Kruk, E. (1993). Divorce and disengagement: Patterns of fatherhood within and beyond marriage. Halifax: Fernwood. Kurdek, L. R. (1991) . The r e l a t i o n s between reported well-being and divorce h i s t o r y , a v a i l a b i l i t y of a proximate adult, and gender. Journal of Marriage and the Family, 53, 71-78. Lawton, L., S i l v e r s t e i n , M., & Bengtson, V. (1994). A f f e c t i o n , s o c i a l contact, and geographic distance between adult c h i l d r e n and t h e i r parents. Journal of Marriage and the Family, 56, 57-68. Lee, G. R., & E l l i t h o r p e , E. (1982). Intergenerational exchange and subjective well-being among the e l d e r l y . Journal of Marriage and the Family, 44 (1) , 217-224. L e v i t t , M. J., Guacci, N., & Weber, R. A. (1992). Intergenerational support, r e l a t i o n s h i p q u a l i t y , and well-being: A b i c u l t u r a l a n a l ysis. Journal of Family Issues, 13(4), 465-481. Lewis, C., & O'Brien, M. (1987). Constraints on fathers: Research, theory and c l i n i c a l p r a c t i c e . In C. Lewis & M. O'Brien (Eds.), Reassessing fatherhood: New observations on fathers and the modern family (pp. 1-19) . London: Sage. Lloyd, S. A., & Zick, C. D. (1986). Divorce at mid and l a t e r l i f e : Does the empirical evidence support the theory? Journal of Divorce, 9(3), 89-102. Mancini, J. A., & Blieszner, R. (1989). Aging parents and adult ch i l d r e n : Research themes i n intergenerational r e l a t i o n s . Journal of Marriage and the Family, 51, 275-290. Marciano, T. D. (1987). Families and r e l i g i o n . In M. B. Sussman & S. K. Steinmetz (Eds.), Handbook of marriage and the family (pp. 285-315). New York: Plenum Press. Marks, N. F. (1991). Remarried and single parents i n middle adulthood: Differences i n psychological well-being and r e l a t i o n s h i p s with adult c h i l d r e n (NSFH Working Paper No. 47). Madison: U n i v e r s i t y of Wisconsin, Center for Demography and Ecology. Marks, N. F. (1995). M i d l i f e marital status differences i n s o c i a l support r e l a t i o n s h i p s with adult c h i l d r e n and psychological well-being. Journal of Family Issues, 16(1), 5-28. Mastekaasa, A. (1994a). The subjective well-being of the previously married: The importance of unmarried cohabitation and time since widowhood or divorce. S o c i a l Forces, 73(2), 665-692. Mastekaasa, A. (1994b). M a r i t a l status, d i s t r e s s , and well-being: An i n t e r n a t i o n a l comparison. Journal of Comparative Family Studies, 25(2), 183-205 . 128 McCall, G. J., & Simmons, J. L. (1978). I d e n t i t i e s and i n t e r a c t i o n s : An examination of human associations i n everyday l i f e . New York: The Free Press. McLanahan, S., & Adams, J. (1987). Parenthood and psychological well-being. Annual Review of Sociology, 5, 237-257. McLanahan, S., & Adams, J. (1989). The e f f e c t s of c h i l d r e n on adults' psychological well-being: 1957-1976. S o c i a l Forces, 68(1), 124-146. McLanahan, S., Wedemeyer, N. V., & Adelberg, T. (1981). Network structure, s o c i a l support, and psychological well-being i n the si n g l e -parent family. Journal of Marriage and the Family, 43(3), 601-612. Menaghan, E. G. (1985) . Depressive a f f e c t and subsequent divorce. Journal of Family Issues, 6(3), 295-306. Menaghan, E. G. (1989). Role changes and psychological well-being: Variations i n e f f e c t s by gender and r o l e r e p e r t o i r e . S o c i a l Forces, 67(3), 693-714. Menaghan, E. G., & Lieberman, M. A. (1986). Changes i n depression following divorce: A panel study. Journal of Marriage and the Family, 48, 319-328 . Mercier, J. M., Paulson, L., & Morris, E. W. (1989). Proximity as a mediating influence on the perceived aging parent-adult c h i l d r e l a t i o n s h i p . The Gerontologist, 29(6), 785-791. Morgan, L. A. (1991) . Af t e r marriage ends: Economic consequences for m i d l i f e women. Newbury Park, CA: Sage. Moss, M. S., & Moss, S. Z. (1992). Themes i n pa r e n t - c h i l d r e l a t i o n s h i p s when e l d e r l y parents move nearby. Journal of Aging Studies, 6(3), 259-271. Moss, M. S., Moss, S. Z., & Moles, E. L. (1985). The q u a l i t y of r e l a t i o n s h i p s between e l d e r l y parents and t h e i r out-of-town c h i l d r e n . The Gerontologist, 25(2), 134-140. Mutran, E., & Reitzes, D. C. (1984). Intergenerational support a c t i v i t i e s and well-being among the e l d e r l y : A convergence of exchange and symbolic i n t e r a c t i o n perspectives. American S o c i o l o g i c a l Review, 49, 117-130. National Center for Health S t a t i s t i c s . Advance report of f i n a l divorce s t a t i s t i c s , 1988. Monthly V i t a l S t a t i s t i c s Report. 39. Washington, DC: U. S. Public Health Service. Nuttbrock, L., & Freudiger, P. (1991). Identity salience and motherhood: A te s t of Stryker's theory. S o c i a l Psychology Quarterly, 54 (2) , 146-157. Peck, J. S. (1988-89). The impact of divorce on c h i l d r e n at various stages of the family l i f e cycle. Journal of Divorce and Remarriage, 12(2-3), 81-106. Peterson, G. W. & R o l l i n s , B. C. (1987) . Parent-child s o c i a l i z a t i o n . In M. B. Sussman & S. K. Steinmetz (Eds.), Handbook of marriage and the family (pp. 471-507). New York: Plenum Press. Pett, M. A., Lang, N., & Gander, A. (1992). L a t e - l i f e divorce: Its impact on family r i t u a l s . Journal of Family Issues, 13(4), 526-552. 129 Pleck, J. H. (1985). Working wives/working husbands. Beverly H i l l s , CA: Sage. Pledge, D. S. (1992). M a r i t a l separation/divorce: A review of i n d i v i d u a l responses to a major l i f e stressor. Journal of Divorce and Remarriage, 17(3/4), 151-181. Price, S. J., & McKenry, P. C. (1988). Divorce. Newbury Park, CA: Sage. Quinn, W. H. (1983). Personal and family adjustment i n l a t e r l i f e . Journal of Marriage and the Family, 45(1), 57-73. Radloff, L. S. (1977). The CES-D scale: A s e l f - r e p o r t depression scale f o r research i n the general population. Applied Psychological Measurement, 1, 385-401. Raschke, H. J. (1987) . Divorce. In M. B. Sussman & S. K. Steinmetz (Eds.), Handbook of marriage and the family (pp. 597-624). New York: Plenum Press. Reskin, B. F., & Coverman, S. (1985). Sex and race i n the determinants of psychophysical d i s t r e s s : A reappraisal of the sex-role hypothesis. S o c i a l Forces, 63(4), 1038-1059. Riessman, C. K., & Gerstel, N. (1985) . M a r i t a l d i s s o l u t i o n and health: Do males or females have greater risk ? S o c i a l Science and Medicine, 20(6), 627-635. Riley, M. W., & Waring, J. (1976). Age and aging. In R. K. Merton & R. Nisbet (Eds.), Contemporary s o c i a l problems (4th ed.). New York: Harcourt Brace Jovanovich. Roach, M. J., & Kitson, G. C. (1989). Impact of forewarning on adjustment to widowhood and divorce. In D. A. Lund (Ed.), Older bereaved spouses: Research with p r a c t i c a l applications (pp. 185-203). New York: Hemisphere. Roberts, R. E. L., & Bengtson, V. L. (1993). Relationships with parents, self-esteem, and psychological well-being i n young adulthood. S o c i a l Psychology Quarterly, 56(4), 263-277. Rook, K. S. (1984). The negative side of s o c i a l i n t e r a c t i o n : Impact on psychological well-being. Journal of Personality and S o c i a l Psychology, 46, 1097-1108. Rosenberg, M. (1989) . Society and the adolescent self-image. Middletowm, CT: Wesleyan Un i v e r s i t y Press. Rosenthal, C. (1985) . Kinkeeping i n the f a m i l i a l d i v i s i o n of labor. Journal of Marriage and the Family, 47(4), 965-974. Rosenthal, C. (1987) . Aging and intergenerational r e l a t i o n s i n Canada. In V. W. Marshall (Ed.), Aging i n Canada: S o c i a l perspectives (2nd ed., pp. 311-342). Markham, ON: Fitzhenry & Whiteside. Rossi, A. S., & Rossi, P. H. (1990). Of human bonding: Parent-child r e l a t i o n s across the l i f e course. New York: Aldine de Gruyter. Rossi, A. S., & Rossi, P. H. (1991). Normative obligations and parent-child help exchange across the l i f e course. In K. Pillemer & K. McCartney (Eds.), Parent-child r e l a t i o n s throughout l i f e (pp. 210-223). H i l l s d a l e , NJ: Lawrence Erlbaum. 130 Seltzer, J . A. (1991) . Relationships between fathers and c h i l d r e n who l i v e apart: The father's r o l e a f t e r separation. Journal of Marriage and the Family, 53, 79-101. Scott, J., & Alwin, D. F. (1989). Gender differences i n parental r o l e s t r a i n : Parental r o l e or gender role? Journal of Family Issues, 10(4), 482-503. Seltzer, J. A., & Brandreth, Y. (1994). What fathers say about involvement with c h i l d r e n a f t e r separation. Journal of Family Issues, 15 (1), 49-77. Serpe, R. T. (1987) . S t a b i l i t y and change i n s e l f : A s t r u c t u r a l symbolic i n t e r a c t i o n i s t explanation. S o c i a l Psychology Quarterly, 50(1), 44-55. S i l v e r s t e i n , M., & Bengtson, V. L. (1991). Do close parent-child r e l a t i o n s reduce the mortality r i s k of older parents? Journal of Health and S o c i a l Behavior, 32, 382-395. Simon, R. W. (1992). Parental role s t r a i n s , salience of parental i d e n t i t y and gender differences i n psychological d i s t r e s s . Journal of Health and S o c i a l Behavior, 33, 25-35. Smyer, M. A., & Hofland, B. F. (1982) . Divorce and family support i n l a t e r l i f e : Emerging concerns. Journal of Family Issues, 3(1), 61-77. Spanier, G. B., & Thompson, L. (1984). Parting: The aftermath of separation and divorce. Beverly H i l l s , CA: Sage. Stryker, S. (1968). Identity salience and r o l e performance: The relevance of symbolic i n t e r a c t i o n theory for family research. Journal of Marriage and the family, 30(4), 558-564. Stryker, S. (1980). Symbolic interactionism: A s o c i a l s t r u c t u r a l version. Menlo Park, CA: Benjamin/Cummings. Stryker, S. (1981). Symbolic interactionism: Themes and v a r i a t i o n s . In M. Rosenberg & R. H. Turner (Eds.), S o c i a l psychology: S o c i o l o g i c a l perspectives (pp. 3-29) . New York: Basic Books. Stryker, S. (1987). The v i t a l i z a t i o n of symbolic interactionism. S o c i a l Psychology Quarterly, 50(1), 83-94. Stryker, S., & Serpe, R. T. (1982). Commitment, i d e n t i t y salience, and r o l e behavior: Theory and research example. In W. Ickes & E. S. Knowles (Eds.), Personality, roles and s o c i a l behavior (pp. 199-218). New York: Springer-Verlag. Sweet, J. A., Bumpass, L. L., & C a l l , V. (1988). The design and content of the National Survey of Families and Households (NSFH Working Paper No. 1). Madison: U n i v e r s i t y of Wisconsin, Center for Demography and Ecology. Tschann, J. M., Johnston, J. R., & Wallerstein, J. S. (1989). Resources, stressors, and attachment as predictors of adult adjustment a f t e r divorce: A l o n g i t u d i n a l study. Journal of Marriage and the Family, 51, 1033-1046. Tcheng-Laroche, F., & Prince, R. (1983). Separated and divorced women compared with married controls: Selected l i f e s a t i s f a c t i o n , stress and health indices from a community survey. S o c i a l Science and Medicine, 17 (2), 95-105 . 131 Thoits, P. A. (1983). Multiple i d e n t i t i e s and psychological well-being: A reformulation and test of the s o c i a l i s o l a t i o n hypothesis. American S o c i o l o g i c a l Review, 48, 174-187. Thoits, P. A. (1991) . On merging i d e n t i t y theory and stress research. S o c i a l Psychology Quarterly, 54(2), 101-112. Thoits, P. A. (1992). Identity structures and psychological well-being: Gender and marital status comparisons. S o c i a l Psychology Quarterly, 55(3), 236-256. Thomas, W. I., & Znaniecki, F. (1927). The p o l i s h peasant i n Europe and America. New York: A l f r e d A. Knopf. T r o l l , L. E. (1982) . Family l i f e i n middle and o l d age: The generation gap. The Annals of the American Academy of P o l i t i c a l and Soci a l Science, 464, 38-46. T r o l l , L. E. (1986). Parent-adult c h i l d r e l a t i o n s . In L. E. T r o l l (Ed.), Family issues i n current gerontology (pp. 75-83). New York: Springer. Uhlenberg, P., & Myers, M. A. P. (1981). Divorce and the e l d e r l y . The Gerontologist, 21(3), 276-282. Uhlenberg, P., Cooney, T., & Boyd, R. (1990). Divorce for women af t e r m i d l i f e . Journal of Gerontology, 45(1), S3-11. Umberson, D. (1989a). Relationships with c h i l d r e n : Explaining parents' psychological well-being. Journal of Marriage and the Family, 51, 999-1012. Umberson, D. (1989b). Parenting and well-being: The importance of context. Journal of Family Issues, 10(4), 427-439. Umberson, D. (1992) . Relationships between adult c h i l d r e n and t h e i r parents: Psychological consequences for both generations. Journal of Marriage and the Family, 54, 664-674. Umberson, D., & Gove, W. R. (1989). Parenthood and psychological well-being: Theory, measurement, and stage i n the family l i f e course. Journal of Family Issues, 10(4), 440-462. Umberson, D., & Williams, C. L. (1993). Divorced fathers: Parental r o l e s t r a i n and psychological d i s t r e s s . Journal of Family Issues, 14(3), 378-400. Waite, L. J., & L i l l a r d , L. A. (1991). Children and marital di s r u p t i o n . American Journal of Sociology, 96(4), 930-953. Walker, A. J., & Thompson, L. (1983). Intimacy and intergenerational aid and contact among mothers and daughters. Journal of Marriage and the Family, 45(4), 841-849. Wallerstein, J. S. (1986). Women a f t e r divorce: Preliminary report from a ten-year follow-up. American Journal of Orthopsychiatry, 56(1), 65-77. Wallerstein, J. S., & Kelly, J. B. (1980). Surviving the breakup: How c h i l d r e n and parents cope with divorce. New York: Basic Books. 132 Ward, R. A., & Spitze, G. (1992). Consequences of parent-adult c h i l d coresidence: A review and research agenda. Journal of Family Issues, 13(4), 553-572. Weingarten, H. R. (1985) . M a r i t a l status and well-being: A national study comparing f i r s t - m a r r i e d , currently divorced, and remarried adults. Journal of Marriage and the Family, 47(3), 653-662. Weingarten, H. R. (1988). Late l i f e divorce and the l i f e review. Journal of Divorce, 12, 21-39. Weingarten, H., & Bryant, F. B. (1987). M a r i t a l status and the meaning of subjective well-being: A s t r u c t u r a l a n a lysis. Journal of Marriage and the Family. 49, 883-892. Weissman, M., & Paykel, E. (1974). The depressed woman: A study of s o c i a l r e l a t i o n s h i p s . Chicago: Un i v e r s i t y of Chicago Press. White, L. K. (1990). Determinants of divorce: A review of research i n the E i g h t i e s . Journal of Marriage and the Family, 52, 904-912. White, L. (1992). The e f f e c t of parental divorce and remarriage on parental support for adult c h i l d r e n . Journal of Family Issues, 13(2), 234-250. White, L. K., & Booth, A. (1991). Divorce over the l i f e course: The r o l e of marital happiness. Journal of Family Issues, 12(1), 5-21. White, L., & Edwards, J. N. (1990). Emptying the nest and parental well-being: an analysis of national panel data. American S o c i o l o g i c a l Review, 55, 235-242. Wingard, D. L. (1984). The sex d i f f e r e n t i a l i n morbidity, mortality, and l i f e s t y l e . Annual Review of Public Health, 5, 433-458. Wright, C. L., & Maxwell, J. W. (1991). S o c i a l support during adjustment to l a t e r - l i f e divorce: How adult c h i l d r e n help parents. Journal of Divorce and Remarriage, 15(3/4), 21-48'. Z i l l , N., Morrison, D. R., & Coiro, M. J. (1993). Long-term e f f e c t s of parental divorce on parent-child r e l a t i o n s h i p s , adjustment, and achievement i n young adulthood. Journal of Family Psychology. 7(1), 91-103. Appendix A Research Questions 134 Appendix A Research Questions A. Psychological Well-being 1. Happiness At Time One the question i s from the self-administered questionnaire, SE-2: Q.l. [The lead-in to t h i s section i s : Next are some questions about how you see yourself and your l i f e . ] F i r s t , taking things a l l together, how would you say things are these days. C i r c l e the number that best describes how you f e e l . At Time Two the question i s SE-2: MT201. Next are some questions about how you see yourself and your l i f e . F i r s t taking things a l l together, how would you say things are these days? Ratings f o r both questions are on a seven point scale, ranging from 1 (very unhappy) to 7 (very happy). 2. Depression The questions at Time One are from the self-administered questionnaire, SE-2: Q.2. Next i s a l i s t of the ways you might have f e l t or behaved during the past week. C i r c l e your answer to each question. Response categories under the heading "Number of days i n past week" are "none, 1, 2, 3, 4, 5, 6, 7." On how many days during the past week d i d you: a. f e e l bothered by things that u s u a l l y don't bother you? b. not f e e l l i k e eating; your appetite was poor? c. f e e l that you could not shake o f f the blues even with help from your family or friends? 135 d. have trouble keeping your mind on what you were doing? e. f e e l depressed? f. f e e l that everything you did was an e f f o r t ? g. f e e l f e a r f u l ? h. sleep r e s t l e s s l y ? i . t a l k less than usual? j . f e e l lonely? k. f e e l sad? 1. f e e l you could not get going? The questions f o r depression at Time Two, also from SE-2, are i d e n t i c a l (except that the i n s t r u c t i o n about c i r c l i n g i s omitted) and are presented i n the same order (MT2 05A-MT2 05L). 3. Self -es teem Self-esteem i s measured by three items i n the self-administered questionnaire. At Time One the items are as follows: (SE-13. Q's 59b, 60e, 60m respectively) [The lead-in f o r the questions i s as follows: Please indicate how much you agree or disagree with each of the following statements.] I f e e l that I'm a person of worth, at least on an equal plane with others. On the whole, I am s a t i s f i e d with myself. I am able to do things as well as other people Response ratings are (1) strongly agree, (2) agree, (3) neither agree nor disagree, (4) disagree, and (5) strongly disagree. At Time Two the actual items are i d e n t i c a l to those at Time One (SE-2. Q'S MT210I, MT210A, MT210F). B. Parent-Adult C h i l d Relationships 1. Time Two Contact Questions on contact are contained i n Module C and are asked f o r a l l the respondent's c h i l d r e n who are l i v i n g elsewhere, including those away at college and i n the m i l i t a r y . Q.MC10P01-MC10P14. During the past 12 months, how often d i d you see (child)? Q.MG11P01-MC11P14. During the past 12 months, how often did you t a l k on the telephone or receive a l e t t e r from (child)? Response categories for both questions are "not at a l l (01), about once a year (02), several times a year (03), 1 to 3 times a month (04), about once a week (05), more than once a week (06), some other frequency (07)." 2. Time Two Support Questions on support are included i n Module G. The q u a l i f y i n g question for respondents was as follows: Q.MG1. Next we are interested i n help and support that you may have given to your c h i l d r e n (or your wife/husband/partner's children) age 19 and older. This includes: [Roster of children's names generated by computer]. Include only help that you, yourself gave, but not help that other members of your family may have given. Q.MG2P01-MG2P06 Which of your c h i l d r e n (or your wife/husband/partner's children) age 19 and older have you given help with shopping, errands, or transportation i n the l a s t month? Q.MG4P01-MG4P06 Which of your c h i l d r e n (or your wife/husband/partner's children) age 19 and older have you given help with housework, yard work, car repai r s , or other work around the house i n the l a s t month? Q.MG6P01-MG6P06 Which of your c h i l d r e n (or your wife/husband/partner's children) age 19 and older have you given advice, encouragement, moral or emotional support i n the l a s t month? Q.MG8P01-MG8P06 Which of your ch i l d r e n (or your wife/husband/partner's children) age 19 and older have you helped with c h i l d care while they were working? Q.MG10P01-MG10P06 Which of your c h i l d r e n (or you wife/husband/partner's children) age 19 and older have you helped with c h i l d care at times other than when they were working? Q.MG13P01-MG13P06 From which of your ch i l d r e n (or your wife/husband/partner's children) age 19 and older have you received help with shopping, errands, or transportation during the l a s t month? Q.MG15P01-MG15P06 From which of your ch i l d r e n (or your wife/husband/partner's children) age 19 and older have you received help with housework, yard work, car repairs, or other work around the house i n the l a s t month? 138 Q.MG17P01-MG17P06 From which of your c h i l d r e n (or your wife/husband/partner's ch i l d r e n age 19 and older have you received advice, encouragement, moral or emotional support i n the l a s t month? C. Control Variables 1. Income/Assets Questions on income, assets and debts are contained i n the interview codebook and responses were f a c i l i t a t e d by the use of cards l i s t i n g the categories. Household income was measured by a va r i a b l e which the NSFH has constructed of household income from a l l sources. At Time One assets and debts are measured by the following questions: Q.661. Do you (or your husband/wife) own (a TO d BELOW)? Q.662 . How much do you think your (ASSET TYPE) would s e l l f o r now? Q.663. How much, i f anything, do you (or your husband/wife) owe on your (ASSET TYPE)? (IF R ISN'T SURE, ASK FOR THE BEST ESTIMATE). [Respondents are also asked i f they personally own the asset]. a. your own home? ( i f home i s part of a farm, include below i n business or farm). b. other r e a l estate? c. a business or farm? d. motor vehicles, including cars, trucks, campers, boats, and other r e c r e a t i o n a l vehicles? 139 Q.66.4. What i s the approximate t o t a l value of your (and your husband's/wife's) savings, including savings accounts, savings bonds, IRAs, money market shares, and CD's? Just t e l l me the l e t t e r from t h i s card. Response categories are (A) None, 01; (B) $1 to $1,499, 02; (C) $1,500 to $2,999, 03; (D) $3,000 to $4,999, 04; (E) $5,000 to $9,999, 05; (F) $10,000 to $19,000, 06; (G) $20,000 to $49,999, 07; (H) $50,000 to $99,999, 08; and (I) $100,000 or more, 09. Q.665. And what i s the approximate t o t a l value of your (and your husband's/wife's) investments, including stocks, bonds, shares i n mutual funds, or other investments? T e l l me the l e t t e r from t h i s card. Response categories are the same as for Q664 above. Q.666 . Next I w i l l read a l i s t of things that people often owe money on. T e l l me i f you (and your husband/wife/partner) owe money f o r : (ASK Q.667 FOR EACH YES) Q.667. How much do you (and your husband/wife/partner) owe on your (DEBT TYPE)? a. c r e d i t card or charge accounts that you're paying o f f gradually. b. installment loans f o r major purchases, such as f u r n i t u r e or appliances, but other than auto loans. c. educational loans. 140 c. educational loans. d. personal loans from banks and other businesses, other than mortgage or auto loans. e. personal loans from friends or r e l a t i v e s . f. other b i l l s you've owed for more than two months. g. home improvement loans. At Time Two assets and debts are measured by the following questions: MQ8 5 . How much do you think your home would s e l l f o r now? 5,000 $5,000 or less 5,001-9,999 10,000-24,999 25,000-49,999 50,000-74,999 75,000-99,999 100,000-149,999 150,000-999,998 999,999 $999,999 or more 9999997 Refused 9999998 Don't know 9999999 Inapplicable/no answer Other MQ91A. How much do you think t h i s other r e a l estate would s e l l f o r now? 1-9999 Remaining categories the same as MQ85. MQ91B. How much do you think t h i s business or farm would s e l l f o r now? 0 Nothing Remaining categories the same as MQ85. MQ91C. How much do you think these vehicles [cars, trucks, campers, boats, other r e c r e a t i o n a l vehicles] would s e l l f o r now? Response categories the same as MQ91B. MQ86 . How much, i f anything, do you (or your wife/or your husband) owe on your home? 0 (owe nothing) Remaining response categories the same as MQ85. MQ92A. How much, i f anything, do you (or your wife/or your husband) owe on your other r e a l estate? Response categories the same as MQ86. MQ92B. How much, i f anything, do you (or your wife/or your husband) owe on your business or farm? Response categories the same as MQ86. MQ92C. How much, i f anything, do you (or your wife/or your husband) owe on these vehicles? Response categories the same as MQ86. 142 MQ93 . What i s the approximate t o t a l value of your (and your wife's/and your husband's) savings, including savings accounts, savings bonds, IRAs, money market funds, and CDs? 00 None 01 $1 to $1,499 02 $1,500 to $2,999 03 $3,000 to $4,999 04 $5,000 to $9,999 05 $10,000 to $19,999 06 $20,000 to $49,999 07 $50,000 to $99,999 08 #100,000 or more 97 Refused 98 Don't know 99 Inapplicable/no anser Other MQ94. In addition to these savings, what i s the approximate t o t a l value of your (and your wife's/and your husband's) other investments, inc l u d i n g stocks, bonds, shares i n mutual funds, or other investments? Response categories the same as MQ93 except f o r the addition of the following: 90 R sa i d investments were reported above. 143 MQ95A. Next I w i l l read a l i s t of things that people often owe money on. How much, i f anything, do you (and your wife/and your husband) owe on c r e d i t cards or charge accounts that you are paying o f f gradually? If you almost always pay o f f your c r e d i t card balance each month, answer "0". 0 Nothing 1-199 200-499 500-999 1000-1999 2000-4999 5000-9999 10000-99998 99999 $99,999 or more 999997 Refused 999998 Don't know 999999 Inapplicable/no answer Other MQ95B. How much, i f anything, do you (and your wife/and your husband) owe on installment loans for major purchases, such as fu r n i t u r e or appliances, but other than auto loans? Response categories than same as MQ95A. MQ95C. How much, i f anything, do you (and your wife/and your husband) owe on educational loans? Response categories the same as MQ95A. 144 MQ95D. How much, i f anything, do you (and your wife/and your husband) owe on personal loans from banks and other businesses, other than mortgage or auto loans you have already t o l d me about? Response categories the same as MQ95A. MQ95E. How much, i f anything, do you (and your wife/and your husband) owe on personal loans from friends or r e l a t i v e s , other than those you have already t o l d me about? Response categories the same as MQ95A. MQ95F. How much, i f anything, do you (and your wife/and your husband) owe on home improvement loans, other than those you have already t o l d me about? Response categories the same as MQ95A. MQ95G. How much, i f anything, do you (and your wife/and your husband) owe on other b i l l s you've owed for more than two months? Response categories the same as MQ95A. MQ95H. How much, i f anything, do you (and your wife/and your husband) owe on any other debts that we have not mentioned? Response categories the same as MQ95A. 145 2. Race Question i s from the interview codebook at Time One. Q.484 . Which of the groups on t h i s card best describes you? Just t e l l me the number. Response categories are "Black 01; White - not of Hispanic o r i g i n 02; Mexican American, Chicano, Mexicano 03; Puerto Rican 04; Cuban 05; Other Hispanic 06; American Indian 07,- Asian 08; Other (specify:) 09." 3. Age of Respondent Question i s from the interview codebook at Time One. Q.485. What i s your date of birt h ? Response categories are "month, day, year." 4. M a r i t a l cohort Q.MI30 Next we want to f i n d out about any changes i n your marital status since NSFH1. At that time you were married to (spouse's name). Are you s t i l l married (and l i v i n g with/to) (him/her)? 1 Yes 2 No 6 R says they were married to someone else 7 Refused 9 Inapplicable/no answer Other Q.MI35 Next we want to f i n d out about any changes i n your marital status since NSFH1. At that time you were married to (spouse's name). Did your marriage end i n : 1 divorce 2 separation with no divorce, or 3 widowhood 5 married continuously since NSFH1 6 R says not married at NSFH1 9 Inapplicable/no answer Other Q.MI38 In what month and year did you and (spouse's name) stop l i v i n g together Date coded i n century months 1045 - 1140 9997 Refused 9998 Don't know 9999 Inapplicable/no answer Other 147 5. Degree of S o c i a l Integration Questions at Time One are from the self-administered questionnaire, SE-2: Q.16 & Q.17). Q.16. About how often do you do the following things. . . HOW OFTEN DO YOU: a. Spend a s o c i a l evening with: - r e l a t i v e s ? - a neighbor? - people you work with? - friends who l i v e outside your neighborhood? b. Attend a s o c i a l event at your church or synagogue? c. Go to a bar or tavern? d. P a r t i c i p a t e i n a group r e c r e a t i o n a l a c t i v i t y such as bowling, golf, square dancing, etc.? Response categories are "never, several times a year, about once a month, about once a week, several times a week." Q.17. Here i s a l i s t of various kinds of organizations. How often, i f at a l l , do you p a r t i c i p a t e i n each type of organization? a. f r a t e r n a l groups b. service clubs c. veterans' groups d. p o l i t i c a l groups e. labor unions f. sports groups g. youth groups 148 h. school r e l a t e d groups 1. hobby or garden clubs school f r a t e r n i t i e s or s o r o r i t i e s k. n a t i o n a l i t y groups 1. farm organizations m. l i t e r a r y , a r t, study, or discussion groups n. pr o f e s s i o n a l or academic s o c i e t i e s o. c h u r c h - a f f i l i a t e d groups Response categories are "never, several times a year, about once a month, about once a week, several times a week." At Time Two the comparable question to Time One Q.17 i s SE-18: MT1806A1-MT1806D. The wording of the items and the order of presentation i s i d e n t i c a l to Time One except that MT1806A1 i s worded as follows: [About how often do you] "Get together s o c i a l l y with: [ r e l a t i v e s , a neighbor . . . ] . Data f o r the var i a b l e "presence of other adults i n the household" was taken from the household roster i n the interview codebook. Q.l. I've recorded (NAMES FROM SCREENING ROSTER) as members of t h i s household. (IF NOT OBVIOUS, ASK:) Is (PERSON) male or female? The form f o r the FULL-TIME HOUSEHOLD ROSTER includes the following columns: "Column 1 - F i r s t Name; Column 2 - Age: Column 3 - M a r i t a l Status (Married, Separated, Divorced, Widowed, Never Married); Column 4 - Sex [Male, Female],- Column 5 - Relationship, Code." 149 Q.2. T e l l me how each person on t h i s l i s t i s rel a t e d to you. Choose your answer from these categories. (HAND R CARD #1. WRITE IN THE RELATIONSHIP AND ENTER THE LETTER CODE IN COL.5. IF CODE B [LOVER/PARTNER] IS GIVEN TO PERSON OF OPPOSITE SEX, CIRCLE "COHABITING" ON LINE 2 OF BOOKMARK) 6. R e l i g i o s i t y Question i s from the interview codebook at Time One. Q.492. How often do you attend r e l i g i o u s services? Response categories are " (# times) per day, week, month, year -or-never." 150 7 . T ime Two G e o g r a p h i c D i s t a n c e The question on geographic distance i s included i n Module C and was asked f o r a l l the respondent's ch i l d r e n l i v i n g elsewhere, including those away at college and i n the m i l i t a r y . Q.MC9P01-MC9P14. Approximately how many miles from here does (he/she) l i v e ? 0000 Less than one mile 0001-0004 0005-0009 0010-0018 0019-0024 0025-0099 0100-0199 0200-0499 0500-0999 1000-1999 2000-9000 9997 Refused 9998 Don't know where (he/she) l i v e s 9999 Inapplicable/no answer. Appendix Tables 152 Table 1 Means/Proportions and Standard Deviations f o r Study Variables Variable means/proport ions SD Independent Variables Divorce (divorce) Gender parent (female) Femcomp Dependent Variables Unhappiness TI (logged) Unhappiness T2 (logged) Depression TI (logged) Depression T2 (logged) Self-esteem TI Self-esteem T2 Contact (squared) Support given (logged) Support received (logged) Control Variables .027 .566 .478 .698 .746 1.883 1.961 6 .402 6 .265 27.841 .623 .410 .161 .496 .330 .561 .549 .204 .185 .610 .576 .782 .486 .447 1584 1584 1584 1412 1359 1561 1533 1513 1527 1568 1584 1584 Age (31-49=0; 50-85=1) Race (White) Household income TI Household income T2 Assets TI Assets T2 Soc i a l a c t i v i t i e s TI Soc i a l a c t i v i t i e s T2 Other adults i n household TI Other adults i n household T2 R e l i g i o s i t y TI (logged) Geographic distance T2 (logged) .625 .838 2.002 ,001 .004 , 101 .531 10.602 .994 .977 2 .812 3.999 .484 .369 .819 .819 .816 . 847 .831 . 941 .075 .151 .739 .346 1584 1584 1215 1571 1481 1547 1494 1520 1584 1584 1552 1572 Table 2 H i e r a r c h i c a l OLS Estimates f o r the E f f e c t s of Divorce on Unhappiness (Understandardized C o e f f i c i e n t s i n Parentheses) Step 1 Step 2 beta beta (b) (b) T l unhappiness .320*** .305*** (.309) (.294) Divorce .055* .054* (.196) (.193) Age Race Gender of parent .080** (.085) Household income T l Household income T2 Assets T l AssetsT2 S o c i a l a c t i v i t i e s T l So c i a l a c t i v i t i e s T2 -.055* (-.006) Other adults i n household T2 Other adults i n household T2 R e l i g i o s i t y T l Geographic distance T2 R2 .106 .122 Number of cases 1172 1172 * p<.05, **p<.01, ***p<.001 Table 3 H i e r a r c h i c a l OLS Estimates for the E f f e c t s of Divorce on Depression (Understandardized C o e f f i c i e n t s i n Parentheses) Step 1 Step 2 beta beta (b) (b) T l depression .438*** .414*** (.424) (.401) Divorce .061** .052* (.466) (.396) Age Race Gender of parent .090*** Household income T l Household income T2 -.118*** (-.168) Assets T l Assets T2 Soc i a l a c t i v i t i e s T l So c i a l a c t i v i t i e s T2 Other adults i n household T l Other adults i n household T2 R e l i g i o s i t y T l Geographic distance T2 R2 .198 .219 Number of cases 1443 1443 *p<.05, **p<.01, ***p<.001 Table 4 H i e r a r c h i c a l OLS Estimates f o r the E f f e c t s of Divorce on Self-Esteem (Understandardized C o e f f i c i e n t s i n Parentheses) Step 1 Step 2 beta beta (b) (b) TI self-esteem .404*** .390*** (.388) (.375) Divorce -.006 .002 (-.066) (.023) Age Race Gender of parent Household income TI Household income T2 Assets TI Assets T2 s o c i a l a c t i v i t i e s TI So c i a l a c t i v i t i e s T2 Other adults i n household TI Other adults i n household T2 R e l i g i o s i t y TI Geographic distance T2 R2 Number of cases - . 090*** (-.277) . 105*** (-198) .163 .182 1456 1456 *p<.05, **p<.01, ***p<.001 Table 5 H i e r a r c h i c a l OLS Estimates f o r the E f f e c t s of Divorce on Contact (Understandardized C o e f f i c i e n t s i n Paranetheses) Step 1 Step 2 beta beta (b) (b) Divorce -.028 -.056* (-1.466) (-2.997) Age -.169*** ( -2.683) Race Gender of parent .075** (1.172) Household income T l .091*** Household income T2 Assets T l .072** T2 self-esteem (.876)   (.687) Assets T2 So c i a l a c t i v i t i e s T l So c i a l a c t i v i t i e s T2 Other adults i n household T l Other adults i n household T2 R e l i g i o s i t y T l Geographic distance T2 -.488*** (-1.630) T2 unhappiness T2 depression .069** (-474) R2 .001 .326 Number of cases 1134 1134 p<.05, **p<.01, ***p<.001 157 Table 6 Mi e r a r c h i c a l OLS Estimates f o r the E f f e c t s of Divorce on Support Given (Understandardized C o e f f i c i e n t s i n Parentheses) Step 1 beta (b) Step 2 beta (b) Divorce .029 095) - .037 (-.123) Age Race Gender of parent Household income TI Household income T2 Assets TI Assets T2 So c i a l a c t i v i t i e s TI So c i a l a c t i v i t i e s T2 Other adults i n household TI Other adults i n household T2 R e l i g i o s i t y TI Geographic distance T2 -.113*** (-.113) .108** (.106) .154*** (-093) .102*** (.060) .091** (.012) - . 136*** (-.029) T2 unhappiness T2 depression T2 self-esteem R2 Number of cases .001 1105 .083** (.035) . 121 1105 *p<.05, **p<.01, ***p<.001 158 Table 7 H i e r a r c h i c a l OLS Estimates f o r the E f f e c t s of Divorce on Support Received (Understandardized C o e f f i c i e n t s i n Parentheses) Step 1 Step 2 beta beta (b) (b) Divorce .013 -.007 (.039) (-.021) Age -.121*** (-!ll2) Race Gender of parent .123*** (-109) Household income T l Household income T2 Assets T l Assets T2 So c i a l a c t i v i t i e s T l So c i a l a c t i v i t i e s T2 Other adults i n household T l Other adults i n household T2 R e l i g i o s i t y T l Geographic distance T2 __108*** T2 unhappiness T2 depression .122*** T2 self-esteem (-.020) . 122* 046) R2 .000 .071 Number of cases 1522 1522 *p<.05, **p<.01( ***p<.001 159 Table 8 H i e r a r c h i c a l OLS Estimates f o r the E f f e c t s of Divorce and Parent-Adult C h i l d Relationships on Depression (Understandardized C o e f f i c i e n t s i n Parentheses) Step 1 beta (b) Step 2 beta (b) Step 3 beta (b) TI depression Divorce .438*** (.424) .058* (.442) .484*** (.401) . 048* (.372) .409*** (.396) .046 ( .350) Age Race Gender of parent Household income TI Household income T2 Assets TI Assets T2 So c i a l a c t i v i t i e s TI So c i a l a c t i v i t i e s T2 Other adults i n household TI Other adults i n household T2 R e l i g i o s i t y TI Geographic distance T2 Contact Support given Support received . 093*** (.216) - . 116*** (-.166) .070** (.162) .135*** (--192) .120*** (.313) R* Number of cases . 197 1431 .218 1431 .236 1431 *p<.05, **p<.01, ***p<.001 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0087213/manifest

Comment

Related Items