Citation
Social change and intergenerational solidarity

Material Information

Title:
Social change and intergenerational solidarity
Creator:
Wakeman, Melanie Ann
Copyright Date:
2005
Language:
English

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Melanie Ann Wakeman. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
0027719461 ( ALEPH )
664556866 ( OCLC )

Downloads

This item has the following downloads:


Full Text













SOCIAL CHANGE AND INTERGENERATIONAL SOLIDARITY:
MOTHER/DAUGHTER RELATIONSHIPS IN THE SECOND HALF OF THE
TWENTIETH CENTURY















By

MELANIE ANN WAKEMAN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2005





























It is with a tremendous amount of love and gratitude
that I dedicate this dissertation
to my mother and my grandmother.
Rebecca T. Roman and Judith A. Jones.
















ACKNOWLEDGMENTS

While writing the dissertation might feel like an isolating and individual process, it is truly a collaborative effort. So many people were involved in the success of this dissertation project, and involved in my progress and success in graduate school in general. First, I would like to acknowledge and thank the members of my doctoral dissertation committee and Institute on Aging (IoA) mentoring team, which include Connie Shehan, Amy Pienta, Barbara Zsembik, Leonard Beeghley. Tanya KoropeckyjCox, and Angel Kwoleck-Folland. As my chair and cochair. Dr. Shehan and Dr. Pienta have been instrumental in helping me to conceptualize this project, spending countless hours talking through ideas over the last several years. Their overall support of me as a graduate student and a colleague has been a constant. In addition- Dr. Pienta has offered her analytic insight and a keen editorial eye, and was instrumental in helping me navigate my way through a very complex data set. I especially appreciate Dr. Zsembik's and Dr. Beeghley's unconditional devotion to me during my time at Florida (and beyond). Their doors were always open when I needed to talk, and they offered valuable insight, both professional and personal, with nothing but my best interest at heart. Finally, I thank Dr. Koropeckyj-Cox and Dr. Kwolek-Folland for joining my committee. Both have contributed a great deal to the progress of this project.

Several people outside of my committee must be thanked for their support during my graduate program. First, Rae Newton, my master's advisor, has always encouraged me, and remains a great source of support, even today. Terry Mills' guidance enabled me


iii










to participate in several programs, and afforded me many great opportunities. Also, many faculty and students at the loA deserve special recognition for their support and belief in me. They are Michael Marsiske, Jeff Dwyer, Ann Horgas, Nicole Alea, Karin McCoy, Angie Semegon, Katherine White, Jennifer Stanley. Tyson Brown, and especially Yvonne Combs.

Members of the office staffs of the Department of Sociology, the Institute on Aging (IoA), the Center for Women's Studies and Gender Research (CWSGR), and the University Center for Excellence in Teaching (UCET) have provided assistance on several levels throughout the years. There have been many people working in this capacity. In the Department of Sociology, Sheran Flowers, Kanitra Perry, and Justin Smith put in endless hours making sure we all get paid and handling the necessary bureaucratic paperwork, among other things. Charlotte Simmons, Richelle Davis, and Anna Abee in the loA always had the answers, even when there were none, and they have done a superb job helping all of us in the first cohort of graduate research trainees. Paula in the CWSGR became a friend, and of all things, a student, during my final year at Florida. And in UCET, Nadine Gillis and Diane were always there to take care of me when I needed it, providing a safe haven, a place to eat, a place to organize, a place to prepare for class, and countless other reasons to hang out and chat.

Many fellow graduate students at the University of Florida deserve special thanks for their friendship and support, especially during that very difficult first year. I thank Leslie Houts, Man DeWees, Jason Davis. Lisa Holland, Brad Tripp, and Clay Hipke. In their own ways, all of them helped me through the good and bad times that invariably are what it is to be a graduate student.


IV










There are also several very special people who have become some of my best

friends. Susan Eichenberger. Sara Crawley, Lara Foley, Helena Alden, Laurel Tripp, Ed Porras, and Barbara Bonano became "Family" for me, and I can't begin to imagine life in Gainesville without them. I guess mention of the solace we all found on my front porch must also be included!

Gator football and the usual football crowd also deserve special mention. Those

Saturdays in the fall were necessary to my sanity. Thanks go to Kevin Fraizee for always being the gracious host, and to Pam, Kathy, Oz, Gabe, and the rest of the gang. GO GATORS!

This acknowledgement would not be complete without mentioning friends, who, although they did not experience graduate school with me, have always been supportive and have loved me unconditionally. Doug Laughton, Jilana Miller, Cristin and Mike Simpson, Kim Anderson, Meg Benjamin, and Steffanie Siebrand are longtime friends from childhood who sometimes know me better than I know myself, and are always there to keep me in line.

I would like to thank my family, without whom I would not be where I am today. My father. Mark Wakeman, who is always the realist, has helped to keep me on track in recent years. My step-father, Don Roman, has also been there for me over the years, to challenge me intellectually and to remind me not to take myself too seriously. In addition, along with his red pen, he served as editor and critic throughout the various stages of the dissertation process. I thank my brother, Andrew, who is my favorite person in the whole world. My mother and Gammy (Rebecca Roman and Judy Jones), have










given me unconditional love and support, both emotional and financial, and are truly the inspiration for this dissertation project. I will never be able to thank them enough.

Finally, I would like to thank Bob Stagnaro. He has become a very significant person in my life. Although I have long since been done with my dissertation and graduate school, he knows why he has been included in this acknowledgement section.


vi

















TABLE OF CONTENTS

~a-ge

ACKNOWLEDGMENTS .ii

LIST OF TABLES ....... . .. .. xi

ABSTRACT ..................... ......... .................... xiii

CHAPTER

I INTRODUCTION TO STUDY......................................I

Introduction .....................................................
Specific Aims ... 2...........................................
Overview of the Study .............................................
Theoretical Framework .. ....................................4
Review of the Literature .. ... ...................................4
Data and Methods .............................................5
Analysis ...............6..................................6
Discussion and Conclusions .....................................6
Conclusion.................................................... 6

2 THEORETICAL FRAMEWORK 9...............................9

Introduction..... ....................... ...........9
Theories of Aging ............................................... 10
Social Exchange Theoy.................................... ... 10
Social Constructionist Theories ...................... ........10
Continuity Theory .................................... II
Age-Stratification Theory ...................................... 12
Age-Integration Theory ................................ ........13
Political Economy of Aging 3.....................................
Theories of Aging Families.......... .14
Family Development Theory .................................... 14
Social Exchange Theory ..... ............ ...................... 17
Intergenerational Stake Hypothesis ............................... 18
Socioemotional Selectivity Theory .................................19
Social Comparison Theory ............................ 19
Understanding Social Change and Women's Intergenerational Relationships 21
Life Course Theory and the Developmental Schism ................... 21


Vii











Intergenerational Family Solidarity Theory ... ......... .. 23
Conclusion 24

3 REVIEW OF THE LITERATURE ...................................26

Introduction ...................................................26
Changes in Social Structural Forces 26
Education and Occupational (Status) Attainment ..................... 27
Changing Work Roles .. 27 Family Formation ............................................29
Marital Stability .............................................30
Ideological Shift 3...................... .......... ........30
Intergenerational Family Relationships ................................3 1
Intergenerational Differences Among Women ...........................32
Statement of Hypotheses ............. ... .... ........... ...... 34
Generation X and Baby Boom Daughters During Young Adulthood........ 34
Baby Boom Generation and World War I Generation Daughters Dunng
Midlife ... .................................................. 35

4 DATA AND METHODS 36

Introduction 36 Sample ... 36 Sample Selection Characteristics 38
Daughters During Young Adulthood .............................. 40
Daughters During Midlife 4..............................40
Measurement............................... ................... 41
Dependent Variables .. 41 Independent Variables ........................................44
Social Distance Measures ......................................9
Analytic Model .................. ............................... 52
Sample Limitations..... ................... ......................54

5 GENERATIONAL DIFFERENCES IN DAUGHTERS' PERCEPTIONS OF
ASSOCIATIONAL, AFFECTUAL. AND CONSENSUAL SOLIDARITY IN
YOUNG ADULTHOOD.......................... .................57

Introduction 57 Descriptive Statistics-Independeni Variables ...........................58
Results 59
Descriptive Statistics and Mean Differences-Dependent Variables ............60
Results- Daughters Perceptions of Associational Solidarity During Young
Adulthood ...... .. ..................... .............. 1
Results Daughters' Perceptions of Affectual Solidarity During Young
Adulthood .. 62
Results--Daughters Perceptions of Consensual Solidarity During Young
Adulthood ................................................ 62


viii











Multivariate Analyses................. .............................. 63
Results-Effects of Daughter Characteristics on Their Perceptions of
Associational Solidantv ..... .................... 64
Results-Effects of Daughter Characteristics on their Perceptions of Affectual
S o lid arity ............................. ...................... 6 6
Results-Effects of Daughter Characteristics on their Perceptions of
Consensual Solidarity ....................................... 67
Social Difference Analyses for Generation X Daughters and their Baby Boom
M others During Young Adulthood ... ..... ...................................... 70
Results-Descriptive Statistics for the Difference Measures Compnsing the
Independent Variables .......................................7 1
Results-Effects of Social Difference Measures on Generation X daughters
Perceptions of Associational, Affectual and Consensual Solidarity with
their Baby Boom Generation Mothers During Young Adulthood......... 72
Conclusions................................................................... 74

6 GENERATIONAL DIFFERENCES IN DAUGHTERS' PERCEPTIONS OF
ASSOCIATIONAL. AFFECTUAL. AND CONSENSUAL SOLIDARITY IN
MID-LIFE..................................................... 85

Introduction .............................................. ....................... . .. 85
Descriptive Statistics-Independent Variables... .... ... ..... .................. .86
Descriptive Statistics and Mean Differences- Dependent Variables ..................... 89
Results-Daughters' Perceptions of Associational Soidarity During Midlife .89 Results-Daughters' Perceptions of Affectual Solidarity During Midlife. .... 90 Results-Daughters' Perceptions of Consensual Solidarity During Midlife 91
MultiNariate Analy ses ......................................... ... 92
Results-Effects of Daughter Characteristics on their Perceptions of
Associatonal Solidarity During Midlife........................... 92
Results-Effects of Daughter Characteristics on their Perceptions of Affectual
Solidarity During Midlife .................................... 95
Results-Effects of Daughter Characteristics on their Perceptions of
Consensual Solidarity During Midlife....... ............................ 97
Social Difference Analyses for Baby Boom Generation Daughters and their
World War II Generation Mothers During Midlife. .... ....................... 99
Results Descriptive Statistics for the Difference Measures Comprising the
Independent Variables ......................................... 100
Results-Effects of Social Difference Measures on Baby Boom Generation
Daughters' Perceptions of Associational. Affectual and Consensual
Solidarity with their World War 11 Generation Mothers During Midlife 10
Conclusions.................................................. 103

7 DISCUSSION AND CONCLUSIONS . 115

Introduction. ... ....................... 115
Review of Empirical Measures and Theoretical Framework l 16


ix











Discussion ................................................... 117
Generational Differences among the Women .............. ..........
Intergenerational Solidarity .... ........... ..................... 120
Conclusions and Implications ......................................122
Future Research ...... 123

LIST OF REFERENCES............... 124

BIOGRAPHICAL SKETCH ............ ..............................................134


x

















LIST OF TABLES


Table page

4.1 Numbers of Generation X and Baby Boom Generation Daughters
During Young Adulthood at Each Stage of Sample Selection .............. 56

4.2 Numbers of Baby Boom Generation and World War II Generation
Daughters during Midlife at Each Stage of Sample Selection............... .56

5.1 Descriptive Statistics- Demographic Characteristics of Generation X Daughters
and Baby Boom Daughters in Young Adulthood. .......................................... .. 78

5.2 Generation X Daughters' and Baby Boom Daughters' Perceptions of
Associational Solidarity with their Mothers in Young Adulthood. ...... ...... 79

5.3 Generation X Daughters' and Baby Boom Daughters' Perceptions of Affectual
Solidarity with their Mothers in Young Adulthood. ..................... 79

5.4 Generation X Daughters' and Baby Boom Daughters' Perceptions of
Consensual Solidarity with their Mothers in Young Adulthood ..................... 79

5.5 OLS Regression Effects of Daughter Characteristics During Young
Adulthood on Associational Solidarity for Generation X and Baby Boom
Generation Daughters....... ............................................. 80

5.6 OLS Regression Effects of Daughter Characteristics on Affectual
Solidarity During Young Adulthood for Generation X and Baby Boom
Generation Daughters............................. ............. ...81

5.7 OLS Regression Effects of Daughter Characteristics During Young
Adulthood on Consensual Solidarity for Generation X and Baby Boom
Generation Daughters....................... ............ ........ 82

5.8 Descriptive Statistics Social Distance Among Generation X Daughters During
Young Adulthood and their Baby Boom Generation Mothers During
Young Adulthood.............................................................. 83

5.9 OLS Regression Effects of Social Distance Between Generation X Daughters
and their Baby Boom Generation Mothers During
Young Adulthood ............................................ ......84


xi











6. 1 Descriptiv e Statistics Demographic Characteristics of Baby Boom Daughters
and W orld W ar 11 D aughters D unng M idlife.........................................................108

6.2 Baby Boom Daughters' and World War 11 Daughters' Perceptions of
Associational Solidarity During M idlife with their M others. ................................109

6.3 Baby Boom Daughters' and World War 11 Daughters' Perceptions of
Affectual Solidarity During M idlife with their M others........................................109

6.4 Baby Boom Daughters' and World War I Daughters' Perceptions of
Consensual Solidarity During M idlife with their M others....................................109

6.5 OLS Regression Effects of Daughter Characteristics During Midlife on
Associational Solidarity for Baby Boom Generation and World War II
Generation Daughters.......................................... I10

6.6 OLS Regression Effects of Daughter Characteristics During Midlife on
Affectual Solidarity for Baby Boom Generation and World War II Generation
Daughters. .... ...................................................II

6.7 OLS Regression Effects of Daughter Characteristics During Midlife on
Consensual Solidarity for Baby Boom Generation and World War 11
Generation Daughters..........................................112

6.8 Descriptive Statistics Social Distance Among Baby Boom Generation
Daughters During Midlife and their World War 11 Generation Mothers
During Midlife. ................ .................................................113

6.9 OLS Regression Effects of Social Distance Between Baby Boom Generation
Daughters and their World War II Generation Mothers
During Midlife................................. .. .......... 114


xii

















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

SOCIAL CHANGE AND INTERGENERATIONAL SOLIDARITY:
MOTHER/DAUGHTER RELATIONSHIPS IN THE SECOND HALF OF THE TWENTIETH CENTURY

By

Melanie Ann Wakeman

May 2005

Chair: Constance L. Shehan
Cochair: Amy Mehraban Pienta
Major Department: Sociology

This study examines generational differences in mother/daughter relationships. As a result of changing opportunities and role expectations, especially regarding education, work, and family, women's life course trajectories have shifted. Using the Longitudinal Study of Generations and Mental Health, this study seeks to gain greater understanding of the indirect impact social and structural changes have had on women's intergenerational solidarity. Using Bengtson's measures of solidarity. daughters' perceptions of associational, affectual, and consensual solidarity with their mothers is examined during both young adulthood and midlife.

Several generations of women from the same families are included, capturing the birth cohorts comprising the WWI Generation, the WWII Generation, the Baby Boom Generation, and Generation X. The perceptions of solidarity of the Generation X daughters as young adults in 1994 are compared with those of their mothers, the Baby


xiii










Boom daughters as young adults in 1971. The perceptions of solidarity of the Baby Boom Generation daughters during midlife in 1994 are compared with those of their mothers, the World War II Generation daughters during midlife in 1971.

Univariate. bivariate, and multivariate analyses are carried out to understand how daughter's perceptions of associational, affectual, and consensual solidarity with their mothers have changed over the last half of the century. Results show several significant differences among the generations of women, both during young adulthood and midlife. Perceptions of associatonal solidarity, aspects of affectual solidarity, and consensual solidarity also differed during young adulthood for the Generation X daughters in 1994 compared with the Baby Boom Generation daughters in 1971. During midlife, associational solidarity and aspects of affectual solidarity were significantly different for Baby Boom Generation daughters in 1994 compared with World War 11 Generation daughters in 1971. Furthermore, ordinary least squares (OLS) regression indicates several significant findings pertaining to the effects of education, work, and family variables on perceptions of associational, affectual, and consensual solidarity for Generation X daughters and Baby Boom Generation daughters during young adulthood, as well as generational differences between the women from 1971 to 1994. This was also the case among the Baby Boom Generation daughters in 1994 and the World War II Generation daughters during midlife in 1971.


xiv
















CHAPTER 1
INTRODUCTION TO STUDY

Introduction

The mother/daughter relationship has become one of the most salient of familial and social relationships in American society in the twentieth century. It is the focus of songs, poems, books. movies, and other cultural and social media. A major integral part of our culture is the bond between mothers and daughters. Understanding this relationship and its recent history is important as, women's lives continue to change. As women's life experiences have been altered from generation to generation, as a result of social change, what happens to intergenerational relationships among women mothers and daughters? Furthermore, how are successive generations of mother/daughter dyads different from each other? This is the subject of this research.

This research sought to identify the stable and dynamic aspects of intergenerational relationships over time, to better understand the relationship between family solidarity changing social and structural forces. This research also attempted to determine whether familial role expectations shape dyadic relationships across generations: and if so. how these relationship differences might change over time, as a result of social distance between the generations.

Social and structural changes in the United States over the second half of the 20th century have led to changes in both family structure and social norms/role expectations for women. Both the social and economic position and attitudes of and about women in U.S. society have changed dramatically over this period (Moen, 1992: Anderson, 2000).


I








2


Women's educational attainment has increased; women's labor force participation has increased; women's economic contribution to the household has increased; women are postponing marriage and having children later; fertility rates have declined; and prevailing ideologies about women's "place" have shifted. For example, changes in social structure are reflected in the beliefs and practices associated with women combining work and family roles, over the life course. These types of changes are expected to transform intergenerational relationships over time.

The profound legislative, economic, demographic, and ideological changes that have occurred have changed the life course trajectories of successive generations of women in families. Figure 1.1 illustrates the timing of each generation's experience of these sociohistorical changes during the twentieth century. It provides a visual representation of women's lives and how the timing of each generation of women's experience of these events has created new and different opportunities. This research examined the impact of these social changes on women's inter-generational solidarity.

Specific Aims

Early research assumed that "familial transmission and not social trends account for generational similarities" (Aldous, 1963; Aldous and Hill, 1965: Farber, 1966). Further, it has been proposed that continuity between the generations within the family is dependent upon the degree of cohesion existing between them (Aldous, 1963; Aldous and Hill, 1965). The assumption was that familial transmission is stronger than social trends. However, this idea may be a bit out dated because social and structural changes have been so profound. Women have encountered rapidly growing opportunities and shifts in expectations, as a result of the rapid social change that occurred in the twentieth century.








3


Thus, the argument can be made that familial relationships have become more complex and dynamic.

This study examined generational differences in daughters' perceptions of

solidarity with their mothers. Three aspects of solidarity were included. These aspects were; 1) associational solidarity, that is the extent to which family members interact with one another; 2) affectual solidarity, that is the extent to which family members are emotionally close to one another; and 3) consensual solidarity, that is the extent to which family members share similar values and world views (Bengtson and Roberts, 1991).

The first aim of this research was to understand how daughters from different

generations perceive their relationships with their mothers during young adulthood. The first research questions is, "How have daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during young adulthood changed, and what factors may be associated with this change?" The second aim of this proposed study seeks to understand how daughters from different birth cohorts perceive their relationships with their mothers during midlife. Thus, the second research question is, "How have daughters' perceptions of associational, affectual. and consensual solidarity with their mothers during midlife changed, and what may be associated with these changes?"

Overview of the Study

A large and growing body of research has focused on understanding

intergenerational family relationships, and family solidarity issues. As individuals continue to live longer, families will increasingly consist of multiple generations. The relationships between the generations, specifically their influence and functions, will also become increasingly more important. With regard to women, as a result of significant








4


social changes in roles and role expectations, women's intergenerational relationships have become all the more intriguing. The following is an outline of this research, focusing on social change and the influence on women's intergenerational relationships. Theoretical Framework

Chapter 2 of this research discusses the major theories used in the study of aging families, intergenerational family relationships, and more specifically, women's intergenerational relationships. This study will focus on Intergenerational Family Solidarity and the Developmental Schism. As a result of the growing diversity in the functions and structure of Family today, as well as an increase in life expectancy, intergenerational relationships have become increasingly important. Intergenerational Family Solidarity Theory (Bengtson et al., 1984) is one of the major theoretical contributions to the area of family gerontology. From this theory, empirical measures have been constructed, allowing researchers to measure the degree of solidarity among different generations. These measures are outlined discussed in Chapter 3. However, more detail as to their use is discussed in Chapter 4.

The Developmental Schism (Fingerman. 1996) is a concept stemming from the life course perspective, and grounded in developmental psychology. This schism specifically addresses the interpersonal problems among intergenerational relationships between females. The Developmental Schism, and is role in this research is discussed in detail in Chapter 3.

Review of the Literature

Chapter 3 is a review of the literature. This chapter outlines changes in social structural forces over the last century, discusses the literature pertaining to intergenerational family relationships and women's intergenerational relationships, and








5


finally, outlines the hypotheses for this study. First, some of the major legislative, economic, demographic, and ideological changes are reviewed. Included in this review of legislative changes are the Equal Pay Act, Title IX, and FDA approval of the birth control pill. Education and occupational (status) attainment, changing work roles, family formation, and marital stability are discussed next. These changes have had an impact not only on women's roles, thus altering their life course trajectories over the successive generations, but they have also had an impact on society's attitudes concerning women as well. Thus ideological shifts occurring in the last half of the twentieth century are also included in this chapter.

Data and Methods

Chapter 4 is a detailed account of the methodological procedures carried out in this study. This study uses the Longitudinal Studv of Generations and Mental Health, 1971 data set [made accessible in 1990, machine-readable data files]. These data were collected by V. Bengtson and M.Gatz and are available through the archive of the Henry A. Murray Research Center of the Radcliffe Institute for Advanced Study, Harvard University, Cambridge, Massachusettes (Producer and Distributor). These data are from an ongoing study of aging families investigating changes in intergenerational family relationships. These data specifically examine intergenerational solidarity, based on perceived and actual levels of cohesion among family members, making it an ideal data set for this research. This chapter includes summary characteristics of the samples used from the data set, as well as the complex sample selection strategy used to create the sample of women for this study. Finally, this chapter outlines the analytic strategies used to better understand how changing opportunities and life course trajectories for








6


successive generations of daughters may have influenced their perceptions of their relationships with their mothers.

Analysis

Two analytic chapters are included in this project. Chapters 5 and 6, employ the same analytic strategy, thus including the results of the univariate, bivaniate, and multivariate statistical analyses carried out in this study. Chapter 5 specifically examines the generational differences in daughters' perceptions of associational, affectual, and consensual solidarity with their mothers while they are young adults. The data analyzed for this chapter include Generation X daughters age 18 to 29 years, with data collected in 1994, and Baby Boom generation daughters age 18 to 29 years, with data collected in 1971. Chapter 6 specifically examines these generational differences among daughters during midlife. This chapter includes Baby Boom Generation daughters, ages 40 to 59 years, in 1994. and World War II Generation daughters, ages 40 to 59 years, in 1971. Discussion and Conclusions

Chapter 7 is the final chapter in this study. Conclusions, implications, and future research potential are discussed in this chapter. First, the generational differences among women are discussed based on the findings from the analyses in chapters 5 and 6. Next, a review of the social distance occurring between the generations of women participating in this research is outlined. Third, levels of intergenerational solidarity, and the effects of the daughter characteristics and social distance measures are reviewed. In addition, a discussion of practical implications and future research are discussed in this chapter.

Conclusion

In sum, this project seeks gain a greater understanding of women's

intergenerational relationships. Specifically, how social structural changes may have








7


influenced these relationships over the last half of the twentieth century as a result of their effect on the life course trajectories of women. Because intergenerational familial relationships are becoming increasingly more prevalent with the increased life expectancies today, it is important to develop a better understanding about women's intergenerational relationships.

This project expands the literature in this area by examining individual

characteristics reflected in the social structural changes of the last century. Stemming from a life course perspective and using Bengtson's measures of intergenerational family solidarity, it is possible for this project to contribute to a growing body of knowledge surrounding aging families and women's intergenerational relationships.

















(n _ ....














I2 I
p

Oda-* e4. 0



o
















(I s .h**, s. 92 7 (2 b j 2 : 4 1.. 6 U. (


Figure 1.1 Major Historical Events Experienced by Each Generation of Women.


'19 ",{ o t 5;A t .\' 5 I y' l k S k 5%W' X.'




RNt s) ii4:'f Iduk "SA 017
-; mAl" edm vRne "'o.,m nri for
isrm Co.r .ks aggsm:haun




(9 X 5(.X 55:5 A ssxas:ni~sm55 drU




TA e X\ pmssi s s ex. .'.5 d %Lcr *:M 55


s:rs %"5s5 sss s'c d shdre im ss:sdj" 3


'ss1- I (m st x chd:n 1ins'.' estsk


*96 s W (y v. wAs 4 i .s 4~~
*"v 5 .5., ~ :'k %5 ssk~ IN p v x .x5j ss s- ~ s
Asf5455 5


ss~.ss ss;ss '* ss ~ ~ %5s>3:'~. t s C,, -c


SN f4. k.5.\5, X -IAM ~



A, kJ5%%5 5 =mbr- 4 -.- l5A% sisz Vs v 41S i 3.
nnsss .1s'~2


00
















CHAPTER 2
THEORETICAL FRAMEWORK

Introduction

Aging families (and specifically, intergenerational relationships) have become an integral part of the field of social gerontology. Sociological theories and many family theories have been reexamined from the perspective of aging studies. As a result, new lines of thought have emerged that consider the effects of age (and the aged in our society) on the family and on family dynamics. This study centered on two guiding theoretical frameworks: Intergenerational Family Solidarity Theory; and the Life Course Perspective (in particular, the Developmental Schism hypothesis).

Intergenerational Family Solidarity Theory (Bengtson et al., 1984) is one of the major theoretical contributions to the area of family gerontology. From this theory, empirical measures for the present study were constructed, allowing researchers to measure the degree of solidarity among different generations. This study also used the Developmental Schism hypothesis (Fingerman, 1996), which stems from the life course perspective, and is grounded in developmental psychology. The Developmental Schism hypothesis is highly relevant to the present study because it addresses interpersonal problems in the intergenerational relationships of females. Before reviewing the guiding theoretical framework for the present study, here is an overview of aging theories, and theories that have been applied to the study of aging families.


9








10


Theories of Aging

There have been several theories developed to better explain aging related

phenomena and relationships. Each of the following theories has made a significant contribution to the field of social gerontology. Social Exchange Theory

Social exchange theory in aging is adopted from the micro-economic premise that interaction or exchange is based on the calculation of the rewards that they will gained. In the study of aging, relationships and interaction, are influenced by the resources of the individuals involved (Bengtson, Parrott, and Burgess. 1997). Resources in terms of social interaction are not just limited to money. Anything that elevates the status of that individual can be a resource. The key question becomes, what resources does each person bring to the table? In other words, do the benefits outweigh the costs of being involved with this person? Therefore, when people are considering interactions with older adults, they weigh the pros and cons of the interaction.

Because most older adults have fewer valued resources, there is less involvement with them. A shortcoming of social exchange theory is that it doesn't account for the benefits of relationships that cannot be logically explained. People get benefits from interactions with people when the obvious reasons for the exchange are not present. In addition, social exchange theory does not account for why some people have more resources than others.

Social Constructionist Theories

Social constructionist theories suggest that individuals negotiate reality through interactions with others. It postulates that people are active subjects, not passive participants. Such an emphasis on human structure means that social structures do not








S1I


impose social rules on us, but rather we create the social structures (Gubrium and Holstein, 1999). Constructionist theorists in the study of aging are interested in how aging individuals create their realities from their experiences and interactions as they go through aging-related phases. Constructionist theories do not fit the definition of a theory per se, as they "are not conceived as causal explanations of the social world, but instead focus attention on problems of meaning in everyday life (Gubrium and Holstein, p. 287). Constructionist theories are similar to social exchange theory because the primaRy focus is on the actions of the individual, and emphasis is placed on interactions with others. A shortcoming of constructionist theories in aging is the overemphasis on the individual at the personal level, and on the individual's constructions of reality. Without an insight into the influences of social forces, there is no context from which to understand social behavior and interaction.

Continuity Theory

This is the idea that the more older adults maintain a sense of continuity in their everyday living and activities, the fewer adjustments they will have to make in later life. Borrowing from the proponents of activity theory, which suggests that if older adults remain positive and active into and throughout "old age," they will experience greater life satisfaction (Havighurst, 1963), Robert Atchley developed continuity theory based on two dimensions of continuity: internal and external. The internal dimension of continuity is a personal dimension, and consists of peoples' values, world views, skills, and general life frameworks. The external dimension is a public dimension- and consists of role performance, status attainment, and everyday activities and hobbies. Atchley suggests that continuing to draw on ideas and activities that were successful in early life will benefit people moving into the stages of later life (Atchley, 1989).







12


Unlike exchange theory and the social constructionist theories, where the individual is the primary focus, the connection in continuity theory is to broader social life. Just as it is in social exchange theory, social interaction is an important focus. Part of external continuity would be for individuals to maintain those interactions that are positive, into older adulthood. Continuity theory posits satisfaction in later life correlates strongly to continuing important, successful interactions with others over time. A major critique of continuity theory is the "normalization" of the aging process (Becker, 1993). Since many of the activities of individuals are shaped by age-related social role expectations, continuity theory suggests the existence of a standard aging process that people go through. Furthermore, feminist researchers argue that this imposes one standard (a male standard) on people as they move through the life course (Browne, 1998; Calasanti, 1996).

Age-Stratification Theory

Matilda White Riley's (1971) age-stratification theory provides a framework for examining age and aging as a social process dependent on the forces of the larger social structures. The focus is on the status of older adults, and specifically on the ways in which statuses and roles are allocated in our society based on age. Age-stratification theory postulates that all societies break themselves down into age categories, ultimately ranking these categories into hierarchies based on larger social notions of importance.

Like social exchange theory, value is placed on the contributable resources people possess. However, age-stratification theory concerns the macro level. A key component of age-stratification theory is the age cohort: a group of people born at the same time, moving through the life course together. The age cohort places a social context around the individual's age, which contributes to the categorization of this particular group of







13


individuals. Often, members of these cohorts share similar viable resources, or perhaps, a lack of resources. Major limitations of age-stratification theory is that it assumes that a general consensus exists among society members about age-related statuses and power. It fails to recognize the power relationships inherent in our social structures. Moreover, there is no way to account for intracohort differences (such as race, class, and gender) that also contribute to inequity in allocating roles, statuses, and power. Age-Integration Theory

Like age-stratification theory, age-integration theory also emphasizes the age

cohort and the hierarchy of groups in our society, based on social factors (including age). Age-integration theory also recognizes the influence of interactions at the micro level. Borrowing from Durkheim's concept of solidarity and integration, gerontologists focus on factors associated with the elderly (Rosow, 1974). Rosow proposes that as a result of status and role losses due to age, the social networks of older adults suffer. If older adults could surround themselves with their peers, they would have a stronger sense of social solidarity, and ultimately greater life satisfaction. A limitation of age-integration theory, however is that (like age-stratification theory) it does not explain the heterogeneity within age cohorts. In addition, because of the arrangement of dominant social structures and the resulting social expectations, there will continue to be age strata in society. All societies divide the human life span into "seasons of life" (Hagestad and Nuegarten, 1985).

Political Economy of Aging

This theory picks up where age-stratification theory leaves off. Like age

stratification theory, political economy of aging highlights structural influences on aging. The political economy of aging theory, however, emphasizes power relationships and







14


social struggles that influence the lives of older adults (Estes, 1991). The political economy of aging theory also helps us to understand how the aged are defined and treated. According to the political economy of aging theory, "old age" is socially constructed, and created through power struggles. One drawback to this theory is that it has no place for positive aspects in the lives of older adults. Focusing on the power and the unequal distribution of status and resources is beneficial (Baltes, 1993). However, advantages come with older age, which the political economy of aging theory does not recognize.

These six theories have contributed a great deal to the study and understanding of aging. In those instances where a particular theory falls short of reflecting some of the patterns of behavior in social life, researchers in aging have been able to build on them and develop other plausible explanations. Examining these theories helps us understand them as independent explanations of age-related phenomena.

Theories of Aging Families

In the 1990s, a shift occurred in the complexity. malleability, and variety of family connections in the second half of life, generating a more complex view of older families. Theoretical advances related to the specific study of older individuals and families brought greater focus to intergenerational relations over the entire life span (DilworthAnderson, 1996. Allen et. al., 2001). Some of the prevailing theories are described next. Family Development Theory

Family development theory has been used to study predictable or normative

changes in families over time. Family development theory relies on basic assumptions about the family, and the processes the family goes through over time. It is a chronological, linear model, based on stages of family life. Family development theory








15


commenced from several theoretical perspectives (life-cycle categories, social-systems theories, human-development theories, life-span and life-course theories, and life-events and life-cycle theories) dating as far back as the early nineteenth century (Mattessisch and Hill, 1987). Family development theory gained strength as a theoretical perspective from which to study the family in the 1950s. Hill and Mattessich (1979; p. 174) proposed the following definition of family development theory:

Family development refers to the process of progressive structural differentiation
and transformation over the family's history, to the active acquisition and selective
discarding of roles by incumbents of family positions as they seek to meet the
changing functional requisites for survival and as they adapt to recurring life
stresses as a family system.

Similar to the life-course perspective in the study of individuals, in order to

investigate the changes families experience over time, a basic set of stages were created to set a framework for the family life span. These are called "family life cycle stages." Duvall and Hill (1948) proposed a three stage family "career," consisting of "changes in family size, changes in age composition, and changes in the occupational status of the breadwinner." Since then, other theorists have further developed this idea. Sociologist Paul Glick was among the first to analyze families in terms of a life-cycle with five stages that spanned the life course from first marriage to the death of a spouse (Glick and Parke, 1965). One of the more commonly used developmental theories for family sociologists is an eight-stage model (developed later also by Duval in 1977) in which attention is paid to life stages of the children.

Family development theory reached its height of development in the 1960s. This theory has given us important insight into the complexities of family life. Family development theory is useful as an indicator of changes that occur within family life, and calling attention to the changing nature of families, family formation, and family








16


relationships (Schwartz and Scott, 2000). However, a long-standing critique is that this framework cannot incorporate the variety of family forms that exist in today's society. Distinguishing among different types of families is difficult with this model. Moreover, this model perpetuates the notion that nontraditional families are pathological in their development. The family has evolved, and family forms are much more diverse. Couples who remain childless. are not represented in this model.

The family development model also assumes that the wife does not work until the children are old enough to attend school. This does not reflect current trends in motherhood. Of all women who return to work after having children, 65% have children less than 6 years old. This percentage is higher for African American women (U.S. Bureau of the Census, 1998). These family life-cycle stages also do not consider the possibility of stay-at-home fathers. While the numbers remain relatively low, a growing number of men choose to stay home and raise children. In 1997 there were almost 3 million father-only families, an increase of 220 % since the mid-1970s (Weldon, 1997).

Families that provide care for their elderly parents are also not represented in the family development model. Often, families that have elderly parents residing with them also have their own children still at home. This "sandwich generation" is becoming more common, as the cost of caring for older family members continues to increase. In addition, adult children are remaining or returning home for economic reasons. This has been termed the "cluttered nest." This arrangement has significant effects on family role expectations and relationships (Bengtson, et. al., 1996).

Finally, the current set of stages allows no way to account for divorced, single headed households, or couples who re-marry navigate the stages in family life. For








17


example, a couple that remarries, already having children from one or both of their previous marriages, never experience time alone with the spouse. At the very least, there is no "stage 1 adjustment period." Several studies have shown that imposing the expectations of the traditional family onto remarriages leads to additional problems and higher rates of divorce (Bray et al. 1998). Other researchers have identified fundamental ways in which the remarried family differs from the traditional nuclear family (Beer, 1989; Papernow, 1993).

Another criticism of family development theory is that this framework does not

allow for intracultural differences in family forms. and the scope of family development theory does not extend to cross-cultural differences in families. For example, distinct catalysts in the transformation and structure of African American families as a result of the shift from rural to urban living have "manifested in spousal relationships, family formation, the socialization of children, and techniques of coping with dissolution, especially divorce" (Wilkinson, p. 42). Mexican American families' concept of family refers to more than just the nuclear family. The average household size of Mexican American families tends to be large, "Mexican Americans are more likely to live in an extended family context" (Sanchez, 66). In addition, the socialization of children is very different for Mexican American families. Adolescence marks different behavior patterns between boys and girls (Locke, 1992).

Social Exchange Theory

While social exchange theory was previously discussed as an aging theory, it

clearly has relevance in family studies. This theory is useful in studying changes in the family over time. The premise of social exchange theory is that all social interaction stems from social exchange among individuals. The work of the prominent social








18


exchange theorists did not originally focus on family issues. However, "since the 1960s, the exchange framework has methodologically worked its way into the mainstream of family studies" (Sabatelli and Shehan, p. 396). The development of social exchange theory can be traced to the works of George Homans and Peter Blau, and the primary assumption is that "humans choose between alternative potential relationships, or behaviors by ranking the actual or expected experiences associated with each and then selecting the best alternative" (Blau, 1964). Individuals are more likely to pursue relationships with those who possess desirable and valued attributes. Different resources and exchange processes are at work throughout family life. Thus, family exchange theory makes a valuable contribution toward helping us understand the variety of family behaviors and relationships over time.

Intergenerational Stake Hypothesis

Making use of both a developmental and a social exchange framework, another ideological structure useful in understanding family relationships over time is the "intergenerational stake hypothesis." This hypothesis was developed by Bengtson and Kuypers in 1971, as a way to understand and explain disparities in perceptions of intergenerational relationships among parents and children. Specifically, the premise of Giarrusso, Stallings and Bengtson's (1995, p. ) hypothesis is:

Contrasts in developmental needs explain differences in the "stake" each generation
has in intergenerational cohesion, continuity and conflict. Furthermore, these
differences in developmental stake explain contrasts in opinions and orientations between the generations, independent of any specific issues that appear to be the
focus.

While a plethora of studies have used the stake hypothesis to explain empirical

findings, clarifications have been made to account for some deficiencies. These include the application of two theories developmental theory and exchange theory. Using the







19


stake hypothesis, these theories provide a longitudinal framework to analyze intergenerational stakes that exist in families over time (Giarusso, Stallings and Bengtson, 1995).

Socioemotional Selectivity Theory

Socioemotial selectivity theory is one of the most important theories to emerge in family gerontology in the last decade. This theory helps to explain the negative correlation between the frequency of social interaction and the degree of emotional closeness among family members as they age (Carstentsen, 1992; Allen et. al., 2001). As the functions of relationships change over the life course, adjustments to the intricacies of social networks among family members help "maximize social and emotional gains" (Allen et al., p. 134). Older people's preferences for emotionally meaningful social relationships are due to perceived limitations of time. Older adults become more selective about their social networks. In other words, older individuals become more selective about whom they spend time with, and with whom they share their emotions, as they age. This has limitations for studying familial relationships. Family relationships are too complex to assume that in later life, older adults have control in choosing to engage in interactions with family members. Social Comparison Theory

One of the more important factors to be found in midlife and later life (in terms of navigating a successful transition into later life stages) is the notion of social comparison (Neugarten, 1968: Levinson & Levinson, 1996; Carr, 1997; Stewart & Vandewater, 1997; Carr, 2003). Social comparison is rooted in psychology and psychosocial behavior, and is a way of self-evaluating based on comparisons with other people, particularly with those individuals most significant in one's life (Festinger, 1954; Suls &








20


Mullen. 1982; Sulls & Wheeler, 2000). This comparison and self-evaluation has important consequences on self-esteem, and this is typically how social comparison is applied in the psychology literature. However, it is important to recognize that an alternative ramification of social comparison may be its effects on relationships between individuals and the people they compare themselves to.

Social comparison theory originally emerged in the early 1950s when researcher Leon Fesginger determined that it was important to people to hold correct assumptions about the world in which they live, but more specifically about their own abilities in this world (Festinger, 1954). Festinger's hypothesis focused on the idea of social comparisons as a need for self-evaluation. However, researchers later revealed other motives for social comparisons (Hakmiller, 1966; Carr, 2003).

In general, the social comparison hypothesis suggests a positively correlated

relationship between evaluation with another person and the effects of this evaluation on self-esteem and emotional well-being (Stouffer, 1948: Festinger, 1954; Suls and Mullen, 1982; Carr, 2003). Thus, a comparison which yields a positive evaluation, will result in positive effects on self-esteem and positive social well-being, and vice versa for a negative evaluation (Carr, 2003).

Ultimately, grounded in the psychological literature, these comparisons and selfevaluations are used to measure self-enhancement outcomes in individuals (Suls & Wheeler, 2000: Carr, 2003). This is beneficial for psychologists. However, from a sociological standpoint, particularly in the interest of intergenerational family relationships. and more specifically, in terms of evaluating social change and women's







21


intergenerational relationships, social comparison could be useful when evaluating its effects on perceptions of interpersonal relationships between generations.

Understanding Social Change and Women's Intergenerational Relationships

While there are several meaningful and useful theories in the aging and family literatures, this study will employ two specific guiding frameworks, Intergenerational Family Solidarity Theory, and The Developmental Schism. Both are important in understanding how social structural change and changing life course trajectories for women over time have effected daughters' perceptions of intergenerational solidarity with their mothers in different life stages. Following is a discussion of each of these frameworks, and how each contributes to this study. Life Course Theory and the Developmental Schism.

There has been a resurgence of the life-course perspective. Elder (1998) explained that this theoretical framework, with its focus on sociohistorical processes, and on individual time, satisfies two increasingly important considerations in studies of families: 1) how individuals change over time, and 2) how their transitions and trajectories are linked across family members. This perspective has been combined with other theories to help explain many facets of aging families (Elder, 1998; Allen et. al., 2001). The life course perspective is a broad multidisciplinary perspective. It is more of a way of thinking, or a paradigm, rather than a specific or explicit explanation of anything (Elder, 1998). Also, the life course perspective allows for the understanding of the impact of a multitude of social and historical factors and experiences on human development. This makes it an ideal perspective to understand the continuous process of aging related phenomena and the impact of social change.








22


Building from a life course perspective, Fingerman (1996) introduced the concept of "developmental schism." A perspective grounded in developmental psychology, this accounts for a key source of interpersonal problems between female intergenerational dyads, which tend to be the strongest and most lasting of filial relationships (Allen, et. al., 2001). Fingerman provides explanations of the mother/daughter bond at three levels of descending analysis: sociopolitical, familial, and psychological. She argues that social and political factors influence the strength of a mother/daughter relationship. However, these factors do not adequately explain the emotional content of their relationship. Moreover, they do not account for the investment of mothers and daughters in their mutual relationship (Fingerman, 2001).

The central idea of the developmental schism hypothesis is that the tensions

between parents and children arise due to the fact that these different generations are at different developmental stages. As a result, each generation has different goals, and varying needs, which result in a difference of perspective. A disparity of roles, status, and cohort membership influence the development of different perspectives as well (Fingerman, 2001). While the developmental schism is psychologically based, it is takes the life course perspective to a more specific level of analysis, focusing on the role, status, and goal needs, which may be specific to cohort membership.

The developmental schism is useful for this project in that it recognizes the variety of aspects that are present in complex interpersonal relationships. Just as Bengtson's intergenerational family solidarity theory reflects behavioral, emotional and cognitive aspects of solidarity through its various empirical measures, Fingerman's hypothesis accounts for aspects of behavior and emotion as well. In addition, Fingerman is clear







23


about the significance of the social context and it effects on the mother/daughter relationship.

Intergenerational Family Solidarity Theory.

Intergenerational Family Solidarity Theory is another major theoretical

contribution to the area of family gerontology. As a result of the diversity in function and structure, intergenerational relationships become more important. Research from the Longitudinal Study of Generations has been used to demonstrate the strengths of multigenerational ties over time, and why it is necessary to look beyond the nuclear family when asking whether families are still functional (Bengtson, 2001).

Interpersonal relationships among family members of different generations are assessed based on six measures of solidarity: associational, affectual, consensual, functional, normative, and structural (Bengtson and Roberts, 1991). These six components of solidarity capture the numerous aspects of family interactions. Included in these aspects of interaction are frequency of contact, emotional closeness, agreement with regard to values and attitudes, helping and exchanging of resources, filial obligations and commitment, and physical family structure. Table 2.1 provides a detailed table describing these six measures.

Using these six dimensions of intergenerational family solidarity, five different family typologies emerge. These typologies are based on where the types of solidarity among intergenerational relationships are strong or weak, and thus help to explain the myriad of families functioning in our society in a variety of ways (Silverstein and Bengtson, 1997). The five types of family relationships are: 1) tight knit, 2) sociable, 3) intimate but distant, 4) obligatory, and 5) detached (Bengtson, 2001). Breaking down the overall idea of solidarity, or cohesion among family members, into manageable aspects,








24


can then be empirically measured and has been a major contribution to the study of aging families. It is important to note that intergenerational family solidarity theory focuses on solidarity, with little understanding of conflict (Connidis, 2001 ).

In this study, intergenerational solidarity forms the basis of analysis for

understanding women's intergenerational relationships, by utilizing the empirical outcomes of this theory. Three of the six measures, associational, affectual, and consensual solidarity, are used to gain a better understanding of how daughters perceive their relationships with their mothers. By using these three measures of solidarity, interaction, emotion and cognitive aspects of the relationship are captured. Analyses will measure these perceptions at two life stages. and in each life stage, at two points in time. Chapter 4 gives a more detailed description of how these measures are used.

Conclusion

In summary, various theories on aging families suggest different outcomes in later life stages. The framework guiding this proposed study is the developmental schism, stemming from a life course perspective. The developmental schism predicts that parentchild relationships converge, as children become adults. To help understand this phenomenon, this study will use Bengtson's measures of intergenerational solidarity, which stem from intergenerational family solidarity theory. The developmental schism will be the foundation for understanding the developmental hypotheses. The impact of social and structural factors on mother/daughter relationships, and the complexity of these relationships is explained by the developmental schism as a result of mothers and daughters being at different developmental stages, and having different goals and needs (Fingerman, 200 1). While the mother/daughter dyads in this proposed study are at the same developmental stage as defined by age young adulthood and midlife social







25


change suggests that the developmental stage has been affected. Each generation has a different perspective, affected by their roles, status, and birth cohort, thus having different goals and needs during the same life stages.















CHAPTER 3
REVIEW OF THE LITERATURE

Introduction

The average life expectancy in the United States has increased a great deal over the last century. This increase has led to multiple generations of families living longer together. As a result, there have been new family dynamics arising as well. Researchers have been interested in how these new dynamics have affected family life. In addition, several social structural changes occurring during the twentieth century have affected the way individuals live out their lives.

This chapter discusses the changes in social structural forces, outlining some of the social ramifications and their impact on individuals as well. The focus then turns to an overview of the research on intergenerational family relationships. then specifically on women's intergenerational relationships. Finally, the hypotheses for this study are outlined at the end of this chapter.

Changes in Social Structural Forces

The focus of this next section is to outline some of the major legislative, economic, demographic, and ideological changes that have occurred. These changes have helped to reshape the life course trajectories of women in this country over the last century. Legislative Changes

The Equal Pay Act of 1963, and Title VII of the Civil Rights Act of 1964, which prohibits sex discrimination in employment, were the first major legislative steps toward ensuring equality in the workplace (Felder, 1999). Congress passed Title IX of the


26








27


Education Amendments of 1972, prohibiting discrimination on the basis of sex in federally funded education programs (Stetson, 1992; Langley, 1994). These laws opened up doors for women, providing expanding opportunities in the educational arena and in the job market. In 1960, the Food and Drug Administration (FDA) approved the birth control pill, and in 1973, abortion was legalized (Stetson, 1992; Langley, 1994). These two landmark decisions afforded women opportunities of choice not previously available in terms of family planning and decisions about child bearing. These legislative changes improved women's access to education and occupational attainment (see discussion below), and are reflected in the changing roles of women in other structural domains (i.e.. marriage and family). These changes have had an impact on American society's attitudes concerning women's roles as well.

Education and Occupational (Status) Attainment

The number of women with a college degree has greatly increased during the second half of the last century. Since mid-century, the number of women graduating from high school and enrolling in college rose by twenty percent. The proportion of higher degrees being awarded to women also increased significantly, by an average of thirty percent for each degree level (U.S. Department of Education, 1999). The social trends in education have led to cohort differences in education, in that the younger groups of women are more likely to be more highly educated than past cohorts of women, often their mothers.

Changing Work Roles

As a result of legislative and social changes women across all ages are more likely to be employed (Moen, 1992). Americans have achieved greater social mobility over time. Biblarz, Bengtson, and Bucur (1996), also find this to be true with occupational







28


attainment within family lineage. However, the rate of upward mobility has slowed down with each succeeding generation. A review of the second half of the century presents the expanded range of life choices available to women (Gerson, 1984; Rexroat and Shehan, 1984). As a result of both continued persistence, as well as ongoing economic growth, women continued to gain ground. Today, the majority of women are employed, and make up close to half of the workforce (U.S. Department of Labor, 2000). As a result of educational opportunities, the occupational choices for women have expanded. Many more women hold jobs, and in a much greater variety of occupations than in the past. A strong presence of women has been established in previously maledominated professions (Opdycke, 2000).

During the second half of the century, women have spent a large proportion of their adult lives employed. Each generation of women return to employment more quickly or never leave it after having children (Moen, 1992). The number of married women with children under age six in the workforce rose to almost 70 percent by 1997. Moreover, over half of all women with infant children are working (Moen, 1992; Hattery, 2001). Divorce, single parenthood, economic need, occupational opportunity and the affordability and availability of childcare are some of the social forces directly associated with maternal employment (Hoschschild, 1997; Brayfield, 1995). More than half of all women have a desire to successfully combine family life and work life (Roper Organization, 1995).

The mid-1970s also saw a severe decline in men's wages. Many women decided to go to work simply to make family financial ends meet, and to maintain their desired standard of living. By the end of the twentieth century, the dual-income family became







29


the norm (Opdvcke. 2000). During the 1970s, there was also a rise in the number femaleheaded households contributing to the growing female workforce. Family Formation

In 2000, the average age at first marriage was 25.1 years for women. In the mid1950s, women were slightly over the age of 20 when they married for the first time (20.2). Just as the age at marriage has increased over the past 50 years, the proportion of women in their twenties and early thirties who aren't married has also increased. In 1960, twenty-eight percent of women aged 20 to 24 had not yet married. By 2000, this had increased to 73 percent (Fields and Casper, 2001).

Fertility rates have declined significantly since the post war, Baby Boom

generation. The number of babies being bom per woman dropped significantly from an average of 3.69 in the 1950s to 1.84 in the 1970s. In 2000, the number rose to an average of 2.13, which was the first time in thirty years that the number of births was above replacement level (Martin et. al., 2000).

Women have begun to delay having children in order to complete education and establish careers (McMahon, 1995). The median age of first time mothers has risen consistently over the past thirty years. The median at first birth increased to 24.6 years by the turn of the century (Martin et. al., 2000). In the 1950s, during the baby boom, most women were having children in their early 20's. This increased to the late 20's in the 1960s. Then in the 1970s and 1980s, a large number of women were giving birth for the first time in their early to mid 30's, with a growing number of women having children in their late 30's and even early 40's (Moen, 1992; Martin et. al., 2000).








30


Marital Stability

The United States has also experienced a dramatic rise in the rate of divorce.

Sociocultural factors, like the increased economic independence of women, changing family functions, the changing nature and high expectations of marriage, fewer social, legal and moral constraints, have all contributed to a greater acceptance of divorce (White, 1991: Kurdek, 1993). From 1965 to the mid-1980s, the divorce rate doubled. By the end of the eighties, the divorce rate has stabilized at this higher level, where it remains today. The implications of divorce are seen in the high percentage of women in the workforce who have no other means of economic support (Moore and Sawhill, 1984; Cherlin and Furstenberg, 1988; Kurdek, 1993). Ideolopgical Shift

The economic period following World War II reemphasized the idea of the stay at home mom, by providing unprecedented economic stability for families. Thus, the intensive motherhood ideology became dominant at mid-twentieth century (Hays, 1996; Reskin and Padavic, 1994; Coontz, 1992). However, there were significant racial and ethnic differences in women's ability to adhere to this ideology (Segura, 1994; HillCollins, 1994). More recently, as a result of changing social and structural forces, attitudes regarding appropriate roles for women in the family have changed to reflect non- egalitarian ideals (Hays, 1996, Margolis, 2000). Today, social perceptions reflect the belief that working women can be good mothers, and especially those women who work out of personal desire, rather than solely out of economic need (Hattery, 2001; Hays, 1996. Klein, 1984).

These social changes have led to disparate life course trajectories for women of

different birth cohorts. In the 1950s, women of the late World War II Generation and the








31


early Baby Boom Generation were in their young adult and adult years, getting married and starting a family at a much younger age compared to the Generation X birth cohort who were entering the young adult years in the 1980s. The Baby Boom Generation was the first birth cohort to take advantage of the educational opportunities presented by Title IX in the 1970s. In addition, women born later in the late Baby Boom years began postponing having children until their late thirties. Women of the Baby Boom generation also saw higher rates of divorce in the 1980s. The majority of Generation X women reach their 25th birthday still single, and are postponing childbirth until they are in their thirties. Educational and occupational opportunities, as well as social pressures, have led to the idea of a "career," where as older generations of women may have held "jobs" if it was financially necessary, but family life was the woman's main priority.

Intergenerational Family Relationships

Similarity of beliefs and values between parents and children has long been

recognized as an important source of stability in society. Indeed, the transmission of sociolpolitical ideologies from one generation to the next permits continuity within families (Miller and Glass. 1989). However, the degree of parent-child similarity at any give point in time for any particular age cohort may be the product of contextual processes (Dannefer, 1984). Period effects that have a different impact on different cohorts and/or age groups may either separate the sociopolitical interests of different generations or draw them closer together (Miller and Glass, 1989).

Rossi and Rossi's (1990) classic study of the life course and intergenerational relationships (specifically parent-child relationships), indicates that as children grow older, new factors become relevant for intimacy. The once central indicators like parental affection, parental authority, and family troubles, shift to role and status








32


attainment markers such as finishing school, occupational attainment, marriage, and children (Rossi and Rossi, 1990). When the younger generations don't share similar roles or statuses, there is more intergenerational strain. For example, if the younger generations are not married, and/or don't have children, the difference in these statuses from their own parents create more of a strain in terms of a general understanding of life experience. When children, or the younger generations, move into normative roles, they share similar experiences, bringing the generations closer (Bengtson and Black, 1973; Aquilino, 1997). While the "role" might be the same, the experience of the "role" or the ramifications of that "role" are different for the succeeding generations as a result of the changes in social and structural forces. This is especially true for women. Thus, even when two generations hold similar roles there maybe intergenerational conflict.

The nature of parent-child relationships in earlier years sets the stage for these relationships in later life. There is evidence of strong continuity in the perception of relationship quality in relationships from early years into adulthood. From adolescence to young adulthood, past relationship pattems continue to be played out. The effects of this are the strongest in measures of emotional closeness (Roberts and Bengtson, 1993; Aquilino, 1997: Thomton et. al., 1995; Tubman and Lemer, 1994; Rossi and Rossi, 1990). That is to say, emotional bonds tend to remain more stable over the life course, compared to other aspects of intergenerational relationships. such as frequency of interaction, shared values and world views, exchange of resources, meeting familial obligations, and geographic proximity of family members.

Intergenerational Differences Among Women

It was previously stated that the mother/daughter relationship is unique and significant. The strength of this bond sets it apart from other social relationships







33


(Fingerman, 2001). There are several factors that contribute to the complexity and the strength of the mother/daughter relationship. Gender socialization and gender differences, a shared status in society, and changes in the larger social structure allow for many distinct differences often apparent in mother/daughter relationships.

As gendered beings, women's relationships tend to be traditionally characterized by stronger emotional bonds. In addition, for women, the maternal role tends to be more psychologically salient for women, as they have been taught the importance of motherhood from a young age. These psycho-emotional elements play a key role in the mother/daughter relationship. Despite geographic obstacles, women tend to maintain strong ties to their parents (Bedford & Blieszner, 1997; 1995; Gottman, 1998).

The strong filial ties daughters maintain to their parents are also a result of gender socialization. While young males are encouraged to gain independence as part of childhood socialization, young girls are socialized to remain close with their mothers (Chodorow, 1994). In addition, daughters rather than sons are more likely to provide later life care for their aging parents. The tum of this century saw women over 60 years of age as far more likely to have an aging parent still alive compared to women at the turn of the last century (Moen, 2001). Life expectancy for women at birth is 4 to 12 years longer for women than for men. With women living longer, the likelihood that women will end up widowed and not remarry is probable. Hence, women who do not have a partner in later life will tum to their children, particularly their daughters for much of the support they need.

Despite a more patriarchal culture, it is the women who tend to be central to the kin networks in our society. Women often have great insight into the larger family structure,







34


and individual family members. The mother/daughter relationship is at the core. Thus, those families with strong mother/daughter connections often foster stronger ties among other family members. With marriage, women tend to bring their husbands into their family of origin.

As a result of the changing life course trajectories for women outlined previously in this chapter, relationships among women in general have been also been affected. With an increase in singlehood and divorce, as well as a decrease in fertility among women in successive generations, women are more apt to face aging alone (Fingerman, 2001; Moen, 2001). Continuing relationships with their own aging parents, particularly their mothers, likely becomes more important. Thus,just as aging mothers will to turn to their daughters in later life, as women age, they in tum are will rely on their mothers as well.

Statement of Hypotheses

This section details this project's hypotheses. The hypotheses HIA through Hlc will each be assessed in chapter five (young adulthood). Hypotheses H2A through H2c will each be assessed in chapter six (midlife). Generation X and Baby Boom Daughters During Young Adulthood

In order to answer the first research question, "How have daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during young adulthood changed, and what may be associated with this change?" several hypotheses will be tested. These hypotheses are outlined below:

* H I: Young adult daughters from Generation X have stronger (more positive)
perceptions of associational and consensual solidarity compared to their Baby
Boom mothers when they were young adults.

* HIB: Affectual solidarity will not be affected by the daughters' social
characteristics the same way associational and consensual solidarity will be.








35


* Hlc: Social distance between the Generation X daughters and their Baby Boom
Generation mothers comprising each mother/daughter dyad is associated with
associational and consensual solidarity, but not with affectual solidarity. Baby Boom Generation and World War II Generation Daughters During Midlife

To answer the second research question, "How have daughters' perceptions of associational. affectual. and consensual solidarity with their mothers during midlife changed, and what may be associated with this change?" the following hypotheses will be tested:

* H2A: During midlife, daughters from the Baby Boom Generation will have
weaker (more negative) perceptions of associational and consensual solidarity
compared to their World War II Generation mothers during midlife.

* H2B: Affectual solidarity will not be affected by the daughters' social
characteristics the same way associational and consensual solidarity will be.

1 H2c: Social distance between the Baby Boom Generation daughters and their
World War II Generation mothers comprising each mother/daughter dyad is
associated with associational and consensual solidarity, but not with affectual
solidarity.

The above hypotheses focus on the importance of the larger social structural

changes, and from a life course perspective, the changing roles of women as they move through the life course stages. These hypotheses reflect the complexity of the mother/daughter relationship over time. In addition, they explore the different aspects of intergenerational relationships and how they may be affected. The next chapter will provide a description of the data, sample, variables, and analytic models used in the analysis of this project.















CHAPTER 4
DATA AND METHODS

Introduction

For this study, analyses focus on generational differences during two life course stages. Generation X daughters' perceptions of their relationships with their mothers during young adulthood in 1994 are compared to Baby Boom daughters' perceptions of their relationships with their mothers during young adulthood in 1971, and Baby Boom Generation daughters' perceptions of their relationships with their mothers during midlife in 1994 are compared to World War II Generation daughters' perceptions of their relationships with their mothers during midlife in 1971. This study uses the Longitudinal Study of Generations and Mental Health, 1971 data set [made accessible in 1990, machine-readable data files]. These data were collected by V. Bengtson and M. Gatz and are available through the archive of the Henry A. Murray Research Center of the Radcliffe Institute for Advanced Study. Harvard University. Cambridge, Massachusetts (Producer and Distributor). These data are from an ongoing study of aging families investigating changes in intergenerational family relationships. These data specifically examine intergenerational solidarity, based on perceived and actual levels of cohesion among family members, making it an ideal data set for this study. Outlined below is a more detailed description of the data set.

Sample

The Longitudinal Study of Generations and Mental Health (LSOG) is an ongoing 31-year longitudinal panel study. The LSOG examines four generations of aging


36








37


families, seeking to investigate several things, including the changes in family social supports and their impact on the mental health of family members. The sample came from a large prepaid health maintenance organization in Southern Califomia. The original participants comprising the first generation in this study were randomly selected on the basis of potential grandparenthood. This sample represents the demographic climate of Southern California at the time of the study's inception. The participants are generally white, and from middle class or economically stable families.

Originally, the study began in 1971 as a cross-sectional study, and included 345

three-generational families. Although it was not initially intended as a longitudinal study, a second wave of data were collected thirteen years later in 1984-85, after the researchers serendipitously came across the participant files from the original study. Additional waves of data were collected every three years after that (1988, 1991, 1994, 1997, and 2000). The three generations were followed through the first three time periods. At the fourth period of data collection in 1991, the fourth generation of these families was included in the study for the first time. Because 1994 was the most recent wave of data available through public archiving at Harvard, this research includes data from 1971 and 1994 time periods of the LSOG.

Using the standard distinction in aging family literature, the generational cohorts of women are comprised as follows: 1) the first generation, Gl's. who are the mothers of the G2's, grandmothers of the G3's and later great-grandmothers of the G4's, 2) the second generation, G2's, the mothers of the G3's and later grandmothers of the G4s as well as being the daughters of the Gi's, 3) the third generation, G3's. the granddaughters of the Gl's and the daughters of the G2's and later mothers themselves of the G4's, and 4) the








38


fourth generation. G4's, who are the great-granddaughters of the Gl's, granddaughters of the G2's. and daughters of the G3's.

Three sets of mother-daughter dyads, representing four generations in the same family, will be examined in this study. The oldest generation, the Gl's, consists of women born prior to the Great Depression some during the late nineteenth century and who were coming of age before World War 11 (1888-1927). The second generation, the G2's, were all born after the turn of the twentieth century, some during the Great Depression, but before World War 11 (1912-1941). The third generation, the G3's, were all born after World War 11, and represent the first of the Baby Boom generation (19441956). The fourth generation, the G4's, somewhat overlap with the Baby Boom women, they were born between the years 1954 and 1979. The age range for the women in the LSOG sample, and their movement as they age, through the social structural changes previously discussed as occurring in the twentieth century, can be visually interpreted by referring to figure 1. 1 in Chapter 1.

The Generation X women were moving through the young adult life stage during the eighties and early nineties. Their Baby Boom mothers experienced young adulthood from the late 1950s to early 1970s. The Baby Boom, as they move through the life course, began experiencing midlife in the late eighties-early nineties, continuing through today. Their World War II generation mothers, however, experienced this life stage as early as 1950 through to the early eighties.

Sample Selection Characteristics

There were several steps involved in preparing the LSOG data for this study. For the analysis of daughters during young adulthood, only those Generation X and Baby Boom generation women age 18 to 29 at their respective time periods (1994, 1971) were







39


selected for this study. For the analysis of daughters during midlife, only those Baby Boom and World War II generation women age 40 to 59 at their respective time periods (1994. 1971) were selected for this study. After the women for each set of analyses were selected based on the appropriate age, they were then selected based on whether or not they responded to the specific questions which comprise the dependent variables, associational, affectual, and consensual solidarity. Those women who did not answer any of the questions pertaining to the solidarity measures were eliminated from the analyses.

In addition, once the women were selected on the basis of having responded to the solidarity questions, it was necessary to ensure that only one daughter from each family, be included in the analyses to eliminate intrafamilial bias. First, all of the women who had no sisters were automatically included. Of the remaining women, one sister from each family was randomly selected. Finally, the last step for the sample selection of women to be included involved matching these women with their mothers. Some women were initially interviewed for the LSOG because their fathers were participants, and some women did not have daughters of their own, but rather sons. If the daughter did not have a mother who also participated in the LSOG, they were excluded from the analyses.

The numbers presented in Tables 4.1 and 4.2 represent the actual number of women in the sample from each generation included in this study. Because there were several criteria for being included in the analyses for this study appropriate age at the time of data collection, necessary responses to the solidarity questions representing the dependent variables, having only one daughter represented from each family, and a match to a "study mom" and the fact that the LSOG sample size is relatively small to begin with, the analytic samples are somewhat small.








40


Daughters During Young Adulthood

Table 4. 1 illustrates the numbers of women from both generations during young adulthood, and how these numbers decrease as a result of the sample selection process. The original sample of Generation X daughters, the G4's in 1994, contains 200 women. Once the women are selected on the basis of age, 18 to 29 years, for the young adulthood analyses, this number drops considerably to 136. Of the Generation X women in this age range, there were not any excluded on the basis of no response to the dependent variable questions regarding solidarity, thus the number remains at this stage at 136. However, when randomly selecting only one daughter from each family, the number of women again drops dramatically to 88. Finally, when selecting on the basis of a match to a study mother, the number drops to 82. Similarly, the Baby Boom daughters, the G3's in 1971, start with an overall sample size of 442. The most significant drop among this sample is when selecting for the young adult analyses. The sample size drops to less than half, 185, when only those women 18 to 29 are included. There were six women who did not answer any of the solidarity questions, thus, the sample dropped to 179 at this stage in sample selection. When randomly selecting only one daughter per family, there is a drop again to 135. Finally, based on matching these daughters to a study mother, the number of Baby Boom daughters in young adulthood included in this study is 124. Daughters During Midlife

The numbers of women in both generations during midlife are presented in Table

4.2. The number of Baby Boom daughters, the G3's in 1994, originally participating in the LSOG is 410 women. For the midlife analyses, this number drops to 230 women age 40 to 59. Only three women were excluded as a result of not answering any of the dependent variable solidarity measures, reducing the sample at this stage to 227. After








41


randomly selecting only one daughter from each family, the number is reduced to 150 women. Ultimately, after matching each daughter to a study mother, an additional 42 are eliminated, and the final sample size for Baby Boom daughters in midlife is 108. There are initially 377 World War II women, G2's in 1971, in the LSOG study. When selecting daughters based on age for the midlife analyses, the sample is reduced to 259 women. Several of the World War II daughters were eliminated from the analyses based on not responding the questions comprising the measures of solidarity, further reducing the sample to 209. The number drops to 182 once only one daughter is randomly selected from each family. After matching the daughters to a study mother, the final sample included 143 World War II generation women in the midlife analyses.

Measurement

Various measures were constructed in order to model the relationship between

sociodemographic variables pertaining to education, work, and family, and perceptions of solidarity. Because certain questions were not asked of all LSOG participants at each time period. specific measures were selected that were comparable across the two time periods, 1971 and 1994, when asked of each generation of women. The measures of solidarity used for the dependent variables are measures previously used in many empirical studies, and are described next.

Dependent Variables

The dependent variables in this study are those variables which measure the three aspects of intergenerational solidarity being examined: associational solidarity, effectual solidarity, and consensual solidarity. Nominal definitions and empirical indicators used to measure these constructs are found in Table 4.2 (Bengtson and Schrader, 1982; McChesney and Bengtson, 1988; Bengtson and Roberts, 1991). The specific measures








42


for this proposed study will use Bengtson's constructs of intergenerational solidarity (Mangen. Bengtson and Landry, 1988) and are outlined in the next section. Appendix A includes a detailed matrix describing the wording for each question asked of each generation at each time period, and includes all response categories. These measures have been selected, in part, because of comparability across the LSOG study waves and the generations of women included in this research. They were also chosen because within the generational dyads, a variety of behavioral components associational, affectual, and cognitive are reflected by these three aspects of family solidarity (Bengtson and Roberts, 1991).

Associational solidarity. Associational solidarity is the amount and frequency of interaction among family members. The more contact or interaction among family members, the more solidarity within the family based on this solidarity aspect (Silverstein and Bengtson, 1977). This study uses the frequency of various types of interaction face to face, telephone, and mail to determine the level of associational solidarity among these four generations of women. For each of the three types of interaction, the following question was asked, "During the past year, how often were you in contact in person, by phone, by mail with your mother?" These questions were asked of the GI's through G3's in 1971, and GI's through G4's in 1994. However, the response categories at each time period differ somewhat (see Appendix A). Each question was based on a scale ranging from 1 to 8, with 1 being never or not at all, and 8 being daily or almost every day. The three measures of associational solidarity combine to make an overall measure with a range of 3 to 24.








43


Affectual solidarity. Affectual solidarity refers to the level of emotional closeness that exists among family members. It is the perception of "positive sentiments" warmth, understanding, trust, etc... family members have for each other (Bengtson and Roberts, 1991). The LSOG asks a set of questions, relating to affect among family members. These questions are as follows: 1) "Taking everything into consideration, how close do you feel is the relationship between you and your mother'?" 2) How is communication between you and your mother, that is how well can you exchange ideas or talk about things that really concern you at this point in your lives'?" 3) Overall, how well do you get along with your mother at this point in your life?" 4) "How well do you feel your mother understands you?" and 5) How well do you feel you understand your mother?" These five questions were asked of the GI's through G3's in 1971, and GI's through G4's in 1994. Each item is measured on a 6-point scale ranging from "1=not very much" to "6=extremely." By incorporating these five questions into one scale, the measure of affectual solidarity becomes a continuous variable with a range of 5 to 30 by simply summing responses across the set of questions.

Consensual solidarity. This aspect of intergenerational solidarity is the degree to which there are similar values and beliefs about world views among family members of different generations. Bengtson's construct of this aspect of solidarity is measured by a single item. The following question was asked of the GI's through G3's in 1971. and the GI's through G4's in 1994. "In general, how similar are your views about life to those of your mother?" This item is measured by a 6-point scale ranging from extremelyy different" to "6=extremelv similar."







44


Independent Variables

The independent variables in this study include those sociodemographic variables, which reflect the generational differences among these women as a result of changing social roles and role expectations with regard to education, work, and family. Not all of the women were asked the same questions at each time period. As a result, the independent variables are limited to those common questions pertaining to education, work, and family indicators. Furthermore, a relatively small set of the most salient variables pertaining to education, work and family have been selected because the sample size for this research has limited the number of covariates which could be included in the OLS regression analyses.

Age. This is the daughter's actual age in years at the time of the LSOG data

collection. For the analyses in this study, the ages range from 17-24 for the daughters during young adulthood, and 40-59 for the daughters during midlife. Even though the age range for the women is the same when comparing across the generations during each life stage, the ages with in these age ranges vary. Therefore, in the multivariate analyses, age is used as a control variable in each of the four models.

Education level. This is the daughter's highest level of education at the time of

LSOG data collection. In 1971, this variable was a categorical variable with a range of 1 to 8. The lowest education category, 1, being "grade school (1" through 6th) and the highest, 8. being "a post-graduate degree (Ph.D., DDS, MD, JD). In 1994, the variable was again an eight category variable, reflecting the level of education at that time, however the answer categories varied from those offered in 1971. The variable ranged from 1 to 8, with the lowest category being "an 8 grade education or less," and 8 being

-a post graduate degree." For this research, the education variable has been reconstructed


I








45


to reflect whether or not the daughter has less than or equal to a high school degree, or more than a high school education. The reference group for the regression analyses is "more than or equal to a high school degree."

Employment status. This variable is indicative of the daughter's employment

status at the time of data collection. In 1971, this variable asked the women if they were working for pay. It was a 5 category, nominal variable, including "I=employed full time," "2=emploved part time," "3=employed full time with additional part time," "4=unemploved, seeking pay," and "5-not employed (homemaker, retired, not seeking)." In 1994, this same variable was a 6 category variable. At this time, the categories included "l=no, retired," "2=no, temporarily out of work," "3= yes, temporary job," "4=yes, part time," "5=yes, full time," and "6=no, housewife." For the purposes of this research, the employment status variables has been recoded as a dichotomous variable, indicating whether or not the daughter is employed or does not work at the time of the LSOG data collection, thus comparing those women who work to those who do not. The reference group for this variable is "does not work."

SEI score for occupation. This score is a numeric value assigned to the different occupations that the women have. The score represents the value for that occupation at each LSOG data collection period. Because not all of the women worked, several women did not have a score assigned to them, further reducing the sample size when this variable was run in the models. Therefore, for the purposes of this research, this variable was recoded as a nominal variable, with three categories, "not working, no score," "low SEI score," and "high SEI score." Low and high SEI scores were calculated using the median SEI score, with all women in the LSOG study included, for each year of data collection.








46


In 1971, the median SEI score for women in the LSOG study was 45.00, therefore, those women included in the World War II Generation sample during midlife, and the Baby Boom Generation sample during young adulthood were coded as low SEI if they had a score below the median, and as high SEI if they had a score equal to the median or above, In 1994, the median SEI score for occupation for women in the LSOG study was 38.51. Women included in the Baby Boom Generation sample during midlife, and Generation X women during young adulthood were coded as low SEI if they fell below this score, and as high SEI if they had a score equal to or above this score. The reference group for this variable in the OLS regression analyses is "low SEI," thus a comparison is made between the high and low scoring women.

Income. This variable has been adjusted to reflect annual income in 1994 dollars, and represents the total household income for the daughters. There were several missing cases with regard to this variable. Therefore, mean scores for income were imputed so that that every woman had an income score represented. Scores were imputed based on education and marital status, using the original LSOG categories for these measures to ensure an appropriate range of variability among the sample of women.

Marital status. This is the daughter's marital status at the time of LSOG data collection, and represents the woman's current marital status. In 1971 this measure consisted of 6 marital status categories. They were "l=single, never married," "2=engaged," -3=married." "4=permanently living with someone, not legally married," "5=widowed." and -6=divorced or separated." In 1994, this same variable consisted of 7 categories, including "'=married," "2=separated," "3=divorced." "4'=widowed," "5=live with as married," "6=stopped living with as married," and "7=single, never married."








47


This variable has been reconstructed to indicate whether or not the daughter is married, living as married, or not married. There were no women among the World War II Generation daughters during midlife who were living together as married. For the multivariate analyses, the reference group for this variable is "not married."

Age at first marriage. This variable reflects the timing of marriage for the

daughters. It is not reflective of the actual age of the daughters at first marriage, but whether or not the daughter married early or married later, relative to the general social trends of the time. Because not all of the women were married, simply using the age created several missing cases, further reducing the sample size. Therefore, this variable is a three category variable including "not married," "married earlier," and "married later." For the multivariate analyses, the reference group "not married." Age breakdowns for early and late were based on the average age of first marriage at the time of marrying age for these women. For the Baby Boom Generation and Generation X women during young adulthood, this was at the time of data collection, 1971 and 1994 respectively. For the World War II Generation and Baby Boom Generation During Midlife, this was the average age at first marriage when they would have been young adults, 1971 for the Baby Boom Generation women, and 1948 for the World War II Generation women. Among those Baby Boom Generation daughters who were married during young adulthood, none were married later. Thus the reference group for the young adult analyses is comprised of those daughters who married earlier, allowing for comparison of Generation X daughters who married early versus late in 1994, as well as those who were not married to those who married early. But for Baby Boom young adult daughters, the comparison is basically between those who were married (early) versus







48


those who were not married. In the midlife analyses the reference group is comprised of those daughters who married later. Because there were no World War II Generation women who had never married, the comparison is among the married women, between those who married earlier versus those who married later.

Children. This variable indicates whether or not the daughter has children at the time of LSOG data collection, thus reflecting a current status for this family measure. It is a dichotomous, yes/no variable. The reference group for this variable is "has children."

Age at first child. This variable reflects the timing of childbearing for the

daughters who have children. It is not reflective of their exact age at first child, but rather did they have their first child earlier or later relative to the general social trends at that time. Because not all of the women had children, simply using the age created several missing cases, further reducing the sample size. Therefore, this was recoded as a nominal variable, including "no children" "had first child earlier," and "had first child later." The reference group for this variable is "does not have children." For the World War II Generation women during midlife, age at first child is not available, therefore, the women without children are compared to women who have children. Age breakdowns for early and late were based on the average age at first child during young adulthood for these women. For the Baby Boom Generation and Generation X women during young adulthood, this was at the time of data collection, 1971 and 1994 respectively. For the World War II Generation and Baby Boom Generation During Midlife, this was the average age at first child when they would have been young adults, 1971 again for the Baby Boom Generation women, and 1948 for the World War II Generation women. For the young adult analyses, the reference group is comprised of those daughters who do not








49


have children. In the midlife analyses, there is no distinction between timing of first child because there were no calculations available for the World War II Generation daughters in 1971. Therefore the reference group in this analysis is comprised of those women who have children, comparing them only to those who do not. Social Distance Measures

For the Generation X daughters during young adulthood, and the Baby Boom

Generation daughters during midlife, additional analyses include examining the effects of social distance between these daughters and their mothers. The presumption here is that the distance between the daughter and their mothers has a unique effect, and not only the daughters' absolute education, work or family status. Again, due to the small samples of women used in this research, the models for the social distance measures are interaction effects between the daughters' characteristics and their mothers' characteristics. The variables used in these analyses are as follows.

Education level. The social distance measure for education level measures an interaction effect between two-two category variables, daughters' education and her mothers' education. This created four categories, both had a higher level of education, both had a lower level of education, the daughter had more education, and the mother had more education. For the young adult analysis, a three category variable was created, including "daughter has more education than her mother," "mother has more education that the daughter," and "both the daughter and mother have the same level of education. The reference group for the young adult analyses examining the effects of social distance on Generation X daughters' perceptions of solidarity during young adulthood is "daughter has more education than her mother." For the midlife analysis, a four category variable was created, including "daughter has more education than her mother," "mother








50


has more education than the daughter."" both the daughter and her mother have a low level of education," and "both the daughter and her mother have a high level of education." The reference group for the midlife analyses examining the effects of social distance on Baby Boom daughters' perceptions of solidarity during midlife is "daughter has more education than her mother."

Work status and SEI score. This measure combines the employment status and SEI score for occupation. After running a crosstabulation to examine the nine possible categories of women emerged. Five were chosen for both the young adult and midlife analyses. These are, "daughter works and has a high sei score and her mother does not work," "mother works and has a high score, daughter does not work and therefore has no score," "both mother and daughter work and have low SEI scores," "both have no score because neither work," and the final category includes "all others." The reference group for this measure for the multivariate analyses is "daughter works, high score, mother does not work." The reference group for both the Generation X daughters in young adulthood, and the Baby Boom daughters during midlife is "daughter works and has a high score, mother does not work and therefore has no score."

Income level. When measuring the social distance between the daughters and their mothers with regard to income two-two category dummy variables were created. Using the median score for income for all women in the LSOG study in 1971 and 1994 as a cut off point, those women who fell below this level of income were considered having lower income, and those women who fell at or above this level were considered to have higher income. This created four categories of women, one in which both daughter and mother had high levels of income, one in which they both had low levels of income, one in which








51


the daughter had a higher level of income than her mother, and one in which the mother had a higher level of income than her daughter. For both the young adult and the midlife analyses, a three category variable was created, and include "daughter's income is high while the mother's is low," "mother's income is high while the daughter's is low," and "both daughter and mother have the same income level." The reference group for both the Generation X daughters during young adulthood and the Baby Boom Generation daughters during midlife is "daughter's income is high, while the mother's income is low.

Marital status and timing of marriage. For this social distance measure, the marital status and timing of marriage were combined. After running a crosstabulation to examine the numbers of women in each of the nine possible categories, four were selected to examine social distance among daughters and their mothers for both the young adult and midlife analyses. They are "daughter is single/never married, mother married early," "daughter married later, mother married early." "both daughter and mother married early," and "all others." The reference group for this variable for both the Generation X daughters during young adulthood and the Baby Boom daughters during midlife is "both daughter and mother married early."

Children. This measure compares daughters to their mothers with regard to who has children. Because only those mothers included were the ones matched to their daughters, thus having children of their own, there are two categories of women included, "both have children," "mother has children, daughter does not." The reference group for both the Generation X daughters during young adulthood and the Baby Boom daughters during midlife is "both have children."








52


Analytic Model

In order to evaluate the research questions and hypotheses previously mentioned, the analysis in this study encompasses several stages. Each stage is described in detail to follow, as well as a brief description of the analytic chapters. As previously illustrated in Figure 4.1, each generation of women experienced the social and structural shifts of the twentieth century during different ages and life course stages. As a result, intergenerational differences among women exist. Figure 4.2 illustrates the conceptual framework for this study.

In order to examine the first research question: "How have daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during young adulthood changed, and what may be associated with this change?" the analyses will begin with a comparison of means across the samples of daughters during young adulthood using the T-test statistic. Perceptions of the Generation X daughters in 1994 will be compared to the perceptions of the Baby Boom daughters in 1971. In addition, OLS regression models will be with the daughter characteristics and social distance measures as independent variables. This analysis will help to understand how social distance between mothers and daughters might affect daughters' perceptions of associational, affectual, consensual solidarity, and how this might change over time.

The same analyses will be done to answer the second research question: "How have daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during midlife changed, and what may be associated with this change'?" Perceptions of the Baby Boom daughters in 1994 will be compared to the perceptions of World War II daughters in 1971.

Following are the statistical models for each set of OLS regression analyses:







53


Model 1: ISASAFCO = N + P1 (age) + P2 (education) + e
Model 2: ISASAFCO = PO + P1 (age) + P2 (work indicators) + e
Model 3: ISASAF,CO =P + P1 (age) + P2 (family indicators) + e
Model 4: ISASAF,CO = P + P1 (age) + P2 (education) + 33 (work indicators) + p4 (family indicators) + e

Where IS represents intergenerational solidarity, sub AS, AF, and CO represent the three measures of solidarity, associational, affectual, and consensual, and indicates that these models are the same for each analysis on these dependent variables. Age is included as a control variable in all four models. Education is included in the first model. The work characteristics included in the second model are work status, SEI for occupation, and income. The family characteristics comprising the third model are marital status, age at first marriage, whether or not the daughter has children, and age at first child. The fourth model includes all three subsets of variables, education, work and family.

For both the young adult and midlife analyses, social difference scores are

examined for one of each of the generations of women in of the analytic chapters. These analyses compare the younger generation of women at each time period to their mothers. Included in the first analytic chapter, are the Generation X daughters during young adulthood and the social distance in relation to their Baby Boom Generation mothers when they were young adults. The social distance analyses included in the second analytic chapter. are Baby Boom Generation daughters during midlife and the social distance in relation to their World War II Generation mothers when they were in midlife.

Two sets of analyses are carried out in each of the analytic chapters. The first set of analyses illustrates the univariate statistics describing the social distance variables for the daughters and their mothers. The second set of analyses in each of the two chapters includes multivariate analyses using OLS regression. These analyses examine the effects







54


of the distances between the daughters and their mothers on the daughters' perceptions of associational, affectual, and consensual solidarity during each of the two life stages young adulthood for the Generation X daughters in 1994, and midlife for the Baby Boom Generation daughters in 1971. The following are the statistical models for each set of OLS regression analyses:

Model 1: ISASAFCO =PO + P1 (education) + P2 (work indicators) + P3 (family indicators) + e

Sample Limitations

While the LSOG data is ideal for this research project in many ways, there are limitations with regard to this research. The study sample is region specific, and the participants are relatively racially and economically homogenous. In addition, as a result of sampling restrictions, the sample may be slightly more advantaged in terms of more positive familial relations. Because successive generations are only considered as participants based on the participation of a prior generation, those descendents who are not in close contact with their family members may have self-selected out of participation in the Longitudinal Study of Generations.

Another limitation of this study is the sample size. The original Longitudinal Study of Generations sample was for the most part reduced by half for this proposed research, to include only women (N=1,071 at Time 1). Furthermore, once the dyads are created based on daughters, who have mothers participating in the study, and attrition is accounted for, the sample size for the analyses gets further reduced. Having a larger sample to begin with would have been beneficial for this proposed research.

The inability to include the mother's perceptions of solidarity with their daughters is also a limitation. Comparing these perceptions would have been beneficial to








55


understanding these mother daughter dyads. This was not possible with these data because not every generation of mothers were asked the same questions, nor were these questions asked at each time period.








56


Table 4. 1. Numbers of Generation X and Baby Boom Generation Daughters during
Young Adulthood at Each Stage of Sample Selection.
Generation X Baby Boom Daughters Daughters (G4's 1994) (G3's 1971)

Original Sample of Women 200 442

Selected Daughters Age 18-29 136 185

Removed Daughters with No Responses on ANY of the Dependent 136 179 Variables

Randomy Selected One Daughter 88 135 from Each Family

Matched Each Daughter with a 82 124 Mother in the Study



Table 4.2. Numbers of Baby Boom Generation and World War II Generation Daughters
during Midlife at Each Stage of Sample Selection.
Baby Boom World War II Daughters Daughters (G3's 1994) (G2's 1971)

Original Sample of Women 410 377

Selected Daughters Age 40-59 230 259

Removed Daughters with No Responses on ANY of the Dependent 227 209 Variables

Randomly Selected One Daughter 150 182 from Each Family

Matched Each Daughter with a 108 143 Mother in the Study















CHAPTER 5
GENERATIONAL DIFFERENCES IN DAUGHTERS' PERCEPTIONS OF
ASSOCIATIONAL, AFFECTUAL, AND CONSENSUAL SOLIDARITY IN YOUNG ADULTHOOD

Introduction

The purpose of this chapter is to examine generational differences in daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during the young adult life stage. This chapter specifically explores the role of education, work, and family in influencing young adult daughters' perceptions of solidarity with their mothers. Two generations are compared young adult women from Generation X in 1994 and young adult women from the Baby Boom Generation in 1971.

This chapter presents the results of several sets of analyses bivariate and multivariate. Descriptive sample characteristics are presented first (a bivariate comparison of two generations) for each generation in young adult. Education, work and family characteristics of each generation are calculated. Tests for differences across the two generations when they were young adults are calculated as well. The second set of analyses calculates levels of associational, affectual, and consensual solidarity for each cohort when in young adulthood. Again, tests for differences in reported solidarity with mothers across the two generations when they were young adults are calculated as well. The next set of analyses included in this chapter are multivaniate models estimating the effects of the independent variables on the three measures of solidarity. We employ a nested modeling strategy where more exogenous variables are entered into the model first (education) and subsequent models explore more proximate factors (work and family


57








58


characteristics). The first model includes age, as a control variable, and one's level of education. The second model includes the work related variables of employment, socioeconomic status for occupation, and income. The third model includes those variables related to family marital status, age at first marriage, and age at first child. The fourth and final model includes all of the variables education, work and family. Thus, we approximate the contribution each set of variables has in explaining levels of solidarity as each set is added to the model.

The final set models are OLS regression models estimated for Generation X

daughters examining the effects of social distance (as compared with their mothers) on perceptions of daughters' solidarity. For example, we examine whether being married later in life than one's mother was married decreases daughters' perceptions of solidarity with their mother. We do not estimate a comparable set of models for the Baby Boom daughters as the data do permit social distance measures to be constructed as there are no data on the mothers of this generation.

Descriptive Statistics-Independent Variables

Table 4.1 includes the descriptive statistics for the demographic characteristics of Generation X and Baby Boom daughters during young adulthood. The quantitative variables are represented by the mean and the standard deviation, while the qualitative variables are represented by the percentage of women, accounting for each of the variable's attributes. T-tests were done to compare the women from each generation during young adulthood across the quantitative variables, and chi square was used for comparison of the qualitative variables.







59


Results

Table 5. 1 presents the descriptive statistics for the independent, sociodemographic variables that comprise the daughter characteristics for each generation of women. There are 82 Generation X daughters and 124 Baby Boom Generation daughters included in the sample of young adults. Results suggest that there are significant differences with regard to sociodemographic characteristics pertaining to education, work and family between the Generation X daughters who were young adults in 1994, and the Baby Boom Generation daughters who were young adults in 1971. Although the age range was the same for both generations (18 to 29), the sample of Generation X daughters are older (mean age = 21.66) than the sample of Baby Boom daughters (mean age = 20.48) (p< .05). With regard to education, there is also a statistically significant difference between the two generations of daughters (p<.01). More Generation X daughters in 1994 had moved beyond a high school degree, 72 %, compared to the Baby Boom Generation daughters in 1971, where only 60% had more than a high school education, and 40% had a high school degree or less.

Turning next to work related characteristics, results show that there are differences across the generations. Fewer women from the Generation X sample of daughters worked in 1994 (64.6 percent) as compared to daughters from the Baby Boom Generation in 1971 (66.1 percent) (p<.05). However, there were more Generation X daughters in 1994 with lower SEI scores for occupation (p<.05). Also, a difference in household income between the two sets of young adult daughters is found (p<.05). The average income level for Generation X daughters in 1994 is much lower, $38,876, than for the Baby Boom Generation in 1971, where the average income was $53,203.








60


The two generations also differ with respect to family circumstances in young

adulthood. For example, in 1994, 26.8% of the Generation X daughters were married as compared to 33.3% of the Baby Boom daughters in 1971. However, the biggest difference between the two generations with regard to marital status is the number of women who were cohabiting in 1994 among the Generation X sample, 20.7%. In 1971, only 3.4% of the Baby Boom sample reported living together. The percentage of not married women was 52.4% of Generation X daughters, and 63.3% of Baby Boom daughters. These differences in marital status between the two samples of young adult women are statistically significant (p<.001 ). In addition, the timing of marriage for the two generations of young women differed (p<. 01). Among Generation X young adult daughters, 19.5% married earlier, and 3.7% married later. In 1971, all of the Baby Boom daughters who had ever been married, 32.5%, married earlier, and none married later. Interestingly, more Generation X daughters had children during young adulthood in 1994 (p<.05). Among the Generation X daughters, 19.5% had children, compared to 13.7% of the Baby Boom Generation daughters in 1971. For the Generation X daughters in 1994, 80.5% had no children, however, 17.1% of these women had children at an earlier age, while 2.4% had children later.

Descriptive Statistics and Mean Differences-Dependent Variables

This section focuses on the descriptive statistics and the mean differences for the three measures of solidarity comprising the dependent variables for each generation of women during young adulthood. In addition to presenting the results of each measure of solidarity, the items used to comprise these measures have also been presented. Mean difference scores using ANOVA were run to determine significant differences in the means for each generation of women. There are some significant differences in the








61


perceptions of solidarity for the Generation X daughters who were young adults in 1994 compared to the Baby Boom Generation daughters who were young adults in 1971. Results-Daughters' Perceptions of Associational Solidarity During Young Adulthood

Table 5.2 illustrates the results of the levels of perceptions of associational

solidarity for each generation of daughters with their mothers during young adulthood. Each measure of solidarity is represented by the mean and standard deviation. T-tests were done to compare the women from each generation. Associatonal solidarity overall, has a range of 3 to 24. Generation X daughters in 1994 have higher perceptions of associational solidarity (mean=13.52) than the Baby Boom daughters in 1971 (mean= 12.42). This is a significant difference in overall perceptions of associational solidarity with their mothers for each generation of women (p<.05). The three items comprising the measure of associational solidarity each have a range of I to 6. Of these items. only one emerges as significantly different for the Generation X and Baby Boom Generation daughters (p<.01). This is the frequency of interaction on the telephone between the daughters and their mothers during a period of a year. Generation X daughters in 1994 have more contact with their mothers (mean=6.19) compared to the Baby Boom Generation daughters in 1971 (mean= 4.63). While the other two items are not significant, it is important to note that the frequency of interaction with their mothers in person during the year decreased for the Generation X daughters in comparison to the Baby Boom daughters (mean=5.28 and 5.71 respectively), where as the frequency of interaction by mail increased (mean=2.50 and 2.02 respectively). While these findings are not significant, they do indicate a trend that would suggest the phone would emerge as a more viable means of interaction for these women. And overall, associational solidarity has increased for the successive generation of women over time.







62


Results-Daughters' Perceptions of Affectual Solidarity During Young Adulthood

The next measure of solidarity, affectual solidarity, is presented Table 5.3. This measure of solidarity has a range of 5 to 30. The overall mean for perceptions of this measure of solidarity increased for the Generation X daughters in 1994 (mean=22.58) compared to the Baby Boom daughters in 1971 (mean=21.90). However, this difference is not significant between these generations of women. Each of the five items comprising this measure have a range of I to 6. When looking at the items in this measure of solidarity, two items emerge as being significantly different between the two generations of young adult women. There is a significant decrease in the daughters' perception of whether or not the daughter feels that her mother understands her (p<.05). For the Generation X daughters in 1994, perceptions decrease (mean=4.00) compared to the Baby Boom Generation daughters' perceptions in 1971 (mean= 4.34). There is also a significant difference in whether or not the daughter feels she understands her mother at this stage (p<. 10). Generation X daughters as young adults have lower perceptions (mean=4.3 1) compared to the Baby Boom Generation daughters (mean= 4.49). Interestingly, while these two variables indicate the perception of some level of understanding between mother and daughter, and both have decreased, each of the other items in this measure closeness, communication, and whether they get along have all increased for the successive generation of women. However, these increases are not statistically significant.

Results-Daughters' Perceptions of Consensual Solidarity During Young Adulthood

The final bivariate table, Table 5.4, presents the findings for the measure of

consensual solidarity. This is a single item measure, with a range of 1 to 6. There is a significant difference in the perceptions of the two generations of daughters with regard








63


to the similarity of world views with their mothers (p<.05). The Generation X daughters in 1994 perceive higher levels of consensual solidarity with their mothers as young adults (mean=3.89) than did the Baby Boom Generation of daughters in 1971 (mean= 1.55).

Multivariate Analyses

This section analyzes the multivariate models, which explore the effects of the education, work, and family measures on each generation of daughters' perceptions of associational. affectual, consensual solidarity with their mothers during young adulthood. Each of the tables includes four models for each of the generation of daughters. Daughter's age is added as a control variable in each of the models. Model 1 includes the education measure, comparing women who have more than a high school education to those with the equivalent of a high school degree or less. Model 2 includes those variables pertaining to work life. These are 1) employment status, which compares women who work to women who do not work, 2) SEI score for occupation, which compares women who do not work and have no score, and women who have a high score to women with lower SEI scores, and 3) income level. Model 3 is the family model which includes 1) marital status, comparing married women to women who are not married and women who are living together, 2) age at first marriage, which compares women who married early to women who are not married and women who married later, and 3) age at first child, which compares women who have no children to women who had their first child early and women who had their first child later. The final model, Model 4, includes all of the covariates from the three domains.







64


Results-Effects of Daughter Characteristics on Their Perceptions of Associational
Solidarity

The first multivariate table (Table 5.5) shows results of the four models, which illustrate the effects of the daughter characteristics on their perceptions of associational solidarity with their mothers during young adulthood. Model 1, which includes the daughter's age and education, yields no significant findings. Furthermore, almost no variability is explained by this model, about 1% for each of the two generations of young adult daughters.

The work measures in Model 2 present several significant findings. First, this model is significant for both the Generation X daughters in 1994 and the Baby Boom Generation daughters in 1971 (p<.10 and p<.05. respectively). In addition, this model explains 14% of the variability in perceptions of associational solidarity for the Generation X daughters, and 11% of the variability for the Baby Boom Generation daughters. For the Generation X daughters in 1994, having a higher SEI score is negatively associated with the daughter's perception of associational solidarity during young adulthood when compared to women with low SEI scores (p<. 10). This is also the case for Baby Boom Generation daughters during young adulthood in 1971 (p<.05). Income also emerges as having a significant effect on daughters' perceptions of associational solidarity with their mothers during young adulthood. For Generation X daughters, a higher level of income is associated with more positive perceptions of associational solidarity (p<. 01). For Baby Boom Generation daughters, the association is negative. Among these women in 1971, a higher level of income was associated with lower levels of associational solidarity with their mothers (p<. 01).







65


Model 3, which includes the family characteristics of the daughters, presents no

significant findings for either generation of daughters. None of the family measures have a significant effect on daughters' perceptions of associational solidarity during young adulthood. However, this model does explain 15% of the variability in perceptions of associational solidarity for Generation X daughters in 1994, and is approaching significance (p<. 10). In addition, in this model, when only the family characteristics are added to age and education, the effect of daughter's age on this measure of solidarity becomes a positively correlated effect. The older the daughter is within the young adult sample, the more positive are her perceptions of association with her mother.

The final model in Table 5.5, Model 4, includes age, and all of the education, work and family variables. This model yields few significant results. With regards to levels of association with their mothers during young adulthood, For the Generation X daughters, this model explains 21% of the variability in this dependent variable. However, neither the model, nor any of the covariates emerge as significant. For Baby Boom Generation daughters, 12% of the variability in perceptions of associational solidarity is explained, and the model is approaching significance (p<. 10). In addition, having a high SEI score, and income emerge as having significant effects on level of association when all of the variables are included in the model. For Baby Boom daughters in 1971, having a higher SEI score is associated with lower levels of association with their mothers compared to women with lower SEI scores (p<.05). For these daughters, income is also negatively associated with associational solidarity (p<.Ol). It is also interesting to note, for the Baby Boom Generation, across the models the relationship between age and perceptions of associational solidarity emerged as a positive correlation. However, when all of the








66


variables are included in the model, this relationship becomes negative. The older the daughter in this young adult sample, the more negative are the perceptions of associational solidarity with their mother.

Results-Effects of Daughter Characteristics on their Perceptions of Affectual Solidarity

Table 5.6 includes the results of all four models on the daughters' levels of

perception of affectional solidarity with their mothers as young adults. The same models are included here as in the previous table. Model 1, which includes daughter's age and education, yields nothing significant. Age and education are not significantly related to perceptions of affectual solidarity for either of these generations during young adulthood. Furthermore, this model is not significant for either generation of women, explaining little variability in this dependent variable.

When the work characteristics are introduced in Model 2, only SEI score for occupation for the Baby Boom Generation daughters in 1971 appears to have a significant effect (p<. 10). For these daughters during young adulthood, having a higher SEI score for occupation is indicative of lower perceptions so affectual solidarity with their mothers during young adulthood compared to those daughters with a low score. Although not statistically significant, for the daughters in Generation X, the relationship between SEI and this measure of solidarity is a positive one. This model is also statistically significant for the Baby Boom Generation daughters (p<. 10), yet only explains about 6% of the variability in the dependent variable. For the Generation X daughters in 1994, about 7% of the variability is explained, and this model is not significant.

When the family characteristics are included in Model 3, little significant value is found. For the Generation X daughters during young adulthood in 1994, less than a high








67


school degree is associated with lower levels of perceptions of affectual solidarity with their mothers compared to those daughters with more than a high school degree (p<. 10). While not statistically significant, with the inclusion of the family variables, it is also important to note that the relationship between daughter's age and affect becomes negative. while it had been positive when looking at education and the work characteristics in the previous two models. This model explains 9% of the variability in the dependent variable for Generation X daughters in 1994, and only 2% for Baby Boom daughters in 1971.

When all of the independent variables are included in Model 4, the relationship between age and affect remains negative for the Generation X daughters during young adulthood in 1994. However, daughter's income is the only covariate that emerges as having a significant effect on affectual solidarity for these women (p<.05). For these daughters, having a higher income level is associated with more positive perceptions of affectual solidarity with their mothers. Interestingly, for the Baby Boom Generation daughters in 1971. the relationship between income and affect is also significant, however, this relationship is negative (p<. 10). In addition, for the Baby Boom daughters in 1971, SEI score for income is also significant (p<. 10). A negative association exists indicating that if the daughter's SEI score is higher. the perceptions of afectual solidarity with her mother during young adulthood are more negative. For the Generation X daughters, this full model explains 16% of the variability in affect, and 8% for Baby Boom Generation daughters.

Results-Effects of Daughter Characteristics on their Perceptions of Consensual Solidarity

Turning attention to Table 5.7, findings from the four models on perceptions of consensual solidarity for the daughters with regard to the daughter characteristics are







68


presented. Model 1 presents significant results in terms of age and education for young adult Baby Boom daughters. Baby Boom daughters in 1971 have higher perceptions of consensual solidarity with their mothers if they are older (p<.05). Age does not have a significant effect on consensus for Generation X daughters in 1994. For Baby Boom Generation women, education also has a significant affect on levels of consensus during young adulthood (p<.05). Women with less than or equal to a high school degree report more similar world views with their mothers compared to those young women who have more education during young adulthood. While not significant. this relationship for the Generation X daughters in 1994 is negative. Model 1 explains almost none of the variability in the dependent variable, 1%, for the Generation X daughters, while explaining 6% for the Baby Boom Generation daughters. For the Baby Boom Generation sample in 1971, this model is significant (p<.05).

Looking at model 2, there are significant findings among the work measures. For the Baby Boom Generation daughters in 1971, age is again significantly associated with consensual solidarity, in that as the daughters age, their perceptions of consensus with their mothers become more similar (p<. 10). With the inclusion of the work measures, education is no longer significant for these daughters. Employment status for Baby Boom Generation daughters during young adulthood in 1971 is significant (p<. 10). For these women, a high SEI score is associated with less similar world views with their mothers compared to those daughters who have a lower SEI score. In addition, for the Baby Boom Generation women a higher level of income has a significant negative association with consensus (p<.05). For the Generation X daughters in 1994, income is also significantly associated with perceptions of consensual solidarity (p<.10). However,








69


for these young adult daughters, a higher income level means more similarity of world views. Model 2 explains 5% of the variability in the dependent variable for the Generation X daughters in 1994, and 9% of the variability for the Baby Boom Generation daughters in 1971. This model is significant for the Baby Boom daughters (p<. 05).

Model 3 includes the family characteristics in the model with age and education. Several significant findings in association with perceptions of consensual solidarity emerge in this model. For Generation X daughters, the inclusion of the family covariates alter the relationship between age and consensus. In the previous two models, this relationship was positive, so that as the daughters got older, their world views became more similar to those of their mothers. This model suggests that this relationship is a negative relationship, although not a significant one. For Baby Boom Generation daughters during young adulthood, being older continues to be significantly associated with more similar world views with their mothers (p<.05). Education again emerges as significant with the family measures added (p<.05). Having less than or equal to a high school education for the Baby Boom Generation daughters in 1971 means more similar world views with their mothers compared to those daughters with more than a high school education. Finally, those daughters who were not married during young adulthood from the Baby Boom sample in 1971 reported more positive perceptions of consensual solidarity with their mothers compared to those women who were married (p<. 10). This model explains 8% of the variability in perceptions of consensual solidarity with ones mother for both generations of women, however the model is only significant in explaining the relationship for the Baby Boom Generation daughters during in 1971 (p<. 05).







70


Turning to Model 4, inclusive of all daughter characteristics, significant findings are present. During young adulthood, age continues to be significant for Baby Boom Generation daughters in 1971 in the same way as the previous models (p<.05). A lower education also continues to be significantly associated with more similar world views with their mothers for these daughters when compared to the daughters with higher education (p<.05). Baby Boom Generation daughters with high SEI scores also reported significantly lower levels of consensus with their mothers during young adulthood than did those daughters with low SEI scores (p<. 10). Income was also negatively associated with perceptions of consensual solidarity for Baby Boom daughters (p<.05). In 1971, daughters with higher levels of income reported less similarity with their mothers in terms of world views. With all of the covariates included in this model, age at first marriage remained significant for the Baby Boom daughters in 1971 (p<. 10). Those women who were not married reported significantly more positive perceptions of consensual solidarity with their mothers than their married counterparts during young adulthood. While Model 3 explains 13% of the variability in consensual solidarity for both generations of women during young adulthood, again, the model is only significant for the Baby Boom Generation daughters in 1971 (p<.05). Social Difference Analyses for Generation X Daughters and their Baby Boom Mothers During Young Adulthood.

This section is targeted toward analyzing the effects of the differences between

daughters and their mothers on the daughters' perceptions of associational, affectual, and consensual solidarity. By comparing the daughters' sociodemographic characteristics in terms of education, work and family, to their mothers, several difference measures were created. Descriptive characteristics are examined first, for differences in education, work







71


status and SEI score for occupation, income level, marital status and timing of marriage, and children. Second, OLS regression analyses are run examining the effects of the difference measures on the three measures of solidarity. Results-Descriptive Statistics for the Difference Measures Comprising the Independent
Variables

Table 5.8 includes the descriptive statistics illustrating the difference measures for Generation X daughters during young adulthood and their Baby Boom Mothers during young adulthood. The three category, education variable reveals that 31% of the Generation X daughters as young adults in 1994 had more education that their Baby Boom mothers did when they were young adults. Almost 14% of the daughters had less education than their mothers. Just over half, 55%, of the Generation X daughters and their mothers have the same level of education, either both low, or both high.

Work status and SEI score for occupation were combined. Five categories were examined. About 12% of the daughters had jobs with high SEI scores while their mothers had no jobs and therefore no scores. There are just over 10% of the dyads where the mothers work and have high SEI scores and the daughters do not work and have no score. Approximately 11% of the dyads are those where both daughters and mothers work and have low SEI scores. Just over 24% of the mother/daughter dyads include those women where both daughters and mothers have no SEI scores because they don't work. Finally, 43% of the dyads are characterized by all of the other categories of SEI score and work status combinations.

Three categories of income differences are examined. For those daughters whose income levels are high and their mother's income is low, about 22% of dyads exist. Almost 26% of the dyads are characterized by a mother with a higher income level and a







72


daughter with a low income level. Finally, approximately 52% of the dyads include daughters and mothers with the same income levels.

Current marital status and timing of marriage were combined and three categories of dads are compared. About 15% of the dyads include those daughters and mothers who are both married, and married early. Daughters who never married and had mothers who married early comprise approximately 49% of the dyads. Those dyads which contain all other categories of marital status and marital timing combinations, make up the remaining 35% of the pairs of mothers and daughters.

With regard to having children, since the mothers are paired with the daughters already, there is no possibility that the mother does not have children. Thus, there are only two categories represented here. Almost 20% of the dads are those where both daughter and mother have children during young adulthood. The remaining 80% are those mother/daughter pairs where the daughter does not have children of her own. Results-Effects of Social Difference Measures on Generation X daughters' Perceptions
of Associational, Affectual and Consensual Solidarity with their Baby Boom
Generation Mothers During Young Adulthood

Table 5.9 illustrates the regression analyses used to explore the social difference variables between the young adult daughters of Generation X with their Baby Boom Generation mothers. All of the variables are included in the model together for associational, affectual, and consensual solidarity. Turning attention to the effects of these difference scores on associational solidarity, several of the difference measures emerge as significant. This model explains 32% of the variability in associational solidarity during young adulthood in 1994 for the Generation X daughters (p<.10). With regard to education, those dyads where the daughters and their mothers had the same level of education during young adulthood had lower perceptions of association with their







73


mothers in 1994 compared to those in which the daughter had more education than the mother (p<. 10).

Several of the marital status and timing of marriage covariates appear to have significant effects on Generation X daughters' perceptions of associational solidarity during young adulthood. When compared to daughters who work and have high SEI scores, and the mother does not work and has no score, those dyads where both worked and had low scores had lower levels of association with their mothers (p<. 10). Those mother/daughter dyads where neither women worked, and therefore had no score had higher levels of association with their mothers during young adulthood in 1994 compared to the reference group (p<. 10).

Moving on to affectual solidarity, with all of the covariates included in the model, about 25% of the variability in Generation X daughters perceptions of affectual solidarity is explained here. This model is not significant. With regard to this measure of solidarity. education differences do not emerge as significant as they had with associational solidarity. Moving on to work status and SEI score, those dyads where the mother works and had a high SEI score, while the daughter did not work and had no score had significantly lower levels of affect when compared to those dyads where the daughters worked and had high SEI scores and the mothers did not work and had no scores (p<.05). Income differences also emerged as significant in this model. For those dyads where the income levels were the same for both the daughter and her mother, higher levels of affectual solidarity were expressed by Generation X daughters during young adulthood in 1994 when compared to those dyads in which the daughters had higher income levels than did their mothers.







74


Finally, looking at consensual solidarity, approximately 27% of the variability in this dependent variable is explained by this model (p<. 10). As was the case for affectual solidarity, dyads where the mothers worked and had high SEI scores and the daughters did not work and had no score had lower levels of consensus with their mothers than did the reference group (p<.05). The only other significant effect on Generation X daughters' perceptions of consensual solidarity during young adulthood in1994 is a marital status and marital timing difference measure. When the daughter is single while her mother married early, lower levels of consensual solidarity exist compared to mother/daughter dyads where both daughter and mother married early (p<. 10).

Conclusions

In order to test hypotheses HIA through Hlc, four steps were taken in this analytic chapter. The first step of the analysis of the daughters during young adulthood, a description of their sociodemographic characteristics (education, work and family) was provided. In the second step of the analysis. a description of the dependent variables associational, affectual, and consensual solidarity was provided. At this time, mean differences between the Generation X daughters and the Baby Boom Generation daughters during young adulthood were presented. Third, the multivariate analyses testing the relationship between daughter characteristics on their perceptions of associational, affectual, and consensual solidarity during young adulthood were run. Finally, the multivariate analyses testing the effects of social distance between the mother daughter dads. which included Generation X daughters and their Baby Boom Generation mothers during young adulthood, on the daughters' perceptions of associational, affectual, and consensual solidarity. Support is provided for each of these three hypotheses.







75


To review, the hypotheses tested in Chapter 5 look at the effects of daughter characteristics and social distance on daughters' perceptions of solidarity with their mothers during young adulthood. The first hypothesis, H11\ looks specifically at the differences between the two generations of women during young adulthood, Generation X daughters in 1994 and Baby Boom Generation daughters in 1971. The second hypothesis, H1B focuses on the effects of the daughters' characteristics, those sociodemographic characteristics pertaining to education, work, and family, on the daughters' perceptions of solidarity with their mothers. With this hypothesis, a distinction is made setting affectual solidarity apart from associational and consensual solidarity. Finally, the third hypothesis, Hlc. focuses on the effects of social distance between the Generation X daughters during young adulthood and their mothers from the Baby Boom Generation on the daughters' perceptions of solidarity with their mothers. Again, a distinction is made here which suggests that affectual solidarity will not be affected in the same ways as associational and consensual solidarity.

Support for H l A can be seen in Tables 5.2 through 5.4. Young adult daughters

from Generation X in 1994 have stronger (more positive) perceptions of associational and consensual solidarity than the Baby Boom Generation daughters in 1971. In Table 5.2, levels of associational solidarity were presented. Overall, associational solidarity was greater for the Generation X daughters in 1994. In terms of each of the individual measures, it was also the case that there was an increase in association for the successive generation. However, the significant difference was found in the increase of association with their mothers by phone.







76


Support for HIe can be seen in Tables 5.5 through 5.7. The effects of daughter

characteristics on perceptions do not affect affectual solidarity in the same way as they do associational and consensual solidarity. Looking at Table 5.5, the effects of daughter characteristics on associational solidarity during young adulthood, Model 2 proves to be significant for both generations of daughters. This model includes the work characteristics, and both SEI for occupation and income are significant for both generations of women. Model 3, which includes the family characteristics, is significant for the Generation X daughters. Finally, the full model, Model 4, is significant for the Baby Boom daughters, and SEI and income again emerge as significant.

Skipping ahead to Table 5.7, looking at the effects of daughter characteristics on perceptions of consensual solidarity with their mothers during young adulthood, several significant findings emerge as well. Each of the four models is significant for the Baby Boom Generation daughters. Model 1 indicates that age and education are in fact associated with levels of consensus. Model 2, which incorporates the work characteristics, suggests that for this generation a high SEI score, and a higher level of income are also associated with consensual solidarity. In Model 3, not being married emerges as significant. Finally, in the full model, age and education remain significant, as well as not being married.

Turning attention now to Table 5.6, the effects of daughter characteristics on

perceptions of affectual solidarity during young adulthood, it can be seen that none of the models show that daughter characteristics are associated with perceptions of affect, with the exception of Model 2 for the Baby Boom Generation approaching significance. Thus, affectual solidarity for daughters during young adulthood is not affected in the same way








77


that associational or consensual solidarity. Furthermore, the fact that these four models in each table reveal more significant associations for the Baby Boom daughters during young adulthood continues to lend support for the first hypothesis, which suggests that there are more negative perceptions of associational and consensual solidarity for this generation of women during young adulthood.

Support for Hic can be seen in Table 5.9. Social distance between Generation X daughters and their Baby Boom mothers during young adulthood is associated with associational and consensual solidarity, but not with affectual solidarity. Looking at Table 5.9, the models for associational and consensual solidarity are significant, suggesting that these social distance measures sufficiently explain variability in these measures of solidarity during young adulthood. However, this is not the case for affectual solidarity.








78


Table 5. 1. Descriptive Statistics-Demographic Characteristics of Generation X
Daughters and Baby Boom Daughters in Young Adulthood.


Generation X Daughters
Young Adulthood
1994 N=82


Baby Boom Daughters
Young Adulthood
1971
N=124


x SD % x SD % Sig.

DAUGHTER
CHARACTERISTICS

Age (mean) 21.66 2.96 20.48 2.30 *

Education Level (%) **
High School Degree or Less 28.0 40.0 More than High School 72.0 60.0

Work Status (%) *
Works 64.6 66.1 Does Not Work 35.4 33.9

SEI (Occupation) (%) *
No Job. No Score 35.4 37.5 SEI Score is Low 45.1 23.3 SEI Score is High 19.5 39.2

Income-1994 Dollars (mean) 38,876 25,601 53,203 36,085 *

Current Marital Status (%)
Married 26.8 33.33 Living as Married 20.7 3.4 Not Married 52.4 63.3

Age at First Marriage (%) **
Never Married 76.8 67.5 Married Younger 19.5 32.5
Married Older 3.7 0

Has Children (%)
Yes 19.5 13.7 *
No 80.5 86.3

Age at First Birth (%) 19.37 2.47 20.05 2.04
No Children 80.5 86.3 Had First Child Younger 17.1 ----Had First Child Older 2.4 ----a. t-tests for significant differences of quantitative variables. chi-square for qualitative variables.








79


Table 5.2. Generation X Daughters' and Baby Boom Daughters' Perceptions of
Associational Solidarity with their Mothers in Young Adulthood.
Generation X Baby Boom Daughters Daughters Young Adulthood Young Adulthood 1994 1971
SD N SD N Sig

Associational Solidarity 13.52 4.25 82 12.42 3.16 124 *
In Person 5.28 2.27 82 5.71 2.10 124
By Phone 6.19 1.85 82 4.63 2.41 124 **
By Mail 2.50 1.62 82 2.02 1.70 124
*p <.05; **p <.01; ***p <.001;~-p < .10 (approaching significance)


Table 5.3. Generation X Daughters' and Baby Boom Daughters' Perceptions of
Affectual Solidarity with their Mothers in Young Adulthood.
Generation X Baby Boom Daughters Daughters Young Adulthood Young Adulthood 1994 1971
SD N SD N Sig

Affectual Solidarity 22.58 6.44 82 21.90 5.03 124
Closeness of Relationship 4.80 1.43 82 4.50 1.24 124 Communication 4.59 1.54 82 4.25 1.41 124 Get Along w/ Each Other 4.97 1.54 82 4.59 1.10 124
Mother Understands Her 4.00 1.53 82 4.04 1.27 124 *
She Understands Mother 4.31 1.47 82 4.49 1.14 124
*p <.05: ** p <.01: *** p <.001;~-p <.10 (approaching significance)


Table 5.4. Generation X Daughters' and Baby Boom Daughters' Perceptions of
Consensual Solidarity with their Mothers in Young Adulthood.
Generation X Baby Boom Daughters Daughters Young Adulthood Young Adulthood 1994 1971
'SD N SD N Sig

Consensual Solidarity 3.89 1.46 82 3.55 1.28 122 *
*p <.05; **p <.01; ***p <.001:~ -p <.10 (approaching significance)














Table 5.5. OLS Regression Effects of Daughter Characteristics During Young Adulthood on Associational Solidarity for Generation
X and Baby Boom Generation Daughters.
Model 1 Model 2 Model 3 Model 4 Gen Baby Gen Baby Boom Gen Baby Gen Baby X Boom X X Boom X Boom


Daughter Characteristics Age
Education (ref = more than high school) Less Than/Equal to High School Employment Status (ref = not work)
Works
SEI Score for Occupation (ref =low SEI)
No Score, Does not Work
High SEI Score Income-I 994 Dollars

Marital Status (ref = married
Not Married
Living as Married
Age at First Marriage(ref= married earlier)
Not Married
Married Later
Age at First Child (ref = no children)
First Child Earlier
First Child Later Constant
R 2


-.09 (.19) .09 (.14) -.09 (.21)


.04 (1.26)


-.00 (.64)


.12 (1.34)


.13 (4.76)

.12 (4.81) .17 (1.51}.35 (.00)*


.02 (.14) .01 (.25) .08 (.19) -.02 (.27) -.10 (.18)


-.09 (.66) .07 (1.30)

-.04 (.63) ------.13 (.82) ----.27 (.83)* ----,29 (.00)* -----


-.00 (.73)


.13 (1.40) -.05 (.72) .50 (5.56) -.07 (.67)


.52 (5.59)
-.12 (1.61)
-.23 (.00)


-.12 (.84)
-.28 (.84)*
-.32
(.00)**


.19(2.58) -.11 (3.27) -.09(2.71) -.07(3.21)
-.22(2.70) -.03 (3.66) -.17 (2.77) .02 (3.54)

-.11 (3.78) .05 (3.33) -.27 (4.13) -.20 (3.23)
.17 (2.97) NA -.35 (3.40) NA


16.48 .01


9.73 .01


13.07
.14-


14.84 11*


-.19 (4.57)
-.26(4.53)
16.28


.16- 01 21 1 2-


-.05 (1.01)
NA
10.41


.16(4.66)
.22 (4.61) 12.87


-.09 (.99)
NA 12.62


a. N= 82, Generation A; 124, Baby Boom Generation..


b. Standardized coefficients are presented; Standard errors appear in parentheses
c.*p <.05; ** p <,01 *** p <.001;~p <.10 (approaching significance)














Table 5.6. OLS Regression Effects of Daughter Characteristics on Affectual Solidarity During Young Adulthood for Generation X
and Baby Boom Generation Daughters.
Model 1 Model 2 Model 3 Model 4 Gen Baby Gen Baby Boom Gen Baby Boom Gen Baby X Boom X X X Boom


Daughter Characteristics Age .11 Education (ref = more than high school) Less Than/Equal to High School -.10( Employment Status(ref=does not work)
Works
SEI Score for Occupation(ref=low SEI)
No Score, Does not Work
High SEI Score ---Income- 1994 Dollars Marital Status (ref = married
Not Married
Living as Married
Age at First Marriage(ref=married earlier)
Not Married
Married Later
Age at First Child (ref = no children)
First Child Earlier
First Child Later --Constant 17. R 2 .0
a. N= 82, Generation X; 124, Baby Boom Generation..


.25) .08(.21) .09(.28)


1.62)


.03 (.22) -.01 (.32)


10(.28) -.11 (.34) .12(.28)


-.05(.96) -.03(1.87) -.12(1.01) -.15 (1.72)- -.02(1.09) -.01 (1.91) -.06(1.11)


------ -.36 (6.93) -.10 (.98) ----------- ----------- .36 (6.96) -.14(1.23) ----------- .07 (2.05) -.19 (1.23)- ----------- .18(.00) -.16(.00) ----------- ------ ------


------ .70 (7.99) -.12 (.99)

------ .58 (7.97) .00 (1.26)
-.01 (2.17) -.21 (1.25)------ .28 (.00)* -.18(.00)-


-.02(3.76) -.27 (5.05) -.20 (3.86) -.22 (5.08)
-.02(3.91) -.89(5.64) -.04(3.93) -.25(5.11)


82
3


18.81 .01


12.06 .07


23.69 .06-


-.24(4.09) .02 (4.25)

.35 (6.64) .36 (6.61)
21.61
.09


2.12(5.13) -.44(5.93) .04(1.22)
NA -.23(4.86) NA


-.10(1.52)
NA
20.45
.02


.24 (6.72) .32 (6.64) 19.25 .16


14(1.52)
NA
22.28 .08


b. Standardized coefficients are presented; Standard errors appear in parentheses
c.*p <.05; ** p <.01; *** p <.001;-p <.10 (approaching significance)














Effects of Daughter Characteristics During Young Adulthood on Consensual Solidarity for Generation


X and Baby Boom Generation Daughters.


Model 1 Model 2 Model 3 Model 4
Gen Baby Gen Baby Boom Gen Baby Boom Gen Baby X Boom X X X Boom


Daughter Characteristics Age
Education (ref = more than high school) Less Than/Equal to High School Employment Status (ref = does not work)
Works
SEI Score for Occupation (ref = low SEI)
No Score, Does not Work
High SEI Score Income-1994 Dollars Marital Status (ref = married
Not Married

Living as Married
Age at First Marriage (ref =married earlier)
Not Married

Married Later
Age at First Child (ref = no children)
First Child Earlier
First Child Later Constant
R2


.02 (.06) .21 (.05)* .02 (.06) .16 (.06)- -.06 (.07) .27 (.07)* -.13 (.07) .28 (.07)*


-.10 (.36) .20 (.25)* -.01 (.41) .14 (.26) -.14 (.38) .24 (.28)*
------ -----1 -.04 (.
------ ------ .47 (l.551) -.04 (.25) ------ ------


.46 (1.56)
.05 (.46) .19 (.00)-


3.75 .01


1.02 .06*


a. N= 82, Generation X; 124, Baby Boom Generation.. b. Standardized coefficients are presented; Standard errors appear in parentheses
c.*p < .05, ** p < .01; ** p < .001; p < .10 (approaching significance)


7.90
.05


-.07 (.32)
-.17 (.32)-.16 (.00)*


2.28
09*


-.02 (.43) .19 (.28)* .64 (1.81) .61 (1.27)


.52 (1.81) .03 (.48) .20 (.00)


-.05 (.32)
-.20 (.32)-.21 (.00)*


00
.11(.84) -.60 (1.28) -.03 (.87) -1.02 (1.29)
-.05(.88) -. 18(1.43) -.08(.89) -.15 (1.42)

-.27(.92) .67 (1.30)- -.30 (1.34) 1.41 (1.30).04(.96) NA -.07(1.10) NA


-.18(1.50) -.03 (.39)
NA
5.83 1.88 .08 .08*


080*.13 13*


.28 (1.53)
-.21 (1.52)
4.93


-.06 (.39)
NA .65


.05 0.


Table 5.7. OLS Regression -








83


Table 5.8. Descriptive Statistics Social Distance Among Generation X Daughters
During Young Adulthood and their Baby Boom Generation Mothers During
Young Adulthood.
Difference Variables For Daughters and Their Mothers %


Education Level
Daughter has More Education than Mother 31.1 Mother has More Education than Daughter 13.8 Both Daughter and Mother have the Same Level of Education 55.2

Work Status and SEI for Occupation
Daughter Works/High SEI, Mother Does Not Work/No SEI 12. 1 Mother Works/High SEI. Daughter Does Not Work/No SEI 10.3 Both Daughter and Mother Work and Have Low SEI 10.7 Both Have No SEI Because Neither Work 24.1 All Others 43.1

Income-1994 Dollars
Daughter's Income Level is High, Mother's is Low 22.4 Mother's Income Level is High, Daughter's is Low 25.9 Daughter and Mother Have Same Income Level 51.7

Marital Status and Timing of Marriage
Both Mother and Daughter Married Early 14.6 Daughter is Single/Never Married and Mother Married Early 48.8 All Others 35.4

Children
Both Daughter and Mother Have Children 19.5 Daughter Does not Have Children, Mother Does 80.5








84


Table 5.9. OLS Regression Effects of Social Distance Between Generation X
Daughters and their Baby Boom Generation Mothers During Young
Adulthood.
Associational Affectual Consensual Solidarity Solidarity Solidarity


Difference Scores


Education Level (ref=daughter has more)
Mother has More Education than Daughter
Daughter and Mother Have Same Education

Work Status and SEI Score (ref = daughter works/high SEI, mother does not work)
Mother Works/High, Daughter Does Not Work
Both Work Have Low SEI
Neither Work and Have No SEI
All Others

Income (ref=Daughter's Income is High, Mother's is Low)
Mother's Income is High, Daughter's is Low
Income Level is the Same for Both

Marital Status and Timing of Marriage (ref=both daughter and mother married early)
Daughter is Single, Mother Married Early
All Others

Children (ref=both have children)
Daughter Has No Children, Mother Does

Constant
R 2


-.30 (2.58)
-.04 (1.93)


-.08 (3.58) .01 (.77) .11 (2.90) .25 (.60)


-.10(3.88) -.46 (4.84) *


-.48 (2.99) .42 (2.68)~ .33 (2.76)


-.15 (2.30)
-.06 (2.03)



.03 (2.08) .23 (1.95)


.01 (1.80)

10.29 .32-


-.13 (4.38)
-.21 (3.82)
-.30 (3.81)


-.34 (1.01) *

.15 (.91)
-.10 (.80)
-.29 (.80)


-.02 (3.40) -.19 (.76) .30 (2.85)~ -.01 (.66)



-.25(2.84) -.23(.61)-.19 (2.86) -.27 (.58)


-.16 (3.03) -.10 (.55)


28.13 .25


4.98 .27-


a. N=82
b. Standardized coefficients are presented; Standard errors appear in parentheses. c. *p<.05; ** p<.01; ***p<.001 -p<.10















CHAPTER 6
GENERATIONAL DIFFERENCES IN DAUGHTERS' PERCEPTIONS OF
ASSOCIATIONAL, AFFECTUAL, AND CONSENSUAL SOLIDARITY IN MID-LIFE Introduction

The purpose of this chapter is to examine generational differences in daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during the midlife stage. This chapter specifically explores the role of education, work, and family in influencing daughters' perceptions of solidarity with their mothers during midlife. Two generations are compared women from the Baby Boom Generation in 1994 in midlife and women from the World War II Generation in 1971 during midlife.

This chapter presents the results of several sets of analyses bivariate and multivariate. Descriptive sample characteristics are presented first (a bivariate comparison of two generations) for each generation during midlife. Education, work and family characteristics of each generation are calculated. Tests for differences across the two generations when they were in midlife wer calculated as well. The second set of analyses calculates levels of associational, affectual, and consensual solidarity for each cohort when in the midlife life stage. Again, tests for differences in reported solidarity with mothers across the two generations when they during midlife are calculated as well. The next set of analyses included in this chapter are multivariate models estimating the effects of the independent variables on the three measures of solidarity. We employ a nested modeling strategy where more exogenous variables are entered into the model first (age and education) and subsequent models explore more proximate factors (work and


85








86


family characteristics). The first model includes age, as a control variable, and one's level of education. The second model includes the work related variables of employment, socioeconomic status for occupation, and income. The third model includes those variables related to family marital status, age at first marriage, and age at first child. The fourth and final model includes all of the variables age, education, work and family. Thus, we approximate the contribution each set of variables has in explaining levels of solidarity as each set is added to the model.

The final set models are OLS regression models estimated for Baby Boom

Generation daughters examining the effects of social distance (as compared with their mothers) on perceptions of daughters' solidarity. For example, we examine whether being married later in life than one's mother was married decreases daughters' perceptions of solidarity with their mother. We do not estimate a comparable set of models for the World War II Generation daughters as the data do permit social distance measures to be constructed as there are no data on the mothers of this generation.

Descriptive Statistics-Independent Variables

The first table in this chapter includes the descriptive statistics for the demographic characteristics of Baby Boom Generation and World War II Generation daughters during midlife. The quantitative variables are represented by the mean and the standard deviation, while the qualitative variables are represented by the percentage of women, accounting for each of the variable's attributes. T-tests were done to compare the women from each generation during young adulthood across the quantitative variables, and chi square was used for comparison of the qualitative variables.




Full Text

PAGE 1

SOCIAL CHANGE AND INTERGENERATIONAL SOLIDARITY: MOTHER/DAUGHTER RELATIONSHIPS IN THE SECOND HALF OF THE TWENTIETH CENTURY By MELANIE ANN WAKEMAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

It is with a tremendous amount of love and gratitude that I dedicate this dissertation to my mother and my grandmother. Rebecca T. Roman and Judith A. Jones.

PAGE 3

ACKNOWLEDGMENTS While wnting the dissertation might feel like an isolating and individual process, it is truly a collaborative effort. So many people were involved in the success of this dissertation project, and involved in my progress and success in graduate school in general. First, I would like to acknowledge and thank the members of my doctoral dissertation committee and Institute on Aging (loA) mentoring team, which include Connie Shehan. Amy Pienta, Barbara Zsembik, Leonard Beeghley, Tanya KoropeckyjCox, and Angel Kwoleck-Folland. As my chair and cochair. Dr. Shehan and Dr. Pienta have been instrumental in helping me to conceptualize this project, spending countless hours talking through ideas over the last several years. Their overall support of me as a graduate student and a colleague has been a constant. In addition. Dr. Pienta has offered her analytic insight and a keen editorial eye, and was instrumental in helping me navigate my way through a very complex data set. I especially appreciate Dr. Zsembik's and Dr. Beeghley "s unconditional devotion to me during my time at Flonda (and beyond). Their doors were always open when I needed to talk, and they offered valuable insight, both professional and personal, with nothing but my best interest at heart. Finally, I thank Dr. Koropeckyj-Cox and Dr. Kwolek-Folland for joining my committee. Both have contributed a great deal to the progress of this project. Several people outside of my committee must be thanked for their support during my graduate program. First. Rae Newton, my master's advisor, has always encouraged me, and remains a great source of support, even today. Terrj' Mills' guidance enabled me

PAGE 4

to participate in several programs, and afforded me many great opportunities. Also, many facult>' and students at the loA deserve special recognition for their support and belief in me. They are Michael Marsiske, Jeff Dwyer, Ann Horgas, Nicole Alea, Karin McCoy, Angle Semegon, Katherine White, Jennifer Stanley, Tyson Brown, and especially Yvonne Combs. Members of the office staffs of the Department of Sociology, the Institute on Aging (loA), the Center for Women's Studies and Gender Research (CWSGR), and the Universitv' Center for Excellence in Teaching (UCET) have provided assistance on several levels throughout the years. There have been many people working in this capacity-. In the Department of Sociology, Sheran Flowers, Kanitra Perry, and Justin Smith put in endless hours making sure we all get paid and handling the necessary bureaucratic paperwork, among other things. Charlotte Simmons, Richelle Davis, and Anna Abee in the loA always had the answers, even when there were none, and they have done a superb job helping all of us in the first cohort of graduate research trainees. Paula in the CWSGR became a friend, and of all things, a student, dunng my final year at Flonda. And in UCET, Nadine Gillis and Diane were always there to take care of me when I needed it, providing a safe haven, a place to eat, a place to organize, a place to prepare for class, and countless other reasons to hang out and chat. Many fellow graduate students at the University of Florida deserve special thanks for their friendship and support, especially during that very difficult first year. I thank Leslie Houts, Man DeWees, Jason Davis. Lisa Holland, Brad Tripp, and Clay Hipke. In their own ways, all of them helped me through the good and bad times that invariably are what it IS to be a graduate student.

PAGE 5

There are also several very special people who have become some of my best fnends. Susan Eichenberger. Sara Crawley, Lara Foley, Helena Alden, Laurel Tripp, Ed Porras, and Barbara Bonano became "Family" for me, and I can't begin to imagine life in Gainesville without them. I guess mention of the solace we all found on my front porch must also be included! Gator football and the usual football crowd also deserve special mention. Those Saturdays in the fall were necessary to my sanity. Thanks go to Kevin Fraizee for always being the gracious host, and to Pam, Kathy, Oz, Gabe, and the rest of the gang. GO GATORS! This acknowledgement would not be complete without mentionmg fnends, who, although they did not experience graduate school with me, have always been supportive and have loved me unconditionally. Doug Laughton, Jilana Miller, Cristin and Mike Simpson, Kim Anderson, Meg Benjamin, and Steffanie Siebrand are longtime friends from childhood who sometimes know me better than I know myself, and are always there to keep me in line. I would like to thank my family, without whom I would not be where I am today. My father, Mark Wakeman, who is always the realist, has helped to keep me on track in recent years. My step-father, Don Roman, has also been there for me over the years, to challenge me intellectually and to remind me not to take myself too seriously. In addition, along with his red pen, he served as editor and critic throughout the various stages of the dissertation process. I thank my brother, Andrew, who is my favorite person in the whole world. My mother and Gammy (Rebecca Roman and Judy Jones), have

PAGE 6

given me unconditional love and support, both emotional and financial, and are truly the mspiration for this dissertation project. I will never be able to thank them enough. Finally, I would like to thank Bob Stagnaro. He has become a very significant person in my life. Although I have long since been done with my dissertation and graduate school, he knows why he has been included in this acknowledgement section.

PAGE 7

TABLE OF CONTENTS page ACKNOWLEDGMENTS m LIST OF TABLES xi ABSTRACT xiii CHAPTER 1 INTRODUCTION TO STUDY 1 Introduction 1 Specific Aims 2 Oveniew of the Study 3 Theoretical Framework 4 Review of the Literature 4 Data and Methods 5 Analysis 6 Discussion and Conclusions 6 Conclusion 6 2 TFTEORETICAL FRAMEWORK 9 Introduction 9 Theories of Aging 10 Social Exchange Theory 10 Social Constructionist Theories 10 Contmuit}' llieorv' 11 Age-Stratification Tlieory 12 Age-lntegration Theor\' 13 Political Economy of Aging 13 Theories of Aging Families 14 Family Development Theory 14 Social Exchange Theoiy 17 Intergenerational Stake Hypothesis 18 Socioemotional Selectivity' Theoiy 19 Social Comparison Theory' 19 Understanding Social Change and Women's Intergenerational Relationships 21 Life Course Theorv' and the Developmental Schism 21

PAGE 8

Intergenerational Family Solidai'it\' Theory23 Conclusion 24 3 REVIEW OF THE LrrER.VrURE 26 Introduction 26 Chajiges in Social Structural Forces 26 Education and Occupational (Status) Attainment 27 Changing Work Roles 27 Family Fonnation 29 Mantal Stabilit>^ 30 Ideological Shift 30 Intergenerational Family Relationships 31 Intergenerational DifTerences Among Women 32 Statement of Hypotheses 34 Generation X and Baby Boom Daughters During Young Adulthood 34 Baby Boom Generation and World War ii Generation Daughters Dunng Midlife 35 4 DATA AND METHODS 36 Introduction 36 Sample 36 Sample Selection Characteristics 38 Daughters During Young Adulthood 40 Daughters During Midlife 40 Measurement 41 Dependent Variables 41 Independent Variables 44 Social Distance Measures 49 Analytic Model 52 Sample Limitations 54 5 GENERATIONAL DIFFERENCES IN DAUGHTERS^ PERCEPTIONS OF ASSOCIATIONAL, AFFECTUAL. AND CONSENSUAL SOLIDARITY IN YOUNG ADULTHOOD 57 Introduction 57 Descriptive Statistics-Independent Variables 58 Results 59 Descriptive Statistics and Mean Differences-Dependent Variables 60 Results Daughters' Perceptions of Associational Solidarity During Young Adulthood 61 Results -Daughters' Perceptions of Affectual Solidaritv During Young Adulthood 62 Results Daughters' Perceptions of Consensual Solidarity During Young Adulthood 62

PAGE 9

Multivariate Analyses 63 Resulls-Effecls of Daughter Characteristics on Their Perceptions of Associational Solidarity 64 Results-Effects of Daughter Characteristics on their Perceptions of Affectual Solidarit\ 66 Resulls-Effecls of Daughter Characteristics on their Perceptions of Consensual Solidarity 67 Social Difference Analyses for Generation X Daughters and their Baby Boom N4others During Young Adulthood 70 Results-Descriptive Statistics for the Difference Measures Compnsing the Independent Vanables 71 Results-Effects of Social Difference Measures on Generation X daughters" Perceptions of Associational, Affectual and Consensual Solidarity with their Baby Boom Generation Mothers During Young Adulthood 72 Conclusions 74 6 GENERATIONAL DIFFERENCES IN DAUGHTERS" PERCEPTIONS OF ASSOCIATIONAL. AFFECTUAL. AND CONSENSUAL SOLIDARITY' IN MID-LIFE 85 Introduction 85 Descriptive Statistics-Independent Variables 86 Descriptive Statistics and Mean Differences-Dependent Variables 89 Results-Daughters" Perceptions of Associational Solidarity During Midlife 89 Results Daughters' Perceptions of Affectual Solidarity During Midlife 90 Results-Daughters" Perceptions of Consensual Solidarity During Midlife 91 Multivariate Analyses 92 Resulls-Effecls of Daughter Characteristics on their Perceptions of Associalonal Solidaiity During Midlife 92 Results-Effects of Daughter Characteristics on their Perceptions of Affectual Solidarity During Midlife 95 Resulls-Effecls of Daughter Characteristics on their Perceptions of Consensual Solidaritv During Midlife 97 Social Difference .Analyses for Baby Boom Generation Daughters and their World War II Generation Mothers During Midlife 99 Results Descriptive Statistics for the Difference Measures Compnsing the Independent Vanables 100 Results-Effects of Social Difference Measures on Baby Boom Generation Daughters" Perceptions of Associational. Affectual and Consensual Solidarity with their World War II Generation Mothers During Midlife 101 Conclusions 103 7 DISCUSSION AND CONCLUSIONS 115 Introduction 1 15 Review of Empirical Measures and Theoretical Framework 116

PAGE 10

Discussion 117 Generaiional Differences among the Women 1 18 Intergenerational Solidarity 120 Conclusions and Implications 122 Future Research 123 LIST OF REFERENCES 124 BIOGRAPHICAL SKETCH 134

PAGE 11

LIST OF TABLES Table page 4. 1 Numbers of Generation X and Baby Boom Generation Daughters Dunng Young Adulthood at Each Stage of Sample Selection 56 4.2 Numbers of Baby Boom Generation and World War II Generation Daughters during Midlife at Each Stage of Sample Selection 56 5. 1 Descriptive Statistics Demographic Characteristics of Generation X Daughters and Baby Boom Daughters in Young Adulthood 78 5.2 Generation X Daughters' and Baby Boom Daughters' Perceptions of Associational Solidarity with their Mothers in Young Adulthood 79 5.3 Generation X Daughters' and Baby Boom Daughters' Perceptions of Affectual Solidarity with their Mothers in Young Adulthood 79 5.4 Generation X Daughters' and Baby Boom Daughters' Perceptions of Consensual Solidarity with their Mothers in Young Adulthood 79 5.5 OLS Regression Effects of Daughter Characteristics During Young Adulthood on Associational Solidanly for Generation X and Baby Boom Generation Daughters 80 5.6 OLS Regression Effects of Daughter Characteristics on Affectual Solidarity During Young Adulthood for Generation X and Baby Boom Generation Daughters 81 5.7 OLS Regression Effects of Daughter Characteristics During Young Adulthood on Consensual Solidarity for Generation X and Baby Boom Generation Daughters 82 5.8 Descriptive Statistics Social Distance Among Generation X Daughters During Young Adulthood and their Baby Boom Generation Mothers During Young Adulthood 83 5.9 OLS Regression Effects of Social Distance Between Generation X Daughters and their Baby Boom Generation Mothers During Young Adulthood 84

PAGE 12

6. 1 Descriptive Statistics Demographic Characteristics of Baby Boom Daughters and World War II Daughters Dunng Midhfe 108 6.2 Baby Boom Daughters" and World War II Daughters' Perceptions of Associational Solidarit>' During Midlife with their Mothers 109 6.3 Baby Boom Daughters" and World War 11 Daughters" Perceptions of Affectual Solidarity During Midlil"e with their Mothers 109 6.4 Baby Boom Daughters" and World War II Daughters" Perceptions of Consensual Solidarity During Midlife with their Mothers 109 6.5 OLS Regression Effects of Daughter Characteristics During Midlife on Associational Solidarity for Baby Boom Generation and World War II Generation Daughters 110 6.6 OLS Regression Effects of Daughter Characteristics During Midlife on Affectual Solidarity for Baby Boom Generation and World War II Generation Daughters Ill 6.7 OLS Regression Effects of Daughter Characteristics During Midlife on Consensual Solidarity for Baby Boom Generation and World War II Generation Daughters 1 12 6.8 Descriptive Statistics Social Distance Among Baby Boom Generation Daughters During Midlife and their World War 11 Generation Mothers During Midlife 1 13 6.9 OLS Regression Effects of Social Distance Between Babv Boom Generation Daughters and their World War II Generation Mothers During Midlife 114

PAGE 13

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SOCIAL CHANGE AND INTERGENERATIONAL SOLIDARITY: MOTHER/DAUGHTER RELATIONSHIPS IN THE SECOND HALF OF THE TWENTIETH CENTURY By Melanie Ann Wakeman May 2005 Chair: Constance L. Shehan Cochair: Amy Mehraban Pienta Major Department: Sociology This study examines generational differences in mother/daughter relationships. As a result of changing opportunities and role expectations, especially regarding education, work, and family, women's life course trajectones have shifted. Using the Longitudinal Study of Generations and Mental Health, this study seeks to gain greater understanding of the indirect impact social and structural changes have had on women's Lntergenerational solidarity. Using Bengtson's measures of solidarity, daughters" perceptions of associational, afl^ectual, and consensual solidarity with their mothers is examined during both young adulthood and midlife. Several generations of women from the same families are included, capturing the birth cohorts comprising the WWI Generation, the WWII Generation, the Baby Boom Generation, and Generation X. The perceptions of solidarity of the Generation X daughters as young adults in 1994 are compared with those of their mothers, the Baby

PAGE 14

Boom daughters as young adults in 1971. The perceptions of solidarity of the Baby Boom Generation daughters during midlife in 1994 are compared with those of their mothers, the World War II Generation daughters during midlife in 1971. Univariate, bivariate, and multivariate analyses are earned out to understand how daughter's perceptions of associational, affectual, and consensual solidarity with their mothers have changed over the last half of the century. Results show several significant differences among the generations of women, both during young adulthood and midlife. Perceptions of associatonal solidarity, aspects of affectual solidarity, and consensual solidarity also differed during young adulthood for the Generation X daughters in 1994 compared with the Baby Boom Generation daughters in 1971. During midlife, associational solidarity and aspects of affectual solidarity were significantly different for Baby Boom Generation daughters in 1994 compared with World War II Generation daughters in 1971. Furthermore, ordinary least squares (OLS) regression indicates several significant findings pertaining to the effects of education, work, and family vanables on perceptions of associational, affectual, and consensual solidanty for Generation X daughters and Baby Boom Generation daughters during young adulthood, as well as generational differences between the women from 1971 to 1994. This was also the case among the Baby Boom Generation daughters in 1994 and the World War II Generation daughters during midlife in 1971

PAGE 15

CHAPTER 1 INTRODUCTION TO STUDY Introduction The mother/daughter relationship has become one of the most salient of familial and social relationships in Amencan society in the twentieth centurv-. It is the focus of songs, poems, books, movies, and other cultural and social media. A major integral part of our culture is the bond between mothers and daughters. Understanding this relationship and its recent history is important as, women's lives contmue to change. As women's life experiences have been altered from generation to generation, as a result of social change, v\hat happens to intergenerational relationships among women mothers and daughters? Furthermore, how are successive generations of mother/daughter dyads different from each other? This is the subject of this research. This research sought to identify the stable and dynamic aspects of intergenerational relationships over time, to better understand the relationship between family solidarity changing social and structural forces. This research also attempted to determine whether familial role expectations shape dyadic relationships across generations; and if so. how these relationship differences might change over time, as a result of social distance between the generations. Social and structural changes in the United States over the second half of the 20"' centurvhave led to changes in both family structure and social norms/role expectations for women. Both the social and economic position and attitudes of and about women in U.S. society have changed dramatically over this period (Moen, 1992, Anderson, 2000). 1

PAGE 16

Women's educational attainment has increased; women's labor force participation has increased; women's economic contribution to the household has increased; women are postponing marriage and having children later; fertility rates have declined; and prevailing ideologies about women's "place" have shifted. For example, changes in social structure are reflected in the beliefs and practices associated with women combining work and family roles, over the life course. These t\'pes of changes are expected to transform intergenerational relationships over time. The profound legislative, economic, demographic, and ideological changes that have occurred have changed the life course trajectories of successive generations of women in families. Figure 1.1 illustrates the timing of each generation's experience of these sociohistorical changes during the twentieth century. It provides a visual representation of women's lives and how the timing of each generation of women's experience of these events has created new and different opportunities. This research examined the impact of these social changes on women's inter-generational solidarity. Specific Aims Early research assumed that "familial transmission and not social trends account for generational similanties" (Aldous, 1963; Aldous and Hill, 1965; Farber, 1966). Further, it has been proposed that continuity between the generations within the family is dependent upon the degree of cohesion existing between them (Aldous, 1963; Aldous and Hill. 1965). The assumption was that familial transmission is stronger than social trends. However, this idea may be a bit out dated because social and structural changes have been so profound. Women have encountered rapidly growing opportunities and shifts in expectations, as a result of the rapid social change that occurred in the twentieth century.

PAGE 17

Thus, the argument can be made that famihal relationships have become more complex and dynamic. This study examined generational differences in daughters' perceptions of solidarity with their mothers. Three aspects of solidarity were included. These aspects were; 1) associational solidarity, that is the extent to which family members interact with one another; 2) affectual solidarity, that is the extent to which family members are emotionally close to one another; and 3) consensual solidarity, that is the extent to which family members share similar values and world views (Bengtson and Roberts, 1991). The first aim of this research was to understand how daughters from different generations perceive their relationships with their mothers during young adulthood. The first research questions is, "How have daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during young adulthood changed, and what factors may be associated with this change?" The second aim of this proposed study seeks to understand how daughters from different birth cohorts perceive their relationships with their mothers during midlife. Thus, the second research question is, "How have daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during midlife changed, and what may be associated with these changes?" Overview of the Study A large and growing body of research has focused on understanding intergenerational family relationships, and family solidarity issues. As individuals continue to live longer, families will increasingly consist of multiple generations. The relationships between the generations, specifically their infiuence and functions, will also become increasingly more important. With regard to women, as a result of significant

PAGE 18

social changes in roles and role expectations, women's intergenerational relationships have become all the more intriguing. The following is an outline of this research, focusing on social change and the influence on women's intergenerational relationships. Theoretical Framework Chapter 2 of this research discusses the major theories used in the study of aging families, intergenerational family relationships, and more specifically, women's intergenerational relationships. This study will focus on Intergenerational Family Solidarity and the Developmental Schism. As a result of the growing diversity in the functions and structure of Family today, as well as an increase in life expectancy, intergenerational relationships have become increasingly important. Intergenerational Family Solidarity Theory (Bengtson et al., 1984) is one of the major theoretical contributions to the area of family gerontology. From this theory, empirical measures have been constructed, allowing researchers to measure the degree of solidarity among different generations. These measures are outlined discussed in Chapter 3. However, more detail as to their use is discussed in Chapter 4. The Developmental Schism (Fingerman, 1996) is a concept stemming from the life course perspective, and grounded in developmental psychology. This schism specifically addresses the interpersonal problems among intergenerational relationships between females. The Developmental Schism, and is role in this research is discussed in detail in Chapter 3. Review of the Literature Chapter 3 is a review of the literature. This chapter outlines changes in social structural forces over the last century, discusses the literature pertaining to intergenerational family relationships and women's intergenerational relationships, and

PAGE 19

finally, outlines the hypotheses for this study. First, some of the major legislative, economic, demographic, and ideological changes are reviewed. Included in this review of legislative changes are the Equal Pay Act, Title IX. and FDA approval of the birth control pill. Education and occupational (status) attainment, changing work roles, family formation, and marital stability are discussed next. These changes have had an impact not only on women's roles, thus altering their life course trajectories over the successive generations, but they have also had an impact on society's attitudes concerning women as well. Thus ideological shifts occurnng in the last half of the twentieth century are also included in this chapter. Data and Methods Chapter 4 is a detailed account of the methodological procedures carried out in this study. This study uses the Longitudinal Study of Generations and Mental Health, 1971 data set [made accessible in 1990, machine-readable data files]. These data were collected by V. Bengtson and M.Gatz and are available through the archive of the Henry A. Murray Research Center of the Radcliffe Institute for Advanced Study, Harvard University. Cambridge, Massachusettes (Producer and Distributor). These data are from an ongoing study of aging families investigating changes in intergenerational family relationships. These data specifically examine intergenerational solidarity, based on perceived and actual levels of cohesion among family members, making it an ideal data set for this research. This chapter includes summary characteristics of the samples used from the data set, as well as the complex sample selection strategy used to create the sample of women for this study. Finally, this chapter outlines the analytic strategies used to better understand how changing opportunities and life course trajectories for

PAGE 20

successive generations of daughters may have influenced their perceptions of their relationships with their mothers. Analysis Two analytic chapters are included in this project. Chapters 5 and 6, employ the same analytic strategy, thus including the results of the univariate, bivariate, and multivariate statistical analyses carried out m this study. Chapter 5 specifically examines the generational differences in daughters' perceptions of associational, affectual, and consensual solidarity with their mothers while they are young adults. The data analyzed for this chapter include Generation X daughters age 18 to 29 years, with data collected in 1994. and Baby Boom generation daughters age 18 to 29 years, with data collected in 1971. Chapter 6 specifically examines these generational differences among daughters during midlife. This chapter includes Baby Boom Generation daughters, ages 40 to 59 years, in 1994, and World War II Generation daughters, ages 40 to 59 years, in 1971. Discussion and Conclusions Chapter 7 is the final chapter in this study. Conclusions, implications, and future research potential are discussed in this chapter. First, the generational differences among women are discussed based on the findings from the analyses in chapters 5 and 6. Next, a review of the social distance occurring between the generations of women participating in this research is outlined. Third, levels of intergenerational solidarity', and the efi'ects of the daughter characten sties and social distance measures are reviewed. In addition, a discussion of practical implications and future research are discussed in this chapter. Conclusion In sum, this project seeks gain a greater understanding of women's intergenerational relationships. Specifically, how social structural changes may have

PAGE 21

influenced these relationships over the last half of the twentieth century as a result of their effect on the life course trajectones of women. Because intergenerationaJ familial relationships are becoming increasingly more prevalent with the increased life expectancies today, it is important to develop a better understanding about women's intergenerational relationships. This project expands the literature in this area by examining individual characteristics reflected in the social structural changes of the last century. Stemming from a life course perspective and usmg Bengtson's measures of intergenerational family solidarity, it is possible for this project to contribute to a growing body of knowledge surrounding aging families and women's intergenerational relationships.

PAGE 22

; = -5= ., S? ^ ^ h ^ '*J '^ y '•' '' -H ?' 5 5 "i S S ^ }~ !X Vt S I Sil 'i 1 1 1 1 .1 1 1 1 1 Is ? ?; J 1" S 3 5? §' s ?; i 5. k. 1.1:

PAGE 23

CHAPTER 2 THEORETICAL FRAMEWORK Introduction Aging families (and specifically, intergenerational relationships) have become an integral part of the field of social gerontology. Sociological theories and many family theories have been reexamined from the perspective of aging studies. As a result, new lines of thought have emerged that consider the effects of age (and the aged m our society) on the family and on family dynamics. This study centered on two guiding theoretical frameworks: Intergenerational Family Solidarit>' Theor>'; and the Life Course Perspective (in particular, the Developmental Schism hypothesis), Intergenerational Family Solidarity Theory (Bengtson et al,, 1984) is one of the major theoretical contributions to the area of family gerontology. From this theory, empirical measures for the present study were constructed, allowing researchers to measure the degree of solidarity among different generations. This study also used the Developmental Schism hypothesis (Fingerman, 1996), which stems from the life course perspective, and is grounded in developmental psycholog}'. The Developmental Schism hypothesis is highly relevant to the present study because it addresses interpersonal problems in the intergenerational relationships of females. Before reviewing the guiding theoretical framework for the present study, here is an overview of aging theories, and theories that have been applied to the study of aging families.

PAGE 24

10 Theories of Aging There have been several theories developed to better explain aging related phenomena and relationships. Each of the following theones has made a significant contribution to the field of social gerontology. Social Exchange Theory Social exchange theory in aging is adopted from the micro-economic premise that interaction or exchange is based on the calculation of the rewards that they will gained. In the study of aging, relationships and interaction, are influenced by the resources of the individuals involved (Bengtson, Parrott, and Burgess, 1997). Resources in terms of social interaction are not just limited to money. Anything that elevates the status of that individual can be a resource. The key question becomes, what resources does each person bring to the table? In other words, do the benefits outweigh the costs of being involved with this person? Therefore, when people are considering interactions with older adults, they weigh the pros and cons of the interaction. Because most older adults have fewer valued resources, there is less involvement with them. A shortcoming of social exchange theory is that it doesn't account for the benefits of relationships that cannot be logically explained. People get benefits from interactions with people when the obvious reasons for the exchange are not present. In addition, social exchange theory does not account for why some people have more resources than others. Social Constructionist Theories Social constructionist theories suggest that individuals negotiate reality through interactions with others. It postulates that people are active subjects, not passive participants. Such an emphasis on human structure means that social structures do not

PAGE 25

11 impose social noles on us, but rather we create the social structures (Gubrium and Holstein. 1999). Constructionist theorists in the study of aging are interested in how aging individuals create their realities from their experiences and interactions as they go through aging-related phases. Constructionist theories do not fit the definition of a theory per se, as they "are not conceived as causal explanations of the social world, but instead focus attention on problems of meaning in ever>day life (Gubrium and Holstein, p. 287). Constructionist theories are similar to social exchange theory because the pnmarv focus is on the actions of the individual, and emphasis is placed on interactions with others. A shortcoming of constructionist theories in aging is the overemphasis on the individual at the personal level, and on the individual's constructions of reality. Without an insight into the influences of social forces, there is no context from which to understand social behavior and interaction. Contmuity Theory This is the idea that the more older adults maintain a sense of continuity in their everyday living and activities, the fewer adjustments they will have to make in later life. Borrowing from the proponents of activity theory, which suggests that if older adults remain positive and active into and throughout "old age," they will experience greater life satisfaction (Havighurst, 1963), Robert Atchley developed continuity theory based on two dimensions of continuity: intemal and external. The internal dimension of continuity is a personal dimension; and consists of peoples" values, world views, skills, and general life frameworks. The external dimension is a public dimension; and consists of role performance, status attainment, and everyday activities and hobbies. Atchley suggests that continuing to draw on ideas and activities that were successful in early life will benefit people moving into the stages of later life (Atchley, 1989).

PAGE 26

12 Unlike exchange theory and the social constructionist theories, where the individual is the primary focus, the connection in continuity theory is to broader social life. Just as it is in social exchange theory, social interaction is an important focus. Part of external continuity would be for individuals to maintain those interactions that are positive, into older adulthood. Continuity theorv' posits satisfaction in later life correlates strongly to continuing important, successful interactions with others over time. A major critique of continuity theory' is the "normalization" of the aging process (Becker, 1993). Since many of the activities of individuals are shaped by age-related social role expectations, continuity theory suggests the existence of a standard aging process that people go through. Furthermore, feminist researchers argue that this imposes one standard (a male standard) on people as they move through the life course (Browne, 1998; Calasanti, 1996). Age-Stratification Theorv Matilda White Riley's (1971) age-stratification theor\' provides aframework for examining age and aging as a social process dependent on the forces of the larger social structures. The focus is on the status of older adults, and specifically on the ways in which statuses and roles are allocated in our society based on age. Age-stratification theory postulates that all societies break themselves down into age categories, ultimately ranking these categories into hierarchies based on larger social notions of importance. Like social exchange theory, value is placed on the contributable resources people possess. However, age-stratification theory concems the macro level. A key component of age-stratification theory is the age cohort: a group of people bom at the same time, moving through the life course together. The age cohort places a social context around the individual's age, which contributes to the categorization of this particular group of

PAGE 27

13 individuals. Often, members of these cohorts share similar viable resources, or perhaps, a lack of resources. Major limitations of age-stratification theorv' is that it assumes that a general consensus exists among society members about age-related statuses and power. It fails to recognize the power relationships inherent in our social structures. Moreover, there is no way to account for intracohort differences (such as race, class, and gender) that also contnbute to inequity in allocating roles, statuses, and power. Age-Integration Theory Like age-stratification theory, age-integration theor\' also emphasizes the age cohort and the hierarchy of groups in our society', based on social factors (including age). Age-integration theory also recognizes the influence of interactions at the micro level. Borrowing from Durkheim's concept of solidarity and integration, gerontologists focus on factors associated with the elderly (Rosow, 1974). Rosow proposes that as a result of status and role losses due to age, the social networks of older adults suffer. If older adults could surround themselves with their peers, they would have a stronger sense of social solidarity, and ultimately greater life satisfaction. A limitation of age-integration theory, however is that (like age-stratification theory) it does not explain the heterogeneity within age cohorts. In addition, because of the arrangement of dominant social structures and the resulting social expectations, there will continue to be age strata in society. All societies divide the human life span into "seasons of life" (Hagestad and Nuegarten, 1985). Political Economy of Aging This theory picks up where age-stratification theory' leaves off Like age stratification theory, political economy of aging highlights structural influences on aging. The political economy of aging theory, however, emphasizes power relationships and

PAGE 28

14 social struggles that influence the lives of older adults (Estes, 1991). The political economy of aging theory also helps us to understand how the aged are defined and treated. According to the political economy of aging theor}', "old age" is socially constructed, and created through power struggles. One drawback to this theory is that it has no place for positive aspects in the lives of older adults. Focusing on the power and the unequal distribution of status and resources is beneficial (Bakes, 1993). However, advantages come with older age, which the political economy of aging theory does not recognize. These six theories have contributed a great deal to the study and understanding of aging. In those instances where a particular theory' falls short of reflecting some of the patterns of behavior in social life, researchers in aging have been able to build on them and develop other plausible explanations. Examining these theories helps us understand them as independent explanations of age-related phenomena. Theories of Aging Families In the 1990s, a shift occurred in the complexit>'. malleability, and variety of family connections in the second half of life, generating a more complex view of older families. Theoretical advances related to the specific study of older individuals and families brought greater focus to intergenerational relations over the entire life span (DilworthAnderson. 1996; Allen et. al., 2001). Some of the prevailing theories are described next. Family Development Theory Family development theory has been used to study predictable or normative changes in families over time. Family development theory' relies on basic assumptions about the family, and the processes the family goes through over time. It is a chronological, linear model, based on stages of family life. Family development theory

PAGE 29

15 commenced from several theoretical perspectives (life-cycle categories, social-systems theories, human-development theories, life-span and life-course theories, and life-events and hfe-cycle theories) dating as far back as the early nineteenth century (Mattessisch and Hill, 1987). Family development theorv' gained strength as a theoretical perspective from which to study the family in the 1950s. Hill and Mattessich (1979; p. 174) proposed the following definition of family development theory: Family development refers to the process of progressive structural differentiation and transformation over the family's history, to the active acquisition and selective discarding of roles by incumbents of family positions as they seek to meet the changing functional requisites for survival and as they adapt to recurring life stresses as a family system. Similar to the life-course perspective in the study of individuals, in order to investigate the changes families experience over time, a basic set of stages were created to set a framework for the family life span. These are called '"family life cycle stages." Duvall and Hill (1948) proposed a three stage family "career," consisting of "changes in family size, changes in age composition, and changes in the occupational status of the breadwinner." Since then, other theorists have further developed this idea. Sociologist Paul Glick was among the first to analyze families in terms of a life-cycle with five stages that spanned the life course from first marriage to the death of a spouse (Glick and Parke, 1965). One of the more commonly used developmental theories for family sociologists is an eight-stage model (developed later also by Duval in 1977) in which attention is paid to life stages of the children. Family development theory reached its height of development in the 1960s. This theory' has given us important insight into the complexities of family life. Family development theory' is useful as an indicator of changes that occur within family life, and calling attention to the changing nature of families, family formation, and family

PAGE 30

16 relationships (Schwartz and Scott, 2000). However, a long-standing critique is that this framework cannot incorporate the variety of family forms that exist in today's society. Distinguishing among different types of families is difficult with this model. Moreover, this model perpetuates the notion that nontraditional families are pathological in their development. The family has evolved, and family forms are much more diverse. Couples who remain childless, are not represented in this model. The family development model also assumes that the wife does not work until the children are old enough to attend school. This does not reflect current trends in motherhood. Of all women who return to work after having children, 65% have children less than 6 years old. This percentage is higher for African American women (U.S. Bureau of the Census, 1998). These family life-cycle stages also do not consider the possibilit>' of stay-at-home fathers. While the numbers remain relatively low, a growing number of men choose to stay home and raise children. In 1997 there were almost 3 million father-only families, an increase of 220 % since the mid-1970s (Weldon, 1997). Families that provide care for their elderly parents are also not represented in the family development model. Often, families that have elderly parents residing with them also have their own children still at home. This ''sandwich generation" is becoming more common, as the cost of caring for older family members continues to increase. In addition, adult children are remaining or returning home for economic reasons. This has been termed the "cluttered nest." This arrangement has significant effects on family role expectations and relationships (Bengtson, et. al., 1996). Finally, the current set of stages allows no way to account for divorced, single headed households, or couples who re-many navigate the stages in family life. For

PAGE 31

17 example, a couple that remarnes, already having children from one or both of their previous marriages, never experience time alone with the spouse. At the very least, there is no "stage 1 adjustment period." Several studies have shown that imposing the expectations of the traditional family onto remarriages leads to additional problems and higher rates of divorce (Bray et al. 1998). Other researchers have identified fundamental ways in which the remarried family differs from the traditional nuclear family (Beer, 1989; Papemow, 1993). Another criticism of family development theoryis that this framework does not allow for intracultural differences in family forms; and the scope of family development theory does not extend to cross-cultural differences in families. For example, distinct catalysts in the transformation and structure of African American families as a result of the shift from rural to urban living have "manifested in spousal relationships, family formation, the socialization of children, and techniques of coping with dissolution, especially divorce" (Wilkinson, p. 42). Mexican American families" concept of family refers to more than just the nuclear family. The average household size of Mexican American families tends to be large, "Mexican Americans are more likely to live in an extended family context" (Sanchez, 66). In addition, the socialization of children is very different for Mexican American families. Adolescence marks different behavior patterns between boys and giris (Locke, 1 992). Social Exchange Theory While social exchange theory was previously discussed as an aging theory, it clearly has relevance in family studies. This theory is useful in studying changes in the family over time. The premise of social exchange theory is that all social interaction stems from social exchange among individuals. The work of the prominent social

PAGE 32

18 exchange theonsts did not originally focus on family issues. However, "since the 1960s, the exchange framework has methodologically worked its way into the mamstream of family studies" (Sabatelli and Shehan, p. 396). The development of social exchange theory can be traced to the works of George Homans and Peter Blau, and the primary assumption is that "humans choose between alternative potential relationships, or behaviors by ranking the actual or expected experiences associated with each and then selecting the best alternative" (Blau, 1964). Individuals are more likely to pursue relationships with those who possess desirable and valued attributes. Different resources and exchange processes are at work throughout family life. Thus, family exchange theory makes a valuable contribution toward helping us understand the variety of family behaviors and relationships over time. Intergenerational Stake Hypothesis Making use of both a developmental and a social exchange framework, another ideological structure useful in understanding family relationships over time is the "intergenerational stake hypothesis." This hypothesis was developed by Bengtson and Kuypers in 1971, as a way to understand and explain disparities in perceptions of intergenerational relationships among parents and children. Specifically, the premise of Giarrusso, Stallings and Bengtson's (1995, p. ) hypothesis is: Contrasts in developmental needs explain differences in the '"stake" each generation has in intergenerational cohesion, continuity and conflict. Furthermore, these differences in developmental stake explain contrasts in opinions and orientations between the generations, independent of any specific issues that appear to be the focus. While a plethora of studies have used the stake hypothesis to explain empirical findings, clarifications have been made to account for some deficiencies. These include the application of two theones developmental theory and exchange theory. Using the

PAGE 33

19 stake hypothesis, these theories provide a longitudinal framework to analyze intergenerational stakes that exist in families over time (Giarusso. Stallings and Bengtson. 1995). SocioemotionaJ Selectivity Theory Socioemotial selectivity theory is one of the most important theories to emerge in family gerontology in the last decade. This theory helps to explain the negative correlation between the frequency of social interaction and the degree of emotional closeness among family members as they age (Carstentsen. 1992; Allen et. al., 2001). As the functions of relationships change over the life course, adjustments to the intricacies of social networks among family members help "maximize social and emotional gains" (Allen et al., p. 134). Older people's preferences for emotionally meaningful social relationships are due to perceived limitations of time. Older adults become more selective about their social networks. In other words, older individuals become more selective about whom they spend time with, and with whom they share their emotions, as they age. This has limitations for studying familial relationships. Family relationships are too complex to assume that in later life, older adults have control in choosing to engage in interactions with family members. Social Comparison Theory One of the more important factors to be found in midlife and later life (in terms of navigating a successful transition into later life stages) is the notion of social comparison (Neugarten, 1968; Levinson & Levinson, 1996; Carr, 1997; Stewart & Vandewater, 1997; Carr, 2003). Social comparison is rooted in psychology and psychosocial behavior; and is a way of self-evaluating based on comparisons with other people, particularly with those individuals most significant in one's life (Festinger, 1954; Suls &

PAGE 34

20 Mullen. 1982; Sulls & Wheeler, 2000). This comparison and self-evaluation has important consequences on self-esteem, and this is typically how social comparison is applied in the psychology literature. However, it is important to recognize that an alternative ramification of social comparison may be its effects on relationships between individuals and the people they compare themselves to. Social comparison theory originally emerged in the early 1950s when researcher Leon Fesginger determined that it was important to people to hold correct assumptions about the worid in which they live, but more specifically about their own abilities in this world (Festinger, 1954). Festinger's hypothesis focused on the idea of social comparisons as a need for self-evaluation. However, researchers later revealed other motives for social compansons (Hakmiller, 1966; Carr, 2003). In general, the social comparison hypothesis suggests a positively correlated relationship between evaluation with another person and the effects of this evaluation on self-esteem and emotional well-being (Stouffer, 1948; Festinger, 1954; Suls and Mullen, 1982; Carr, 2003). Thus, a comparison which yields a positive evaluation, will result m positive effects on self-esteem and positive social well-being, and vice versa for a negative evaluation (Carr, 2003). Ultimately, grounded in the psychological literature, these comparisons and selfevaluations are used to measure self-enhancement outcomes in individuals (Suls & Wheeler. 2000; Carr, 2003). This is beneficial for psychologists. However, from a sociological standpoint, particulariy in the interest of intergenerational family relationships, and more specifically, in terms of evaluating social change and women's

PAGE 35

21 intergenerational relationships, social comparison could be useful when evaluating its effects on perceptions of interpersonal relationships between generations. Understanding Social Change and Women's Intergenerational Relationships While there are several meaningful and useful theories in the aging and family literatures, this study will employ two specific guiding frameworks, Intergenerational Family Solidarity Theory, and The Developmental Schism. Both are important in understanding how social structural change and changing life course trajectories for women over time have effected daughters" perceptions of intergenerational solidarity with their mothers in different life stages. Following is a discussion of each of these frameworks, and how each contributes to this study. Life Course Theory and the Developmental Schism. There has been a resurgence of the life-course perspective. Elder (1998) explained that this theoretical framework, with its focus on sociohistorical processes, and on individual time, satisfies two increasingly important considerations in studies of families: 1) how individuals change over time, and 2) how their transitions and trajectories are linked across family members. This perspective has been combined with other theones to help explain many facets of aging families (Elder, 1998; Allen et. al., 2001). The life course perspective is a broad multidisciplinary perspective. It is more of a way of thinking, or a paradigm, rather than a specific or explicit explanation of anything (Elder, 1998). Also, the life course perspective allows for the understanding of the impact of a multitude of social and historical factors and expenences on human development. This makes it an ideal perspective to understand the continuous process of aging related phenomena and the impact of social change.

PAGE 36

22 Building from a life course perspective, Fingerman (1996) introduced the concept of "developmental schism." A perspective grounded in developmental psychology, this accounts for a key source of interpersonal problems between female intergenerational dyads, which tend to be the strongest and most lasting of filial relationships (Allen, et. al., 2001). Fingerman provides explanations of the mother/daughter bond at three levels of descending analysis: sociopolitical, familial, and psychological. She argues that social and political factors influence the strength of a mother/daughter relationship. However, these factors do not adequately explain the emotional content of their relationship. Moreover, they do not account for the investment of mothers and daughters in their mutual relationship (Fingerman, 2001). The central idea of the developmental schism hypothesis is that the tensions between parents and children arise due to the fact that these different generations are at different developmental stages. As a result, each generation has different goals, and varN'ing needs, which result in a difference of perspective. A disparity of roles, status, and cohort membership influence the development of different perspectives as well (Fingerman, 2001). While the developmental schism is psychologically based, it is takes the life course perspective to a more specific level of analysis, focusing on the role, status, and goal needs, which may be specific to cohort membership. The developmental schism is useful for this project in that it recognizes the variety of aspects that are present in complex interpersonal relationships. Just as Bengtson's intergenerational family solidarity theory reflects behavioral, emotional and cognitive aspects of solidarit>through its various empirical measures, Fingerman's hypothesis accounts for aspects of behavior and emotion as well. In addition, Fingerman is clear

PAGE 37

23 about the significance of the social context and it effects on the mother/daughter relationship. Interaenerational Family Solidarity Theory. Intergenerational Family SoHdarity Theory is another major theoretical contribution to the area of family gerontology. As a result of the diversity in function and structure, intergenerational relationships become more important. Research from the Longitudinal Study of Generations has been used to demonstrate the strengths of multigenerational ties over time, and why it is necessarvto look beyond the nuclear family when asking whether families are still functional (Bengtson, 2001). Interpersonal relationships among family members of different generations are assessed based on six measures of solidarity: associational, affectual, consensual, functional, normative, and structural (Bengtson and Roberts. 1991). These six components of solidarity capture the numerous aspects of family interactions. Included in these aspects of interaction are frequency of contact, emotional closeness, agreement with regard to values and attitudes, helping and exchanging of resources, filial obligations and commitment, and physical family structure. Table 2. 1 provides a detailed table describing these six measures. Using these six dimensions of intergenerational family solidarity, five different family typologies emerge. These typologies are based on where the types of solidarity among intergenerational relationships are strong or weak, and thus help to explain the myriad of families functioning in our society in a variety of ways (Silverstein and Bengtson, 1997). The five types of family relationships are: 1) tight knit, 2) sociable, 3) intimate but distant, 4) obligatory, and 5) detached (Bengtson, 2001). Breaking down the overall idea of solidarity, or cohesion among family members, into manageable aspects,

PAGE 38

24 can then be empirically measured and has been a major contribution to the study of aging families. It is important to note that intergenerational family solidanty theory focuses on solidarit)', with little understanding of conflict (Connidis, 2001). In this study, intergenerational solidarity forms the basis of analysis for understanding women's intergenerational relationships, by utilizmg the empirical outcomes of this theory. Three of the six measures, associational. Effectual, and consensual solidarity', are used to gain a better understanding of how daughters perceive their relationships with their mothers. By using these three measures of solidarity, interaction, emotion and cognitive aspects of the relationship are captured. Analyses will measure these perceptions at two life stages, and in each life stage, at two points in time. Chapter 4 gives a more detailed description of how these measures are used. Conclusion In summary', various theones on aging families suggest different outcomes in later life stages. The framework guiding this proposed study is the developmental schism, stemming from a life course perspective. The developmental schism predicts that parentchild relationships converge, as children become adults. To help understand this phenomenon, this study will use Bengtson's measures of intergenerational solidarity, which stem from intergenerational family solidanty theory. The developmental schism will be the foundation for understanding the developmental hypotheses. The impact of social and structural factors on mother/daughter relationships, and the complexity of these relationships is explained by the developmental schism as a result of mothers and daughters being at different developmental stages, and having different goals and needs (Fingerman, 2001). While the mother/daughter dyads in this proposed study are at the same developmental stage as defined by age young adulthood and midlife social

PAGE 39

25 change suggests thai the developmental stage has been affected. Each generation has a different perspective, affected by their roles, status, and birth cohort, thus having different goals and needs during the same life stages.

PAGE 40

CHAPTER 3 REVIEW OF THE LITERATURE Introduction The average life expectancy in the United States has increased a great deal over the last century. This increase has led to multiple generations of families living longer together. As a result, there have been new family dynamics ansing as well. Researchers have been interested in how these new dynamics have affected family life. In addition, several social structural changes occurring dunng the twentieth century have affected the way individuals live out their lives. This chapter discusses the changes in social structural forces, outlining some of the social ramifications and their impact on individuals as well. The focus then turns to an overview of the research on intergenerational family relationships, then specifically on women's intergenerational relationships. Finally, the hypotheses for this study are outlined at the end of this chapter. Changes in Social Structural Forces The focus of this next section is to outline some of the major legislative, economic, demographic, and ideological changes that have occurred. These changes have helped to reshape the life course trajectories of women in this country over the last century. Legislative Changes The Equal Pay Act of 1963, and Title VII of the Civil Rights Act of 1964, which prohibits sex discnmination in employment, were the first major legislative steps toward ensuring equality in the workplace (F elder, 1 999). Congress passed Title IX of the 26

PAGE 41

27 Education Amendments of 1972, prohibiting discrimination on the basis of sex in federally funded education programs (Stetson, 1992; Langley, 1994). These laws opened up doors for women, providing expanding opportunities in the educational arena and in thejob market. In 1960, the Food and Drug Administration (FDA) approved the birth control pill, and in 1973, abortion was legalized (Stetson, 1992; Langley, 1994). These two landmark decisions afforded women opportunities of choice not previously available in terms of family planning and decisions about child bearing. These legislative changes improved women's access to education and occupational attainment (see discussion below), and are reflected in the changing roles of women in other structural domains (i.e., marriage and family). These changes have had an impact on American society's attitudes concerning women's roles as well. Education and Occupational (Status) Attainment The number of women with a college degree has greatly increased during the second half of the last century. Since mid-century, the number of women graduating from high school and enrolling in college rose by twenty percent. The proportion of higher degrees being awarded to women also increased significantly, by an average of thirtv' percent for each degree level (U.S. Department of Education, 1999). The social trends in education have led to cohort differences in education, in that the younger groups of women are more likely to be more highly educated than past cohorts of women, often their mothers. Changing Work Roles As a result of legislative and social changes women across all ages are more likely to be employed (Moen, 1992). Americans have achieved greater social mobility over time. Biblarz, Bengtson, and Bucur (1996), also find this to be true with occupational

PAGE 42

28 attainment within family lineage. However, the rate of upward mobility has slowed down with each succeeding generation. A review of the second half of the century presents the expanded range of life choices available to women (Gerson, 1984; Rexroat and Shehan, 1984). As a result of both continued persistence, as well as ongoing economic growth, women continued to gain ground. Today, the majority of women are employed, and make up close to half of the workforce (U.S. Department of Labor, 2000). As a result of educational opportunities, the occupational choices for women have expanded. Many more women hold jobs, and in a much greater variety of occupations than in the past. A strong presence of women has been established in previously maledominated professions (Opdycke, 2000). During the second half of the century, women have spent a large proportion of their adult lives employed. Each generation of women return to employment more quickly or never leave it after having children (Moen, 1992). The number of married women with children under age six in the workforce rose to almost 70 percent by 1997. Moreover, over half of all women with infant children are working (Moen, 1992; Hattery, 2001). Divorce, single parenthood, economic need, occupational opportunity and the afifordabilitv' and availability of childcare are some of the social forces directly associated with maternal employment (Hoschschild, 1 997; Brayfield, 1 995). More than half of all women have a desire to successfully combine family life and work life (Roper Organization, 1995). The mid-1970s also saw a severe decline in men's wages. Many women decided to go to work simply to make family financial ends meet, and to maintain their desired standard of living. By the end of the twentieth century, the dual-income family became

PAGE 43

29 the norm (Opdycke, 2000). During the 1970s, there was also a rise in the number femaleheaded households contributing to the growing female workforce. Family Formation In 2000, the average age at first marriage was 25. 1 years for women. In the mid1950s, women were slightly over the age of 20 when they married for the first time (20.2). Just as the age at marriage has increased over the past 50 years, the proportion of women in their twenties and early thirties who aren't married has also increased. In 1960. twenty-eight percent of women aged 20 to 24 had not yet married. By 2000, this had increased to 73 percent (Fields and Casper, 2001). Fertilitv' rates have declined significantly since the post war. Baby Boom generation. The number of babies being bom per woman dropped significantly from an average of 3.69 in the 1950s to 1 .84 in the 1970s. In 2000, the number rose to an average of 2. 1 3, which was the first time in thirty years that the number of births was above replacement level (Martin et. al., 2000). Women have begun to delay having children in order to complete education and establish careers (McMahon, 1995). The median age of first time mothers has risen consistently over the past thirty years. The median at first birth increased to 24.6 years by the turn of the century (Martin et. al., 2000). In the 1950s, dunng the baby boom, most women were having children in their eariy 20's. This increased to the late 20's in the 1960s. Then m the 1970s and 1980s, a large number of women were giving birth for the first time in their eariy to mid 30"s, with a growing number of women having children in their late 30"s and even early 40"s (Moen, 1992; Martin et. al., 2000).

PAGE 44

30 Mantal Stability The United States has also experienced a dramatic rise in the rate of divorce. Sociocultural factors, like the increased economic independence of women, changing family functions, the changing nature and high expectations of marriage, fewer social. legal and moral constraints, have all contributed to a greater acceptance of divorce (White. 1991; Kurdek, 1993). From 1965 to the mid-1980s, the divorce rate doubled. By the end of the eighties, the divorce rate has stabilized at this higher level, where it remains today. The implications of divorce are seen in the high percentage of women in the workforce who have no other means of economic support (Moore and Sawhill, 1984; Cherlin and Furstenberg, 1988; Kurdek, 1993). Ideological Shift The economic period following World War II reemphasized the idea of the stay at home mom, by providing unprecedented economic stabilit>for families. Thus, the intensive motherhood ideology became dominant at mid-twentieth century (Hays, 1996; Reskin and Padavic, 1994; Coontz, 1992). However, there were significant racial and ethnic differences in women^s ability to adhere to this ideology (Segura, 1994; HillCollins, 1994). More recently, as a result of changing social and structural forces, attitudes regarding appropriate roles for women in the family have changed to reflect nonegalitarian ideals (Hays, 1996; Margolis, 2000). Today, social perceptions reflect the belief that working women can be good mothers, and especially those women who work out of personal desire, rather than solely out of economic need (Hattery, 2001; Hays. 1996; Klein, 1984). These social changes have led to disparate life course trajectories for women of different birth cohorts. In the 1950s, women of the late Worid War II Generation and the

PAGE 45

31 early Baby Boom Generation were in their young adult and adult years, getting married and starting a family at a much younger age compared to the Generation X birth cohort who were entenng the young adult years in the 1980s. The Baby Boom Generation was the first birth cohort to take advantage of the educational opportunities presented by Title IX in the 1 970s. in addition, women bom later in the late Baby Boom years began postponing having children until their late thirties. Women of the Baby Boom generation also saw higher rates of divorce in the 1980s. The majontv' of Generation X women reach their 25* birthday still single, and are postponing childbirth until they are in their thirties. Educational and occupational opportunities, as well as social pressures, have led to the idea of a "career," where as older generations of women may have held 'jobs" if it was financially necessar>', but family life was the woman's mam priority. Intergenerational Family Relationships Similarit\' of beliefs and values between parents and children has long been recognized as an important source of stability in society. Indeed, the transmission of sociolpolitical ideologies from one generation to the next permits continuity within families (Miller and Glass, 1989). However, the degree of parent-child similarit>' at any give point in time for any particular age cohort may be the product of contextual processes (Dannefer, 1984). Period effects that have a different impact on different cohorts and/or age groups may either separate the sociopolitical interests of different generations or draw them closer together (Miller and Glass, 1989). Rossi and Rossi's (1990) classic study of the life course and intergenerational relationships (specifically parent-child relationships), indicates that as children grow older, new factors become relevant for intimacy. The once central indicators like parental affection, parental authority, and family troubles, shift to role and status

PAGE 46

32 attainment markers such as finishing school, occupational attainment, marriage, and children (Rossi and Rossi, 1990). When the younger generations don't share similar roles or statuses, there is more intergenerational strain. For example, if the younger generations are not married, and/or don't have children, the difference in these statuses from their own parents create more of a strain in terms of a general understanding of life experience. When children, or the younger generations, move into normative roles, they share similar experiences, bringing the generations closer (Bengtson and Black, 1973; Aquilino, 1997). While the "role" might be the same, the experience of the "role" or the ramifications of that "role" are different for the succeeding generations as a result of the changes in social and structural forces. This is especially true for women. Thus, even when two generations hold similar roles there maybe intergenerational conflict. The nature of parent-child relationships in earlier years sets the stage for these relationships in later life. There is evidence of strong continuity in the perception of relationship quality in relationships from early years into adulthood. From adolescence to young adulthood, past relationship patterns continue to be played out. The effects of this are the strongest in measures of emotional closeness (Roberts and Bengtson, 1993; Aquilino, 1997; Thornton et. al., 1995; Tubman and Lemer, 1994; Rossi and Rossi, 1990). That is to say, emotional bonds tend to remain more stable over the life course, compared to other aspects of intergenerational relationships, such as frequency of interaction, shared values and world views, exchange of resources, meeting familial obligations, and geographic proximit>' of family members. Intergenerational Differences Among Women It was previously stated that the mother/daughter relationship is unique and significant. The strength of this bond sets it apart from other social relationships

PAGE 47

33 (Fingerman, 2001). There are several factors that contribute to the complexity and the strength of the mother/ daughter relationship. Gender socialization and gender differences, a shared status in society, and changes in the larger social structure allow for many distinct differences often apparent in mother/daughter relationships. As gendered beings, women's relationships lend to be traditionally characterized by stronger emotional bonds. In addition, for women, the matemal role tends to be more psychologically salient for women, as they have been taught the importance of motherhood from a young age. These psycho-emotional elements play a key role in the mother/daughter relationship. Despite geographic obstacles, women tend to maintain strong ties to their parents (Bedford & Blieszner, 1997; 1995; Gottman, 1998). The strong filial ties daughters maintain to their parents are also a result of gender socialization. While young males are encouraged to gain independence as part of childhood socialization, young girls are socialized to remain close with their mothers (Chodorow, 1994). In addition, daughters rather than sons are more likely to provide later life care for their aging parents. The turn of this century' saw women over 60 years of age as far more likely to have an aging parent still alive compared to women at the turn of the last century (Moen, 2001). Life expectancy for women at birth is 4 to 12 years longer for women than for men. With women living longer, the likelihood that women will end up widowed and not remarry is probable. Hence, women who do not have a partner in later life will tum to their children, particularly their daughters for much of the support they need. Despite a more patnarchal culture, it is the women who tend to be central to the kin networks in our societv'. Women often have great insight into the larger family structure.

PAGE 48

34 and individual family members. The mother/daughter relationship is at the core. Thus, those families with strong mother/daughter connections often foster stronger ties among other family members. With marriage, women tend to bnng their husbands into their family of origin. As a result of the changing life course trajectories for women outlined previously in this chapter, relationships among women in general have been also been affected. With an increase in singlehood and divorce, as well as a decrease in fertility among women in successive generations, women are more apt to face aging alone (Fingerman, 2001; Moen, 2001). Continuing relationships with their own aging parents, particularly their mothers, likely becomes more important. Thus, just as aging mothers will to turn to their daughters in later life, as women age, they in turn are will rely on their mothers as well. Statement of Hypotheses This section details this project's hypotheses. The hypotheses HIa through Hlc will each be assessed in chapter five (young adulthood). Hypotheses H2a through H2c will each be assessed in chapter six (midlife). Generation X and Babv Boom Daughters During Young Adulthood In order to answer the first research question, "How have daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during young adulthood changed, and what may be associated with this change?" several hypotheses will be tested. These hypotheses are outlined below: • HIa: Young adult daughters from Generation X have stronger (more positive) perceptions of associational and consensual solidarity compared to their Baby Boom mothers when they were young adults. • HIb: Affectual solidarity will not be affected by the daughters" social characteristics the same way associational and consensual solidarity will be.

PAGE 49

35 • Hlc: Social distance between the Generation X daughters and their Baby Boom Generation mothers comprising each mother/daughter dyad is associated with associational and consensual solidarity, but not with affectual solidarity. Baby Boom Generation and World War II Generation Daughters During Midlife To answer the second research question, "How have daughters' perceptions of associational. affectual. and consensual solidarity with their mothers during midlife changed, and what may be associated with this change?" the following hypotheses will be tested: • H2a: Dunng midlife, daughters from the Baby Boom Generation will have weaker (more negative) perceptions of associational and consensual solidarity compared to their World War II Generation mothers during midlife. • H2b: Affectual solidarity will not be affected by the daughters" social characteristics the same way associational and consensual solidarity will be. • H2c: Social distance between the Baby Boom Generation daughters and their World War II Generation mothers comprising each mother/daughter dyad is associated with associational and consensual solidaritv', but not with affectual solidarity. The above hypotheses focus on the importance of the larger social structural changes, and from a life course perspective, the changing roles of women as they move through the life course stages. These hypotheses reflect the complexity of the mother/daughter relationship over time. In addition, they explore the different aspects of intergenerational relationships and how they may be affected. The next chapter will provide a descnption of the data, sample, vanables. and analytic models used in the analysis of this project.

PAGE 50

CHAPTER 4 DATA AND METHODS Introduction For this study, analyses focus on generational differences during two life course stages. Generation X daughters' perceptions of their relationships with their mothers during young adulthood in 1994 are compared to Baby Boom daughters" perceptions of their relationships with their mothers during young adulthood in 1971. and Baby Boom Generation daughters" perceptions of their relationships with their mothers during midlife in 1994 are compared to World War II Generation daughters" perceptions of their relationships with their mothers during midlife in 1971 This study uses the Longitudinal Study of Generations and Mental Health. 1971 data set [made accessible in 1990, machine-readable data files]. These data were collected by V. Bengtson and M. Gatz and are available through the archive of the Henry A. Murray Research Center of the Radcliffe Institute for Advanced Study, Harvard University. Cambndge, Massachusetts (Producer and Distributor). These data are from an ongoing study of aging families investigating changes in intergenerational family relationships. These data specifically examine intergenerational solidarity, based on perceived and actual levels of cohesion among family members, making it an ideal data set for this study. Outlined below is a more detailed description of the data set. Sample The Longitudinal Study of Generations and Mental Health (LSOG) is an ongoing 3 1 -year longitudinal panel study. The LSOG examines four generations of aging 36

PAGE 51

37 families, seeking to investigate several things, including the changes in family social supports and their impact on the mental health of family members. The sample came from a large prepaid health maintenance organization in Southern California. The original participants comprising the first generation in this study were randomly selected on the basis of potential grandparenthood. This sample represents the demographic climate of Southern Califomia at the time of the study's inception. The participants are generally white, and from middle class or economically stable families. Originally, the study began in 1971 as a cross-sectional study, and included 345 three-generational families. Although it was not mitially intended as a longitudinal study, a second wave of data were collected thirteen years later in 1984-85, after the researchers serendipitously came across the participant files from the original study. Additional waves of data were collected every three years after that (1988, 1991, 1994, 1997, and 2000). The three generations were followed through the first three time periods. At the fourth period of data collection in 1991, the fourth generation of these families was included in the study for the first time. Because 1994 was the most recent wave of data available through public archiving at Harvard, this research includes data from 1971 and 1994 time periods of the LSOG. Using the standard distinction in aging family literature, the generational cohorts of women are composed as follows; 1) the first generation, GTs, who are the mothers of the G2"s, grandmothers of the G3's and later great-grandmothers of the G4's, 2) the second generation, G2"s, the mothers of the G3"s and later grandmothers of the G4's as well as being the daughters of the GTs, 3) the third generation, G3's, the granddaughters of the Gl 's and the daughters of the G2's and later mothers themselves of the G4's, and 4) the

PAGE 52

38 fourth generation. G4's, who are the great-granddaughters of the Gl s, granddaughters of the G2's, and daughters of the G3's. Three sets of mother-daughter dyads, representing four generations in the same family, will be examined in this study. The oldest generation, the GFs, consists of women bom prior to the Great Depression some during the late nineteenth century and who were coming of age before World War II (1888-1927). The second generation, the G2"s, were all bom after the turn of the twentieth century, some during the Great Depression, but before World War II (1912-1941). The third generation, the G3's, were all bom after World War II, and represent the first of the Baby Boom generation (19441956). The fourth generation, the G4's, somewhat overlap with the Baby Boom women, they were bom between the years 1954 and 1979. The age range for the women in the LSOG sample, and their movement as they age, through the social structural changes previously discussed as occurring in the twentieth century, can be visually interpreted by refernng to figure 1 1 in Chapter 1 The Generation X women were moving through the young adult life stage during the eighties and early nineties. Their Baby Boom mothers expenenced young adulthood from the late 1950s to early 1970s. The Baby Boom, as they move through the life course, began experiencing midlife in the late eighties-early nineties, continuing through today. Their World War II generation mothers, however, experienced this life stage as early as 1950 through to the early eighties. Sample Selection Characteristics There were several steps involved in preparing the LSOG data for this study. For the analysis of daughters during young adulthood, only those Generation X and Baby Boom generation women age 1 8 to 29 at their respective time periods (1994, 1971) were

PAGE 53

39 selected for this study. For the analysis of daughters during midlife, only those Baby Boom and World War II generation women age 40 to 59 at their respective time periods (1994, 1971) were selected for this study. After the women for each set of analyses were selected based on the appropriate age, they were then selected based on whether or not they responded to the specific questions which comprise the dependent variables, associational affectual, and consensual solidarity. Those women who did not answer any of the questions pertaining to the solidarity measures were eliminated from the analyses. In addition, once the women were selected on the basis of having responded to the solidaritv' questions, it was necessary' to ensure that only one daughter from each family, be included in the analyses to eliminate intrafamilial bias. First, all of the women who had no sisters were automatically included. Of the remaining women, one sister from each family was randomly selected. Finally, the last step for the sample selection of women to be included involved matching these women with their mothers. Some women were initially interviewed for the LSOG because their fathers were participants, and some women did not have daughters of their own, but rather sons. If the daughter did not have a mother who also participated in the LSOG, they were excluded from the analyses. The numbers presented in Tables 4. 1 and 4.2 represent the actual number of women m the sample from each generation included in this study. Because there were several critena for being included in the analyses for this study appropnate age at the time of data collection, necessary responses to the solidarity questions representing the dependent variables, having only one daughter represented from each family, and a match to a "study mom" ~ and the fact that the LSOG sample size is relatively small to begin with, the analytic samples are somewhat small.

PAGE 54

40 Daughters During Young Adulthood Table 4. 1 illustrates the numbers of women from both generations during young adulthood, and how these numbers decrease as a result of the sample selection process. The original sample of Generation X daughters, the G4's in 1994, contains 200 women. Once the women are selected on the basis of age, 18 to 29 years, for the young adulthood analyses, this number drops considerably to 136. Of the Generation X women in this age range, there were not any excluded on the basis of no response to the dependent variable questions regarding solidarity, thus the number remains at this stage at 136. However, when randomly selecting only one daughter from each family, the number of women again drops dramatically to 88. Finally, when selecting on the basis of a match to a study mother, the number drops to 82. Similarly, the Baby Boom daughters, the G3"s in 1971, start with an overall sample size of 442. The most significant drop among this sample is when selecting for the young adult analyses. The sample size drops to less than half, 185, when only those women 18 to 29 are included. There were six women who did not answer any of the solidarity questions, thus, the sample dropped to 179 at this stage in sample selection. When randomly selecting only one daughter per family, there is a drop again to 135. Finally, based on matching these daughters to a study mother, the number of Baby Boom daughters in young adulthood included in this study is 124. Daughters During Midlife The numbers of women in both generations during midlife are presented in Table 4.2. The number of Baby Boom daughters, the G3's in 1994, onginally participating in the LSOG is 410 women. For the midlife analyses, this number drops to 230 women age 40 to 59. Only three women were excluded as a result of not answering any of the dependent vanable solidarity measures, reducing the sample at this stage to 227. After

PAGE 55

41 randomly selecting only one daughter from each family, the number is reduced to 150 women. Ultimately, after matching each daughter to a study mother, an additional 42 are eliminated, and the final sample size for Baby Boom daughters in midlife is 108. There are mitially 377 World War II women, G2's in 1971, in the LSOG study. When selecting daughters based on age for the midlife analyses, the sample is reduced to 259 women. Several of the World War II daughters were eliminated from the analyses based on not responding the questions comprising the measures of solidarit\', further reducing the sample to 209. The number drops to 1 82 once only one daughter is randomly selected from each family. After matching the daughters to a study mother, the final sample included 143 World War II generation women in the midlife analyses. Measurement Various measures were constructed in order to model the relationship between sociodemographic variables pertaining to education, work, and family, and perceptions of solidanty. Because certain questions were not asked of all LSOG participants at each time period, specific measures were selected that were comparable across the two time periods, 1971 and 1994, when asked of each generation of women. The measures of solidarity used for the dependent variables are measures previously used in many empirical studies, and are described next. Dependent Variables The dependent variables in this study are those variables which measure the three aspects of intergenerational solidarity being examined: associational solidarity, affectual solidarity, and consensual solidarity. Nominal definitions and empirical indicators used to measure these constructs are found in Table 4.2 (Bengtson and Schrader, 1982; McChesney and Bengtson, 1988; Bengtson and Roberts, 1991). The specific measures

PAGE 56

42 for this proposed study will use Bengtson's constructs of intergenerational solidarity (Mangen. Bengtson and Landry, 1988) and are outlined in the next section. Appendix A includes a detailed matrix describing the wording for each question asked of each generation at each time period, and mcludes all response categories. These measures have been selected, in part, because of comparability across the LSOG study waves and the generations of women included in this research. They were also chosen because within the generational dyads, a variety of behavioral components associational, affectual. and cognitive are reflected by these three aspects of family solidarity (Bengtson and Roberts, 1991). Associational solidarity Associational solidarity is the amount and frequency of interaction among family members. The more contact or interaction among family members, the more solidarity within the family based on this solidarity aspect (Silverstein and Bengtson, 1977). This study uses the frequencv' of vanous types of interaction face to face, telephone, and mail to determine the level of associational solidarity among these four generations of women. For each of the three types of interaction, the following question was asked, "During the past year, how often were you in contact in person, by phone, by mail with your mother?" These questions were asked of the Gl 's through G3"s in 1971, and GTs through G4's in 1994. However, the response categories at each time penod differ somewhat (see Appendix A). Each question was based on a scale ranging from 1 to 8, with 1 being never or not at all, and 8 being daily or almost every day. The three measures of associational solidarity combine to make an overall measure with a range of 3 to 24.

PAGE 57

43 Affectual solidarity Affectual solidarity refers to the level of emotional closeness that exists among family members. It is the perception of "positive sentiments" warmth, understanding, trust, etc... family members have for each other (Bengtson and Roberts, 1991). The LSOG asks a set of questions, relating to affect among family members. These questions are as follows: 1) "Taking everything into consideration, how close do you feel is the relationship between you and your mother?" 2) How is communication between you and your mother, that is how well can you exchange ideas or talk about thmgs that really concern you at this pomt in your lives?" 3) Overall, how well do you get along with your mother at this point in your life?" 4) "How well do you feel your mother understands you?" and 5) How well do you feel you understand your mother?" These five questions were asked of theGl's through G3"s in 1971. and Gl's through G4"s in 1994. Each item is measured on a 6-point scale ranging from "l=not very much" to "6=extremely." By incorporating these five questions into one scale, the measure of affectual solidanty becomes a continuous variable with a range of 5 to 30 by simply summing responses across the set of questions. Consensual solidarity. This aspect of intergenerational solidarity is the degree to which there are similar values and beliefs about world views among family members of different generations. Bengtson's construct of this aspect of solidarity is measured by a single item. The following question was asked of the Gl's through G3's in 1971, and the Gl's through G4's in 1994. "In general, how similar are your views about life to those of your mother?" This item is measured by a 6-point scale ranging from "l=extremely different" to "6=extremely similar."

PAGE 58

44 Independent Variables The independent variables in this study include those sociodemographic variables, which reflect the generational differences among these women as a result of changing social roles and role expectations with regard to education, work, and family. Not all of the women were asked the same questions at each time period. As a result, the independent variables are limited to those common questions pertaining to education, work, and family indicators. Furthermore, a relatively small set of the most salient variables pertaining to education, work and family have been selected because the sample size for this research has limited the number of covariates which could be included in the OLS regression analyses. Age. This is the daughter's actual age in years at the time of the LSOG data collection. For the analyses in this study, the ages range from 1 7-24 for the daughters during young adulthood, and 40-59 for the daughters during midlife. Even though the age range for the women is the same when comparing across the generations during each life stage, the ages with in these age ranges vary. Therefore, in the multivanate analyses, age is used as a control variable in each of the four models. Education level. This is the daughter's highest level of education at the time of LSOG data collection. In 1971, this variable was a categoncal variable with a range of 1 to 8. The lowest education category, 1, being "grade school (T' through 6^), and the highest, 8. being "a post-graduate degree (Ph.D., DDS, MD, JD). In 1994, the variable was again an eight category variable, reflecting the level of education at that time, however the answer categories varied from those offered in 1971. The variable ranged from 1 to 8, with the lowest category being "an 8"' grade education or less," and 8 being "a post graduate degree." For this research, the education vanable has been reconstructed

PAGE 59

45 to reflect whether or not the daughter has less than or equal to a high school degree, or more than a high school education. The reference group for the regression analyses is "more than or equal to a high school degree." Employment status. This variable is indicative of the daughter's employment status at the time of data collection. In 1971, this variable asked the women if they were workmg for pay. It was a 5 category, nominal variable, including "l=employed full time," "2=employed part lime," "3=employed full time with additional part time," "4=unemployed, seekmg pay." and "5=not employed (homemaker, retired, not seeking)." In 1994, this same variable was a 6 category variable. At this time, the categories mcluded "l=no, retired," "2=no, temporarily out of work," "3= yes, temporary job," "4=yes, part time," "5=yes, full time." and "6=no, housewife." For the purposes of this research, the employment status variables has been receded as a dichotomous variable, indicating whether or not the daughter is employed or does not work at the time of the LSOG data collection, thus comparing those women who work to those who do not. The reference group for this variable is "does not work." SEI score for occupation. This score is a numeric value assigned to the different occupations that the women have. The score represents the value for that occupation at each LSOG data collection period. Because not all of the women worked, several women did not have a score assigned to them, further reducing the sample size when this variable was run in the models. Therefore, for the purposes of this research, this variable was recoded as a nominal variable, with three categories, "not working, no score," "low SEI score," and "high SEI score." Low and high SEI scores were calculated using the median SEI score, with all women in the LSOG study included, for each year of data collection.

PAGE 60

46 In 1971, the median SEI score for women in the LSOG study was 45.00, therefore, those women included in the World War 11 Generation sample during midlife, and the Baby Boom Generation sample during young adulthood were coded as low SEI if they had a score below the median, and as high SEI if they had a score equal to the median or above. In 1994. the median SEI score for occupation for women in the LSOG study was 38.51. Women included in the Baby Boom Generation sample dunng midlife, and Generation X women during young adulthood were coded as low SEI if they fell below this score, and as high SEI if they had a score equal to or above this score. The reference group for this variable in the OLS regression analyses is "low SEI," thus a companson is made between the high and low sconng women. Income. This variable has been adjusted to reflect annual income in 1994 dollars, and represents the total household income for the daughters. There were several missing cases with regard to this variable. Therefore, mean scores for income were imputed so that that ever\' woman had an income score represented. Scores were imputed based on education and marital status, using the onginal LSOG categones for these measures to ensure an appropriate range of variability among the sample of women. Mantal status. This is the daughters marital status at the time of LSOG data collection, and represents the woman's current mantal status. In 1971 this measure consisted of 6 mantal status categories. They were "l=single, never married," "2=engaged," "3=married," "4=permanently living with someone, not legally married," "5=widowed." and "6=divorced or separated." In 1994, this same variable consisted of 7 categones. including "l=married," "2=separated," "3=divorced." '4=widowed," "5=live with as married," "6=stopped living with as married," and '7=single. never married."

PAGE 61

47 This variable has been reconstructed to indicate whether or not the daughter is mamed, hving as marned, or not married. There were no women among the World War II Generation daughters durmg midUfe who were hving together as married. For the multivariate analyses, the reference group for this variable is "not married." Age at first marriage. This variable reflects the timing of marnage for the daughters. It is not reflective of the actual age of the daughters at first marriage, but whether or not the daughter married early or mamed later, relative to the general social trends of the time. Because not all of the women were married, simply using the age created several missing cases, further reducing the sample size. Therefore, this variable is a three category variable including "not married," "married earlier," and "married later." For the multivariate analyses, the reference group "not marned." Age breakdowns for early and late were based on the average age of first marriage at the time of marrying age for these women. For the Baby Boom Generation and Generation X women during young adulthood, this was at the time of data collection, 1971 and 1994 respectively. For the World War II Generation and Baby Boom Generation During Midlife, this was the average age at first marriage when they would have been young adults, 1 97 1 for the Baby Boom Generation women, and 1 948 for the World War II Generation women. Among those Baby Boom Generation daughters who were married during young adulthood, none were married later. Thus the reference group for the young adult analyses is comprised of those daughters who married earlier, allowing for comparison of Generation X daughters who married early versus late in 1994, as well as those who were not married to those who married early. But for Baby Boom young adult daughters, the comparison is basically between those who were married (early) versus

PAGE 62

48 those who were not married. In the midlife analyses the reference group is comprised of those daughters who married later. Because there were no World War II Generation women who had never married, the comparison is among the married women, between those who married earlier versus those who marned later. Children. This variable indicates whether or not the daughter has children at the time of LSOG data collection, thus reflectmg a current status for this family measure. It is a dichotomous, yes/no variable. The reference group for this variable is "has children." Age at first child. This variable reflects the timing of childbeanng for the daughters who have children. It is not reflective of their exact age at first child, but rather did they have their first child earlier or later relative to the general social trends at that time. Because not all of the women had children, simply using the age created several missing cases, further reducing the sample size. Therefore, this was recoded as a nominal variable, including "no children" "had first child earlier," and "had first child later." The reference group for this variable is "does not have children." For the Wodd War II Generation women during midlife, age at first child is not available, therefore, the women without children are compared to women who have children. Age breakdowns for early and late were based on the average age at first child during young adulthood for these women. For the Baby Boom Generation and Generation X women during young adulthood, this was at the time of data collection, 1971 and 1994 respectively. For the World War II Generation and Baby Boom Generation During Midlife, this was the average age at first child when they would have been young adults, 1971 again for the Baby Boom Generation women, and 1948 for the World War II Generation women. For the young adult analyses, the reference group is comprised of those daughters who do not

PAGE 63

49 have children. In the midlife analyses, there is no distinction between timing of first child because there were no calculations available for the World War II Generation daughters in 1971. Therefore the reference group in this analysis is comprised of those women who have children, companng them only to those who do not. Social Distance Measures For the Generation X daughters during young adulthood, and the Baby Boom Generation daughters during midlife, additional analyses include examining the effects of social distance between these daughters and their mothers. The presumption here is that the distance between the daughter and their mothers has a unique effect, and not only the daughters" absolute education, work or family status. Again, due to the small samples of women used in this research, the models for the social distance measures are interaction effects between the daughters" characteristics and their mothers" characteristics. The variables used in these analyses are as follows. Education level. The social distance measure for education level measures an interaction effect between two-two category variables, daughters" education and her mothers" education. This created four categories, both had a higher level of education, both had a lower level of education, the daughter had more education, and the mother had more education. For the young adult analysis, a three categorv' variable was created, including "daughter has more education than her mother," "mother has more education that the daughter," and "both the daughter and mother have the same level of education." The reference group for the young adult analyses examining the effects of social distance on Generation X daughters" perceptions of solidarity during young adulthood is "daughter has more education than her mother." For the midlife analysis, a four category variable was created, including "daughter has more education than her mother," "mother

PAGE 64

50 has more education than the daughter." both the daughter and her mother have a low level of education," and "both the daughter and her mother have a high level of education." The reference group for the midlife analyses examining the effects of social distance on Baby Boom daughters' perceptions of solidarity dunng midlife is "daughter has more education than her mother." Work status and SEI score. This measure combines the employment status and SEI score for occupation. After running a crosstabulation to examme the nine possible categories of women emerged. Five were chosen for both the young adult and midlife analyses. These are, "daughter works and has a high sei score and her mother does not work," "mother works and has a high score, daughter does not work and therefore has no score," "both mother and daughter work and have low SEI scores," "both have no score because neither work," and the final category includes "all others." The reference group for this measure for the multivariate analyses is "daughter works, high score, mother does not work." The reference group for both the Generation X daughters in young adulthood, and the Baby Boom daughters during midlife is "daughter works and has a high score, mother does not work and therefore has no score." Income level. When measuring the social distance between the daughters and their mothers with regard to income two-two category dummy variables were created. Using the median score for income for all women in the LSOG study in 1971 and 1994 as a cut off point, those women who fell below this level of income were considered having lower income, and those women who fell at or above this level were considered to have higher income. This created four categories of women, one in which both daughter and mother had high levels of income, one in which they both had low levels of income, one in which

PAGE 65

51 the daughter had a higher level of income than her mother, and one in which the mother had a higher level of income than her daughter. For both the young adult and the midlife analyses, a three category variable was created, and include "daughter's income is high while the mother's is low," "mother's income is high while the daughter's is low," and "both daughter and mother have the same income level." The reference group for both the Generation X daughters during young adulthood and the Baby Boom Generation daughters during midlife is "daughter's income is high, while the mother's income is low." Marital status and timing of marriage. For this social distance measure, the marital status and timing of marriage were combined. After running a crosstabulation to examine the numbers of women in each of the nine possible categories, four were selected to examine social distance among daughters and their mothers for both the young adult and midlife analyses. They are "daughter is single/never married, mother married early," "daughter married later, mother married eariy." "both daughter and mother married eariy," and "all others." The reference group for this variable for both the Generation X daughters during young adulthood and the Baby Boom daughters dunng midlife is "both daughter and mother married early." Children. This measure compares daughters to their mothers with regard to who has children. Because only those mothers included were the ones matched to their daughters, thus having children of their own, there are two categories of women included, "both have children," "mother has children, daughter does not." The reference group for both the Generation X daughters during young adulthood and the Baby Boom daughters during midlife is "both have children."

PAGE 66

52 Analytic Model In order to evaluate the research questions and hypotheses previously mentioned, the analysis in this study encompasses several stages. Each stage is described in detail to follow, as well as a brief description of the analytic chapters. As previously illustrated in Figure 4. 1, each generation of women expenenced the social and structural shifts of the twentieth centurv' during different ages and life course stages. As a result, intergenerational differences among women exist. Figure 4.2 illustrates the conceptual framework for this study. In order to examine the first research question: "How have daughters' perceptions of associational. affectual. and consensual solidarity with their mothers during young adulthood changed, and what may be associated with this change?'" the analyses will begin with a comparison of means across the samples of daughters during young adulthood using the T-test statistic. Perceptions of the Generation X daughters in 1994 will be compared to the perceptions of the Baby Boom daughters in 1971. In addition, OLS regression models will be with the daughter characteristics and social distance measures as independent variables. This analysis will help to understand how social distance between mothers and daughters might affect daughters" perceptions of associational. affectual. consensual solidarit)', and how this might change over time. The same analyses will be done to answer the second research question: "How have daughters" perceptions of associational, affectual, and consensual solidarity with their mothers during midlife changed, and what may be associated with this change?" Perceptions of the Baby Boom daughters in 1994 will be compared to the perceptions of World War II daughters in 1971. Following are the statistical models for each set of OLS regression analyses:

PAGE 67

53 Model 1: ISas.af.co = Po + Pi (age) + Ps (education) + e Model 2. ISas,af,co = Po + Pi (age) + p2 (work indicators) + e Model 3: ISasafco = Po + Pi (age) + p2 (family indicators) + e Model 4: ISasafco = Po + Pi (age) + P2 (education) + (33 (work indicators) + P4 (family indicators) + e Where IS represents intergenerational solidarity, sub AS, AF. and CO represent the three measures of solidarity, associational, affectual, and consensual, and indicates that these models are the same for each analysis on these dependent variables. Age is included as a control variable in all four models. Education is included in the first model. The work characteristics included in the second model are work status, SEI for occupation, and income. The family characteristics comprising the third model are marital status, age at first marriage, whether or not the daughter has children, and age at first child. The fourth model includes all three subsets of variables, education, work and family. For both the young adult and midlife analyses, social difference scores are examined for one of each of the generations of women in of the analytic chapters. These analyses compare the younger generation of women at each time period to their mothers. Included in the first analytic chapter, are the Generation X daughters during young adulthood and the social distance in relation to their Baby Boom Generation mothers when they were young adults. The social distance analyses included in the second analytic chapter, are Baby Boom Generation daughters during midlife and the social distance in relation to their World War II Generation mothers when they were in midlife. Two sets of analyses are carried out in each of the analytic chapters. The first set of analyses illustrates the univariate statistics descnbing the social distance variables for the daughters and their mothers. The second set of analyses in each of the two chapters includes multivariate analyses using OLS regression. These analyses examine the effects

PAGE 68

54 of the distances between the daughters and their mothers on the daughters' perceptions of associational, affectual, and consensual solidarity dunng each of the two life stages young adulthood for the Generation X daughters in 1994. and midlife for the Baby Boom Generation daughters in 1971. The following are the statistical models for each set of OLS regression analyses: Model 1. ISas.af.co = Po + Pi (education) + P2 (work indicators) + p3 (family indicators) + e Sample Limitations While the LSOG data is ideal for this research project in many ways, there are limitations with regard to this research. The study sample is region specific, and the participants are relatively racially and economically homogenous. In addition, as a result of sampling restrictions, the sample may be slightly more advantaged in terms of more positive familial relations. Because successive generations are only considered as participants based on the participation of a pnor generation, those descendents who are not in close contact with their family members may have self-selected out of participation in the Longitudinal Study of Generations. Another limitation of this study is the sample size. The onginal Longitudinal Study of Generations sample was for the most part reduced by half for this proposed research, to include only women (N=l,071 at Time 1). Furthermore, once the dyads are created based on daughters, who have mothers participating in the study, and attrition is accounted for, the sample size for the analyses gets further reduced. Having a larger sample to begin with would have been beneficial for this proposed research. The inability to include the mother's perceptions of solidarity with their daughters is also a limitation. Comparing these perceptions would have been beneficial to

PAGE 69

55 understanding these mother daughter dyads. This was not possible with these data because not every' generation of mothers were asked the same questions, nor were these questions asked at each time period.

PAGE 70

56 Table 4. 1 Numbers of Generation X and Baby Boom Generation Daughters during Young Adulthood at Each Stage of Sample Selection.

PAGE 71

CHAPTER 5 GENERATIONAL DIFFERENCES IN DAUGHTERS' PERCEPTIONS OF ASSOCIATIONAL, AFFECTUAL, AND CONSENSUAL SOLIDARITY IN YOUNG ADULTHOOD Introduction The purpose of this chapter is to examine generational differences in daughters' perceptions of associational, affectual, and consensual solidarity with their mothers during the young adult life stage. This chapter specifically explores the role of education, work, and family in mfluencing young adult daughters" perceptions of solidarity with their mothers. Two generations are compared young adult women from Generation X m 1994 and young adult women from the Baby Boom Generation m 1971. This chapter presents the results of several sets of analyses bivariate and multivariate. Descriptive sample characteristics are presented first (a bivariate companson of two generations) for each generation in young adult. Education, work and family characteristics of each generation are calculated. Tests for differences across the two generations when they were young adults are calculated as well. The second set of analyses calculates levels of associational, affectual, and consensual solidarity for each cohort when in young adulthood. Again, tests for differences in reported solidarity with mothers across the two generations when they were young adults are calculated as well. The next set of analyses included in this chapter are multivanate models estimating the effects of the independent variables on the three measures of solidarity. We employ a nested modeling strategy where more exogenous variables are entered into the model first (education) and subsequent models explore more proximate factors (work and family 57

PAGE 72

58 characteristics). The first model includes age, as a control variable, and one's level of education. The second model includes the work related variables of employment, socioeconomic status for occupation, and income. The third model includes those variables related to family marital status, age at first marriage, and age at first child. The fourth and final model includes all of the variables education, work and family. Thus, we approximate the contribution each set of variables has in explaining levels of solidanty as each set is added to the model. The final set models are OLS regression models estimated for Generation X daughters examining the effects of social distance (as compared with their mothers) on perceptions of daughters' solidarity. For example, we examine whether being married later in life than one's mother was married decreases daughters" perceptions of solidarity with their mother. We do not estimate a comparable set of models for the Baby Boom daughters as the data do permit social distance measures to be constructed as there are no data on the mothers of this generation. Descriptive Statistics-Independent Variables Table 4. 1 includes the descriptive statistics for the demographic characteristics of Generation X and Baby Boom daughters during young adulthood. The quantitative variables are represented by the mean and the standard deviation, while the qualitative variables are represented by the percentage of women, accounting for each of the variable's attributes. T-tests were done to compare the women from each generation dunng young adulthood across the quantitative variables, and chi square was used for comparison of the qualitative variables.

PAGE 73

59 Results Table 5.1 presents the descriptive statistics for the independent, sociodemographic variables that comprise the daughter characteristics for each generation of women. There are 82 Generation X daughters and 124 Baby Boom Generation daughters included in the sample of young adults. Results suggest that there are significant differences with regard to sociodemographic characteristics pertaining to education, work and family between the Generation X daughters who were young adults in 1994, and the Baby Boom Generation daughters who were young adults in 1 971 Although the age range was the same for both generations (18 to 29), the sample of Generation X daughters are older (mean age = 21.66) than the sample of Baby Boom daughters (mean age = 20.48) (p< .05). With regard to education, there is also a statistically significant difference between the two generations of daughters (p<.01). More Generation X daughters in 1994 had moved beyond a high school degree, 72 %, compared to the Baby Boom Generation daughters in 1971, where only 60% had more than a high school education, and 40% had a high school degree or less. Turning next to work related characteristics, results show that there are differences across the generations. Fewer women from the Generation X sample of daughters worked in 1 994 (64.6 percent) as compared to daughters from the Baby Boom Generation in 1971 (66. 1 percent) (p<.05). However, there were more Generation X daughters in 1994 with lower SEI scores for occupation (p<.05). Also, a difference in household income between the two sets of young adult daughters is found (p<.05). The average income level for Generation X daughters in 1994 is much lower, $38,876, than for the Baby Boom Generation in 1971, where the average income was $53,203.

PAGE 74

60 The two generations also differ with respect to family circumstances in young adulthood. For example, in 1994, 26.8% of the Generation X daughters were married as compared to 33.3% of the Baby Boom daughters in 1971. However, the biggest difference between the two generations with regard to marital status is the number of women who were cohabiting in 1994 among the Generation X sample, 20.7%, In 1971, only 3.4% of the Baby Boom sample reported livmg together. The percentage of not married women was 52.4% of Generation X daughters, and 63,3% of Baby Boom daughters. These differences in marital status between the two samples of young adult women are statistically significant (p<,001). In addition, the timing of marriage for the two generations of young women differed (p<.01). Among Generation X young adult daughters, 19.5% married eariier, and 3.7% married later. In 1971, all of the Baby Boom daughters who had ever been married, 32,5%, marned earlier, and none married later. Interestingly, more Generation X daughters had children dunng young adulthood in 1994 (p<,05). Among the Generation X daughters, 19,5%) had children, compared to 13,7% of the Baby Boom Generation daughters in 1971. For the Generation X daughters in 1994, 80.5% had no children; however, 17. 1% of these women had children at an earlier age, while 2.4%) had children later, Descnptive Statistics and Mean Differences-Dependent Variables This section focuses on the descriptive statistics and the mean differences for the three measures of solidarity comprising the dependent variables for each generation of women during young adulthood. In addition to presenting the results of each measure of solidarity, the items used to comprise these measures have also been presented. Mean difference scores using ANOVA were run to determine significant differences in the means for each generation of women. There are some significant differences in the

PAGE 75

61 perceptions of solidaritv' for the Generation X daughters who were young adults in 1994 compared to the Baby Boom Generation daughters who were young adults in 1971. Results-Daughters^ Perceptions of Associational Solidarity During Young Adulthood Table 5.2 illustrates the results of the levels of perceptions of associational solidarity for each generation of daughters with their mothers during young adulthood. Each measure of solidarity is represented by the mean and standard deviation. T-tests were done to compare the women from each generation. Associatonal solidarity overall, has a range of 3 to 24. Generation X daughters in 1994 have higher perceptions of associational solidant\' (mean=13.52) than the Baby Boom daughters in 1971 (mean=12.42). This is a significant difference in overall perceptions of associational solidarity with their mothers for each generation of women (p<.05). The three items comprising the measure of associational solidanty each have a range of 1 to 6. Of these items, only one emerges as significantly different for the Generation X and Baby Boom Generation daughters (p<.01). This is the frequency of interaction on the telephone between the daughters and their mothers during a period of a year. Generation X daughters in 1994 have more contact with their mothers (mean=6. 19) compared to the Baby Boom Generation daughters in 1971 (mean= 4.63). While the other two items are not significant, it is important to note that the frequency of interaction with their mothers in person dunng the year decreased for the Generation X daughters in comparison to the Baby Boom daughters (mean=5.28 and 5.71 respectively), where as the frequency of interaction by mail increased (mean=2.50 and 2.02 respectively). While these findings are not significant, they do indicate a trend that would suggest the phone would emerge as a more viable means of interaction for these women. And overall, associational solidarity has increased for the successive generation of women over time.

PAGE 76

62 Results-Daughters' Perceptions of Affectual Solidarity During Young Adulthood The next measure of solidarity, affectual solidarity, is presented Table 5.3. This measure of solidarity has a range of 5 to 30. The overall mean for perceptions of this measure of solidarity increased for the Generation X daughters in 1994 (mean=22.58) compared to the Baby Boom daughters m 1971 (mean=21.90). However, this difference is not significant between these generations of women. Each of the five items comprising this measure have a range of 1 to 6. When looking at the items in this measure of solidarity, two items emerge as being significantly different between the two generations of young adult women. There is a significant decrease in the daughters' perception of whether or not the daughter feels that her mother understands her (p<.05). For the Generation X daughters in 1994, perceptions decrease (mean=4.00) compared to the Baby Boom Generation daughters" perceptions in 1971 (mean= 4,34). There is also a significant difference in whether or not the daughter feels she understands her mother at this stage (p<. 10). Generation X daughters as young adults have lower perceptions (mean=4.3 1) compared to the Baby Boom Generation daughters (mean= 4,49). Interestingly, while these two variables indicate the perception of some level of understanding between mother and daughter, and both have decreased, each of the other items in this measure closeness, communication, and whether they get along have all increased for the successive generation of women. However, these increases are not statistically significant. Results-Daughters" Perceptions of Consensual Solidanty During Young Adulthood The final bivariate table. Table 5.4. presents the findings for the measure of consensual solidarity'. This is a single item measure, with a range of 1 to 6. There is a significant difference in the perceptions of the two generations of daughters with regard

PAGE 77

63 to the similarity of world views with their mothers (p<.05). The Generation X daughters in 1994 perceive higher levels of consensual solidarity with their mothers as young adults (mean=3.89) than did the Baby Boom Generation of daughters in 1971 (mean= 1.55). Multivariate Analyses This section analyzes the multivariate models, which explore the effects of the education, work, and family measures on each generation of daughters' perceptions of associational, affectual, consensual solidarity with their mothers during young adulthood. Each of the tables includes four models for each of the generation of daughters. Daughter's age is added as a control variable in each of the models. Model 1 includes the education measure, comparing women who have more than a high school education to those with the equivalent of a high school degree or less. Model 2 includes those variables pertaining to work life. These are 1) employment status, which compares women who work to women who do not work. 2) SEI score for occupation, which compares women who do not work and have no score, and women who have a high score to women with lower SEI scores, and 3) income level. Model 3 is the family model which includes 1) marital status, comparing married women to women who are not married and women who are living together, 2) age at first marriage, which compares women who married early to women who are not married and women who married later, and 3) age at first child, which compares women who have no children to women who had their first child early and women who had their first child later The final model. Model 4, includes all of the covariates from the three domains.

PAGE 78

64 Results-Effects of Daughter Characteristics on Their Perceptions of Associational Sohdanty The first multivariate table (Table 5.5) shows results of the four models, which illustrate the effects of the daughter characteristics on their perceptions of associational solidarity with their mothers during young adulthood. Model 1, which includes the daughter's age and education, yields no significant findings. Furthermore, almost no variabilit\' is explained by this model, about 1% for each of the two generations of young adult daughters. The work measures in Model 2 present several significant findings. First, this model is significant for both the Generation X daughters in 1994 and the Baby Boom Generation daughters in 1971 (p<.10 and p<.05, respectively). In addition, this model explains 14% of the variability in perceptions of associational solidarity for the Generation X daughters, and 11% of the variability for the Baby Boom Generation daughters. For the Generation X daughters in 1 994, having a higher SEl score is negatively associated with the daughter's perception of associational solidarity during young adulthood when compared to women with low SEl scores (p<. 10). This is also the case for Baby Boom Generation daughters during young adulthood in 1971 (p<.05). Income also emerges as having a significant effect on daughters' perceptions of associational solidarity with their mothers during young adulthood. For Generation X daughters, a higher level of income is associated with more positive perceptions of associational solidarity (p<.01). For Baby Boom Generation daughters, the association is negative. Among these women in 197], a higher level of income was associated with lower levels of associational solidarity with their mothers (p<.01).

PAGE 79

65 Model 3. which includes the family characteristics of the daughters, presents no significant findings for either generation of daughters. None of the family measures have a significant effect on daughters" perceptions of associational solidarity during young adulthood. However, this model does explain 15% of the variability in perceptions of associational solidarity for Generation X daughters in 1994, and is approaching significance (p<. 10). In addition, in this model, when only the family characteristics are added to age and education, the effect of daughter's age on this measure of solidarity becomes a positively correlated effect. The older the daughter is within the young adult sample, the more positive are her perceptions of association with her mother. The final model in Table 5.5, Model 4, includes age, and all of the education, work and family vanables. This model yields few significant results. With regards to levels of association with their mothers during young adulthood. For the Generation X daughters, this model explains 21% of the variability in this dependent variable. However, neither the model, nor any of the covariates emerge as significant. For Baby Boom Generation daughters. 12% of the vanability in perceptions of associational solidanty is explained, and the model is approaching significance (p<. 10). In addition, having a high SEI score, and income emerge as having significant effects on level of association when all of the vanables are included in the model. For Baby Boom daughters in 1971, having a higher SEI score is associated with lower levels of association with their mothers compared to women with lower SEI scores (p<.05). For these daughters, income is also negatively associated with associational solidanty (p<.01). It is also interesting to note, for the Baby Boom Generation, across the models the relationship between age and perceptions of associational solidarity emerged as a positive correlation. However, when all of the

PAGE 80

66 vanables are included in the model, this relationship becomes negative. The older the daughter in this young adult sample, the more negative are the perceptions of associational solidarity' with their mother. Results-Effects of Daughter Charactenstics on their Perceptions of AfFectual Solidantv Table 5.6 includes the results of all four models on the daughters' levels of perception of affectional solidarity with their mothers as young adults. The same models are included here as in the previous table. Model 1, which includes daughter's age and education, yields nothing significant. Age and education are not significantly related to perceptions of affectual solidarity for either of these generations during young adulthood. Furthermore, this model is not significant for either generation of women, explaining little vanability in this dependent variable. When the work characteristics are introduced in Model 1, only SEI score for occupation for the Baby Boom Generation daughters in 1971 appears to have a significant effect (p<. 10). For these daughters during young adulthood, having a higher SEI score for occupation is indicative of lower perceptions so affectual solidarity with their mothers during young adulthood compared to those daughters with a low score. Although not statistically significant, for the daughters in Generation X, the relationship between SEI and this measure of solidarity is a positive one. This model is also statistically significant for the Baby Boom Generation daughters (p<. 10), yet only explains about 6% of the variability in the dependent variable. For the Generation X daughters in 1994, about 7% of the vanability is explained, and this model is not significant. When the family characteristics are included in Model 3, little significant value is found. For the Generation X daughters during young adulthood in 1994, less than a high

PAGE 81

67 school degree is associated with lower levels of perceptions of affectuai solidarity with their mothers compared to those daughters with more than a high school degree (p<. 1 0). While not statistically significant, with the inclusion of the family variables, it is also important to note that the relationship between daughters age and affect becomes negative, while it had been positive when looking at education and the work characten sties in the previous two models. This model explains 9% of the variability in the dependent variable for Generation X daughters in 1994, and only 2% for Baby Boom daughters in 1 97 1 When all of the independent variables are included in Model 4, the relationship between age and affect remains negative for the Generation X daughters during young adulthood in 1994. However, daughter's income is the only covariate that emerges as having a significant effect on affectuai solidarity for these women (p<.05). For these daughters, having a higher income level is associated with more positive perceptions of affectuai solidarity with their mothers. Interestingly, for the Baby Boom Generation daughters in 1971. the relationship between income and affect is also significant, however, this relationship is negative (p<. 10). In addition, for the Baby Boom daughters in 1971. SEI score for income is also significant (p<. 10). A negative association exists indicating that if the daughters SEI score is higher, the perceptions of afectual solidarity with her mother during young adulthood are more negative. For the Generation X daughters, this full model explains 16% of the variability in affect, and 8% for Baby Boom Generation daughters. Results-Effects of Daughter Characteristics on their Perceptions of Consensual Solidarity Turning attention to Table 5.7, findings from the four models on perceptions of consensual solidarity for the daughters with regard to the daughter characteristics are

PAGE 82

68 presented. Model 1 presents significant results in terms of age and education for young adult Baby Boom daughters. Baby Boom daughters m 1971 have higher perceptions of consensual solidarity with their mothers if they are older (p<.05). Age does not have a significant effect on consensus for Generation X daughters in 1994. For Baby Boom Generation women, education also has a significant affect on levels of consensus during young adulthood (p<.05). Women with less than or equal to a high school degree report more similar world views with their mothers compared to those young women who have more education during young adulthood. While not significant, this relationship for the Generation X daughters in 1994 is negative. Model 1 explains almost none of the vanability in the dependent variable, 1%, for the Generation X daughters, while explaining 6% for the Baby Boom Generation daughters. For the Baby Boom Generation sample in 1971, this model is significant (p<.05). Looking at model 2, there are significant findings among the work measures. For the Baby Boom Generation daughters in 1971, age is again significantly associated with consensual solidanty, in that as the daughters age, their perceptions of consensus with their mothers become more similar (p<. 1 0). With the inclusion of the work measures, education is no longer significant for these daughters. Employment status for Baby Boom Generation daughters dunng young adulthood in 1971 is significant (p<. 10). For these women, a high SEI score is associated with less similar world views with their mothers compared to those daughters who have a lower SEI score. In addition, for the Baby Boom Generation women a higher level of income has a significant negative association with consensus (p<.05). For the Generation X daughters in 1994, income is also significantly associated with perceptions of consensual solidarity (p<. 10). However,

PAGE 83

69 for these young adult daughters, a higher income level means more similarity of world views. Model 2 explains 5% of the variability in the dependent variable for the Generation X daughters in 1994, and 9% of the variability for the Baby Boom Generation daughters in 1971. This model is significant for the Baby Boom daughters (p<.05). Model 3 includes the family characteristics in the model with age and education. Several significant findings in association with perceptions of consensual solidarity emerge in this model. For Generation X daughters, the inclusion of the family covariates alter the relationship between age and consensus. In the previous two models, this relationship was positive, so that as the daughters got older, their world views became more similar to those of their mothers. This model suggests that this relationship is a negative relationship, although not a significant one. For Baby Boom Generation daughters during young adulthood, being older continues to be significantly associated with more similar world views with their mothers (p<.05). Education again emerges as significant with the family measures added (p<.05). Having less than or equal to a high school education for the Baby Boom Generation daughters in 1971 means more similar world views with their mothers compared to those daughters with more than a high school education. Finally, those daughters who were not married during young adulthood from the Baby Boom sample in 1971 reported more positive perceptions of consensual solidarity with their mothers compared to those women who were married (p<. 10). This model explains 8% of the variability in perceptions of consensual solidarity with ones mother for both generations of women, however the model is only significant in explaining the relationship for the Baby Boom Generation daughters during in 1971 (P<.05).

PAGE 84

70 Turning to Model 4, inclusive of all daughter charactenstics, significant findings are present. During young adulthood, age continues to be significant for Baby Boom Generation daughters in 1971 in the same way as the previous models (p<.05). A lower education also continues to be significantly associated with more similar world views with their mothers for these daughters when compared to the daughters with higher education (p<.05). Baby Boom Generation daughters with high SEI scores also reported significantly lower levels of consensus with their mothers during young adulthood than did those daughters with low SEI scores (p<. 10). Income was also negatively associated with perceptions of consensual solidarity for Baby Boom daughters (p<.05). In 1971, daughters with higher levels of income reported less similarity with their mothers in terms of world views. With all of the covariates included in this model, age at first marriage remained significant for the Baby Boom daughters in 1971 (p<. 10). Those women who were not marned reported significantly more positive perceptions of consensual solidarity with their mothers than their married counterparts during young adulthood. While Model 3 explains 13% of the variability in consensual solidarity for both generations of women during young adulthood, again, the model is only significant for the Baby Boom Generation daughters in 1971 (p<.05). Social Difi'erence Analyses for Generation X Daughters and their Baby Boom Mothers During Young Adulthood. This section is targeted toward analyzing the effects of the differences between daughters and their mothers on the daughters' perceptions of associational, aff'ectual, and consensual solidarit>'. By comparing the daughters' sociodemographic characteristics in terms of education, work and family, to their mothers, several difference measures were created. Descriptive characteristics are examined first, for differences in education, work

PAGE 85

71 status and SEI score for occupation, income level, marital status and timing of marriage, and children. Second, OLS regression analyses are run examining the effects of the difference measures on the three measures of solidarity. Results -Descriptive Statistics for the Difference Measures Comprising the Independent Variables Table 5.8 includes the descriptive statistics illustrating the difference measures for Generation X daughters during young adulthood and their Baby Boom Mothers during young adulthood. The three category, education variable reveals that 31% of the Generation X daughters as young adults in 1994 had more education that their Baby Boom mothers did when they were young adults. Almost 14% of the daughters had less education than their mothers. Just over half, 55%, of the Generation X daughters and their mothers have the same level of education, either both low, or both high. Work status and SEI score for occupation were combined. Five categories were examined. About 12% of the daughters had jobs with high SEI scores while their mothers had no jobs and therefore no scores. There are just over 10% of the dyads where the mothers work and have high SEI scores and the daughters do not work and have no score. Approximately 1 1% of the dyads are those where both daughters and mothers work and have low SEI scores. Just over 24% of the mother/daughter dyads include those women where both daughters and mothers have no SEI scores because they don't work. Finally, 43% of the dyads are characterized by all of the other categories of SEI score and work status combinations. Three categones of income differences are examined. For those daughters whose income levels are high and their mother's income is low, about 22% of dyads exist. Almost 26% of the dyads are characterized by a mother with a higher income level and a

PAGE 86

72 daughter with a low income level. Finally, approximately 52% of the dyads include daughters and mothers with the same income levels. Current marital status and timing of marriage were combined and three categories of dyads are compared. About 15% of the dyads include those daughters and mothers who are both married, and marned early. Daughters who never mamed and had mothers who married early comprise approximately 49% of the dyads. Those dyads which contain all other categories of marital status and marital timing combinations, make up the remaining 35% of the pairs of mothers and daughters. With regard to having children, since the mothers are paired with the daughters already, there is no possibility that the mother does not have children. Thus, there are only two categones represented here. Almost 20% of the dyads are those where both daughter and mother have children during young adulthood. The remaining 80% are those mother/daughter pairs where the daughter does not have children of her own. Results-Effects of Social DifFerence Measures on Generation X daughters' Perceptions of Associational, Affectual and Consensual Solidarity with their Baby Boom Generation Mothers During Young Adulthood Table 5.9 illustrates the regression analyses used to explore the social difference variables between the young adult daughters of Generation X with their Baby Boom Generation mothers. All of the variables are included m the model together for associational, affectual, and consensual solidarity. Turning attention to the effects of these difference scores on associational solidarity, several of the difference measures emerge as significant. This model explains 32% of the variability in associational solidarity during young adulthood in 1994 for the Generation X daughters (p<. 10). With regard to education, those dyads where the daughters and their mothers had the same level of education during young adulthood had lower perceptions of association with their

PAGE 87

73 mothers in 1994 compared to those in which the daughter had more education than the mother (p<. 10). Several of the marital status and timing of marriage covanates appear to have significant effects on Generation X daughters' perceptions of associational solidarity during young adulthood. When compared to daughters who work and have high SEl scores, and the mother does not work and has no score, those dyads where both worked and had low scores had lower levels of association with their mothers (p<. 10). Those mother/daughter dyads where neither women worked, and therefore had no score had higher levels of association with their mothers during young adulthood in 1994 compared to the reference group (p<. 10). Moving on to affectual solidarity, with all of the covariates included in the model, about 25% of the variability in Generation X daughters perceptions of affectual solidarity is explained here. This model is not significant. With regard to this measure of solidarity, education differences do not emerge as significant as they had with associational solidarity. Moving on to work status and SEI score, those dyads where the mother works and had a high SEI score, while the daughter did not work and had no score had significantly lower levels of affect when compared to those dyads where the daughters worked and had high SEI scores and the mothers did not work and had no scores (p<.05). Income differences also emerged as significant in this model. For those dyads where the income levels were the same for both the daughter and her mother, higher levels of affectual solidarity were expressed by Generation X daughters during young adulthood in 1994 when compared to those dyads in which the daughters had higher income levels than did their mothers.

PAGE 88

74 Finally, looking at consensual solidarity, approximately 27% of the variability in this dependent variable is explained by this model (p<. 10). As was the case for affectual solidanty, dyads where the mothers worked and had high SEI scores and the daughters did not work and had no score had lower levels of consensus with their mothers than did the reference group (p<.05). The only other significant effect on Generation X daughters' perceptions of consensual solidarity during young adulthood in 1994 is a marital status and marital timing difference measure. When the daughter is single while her mother married early, lower levels of consensual solidarity exist compared to mother/daughter dyads where both daughter and mother married early (p<. 10). Conclusions In order to test hypotheses HI a through Hlc. four steps were taken in this analytic chapter. The first step of the analysis of the daughters during young adulthood, a description of their sociodemographic characteristics (education, work and family) was provided. In the second step of the analysis, a description of the dependent variables associational, affectual, and consensual solidarity was provided. At this time, mean differences between the Generation X daughters and the Baby Boom Generation daughters during young adulthood were presented. Third, the multivariate analyses testing the relationship between daughter charactenstics on their perceptions of associational, affectual, and consensual solidarity during young adulthood were run. Finally, the multivariate analyses testing the effects of social distance between the mother daughter dyads, which included Generation X daughters and their Baby Boom Generation mothers dunng young adulthood, on the daughters' perceptions of associational, affectual, and consensual solidarity. Support is provided for each of these three hypotheses.

PAGE 89

75 To review, the hypotheses tested in Chapter 5 look at the effects of daughter characten sties and social distance on daughters" perceptions of solidarity with their mothers during young adulthood. The first hypothesis, HI a. looks specifically at the differences between the two generations of women during young adulthood. Generation X daughters in 1994 and Baby Boom Generation daughters in 1971. The second hypothesis, HIb, focuses on the effects of the daughters' characteristics, those sociodemographic characteristics pertaining to education, work, and family, on the daughters" perceptions of solidarity with their mothers. With this hypothesis, a distinction is made setting affectual solidarity apart from associational and consensual solidarity. Finally, the third hypothesis, Hlc. focuses on the effects of social distance between the Generation X daughters during young adulthood and their mothers from the Baby Boom Generation on the daughters" perceptions of solidarity with their mothers. Again, a distinction is made here which suggests that affectual solidarity will not be affected in the same ways as associational and consensual solidanty. Support for HI A can be seen in Tables 5.2 through 5.4. Young adult daughters from Generation X in 1994 have stronger (more positive) perceptions of associational and consensual solidaritv' than the Baby Boom Generation daughters in 1971. hi Table 5.2, levels of associational solidanty were presented. Overall, associational solidarity was greater for the Generation X daughters in 1 994. In terms of each of the individual measures, it was also the case that there was an increase in association for the successive generation. However, the significant difference was found in the increase of association with their mothers by phone.

PAGE 90

76 Support for HIb can be seen in Tables 5.5 through 5.7. The effects of daughter characteristics on perceptions do not affect affectual sohdarity in the same way as they do associational and consensual solidarity. Looking at Table 5.5, the effects of daughter characten sties on associational solidarity during young adulthood. Model 2 proves to be significant for both generations of daughters. This model includes the work characteristics, and both SEI for occupation and income are significant for both generations of women. Model 3, which includes the family characteristics, is significant for the Generation X daughters. Finally, the full model, Model 4, is significant for the Baby Boom daughters, and SEI and income again emerge as significant. Skipping ahead to Table 5.7, looking at the effects of daughter characteristics on perceptions of consensual solidarity with their mothers during young adulthood, several significant findings emerge as well. Each of the four models is significant for the Baby Boom Generation daughters. Model 1 indicates that age and education are in fact associated with levels of consensus. Model 2, which incorporates the work charactenstics. suggests that for this generation a high SEI score, and a higher level of income are also associated with consensual solidarity. In Model 3, not being married emerges as significant. Finally, in the full model, age and education remain significant, as well as not being married. Turning attention now to Table 5.6, the effects of daughter characteristics on perceptions of affectual solidarity during young adulthood, it can be seen that none of the models show that daughter characteristics are associated with perceptions of affect, with the exception of Model 2 for the Baby Boom Generation approaching significance. Thus, affectual solidarity for daughters dunng young adulthood is not affected in the same way

PAGE 91

77 that associational or consensual solidarity. Furthermore, the fact that these four models in each table reveal more significant associations for the Baby Boom daughters during young adulthood continues to lend support for the first hypothesis, which suggests that there are more negative perceptions of associational and consensual solidarity for this generation of women during young adulthood. Support for HI c can be seen in Table 5.9. Social distance between Generation X daughters and their Baby Boom mothers during young adulthood is associated with associational and consensual solidarity, but not with affectual solidarity. Looking at Table 5.9, the models for associational and consensual solidarit)' are significant, suggesting that these social distance measures sufficiently explain variability in these measures of solidarity during young adulthood. However, this is not the case for affectual solidaritv.

PAGE 92

78 Table 5. 1 Descriptive Statistics-Demographic Characteristics of Generation X Daughters and Baby Boom Daughters in Young Adulthood. Generation X Daughters Young Adulthood 1994 N=82 Baby Boom Daughters Young Adulthood 1971 N=124 SD 0/ /o SD % Sig. DAUGHTER CHARACTERISTICS Age (mean) 21.66 2.96 20.48 2.30 Education Level (%) High School Degree or Less More than High School 28.0 72.0 40.0 60.0 Work Status (%) Works Does Not Work 64.6 35.4 66.1 33.9 SEI (Occupadon) (%) No Job. No Score SEI Score is Low SEI Score is High 35.4 45.1 19.5 37.5 23.3 39.2 Income1994 Dollars (mean) 38,876 25,601 53,203 36.085 Current Marital Status (%) Married Living as Married Not Married 26.8 20.7 52.4 33.33 3.4 63.3 Age at First Marriage (%) Never Married Married Younger Married Older 76.8 19.5 3.7 67.5 32.5 Has Children (%) Yes No 19.5 80.5 13.7 86.3 Age at First Birth (%) No Chddren Had First Child Younger Had First Child Older 19.37 2.47 80.5 17.1 2.4 20.05 2.04 86.3 a. t-tests for significant differences of quantitadve variables, chi-square for qualitative variables.

PAGE 93

79 Table 5.2. Generation X Daughters' and Baby Boom Daughters' Perceptions of Associational Solidarity with their Mothers in Young Adulthood. Generation X Daughters Young Adulthood 1994 Baby Boom Daughters Young Adulthood 1971 SD N SD N Sig Associational Solidarity 13.52 In Person 5.28 By Phone 6.19 By Mail 2.50

PAGE 94

mcS O OB Q S c .2 m (1) XI OX .^ X .'^ X VD t; ^00 ? S rs *^ ^ • •• '^ o 80 (N ^ — m F^S O — ON ^
PAGE 95

X 5 mm r'^ X m mm ^ Ao

PAGE 96

82 O ^ u 00 a> JC ^ O w O I E o o on

PAGE 97

83 Table 5.8. Descriptive Statistics Social Distance Among Generation X Daughters During Young Adulthood and their Baby Boom Generation Mothers During Young Adulthood. Difference Variables For Daughters and Their Mothers % Education Level Daughter has More Education than Mother 31.1 Mother has More Education than Daughter 13.8 Both Daughter and Mother have the Same Level of Education 55.2 Work Status and SEI for Occupation Daughter Works/High SEI, Mother Does Not Work/No SEI 1 2. 1 Mother Works/High SEI. Daughter Does Not Work/No SEI 1 0, 3 Both Daughter and Mother Work and Have Low SEI 10.7 Bodi Have No SEI Because Neitiier Work 24. 1 All Others 43. 1 Income1 994 Dollars Daughter's Income Level is High, Mother's is Low 22.4 Mother's Income Level is High, Daughter's is Low 25.9 Daughter and Mother Have Same Income Level 51.7 Marital Status and Timing of Marriage Both Mother and Daughter Married Early 14.6 Daughter is Single/Never Married and Mother Married Early 48.8 All Others 35.4 Children Both Daughter and Mother Have Children 1 9.5 Daughter Does not Have Children, Mother Does 80.5

PAGE 98

84 Table 5.9. OLS Regression Effects of Social Distance Between Generation X Daughters and their Baby Boom Generation Mothers During Young Adulthood. Associational Affectual Consensual Solidarit> Solidarity Solidarity Difference Scores Education Level (ref=daughter has more) Mother has More Education than Daughter -.30(2.58) -.08(3.58) .01 (.77) Daughter and Mother Have Same Education -.04 (1 .93) ~ .11 (2.90) .25 (.60) Work Status and SEI Score (ref = daughter works/high SEI, mother does not work) Modier Works/High, Daughter Does Not -.10 (3.88) 46 (4 84) 34 (1 01) Work Both Work Have Low SEI Neither Work and Have No SEI All Others Income (ref= Daughter's Income is High, Mother's is Low) Mother's Income is High. Daughter's is Low Income Level is the Same for Both Marital Status and Timing of Marriage (ref=both daughter and mother married early) Daughter is Single, Mother Married Early All Others Children (ref=both have children) Daughter Has No Children, Mother Does .01(1.80) -.16(3.03) -.10(55) Constant 10.29 28.13 4.98 R' .32.25 .27a. N=82 ^ b. Standardized coefficients are presented; Standard errors appear in parentheses. c. *p<.05; *• p<.01; ***p<.001;~p<.10 .48 (2.99) 42 (2.68) .33 (2.76)

PAGE 99

CHAPTER 6 GENERATIONAL DIFFERENCES IN DAUGHTERS" PERCEPTIONS OF ASSOCIATIONAL. AFFECTUAL, AND CONSENSUAL SOLIDARITY IN MID-LIFE Introduction The purpose of this chapter is to examine generational differences in daughters" perceptions of associational, affectual, and consensual solidarity with their mothers during the midlife stage. This chapter specifically explores the role of education, work, and family in influencing daughters" perceptions of solidarity with their mothers during midlife. Two generations are compared women from the Baby Boom Generation in 1994 in midlife and women from the Worid War II Generation in 1971 during midlife. This chapter presents the results of several sets of analyses bivariate and multivariate. Descriptive sample characteristics are presented first (a bivariate companson of two generations) for each generation during midlife. Education, work and family characteristics of each generation are calculated. Tests for differences across the two generations when they were in midlife wer calculated as well. The second set of analyses calculates levels of associational, affectual, and consensual solidarity for each cohort when in the midlife life stage. Again, tests for differences in reported solidarity with mothers across the two generations when they during midlife are calculated as well. The next set of analyses included in this chapter are multivariate models estimating the effects of the independent variables on the three measures of solidarity. We employ a nested modeling strategy where more exogenous variables are entered into the model first (age and education) and subsequent models explore more proximate factors (work and 85

PAGE 100

86 family characteristics). The first model includes age, as a control variable, and one's level of education. The second model includes the work related variables of employment, socioeconomic status for occupation, and income. The third model includes those variables related to family marital status, age at first marriage, and age at first child. The fourth and final model includes all of the variables age, education, work and family. Thus, we approximate the contribution each set of variables has in explaining levels of solidarit>' as each set is added to the model. The final set models are OLS regression models estimated for Baby Boom Generation daughters examining the effects of social distance (as compared with their mothers) on perceptions of daughters' solidarity. For example, we examine whether being married later in life than one's mother was married decreases daughters' perceptions of solidarity with their mother. We do not estimate a comparable set of models for the World War II Generation daughters as the data do permit social distance measures to be constructed as there are no data on the mothers of this generation. Descriptive Statistics-Independent Variables The first table in this chapter includes the descriptive statistics for the demographic characteristics of Baby Boom Generation and World War II Generation daughters during midlife. The quantitative variables are represented by the mean and the standard deviation, while the qualitative variables are represented by the percentage of women, accounting for each of the variable's attributes. T-tests were done to compare the women from each generation during young adulthood across the quantitative variables, and chi square was used for comparison of the qualitative variables.

PAGE 101

87 Results Table 6. 1 presents the descriptive statistics for the independent sociodemographic variables that comprise the daughter characteristics for each generation of women. There are 108 Baby Boom Generation daughters and 143 World War II Generation daughters included in the sample of women dunng midlife. Results suggest that there are significant differences with regard to sociodemographic characteristics pertaining to education, work and family between the Baby Boom Generation daughters during midlife in 1994 and the World War II Generation daughters during midlife in 1971. Although the age range was the same for both generations (40 to 59), the Baby Boom Generation daughters are younger (mean age = 43. 16) than the sample of World War II daughters (mean age = 44.23) (p< .05). With regard to education, there is also a statistically significant difference (p<.001). Almost all of the Baby Boom daughters (95 percent) have more than a high school education during midlife in 1994, whereas little more than half (56 percent) of the World War II Generation daughters fall into this category in 1971. Turning next to work related characteristics, results show that there are differences across the generations. More women from the Baby Boom Generation sample of daughters worked in 1994 (80 percent) as compared to daughters from the World War II Generation in 1971 (48 percent). This is a significant increase in 1994 (p<.001). In addition, for the Baby Boom Generation women during midlife in 1994, SEI scores for occupation also increased compared to their World War II Generation counterparts (p<.01). Of those women working in 1994 who had SEI scores, approximately 35% of them had low SEI scores, while 44% had high SEI scores. In 1971 for working World

PAGE 102

88 War II Generation women during midlife, approximately 19% had low scores, while almost 29% had high SEI scores. Also, a difference in household income between the two sets of midlife women is found (p<.001). The average income level for Baby Boom daughters during midlife in 1994 is much lower. $55,685. than for the World War II Generation m 1971, where the average income was $68,304. The two generations also differ with respect to family circumstances in midlife. For example, in 1994. only 64.8% of the Baby Boom daughters were married as compared to 93% of the World War II daughters in 1 97 1 Three other important differences between the two generations with regard to marital status were the number of women who were cohabitmg in 1994 among the Baby Boom sample. 8.3%. In 1971. no World War II Generation women reported living together. Also, the number of divorced women in 1994 is much higher, 16%. compared to only 5% in 1971 Finally, the percentage of single-never married women was almost 9% for the Baby Boom Generation daughters during midlife, while none of the World War II daughters had never married. In both of the samples of women, about 2% of the women were widowed at the midlife stage. These differences in marital status between the two samples of young adult women are statistically significant (p<.001). In addition, the timing of marriage for the two generations of women during midlife differed (p<.05). Among Baby Boom daughters, 11% had never married, about 30% marned earlier, and almost 59% married later. In 1971, all of the World War II daughters had been mamed. Approximately 53% married earlier, and 47%o married later. Interestingly, more Baby Boom daughters had children in 1994 than did World War II Generation women in 1971 (p<.05). Among the Baby Boom daughters. 83.3% had children, compared to 76.2% of

PAGE 103

89 the World War II Generation daughters. For the Baby Boom daughters in 1994, 16.6% had no children; however, 64.8% of these women had children at an earlier age, while 1 8.6% had children later. Age at first child data was not available for World War II daughters in 1971. Descriptive Statistics and Mean Differences-Dependent Variables This section focuses on the descnptive statistics and the mean differences for the three measures of solidarity comprising the dependent variables for each generation of women during midlife. In addition to presenting the results of each measure of solidarity, the items used to comprise these measures have also been presented. Mean difference scores using ANOVA were run to determine significant differences in the means for each generation of women. There are some significant differences in the perceptions of solidarit\' for the Baby Boom Generation daughters during midlife in 1994 compared to the World War II Generation daughters during midlife in 1971. Results-Daughters' Perceptions of Associational Solidarity Durina Midlife Table 6.2 illustrates the results of the levels of perceptions of associational solidanty for each generation of daughters with their mothers during midlife. This measure of associational solidarity has a range of 3 to 24. Baby Boom Generation daughters in 1994 had lower levels of associational solidarity with their mothers in 1994 (mean= 1 1.83) than did the World War II Generation daughters in 1971 (mean=I2.49). This is statistically significant for a difference in overall perceptions of associational solidarity with their mothers during midlife (p<. 10). The three items comprising the measure of associational solidarity, frequency of contact in person, by phone, and by mail, each have a range of 1 to 8. Of these items, only one emerges as significantly different for the Baby Boom and World War II Generation daughters (p<.01). This is the

PAGE 104

90 frequency of interaction by mail between the daughters and their mothers over the course of a year. For the World War II Generation women in 1971, there was a greater frequency of contact with their mothers by mail (mean= 3.29) than for the Baby Boom Generation daughters in 1971 (meaji= 2.58). While not significant on an individual level, however, contributing nonetheless to the overall levels of perceptions of associational solidarity, is the frequency of contact in person and by phone over the course of the year. Frequency of contact in person for Baby Boom daughters in 1994 was lower (mean =3.99) than it was for the World War II Generation daughters during midlife in 197 1 (mean= 4.22). Frequency of contact by phone for the successive generation of women did increase however (mean=5.46) for the Baby Boom Generation daughters in 1994. This average is up from the World War II Generation daughters (mean=5.30). Again, while these findings are not significant, they do indicate a trend that would suggest the phone would emerge as a more viable means of interaction for these women, whose life course trajectones have taken them in a different direction than their mothers. Furthermore, this is important given that overall, associational solidarity has decreased for the successive generation of women over time during midlife. Results-Daughters' Perceptions of AffectuaJ Solidarity Dunng Midlife. Table 6.3 includes the means for perceptions of affectual solidarity for each generation of daughters included in the midlife analysis. This measure has a range of 5 to 30. The overall mean for perceptions of this measure of solidarit>' decreased for the Baby Boom daughters during midlife in 1994 (mean=21.2I) compared to the World War II Generation daughters in 1 97 1 (mean=22. II). When comparing the two generations of daughters, this decrease is not significant between these two generations. However, when looking individually at the items that comprise this measure, some significant

PAGE 105

91 differences do emerge. Each of the items comprising this measure, closeness of relationship, communication, get along with each other, understanding of mom, understanding of daughter, has a range of 1 to 6. Two of the items in this measure of affectional solidarity emerge as significant. For the Baby Boom Generation daughters in 1994, there is a decrease in the perceptions of how well the daughter feels her mother understands her in midlife (mean=3.80). This is down compared to the World War II Generations perceptions of how well they feel their mother understands them during midlife in 1971 (mean=4.13). This difference is significantly lower (p<.05). Another aspect of this measure of solidanty that decreases for the successive generation is how well the daughters feel they understand their mothers during midlife. This decrease is also statistically significant (p<.05). For the Baby Boom Generation daughters in 1994, the average is lower (mean=4.42), compared to the World War II Generation daughters in 1971 (mean=4.67). While none of the other individual items are significantly different between the two generations of daughters during midlife, perceptions of closeness and communication both decreased for the successive generation of women. Interestingly, daughters' ideas about whether or not they get along with their mothers remained exactly the same, with identical means and virtually no difference in vanability. Results-Dau.ghters' Perceptions of Consensual Solidarity^ During Midlife The final bivariate table. Table 6.5, presents the findings for the measure of consensual solidarity. This is a single item measure assessing the level of similarity of worid views among family members from different generations. This measure has a range of 1 to 6. There is no significant difference in the perceptions of the two generations of daughters, and their similarity of world views with their mothers during midlife. The Baby Boom Generation daughters in 1994 have virtually the same mean

PAGE 106

92 perceptions (mean=3.79) as the World War II Generation daughters in 1971 (mean=3.77). Multivariate Analyses This section analyzes the multivariate models, which explore the effects of the education, work, and family measures on each generation of daughters' perceptions of associational, affectual, consensual solidarity with their mothers during young adulthood. Each of the tables includes four models for each of the generation of daughters. Daughter's age is added as a control variable in each of the models. Model 1 includes the education measure, comparing women who have more than a high school education to those with the equivalent of a high school degree or less. Model 2 includes those variables pertaining to work life. These are 1) employment status, which compares women who work to women who do not work, 2) SEI score for occupation, which compares women who do not work and have no score, and women who have a high score to women with lower SEI scores, and 3) income level. Model 3 is the family model which includes I) marital status, comparing married women to women who are not married and women who are living together, 2) age at first marriage, which compares women who married early to women who are not married and women who married later, and 3) age at first child, which compares women who have no children to women who had their first child early and women who had their first child later. The final model. Model 4, includes all of the covariates from the three domains. Results-Effects of Daughter Characteristics on their Perceptions of Associatonal Solidarity During Midlife The first multivariate table. Table 6.5, presents the results of the four models, which illustrate the effects of the daughter characteristics on their perceptions of

PAGE 107

93 associational solidarity with their mothers during young aduhhood. Model 1 includes only age and education. This model indicates that for the Baby Boom daughters during midlife in 1994, less than or equal to a high school education, levels of perceptions of associational solidarity with their mothers were less than those daughters who had more than a high school education (p<. 10). There are no other significant findings produced by this model. Only 4% of the variability is explained for the Baby Boom Generation daughters in 1994. and 2% of the variability in association solidarity for the World War II Generation daughters in 1971. In Model 2, work characteristics are added to the model. For both generations of women, the effects of age and education remain the same once these variables are added. Less education continues to have a negative effect on associational solidarity for the Baby Boom daughters dunng midlife in 1994. Not one of the work characteristics has a significant effect on levels of association for either generation of daughters during midlife. For the Baby Boom daughters, the explanatory value of this model increases to about 7%. However, the model is still not significant. For the World War II daughters, ver>' little is explained, only 3%, by this model. In Model 3, the family characteristics are included with the control variables, age and education. The inclusion of these variables alters the effects of education for the Baby Boom daughters in 1994, and having less education is no longer significant. In fact, for the Baby Boom Generation, there are no significant effects on associational solidarity with their mothers during midlife in 1994. For the World War II Generation daughters, the effects of age and education remain the same once the family charactenstics are introduced. However, for this generation of daughters, not having

PAGE 108

94 children is associated with higher levels of association with their mothers during midlife (p<.05). Approximately 12% of the variability' in levels of association is explained for the Baby Boom daughters during midlife in 1994, and about 6% for the World War II Generation daughters in 1971. The final model. Model 4, is the full model. With the inclusion of all of the covariates, education again appears to be significant for Baby Boom Generation daughters during midlife (p<.05). The daughters with less education have less contact with their mothers than do those daughters with more than a high school education. Also, for the Baby Boom daughters, once all of the variables are included family measures become significant. Mantal status is the first variable to emerge as significant (p<. 10). For this generation of women during midlife in 1994, those daughters who are not married, have a higher frequency of contact with their mothers when compared to daughters who married later. In addition to marital status, not having children is also significant (p<. 10). Baby Boom Generation daughters during midlife who do not have children have lower perceptions of associational solidarity with their mothers compared to those daughters who have children. Once all of the variables are included in the full model for World War II Generation daughters, none of the effects are significant. This model is significant for the Baby Boom Generation daughters, explaining 14% of the variability in associational solidarity dunng midlife in 1971 (p<.05). It is not significant for the World War II Generation daughters in 1971, explaining 7% of the variability in the dependent variable.

PAGE 109

95 Results-Effects of Daughter Characteristics on their Perceptions of Affectual Solidarity During Midhfe Table 6.6 includes the results of all four models on the daughters' levels of perception of affectional solidarity with their mothers during midlife. The same models are included here as in the previous table. Model 1, inclusive of age and education of the daughters, yields no significant findings for either generation of daughters dunng midlife. Furthermore, this model has little explanatory value for the Baby Boom Generation daughters in 1994 (3 percent), and almost no explanatory value for the Worid War II Generation daughters in 1971 (1 percent). When the work characteristics are introduced into the model in Model 2, significant findings appear for both generations of women. The effects of age and education remain the same for both generations of women in terms of their effect on affectual solidanty, and are insignificant. For the Baby Boom Generation in 1994, those daughters who were not working and therefore did not have an SEI score for occupation indicates lower levels of affect when compared to those who had low SEI scores for their jobs (p<. 10). For the World War II Generation women in 197 1 higher levels of income were associated with lower levels of affectual solidarity (p<.01). Model 2 is not significant for the Baby Boom Generation daughters dunng midlife in 1994 in explaining effects of daughter characteristics on affectual solidarity with their mother, but explains about 8% of the variability in the dependent variable. This model is significant for the Worid War II Generation daughters, however, explaining about 7% of the variability. In Model 3, the family characteristics have been included. For the Baby Boom Generation daughters in 1994, the introduction of these variables changes the effect education has on perceptions of affectual solidarity, and education becomes significant.

PAGE 110

96 For this generation of daughters during midhfe, having less than or equal to or less than a high school education is associated with lower levels of affect compared to those Baby Boom daughters who have more than a high school education (p<. 10). For the Baby Boom daughters, two of the family charactenstics have significant effects as well. Not being married during midlife in 1994 for the Baby Boom daughters yields more positive levels of association compared to those women who are married and married later (p<. 10). In addition, for these daughters, not having children is associated with significantly lower levels of affect compared to those daughters who do have children (p<.01). For the World War II Generation daughters during midlife in 1971, the introduction of the family characteristics does nothing to change the effects of age or education on perceptions of affectual solidarity with their mothers. The only significant finding in this model for the World War II Generation daughters is with regard to age at first marriage (p<. 10). All of the women from this generation have married by midlife, and those women who married earlier have more positive perceptions of affectual solidarit>' with their mothers than those who married later. In the final model, Model 4, with all of the covariates included, significant findings emerge for both generations of daughters. For the Baby Boom Generation during midlife in 1994, previous significant findings remain significant. Those daughters with less than a high school education have lower levels of affectual solidarit>' with their mothers compared to those daughters with a higher level of education (p<.05). A higher SEI score for these daughters in 1994 is associated with lower levels of affect compared to those women with a low SEI score (p<. 10). Finally, not having children is associated with more negative perceptions of affectual solidarity levels for the Baby Boom

PAGE 111

97 daughters in 1994 (p<.01). The full model illustrates significant findings with regard to affectual solidarity among the World War II Generation daughters during midlife. In this model, age emerges as a significant variable for the first time in this model (p<. 10). Older daughters among this sample during midlife in 1971 have more positive perceptions of associational solidarity with their mothers. As was the case in Model 2, income is significant for the World War II daughters (p<.01). Those with higher levels of income have more negative perceptions of affect with their mothers. Finally, as was the case in the family model, having married earlier means more positive perceptions of affect compared to having married later (p<. 10). Results-Effects of Daughter Characteristics on their Perceptions of Consensual Solidarity During Midlife Table 6.7 illustrates the findings from all four models on perceptions of consensual solidanty. Model 1 includes the effects of age and education on the daughters' perceptions of consensual solidarity with their mothers during midlife. For the Baby Boom daughters in 1994, having less than a high school degree is associated with lower levels of consensus when compared to those daughters who have more than a high school degree (p<.05). Neither age nor education is significant in this model for the World War II Generation women in 1971. Model 1 explains 3% of the vanability in levels of consensual solidarity for the Baby Boom daughters, and is significant (p<.05). This model is not significant for the World War II Generation daughters. Model 2 adds the work characteristics to age and education. Education continues to be significant for the Baby Boom daughters in 1994 (p<.05). Lower levels of education have a negative effect on consensus for these daughters. There are no significant findings among the work variables for the Baby Boom daughters during

PAGE 112

98 midlife. For the World War II Generation daughters during midlife, the inclusion of the family variables does nothing to change the effects of age or education. In addition, the family variables yield no significant results for this generation either. Model 2 explains approximately 7% of the vanability in consensual solidarity for the Baby Boom daughters, and 3% for the Worid War II Generation daughters. The model is not significant for either generation. The family characteristics have been included with age and education in Model 3. For the Baby Boom Generation daughters in 1994, less than a high school education continues to have a negative effect on consensual solidanty with their mothers compared to the daughters with more than a high school education (p<.05). While age is not significant in this model, the family characteristics have changed the direction of the relationship between age and perceptions of consensus for the Baby Boom Generation daughters during midlife. In this model, the older daughters have less similarity with their mothers in terms of world views. Baby Boom daughters in 1994 with no children have lower perceptions of consensual solidarity with their mothers during midlife (p<.01). For the Worid War II Generation daughters dunng midlife in 1971, the inclusion of the family characteristics does not yield significant findings among age and education. However, as it was for the Baby Boom Generation daughters, these characteristics change the effects of age on perceptions of consensus. While not significant, in this model, the older daughters report having more similar world views with their mothers during midlife. Age at first mamage is also significant among the World War II Generation daughter in 1971 (p<.01). Those daughters who married early reported more having more similar world views with their mothers during midlife

PAGE 113

99 compared to those daughters who married later. Model 3 explains 12% of the variability in perceptions of consensual solidarity for the Baby Boom Generation daughters in 1994, and is a significant in explaining the effects of daughter characteristics on the dependent variable (p<. 10). This model explains 9% of the variabilitv' in the dependent variable for the World War 11 Generation daughters in 1971, and is also significant (p<.05). Once all of the variables have been included in Model 4, significant effects for the Baby Boom Generation daughters remain consistent with previous models. Education remains significant, with less education promoting more negative perceptions of consensual solidarity (p<.05). As in Model 3, for this generation of daughters, not having children has a negative effect on perceptions of solidarity compared to daughters with children (p<.01). For the World War 11 Generation daughters during midlife in 1971, none of the covariates in the full model have a significant effect on the daughters' perceptions of consensual solidarity with their mothers. The full model explains 17% of the variability in perceptions of consensual solidarity for the Baby Boom daughters in 1994, and is significant (p<.05). While there are no significant effects found among the vanables in this model for the World War 11 Generation daughters during midlife in 1971, this model is significant, and explains 1 1% of the variability in the dependent variable (p<. 10). Social Difference Analyses for Baby Boom Generation Daughters and their World War II Generation Mothers DurinR Midlife. This section is targeted toward analyzing the effects of the differences between daughters and their mothers on the daughters' perceptions of associational, affectual, and consensual solidarity. By comparing the daughters' sociodemographic characteristics in terms of education, work and family, to their mothers, several difference measures were

PAGE 114

100 created. Descriptive characteristics are examined first for differences in education, work status and SEI score for occupation, income level, marital status and timing of marriage, and children. Second, OLS regression analyses are run examining the effects of the difference measures on the three measures of solidarity. Results Descriptive Statistics for the Difference Measures Comprising the Independent Variables Table 6.8 includes the descriptive statistics illustrating the difference measures for Baby Boom Generation daughters during midlife and their Worid War II mothers during midlife. The four category, education variable shows that almost 46% of the Baby Boom daughters during midlife in 1994 have more education than their Worid War II Generation mothers did during midlife in 1971. Less than 3% of the mothers had more education than their daughters during midlife. Just over 4% of the dyads included daughters and mothers who both had low levels of education during midlife. Finally, a little more than 47% of these dyads were those with both daughters and mothers had more than a high school degree at midlife. Work status and SEI score for occupation were combined, and five categories were examined and used in the analyses for this chapter. Approximately 24%) of the Baby Boom daughters worked and had jobs with high SEI scores during midlife, while their Worid War II mothers did not work and therefore had no scores during midlife. Less than 3%) of the dyads were the reverse, with mothers that worked and had high scores and daughters that did not work and had no scores. About 1 1% of included daughters and mothers both working during midlife, with low SEI scores. Almost 1 1% of the dyads were comprised of daughters and mothers that both did not work and therefore neither

PAGE 115

101 had SEI scores. Finally, about 54% of the dyads are characterized by aJl of the other categories of SEI score and work status combinations. Three categories of income differences are examined. For those daughters whose income levels are high and their mother's income is low, about 13% of dyads exist. Approximately 34% of the dyads are characterized by a mother with a higher income level and a daughter with a low income level. Finally, about 53% of the dyads include daughters and mothers with the same income levels. Current marital status and timing of marriage were combined and three categories of dyads are compared. Seven percent of the dyads include those daughters and mothers who are both marned, and married early. Daughters who never married and had mothers who married early comprise 21% of the dyads. Those dyads which contain all other categories of mantal status and marital timing combinations, make up the remaining 72% of the pairs of mothers and daughters used in the analyses for this chapter. With regard to having children, since the mothers are paired with the daughters already, there is no possibihty that the mother does not have children. Thus, there are only two categones represented here. About 68% of the dyads are those where both daughter and mother have children during midlife. The remaimng 32% are those mother/daughter pairs where the daughter does not have children of her own. Results-Effects of Social Difference Measures on Baby Boom Generation Daughters' Perceptions of Associational, Affectual and Consensual Solidantv with their Worid War II Generation Mothers During Midlife Table 6.9 illustrates the regression analyses used to explore the social difference variables between the Baby Boom Generation daughters during midlife and their World War II Generation mothers. All of the variables are included in the model together for associational, affectual, and consensual solidarity. Turning attention to the effects of

PAGE 116

102 these difference scores on associational solidarity, sigtuficant effects emerge. Work status and SEI score for occupation has a significant effect on levels of association for Baby Boom daughters during midlife. For those daughters who have mothers who work and have high SEI scores, yet they do not work and have no score, levels of association are reportedly more positive when compared to those dyads where the daughter works and has a high SEI score, and the mother does not work (p<. 10). Baby Boom daughters who are among dyads where neither the daughter nor the mother works and therefore have no SEI scores also report having higher levels of association when compared to the reference group (p<.05). This model is significant, and explains 28% of the variability in associational solidarity during midlife in 1994 for the Baby Boom Generation daughters (P<10). Moving on to affectual solidarity, with all of the covanates included in the model, several social difference variables emerge as significant. With regard to educational differences, those Baby Boom daughters who at midlife, have mothers with more education, report more negative perceptions of affectual solidarity with their mothers when compared to daughters who have more education than their mothers during midlife (p<. 10). Turning to the work status and SEI score dyads, those daughters during midlife who did not work, but had mothers who did work, and had high SEI scores reported significantly higher levels of affect compared to those daughters who worked and had high scores, and had mothers who did not work and had no score (p<.05). Differences in income were also significant. Daughters who had mothers with high incomes while theirs was low reported lower levels of affection compared to daughters in dyads where the daughters had high incomes while their mothers had low incomes (p<. 1 0). None of

PAGE 117

103 the social difference measures based on family charactenstics had a sigmficant effect on affectual soUdarity for Baby Boom women during midhfe in 1994. This model is significant, and about 26% of the variability in Baby Boom daughters' perceptions of affectual solidarity is explained here (p<. 1 0). Fmally, looking at consensual solidant\', there are several significant findings. With regard to education, those dyads made up by daughters who had more education than their mothers comprise the reference group. Daughters who had mothers with more education that they did reported lower levels of consensus with their mothers than the daughters in the reference group (p<.5). Lower levels of consensus were also reported daughters in dyads where both daughters' and mothers' education was low (p<.05). The same low levels of consensus were found among daughters with higher levels of education who had mothers who also had high levels of education (p<.05). Work status and SEl score is the only other significant finding in this model. Those Baby Boom daughters who have mothers who work and have high scores, while they do not work and therefore have no score reported more positive levels of consensus compared to daughters who work and have high scores, and had mothers who did not work and had no scores (p<.01). This model is significant, and explains 35% of the variability in the perceptions of consensual solidarity of Baby Boom Generation daughters with their mothers during midlife m 1994. Conclusions To test hypotheses H2a through H2c. four analytic steps were taken. The first step of the analysis of the daughters during midlife, a description of their sociodemographic characteristics (education, work and family) was provided. In the second step of the analysis, a descnption of the dependent vanables association^, affectual, and

PAGE 118

104 consensual solidarity was provided. At this time, mean differences between the Baby Boom Generation daughters and the World War 11 Generation daughters during midlife were presented. Third, the multivariate analyses testing the relationship between daughter characteristics on their perceptions of associational, affectual, and consensual solidaritv' dunng midlife were run. Finally, the multivariate analyses testing the effects of social distance between the mother daughter dyads, which included Baby Boom Generation daughters and their World War II Generation mothers during midlife, on the daughters' perceptions of associational, affectual, and consensual solidarity. Mixed support is provided for these hypotheses. To review, the hypotheses tested in Chapter 6 look at the effects of daughter characten sties and social distance on daughters" perceptions of solidarity with their mothers dunng midlife. The first hypothesis, H2a, looks specifically at the differences between the two generations of women during midlife. Baby Boom Generation daughters in 1994 and Worid War II Generation daughters in 1971. The second hypothesis, H2b, focuses on the effects of the daughters' charactenstics. those sociodemographic characteristics pertaining to education, work, and family, on the daughters' perceptions of solidarity' with their mothers. With this hypothesis, a distinction is made setting affectual solidantv' apart from associational and consensual solidarity. Finally, the third hypothesis, Hlc, focuses on the effects of social distance between the Baby Boom daughters during midlife and their mothers from the Worid War II Generation on the daughters' perceptions of solidarity with their mothers. Again, a distinction is made here which suggests that affectual solidarity will not be affected in the same ways as associational and consensual solidarity.

PAGE 119

105 The first hypothesis for the midhfe analysis can be examined by looking at Tables 6.2 through 6.4. Partial support is provided for this hypothesis. In Table 6.2, we can see that with regard to associational solidarity, the hypothesis is supported. Baby Boom Generation daughters have significantly weaker perceptions of associational solidarity with their mothers than do the World War II Generation daughters during midlife. Also, perceptions of aflfectual solidarity with their mothers are not significantly different for these two generations of women during midlife. However, with regard to consensual solidarity, this hypothesis is not supported. Table 6.4 shows that Baby Boom Generation daughters do not have lower levels of consensus compared to the World War II Generation daughters during midlife. In fact, while not significant, they have slightly more positive perceptions. However, the variabilitv' among the Baby Boom Generation daughters is much greater than for the World War II Generation daughters with regard to consensus. The second hypothesis for the midlife analysis, H2A.is not supported. By looking at Tables 6.5 through 6.7, it can be determined that daughter characteristics affect associational solidarity in ways similar to associational and consensual solidarity. Table 6.5, shows that out of the three measures of solidanty, associational solidarity is actually the least affected by daughter characteristics for these two generations of women during midlife. Model 1 is significant for the Baby Boom Generation, revealing that education level has an effect on associational solidarity during midlife. Model 4, the full model, is also significant only for the Baby Boom Generation. Education remains significant in this model, also, not being married, and not having children are also significant.

PAGE 120

106 Turning attention to Table 6.6, effects of daughter characteristics on perceptions of affectual sohdanty during midlife, many of the same effects emerge as significant. Model 2 is significant for the World War II Generation daughters, indicating that income level is associated with affect for this generation of women during midlife. Model 3 is significant for the Baby Boom Generation daughters. Here, both education and not being married are associated with perceptions of affectual solidarity. Finally, Model 4 is significant for the Baby Boom Generation. For this generation of daughters during young adulthood, affect is associated with education, SEI score for occupation, and not having children. The final measure of solidarity for this midlife analysis, consensual solidarity, can be seen in Table 6.7. Model 1 is significant for the Baby Boom Generation, with education being associated with levels of consensus. Model 3 is significant for both the Baby Boom and World War II Generations of daughters. However, different covariates emerge as significant for each generation. For the Baby Boom Generation daughters during midlife, education and not having children are associated with consensus. For the World War II Generation, it is the timing of marriage. The final model. Model 4, is also significant for both generations. Again, for Baby Boom daughters during midlife, it is education and not having children that affects their perceptions of consensual solidarity with their mothers. For the World War II Generation, there are no significant covariates found in a significant model. Support for H2c is not found. Table 6.9 reveals that social distance between Baby Boom daughters and their World War II Generation mothers during midlife is associated with affectual solidarity in much the same way as it is for associational and consensual

PAGE 121

107 solidarity. Looking at Table 5.9, the models for all three measures of solidarity associational, affectual, and consensual solidarity are significant. This suggests that these social distance measures sufficiently explain variability in these measures of solidarit\' during midlife.

PAGE 122

108 Table 6. 1 Descriptive Statistics Demographic Characteristics of Baby Boom Daughters and World War II Daughters During Midlife. Babv Boom

PAGE 123

109 Table 6,2. Baby Boom Daughters' and World War II Daughters' Perceptions of Associational Solidarity During Midlife with their Mothers. Baby Boom Daughters Midlife 1994 SD World War II Daughters Midlife 1971 SD *p <.05:**p
PAGE 124

^^ I S3 o g U o go W o

PAGE 125

ii: 3 ^

PAGE 126

112 3 o a O o — 03 -^^ 2 E ^-^ n .0 c S .>. ^ -2; 2 -' ? "r^ ,0 3 .H 51 2 2.1 "u:2 .=3 -=•--••=' r:i Z K Zj a,Z2 jsZ c_ S -S '^ o S 5 o 5 ? 2 5 ^ o £ ? S -I s -s -^ 5. 2 r> li) ^8
PAGE 127

113 Table 6.8 Descriptive Statistics Social Distance Among Baby Boom Generation Daughters During Midlife and their Worid War II Generation Mothers During Midlife. Difference Variables For Daughters and Their Mothers % Education Level Daughter has More Education than Mother Mother has More Education than Daughter Both Daughter and Mother have a Low Level of Education Both Daughter and Mother have a High Level of Education Work Status and SEI for Occupation Daughter Works/High SEI, Mother Does Not Work/No SEI Mother Works/High SEI, Daughter Does Not Work/No SEI Both Daughter and Mother Work and Have Low SEI Both Have No SEI Because Neither Work All Others Income1 994 Dollars Daughter's Income Level is High, Mother's is Low Mother's Income Level is High, Daughter's is Low Daughter and Mother Have Same Income Level Marital Status and Timing of Marriage Both Mother and Daughter Married Early Daughter is Single/Never Married and Mother Married Early All Others Children Both Daughter and Mother Have Children Daughter Does not Have Children, Mother Does 45.9

PAGE 128

07(3.05)

PAGE 129

CHAPTER 7 DISCUSSION AND CONCLUSIONS Introduction The overall aim of this study is to gain a better understanding of women's intergenerational relationships. Specifically, how mother/daughter relationships have been affected by the rapid social change that occurred in the second half of the twentieth century. Using Bengtson's measures of associational, affectual, and consensual solidarity as the basis for measuring the quality of mother/daughter relationships several hypotheses were tested. Chapter three gave a detailed outline of the social structural changes that occurred in the twentieth century. As a result of these changes, it has been the case that for many women, the life course trajectories have also changed. A central expectation was that as a result of education, work, and family roles shaping different life course trajectories for women, mother/daughter relationships would be affected in various ways over the course of successive generations. Previous studies suggest that several factors have an influence on intergenerational family relationships. The analysis in this project focuses on two specific influences: 1) the effects of daughters individual characten sties on their perceptions of solidarity with their mothers, and 2) the differences in daughter's and mothers achievements and their effects on daughters' perceptions of solidarity with their mothers. Using the LSOG data, several generations of women from the same families are compared in two sets of analyses, the first at the young adult life stage, and the second at the midlife stage. 115

PAGE 130

116 The remainder of this chapter will accomplish several goals. First, a brief review, outlining the three measures of solidarity, as well as the life course perspective and the developmental schism is presented. Second, a brief discussion of the generational differences among the women, intergenerational solidarity, and social distance is presented, highlighting the major findings from this project. Next, conclusions and implications are drawn summarizing how this project contributes to the existing literature. Fmally, future research in the area of women's intergenerational relationships is outlined in detail. Review of Empirical Measures and Theoretical Framework Broadly speaking, Intergenerational Solidarity Theory is a theory of aging families stemming from empirical research done on multiple generations from the same family. There are six measures of intergenerational solidarity, of which this project uses three. To review, they are associational, affectual, and consensual solidarity. Associational solidarity is the first construct, measuring the frequency of contact occurring between the different generations in a family. There are three items used to create this construct: 1) contact m person, 2) contact by phone, and 3) contact by mail. The second measure of solidarity is a construct, which measures the degree of closeness between members of different generations. This is a five-item measure that includes variables addressing the following: 1) closeness of relationship, 2) communication between the individuals, 3) how well the individuals get along, 4) how well the individual understands the other, and 5) how well the other understands the individual. Finally, the third measure of solidarity used in this research is consensual solidarity. This is a one-item measure, which accounts for a similarity of world views.

PAGE 131

117 The main idea is that the stronger, or more positive the perceptions of these measures, the greater the degree of sohdarity between the two generations. Several hypotheses in this study suggest that not aspects of intergenerational sohdarity would be affected in the same ways. Specifically, this study hypothesized that affectual solidarity would not be influenced by social structural factors, or social distance between mothers and daughters in the same way that associational and consensual solidarity would be. Using these three measures, this project used Fingerman's Developmental Schism as the theoretical basis for understanding women's intergenerational relationships, comparing two generations of women at different periods of time from the same life stage young adulthood, and midlife. The Developmental Schism is a concept stemming from the life course perspective. In general, the life course perspective satisfies two key elements of family studies. These are how individuals change over time, and how their transitions and trajectories are linked across family members. Building from this, Fingerman introduced her concept, grounded in developmental psychology. Her explanations of mother/daughter relationships are multifaceted. The Developmental Schism suggests that tensions exist between generations due as a result of the different generations being at different developmental stages, and differing goals and varying needs result in different perspectives. Discussion The aforementioned points form the basis for this project. This framework was not used as a formal theory. Instead it served as a guiding framework from which to inform specific research hypotheses. In turn, the results of this study provide evidence for utilizing Bengtson's measures of solidarity to analyze the effects of daughter charactenstics and social distance between daughters and their mothers and perceptions

PAGE 132

118 of solidarity. The life course perspective, and more specifically Fingerman's Developmental Schism, provides a unique perspective that influences this project. The next section highlights some of the important findings from the analyses in Chapters 5 and 6. Generational Differences among the Women. Results reveal that for the daughters in both life stages (young adulthood and midlife), generational differences exist between the women with regard to the sociodemographic characteristics pertaining to education, work, and family. Many of these sociodemographic characteristics have been impacted as a result of the social structural changes that have occurred. Thus, the changes in the larger social structure have an indirect impact on women's intergenerational solidarity, as the effect is filtered through the changes in the women's individual characteristics. Young adulthood. For the Generation X daughters in 1994 and the Baby Boom Generation Daughters in 1971, several generational differences were found. First of all, the women in the Generation X sample were older, despite the sample selection criteria which governed the age range for young adulthood. As a result, some of the other individual characteristics may have been impacted by age. The young adult life stage is the most unstable, given that the young women during this life stage are still in the decision making phase for many of the education, work, and family outcomes. On the other hand, daughters during midlife have already '"chosen" and lived out their decisions. Educational differences were evident, with more of the daughters from Generation X in 1994. having moved beyond the minimum of a high school degree. Most of the women from both generations worked, however results from this study show that fewer daughters from Generation X in 1994 were working during young adulthood. More of

PAGE 133

119 the working women from the Baby Boom Generation had high SEI scores, compared to those women from Generation X. Also, the average household income was greater for Baby Boom daughters in 1971. Because the income measure is a household measure, differences in income between the generations is likely a result of the number of women who are either single or living with someone. If they are single and living alone, income will be lower. If they are cohabiting, income levels for individuals during young adulthood will be much lower than the income levels of their parents. Differences in timmg of marriage also exist, with all of the daughters married in 1971 in this sample marrying at an age younger than the national average at the time. While few women from Generation X married older, more women from this sample in 1994 were still single. Interestingly, while the majority of daughters in both generations did not have children as young adults, more women from Generation X had children compared to the Baby Boom Generation. This can possibly be explained by an increase in teen pregnancy in 1994 compared to 1971. Also the slightly older age of the Generation X daughters might capture some of this difference. Midlife. Among the two generations of daughters during midlife, differences exist in terms of daughter social charactenstics pertaining to education, work, and family. Again, despite sample selection criteria for those included in the midlife analysis, there is a difference in age. The daughters from the World War II Generation of women are slightly older. However, unlike the young adults, this difference will have little bearing on the outcomes of the other measures during midlife. Educational differences are most evident between these two generations, with almost all of the women form the Baby

PAGE 134

120 Boom Generation moving beyond a high school degree compared to just over half of the women in the World War II Generation. Stark differences in the numbers of women who work are also evident. Significantly more women from the Baby Boom Generation held jobs. In addition, SEI scores were also higher for the daughters in this generation. However, income levels were higher for the World War II Generation of women during young adulthood in 1971 Some of the difference in income can likely be captured in the number of single and divorced women from the Baby Boom Generation in 1994, compared to no single and very few divorced women in 1971. In addition, there were no cohabiting daughters from the World War 11 Generation during midlife. Fewer women from the Baby Boom Generation married at an earlier age compared to the women from the World War II Generation, where just over half were marned at an age younger than the national average at the time. Surprisingly, fewer women from World War II Generation had children by midlife. Intergenerational Solidarity. The first hypothesis for each of the analytic chapters centers around the differences in the daughters' perceptions of solidarity with their mothers during the different life stages. During young adulthood, it is hypothesized that the successive generation. Generation X, will have stronger perceptions of solidarity with their mothers compared to the Baby Boom Generation. During midlife, it is hypothesized that the successive generation, the Baby Boom Generation, will weaker perceptions of solidarity with their mothers. The second hypothesis for each of the analytic chapters has to do with the effects of the daughter charactenstics on levels of solidarity with their mothers. For each life stage

PAGE 135

121 (young adult and midlife) it is assumed that affectual solidaritywill not be affected in the same ways that associational and consensual solidarity will be. Measures of solidarit>'. Results in this study show that in fact during young adulthood, the Generation X daughters had more positive perceptions of all associational, affectual, and consensual solidarity with their mothers. Thus, support for the first hypothesis in the young adulthood analysis is found. However, there is only partial support for the first hypothesis in the midlife analysis. Not only were Baby Boom Daughters perceptions of consensual solidarity not significantly different than the Worid War II Generation daughters, they were slightly higher. Effects of daughter characteristics. There is mixed support for the hypotheses focusing around the effects of daughter characteristics on the specific measures of solidarity. In the young adult analysis, this hypothesis is supported. There are several characteristics, which emerge as having a significant effect on perceptions of associational and consensual solidarity for the daughters during young adulthood for both the Generation X and Baby Boom Generation Women. At the same time, it seems that daughter characteristics have little effect on affectual solidarity for women as young adults both in 1994 and in 1971. However, for the women during midlife, this second hypothesis is not supported. Results from this study show that affectual solidarity during midlife is affected in much the same way as associational and consensual solidarity are. This seems to be the case for both the Baby Boom Generation and the Worid War II Generation daughters. In fact, during midlife, the measure of solidarity least affected by daughter characteristics among

PAGE 136

122 these samples of women is associational solidarity. This appears to hold true for both generations of women. Social distance. For each of the analytic chapters in this study involves the effects of social distance between the daughters from the successive generations and their mothers on the daughters' perceptions of solidarity with their mothers. During young adulthood, this analysis included the Generation X daughters and the differences with their Baby Boom Generation mothers. During midlife, this analysis included the Baby Boom Generation daughters and the differences with their World War II Generation mothers. These hypotheses again suggest that afifectual solidarity will not be affected in the same way that associational and consensual solidarit>' are. There is mixed support for this idea. During young adulthood, the third hypothesis is supported. Results from this study show that the social distance measures did not have an effect on the daughters" perceptions of affectual solidaritv', while explaining significant amounts of variability in the perceptions of associational and consensual solidarit)'. During midlife, this hypothesis is not supported. For the daughters during midlife, the social distance measures affect all three measures of solidarity, explaining significant amounts of each. Because there was no data available on the mothers for the prior generations of women in each life stage, a comparison of social distance between the generations was not available. Thus, this hypothesis is limited. Conclusions and Implications Several implications are relevant to this project. In terms of academic scholarship, this research will contribute to the sociological literature of aging families in several ways. A great deal of the literature on intergenerational relationships centers around caregiving, either adult children caring for their aging parents, or grandparents as

PAGE 137

123 surrogate parents (Connidis, 2001; Mills, Wakeman and Fea, 2001). It is important to understand other aspects of intergenerational relationships, both to enhance the understanding of caregiving and support, but also to uncover the qualitative dimensions central to intergenerational bonds among family members, particularly as they age. This study focuses on social and structural factors, and how they might affect these dimensions. Specifically, the perceptions of the quality of solidarit>', as perceived by the daughters, as opposed to the mothers' perception of the relationship. The "verticalization" of the family structure means that there are more generations existing in family, but fewer family members. This is a result of increasing life expectancy and the decline in fertility (Martm-Matthews, 2000; McPherson, 1998). With fewer family ties, greater emphasis is placed upon those familial relationships which do exist. Intergenerational relationships must be negotiated and renegotiated over time in order for family members to sustain these important family ties. This research focuses on the family relationship central to most women, the mother/daughter relationship. Future Research This study is the first step in a long-term research agenda that will continue to focus on the dynamics of intergenerational relationships as family members move across the life course. The next step will be to follow up this research with a qualitative study of mothers and daughters, interviewing mothers and daughters separately, as well as together as a dyad. In addition, further research will include the need to incorporate a more diverse study population. Diversity in areas such as race, the socioeconomic status of the family, and other non-traditional "family" forms.

PAGE 138

LIST OF REFERENCES Albers, C. (ed.). (1999). Sociology of Families: Readings. Thousand Oaks. CA; Pine Forge Press Allen, K.R., Blieszner, R., and Roberto, K.A. (2001). Families in the Middle and Later Years: A Review and Critique of Research in the 1990's In Understanding Families Into the New Millennium: A Decade in Review. Milardo, R.M. (Ed.). Minneapolis, MN: National Council on Family Relations. Anderson, M.L. (2000). Thinking about Women: Sociological Perspectives on Sex and Gender. Boston; Allyn and Bacon. Aquilino, W.S. (1997). From Adolescent to YounR Adult: A Prospective Study of Parent-Child Relations During the Transition to Adulthood. Journal of Marriage and the Family. 59:670-686. Atchlev, R.C. (1989). A Continuity Theors' of Normal Aging. The Gerontologist. 29: 183-190. Baltes, P.B. (1993). 'The Aging Mind: Potential and Limits." The Gerontologist. 33: 580-594. Baltes, P.B, and Brim, O. (eds). (1979). Life Span Development and Behavior. Vol. 3. New York: Academic Press. Bedford, V.H., and Blieszner, R. (1997). Personal Relationships in Later Life Families. In Handbook of Personal Relationships. 2"'' Ed. Duck, S, (Ed.). New York, NY: Wiley. Beer, W.R. (1989). Strangers in the House: The World of Stepsiblings and Halfsiblings. New Brunswick, N.J.: Transaction. Bengtson. V.L. (2001). Beyond the Nuclear Family: Increasing Importance of Multi generational Bonds. Journal of Marriage and the Family. 63 : 1 1 6. Bengtson, V.L. (1975). Generation and Family Effects in Value Socialization. American Sociological Review. 40: 358-37 1 Bengtson, V.C, and Kuypers, J. A. (1971). Generational Difference and the 'Developmental Stake. Aging and Human Development. 2:249-260. 124

PAGE 139

125 Bengtson, V.L., Mangen, D.J. and Landry. PH. (1984). The MultiReneration Family: Concepts and Findings. \n Intergeneralional Relationships. Hoeming, E.M. and Schaeffer. D. (Eds.). New York: C.J. Hogrefe. Inc. Bengtson. V.L.. Parrott, T., and Burgess, E. (1997). Theory, Explanation, and a Third Generation of Theoretical Developments in Social Gerontology. Journal of Gerontology. 52B: S72-88. Bengtson, V.L.. and Roberts, RE. L. (1991). Intergeneralional Solidarity in Aging Families: An Example of Formal Theory Construction. Journal of Marriage and the Family. 53:856-870. Bengtson. V.C. Rosenthal. C.J. and Burton. C. (1996). Paradoxes of Families and aging. In R.H. Binstock and L.K. George, Handbook of Aging and the Social Sciences. Sand Diego: Academic Press. Bengtson. V.L.. and Schaie, K.W. (eds.). (1999). Handbookof Theories of Aging. New York: Springer Publishing. Bengtson. V.C. Schaie, K.W.. and Burton, L.M. (eds.). (1995). Adult Intergenerational Relations: Effects of Social Change. New York: Springer Publishing. Bengtson. V.L., and Schrader, S.S. (1982). Parent-Child Relations. \x\ Handbook of Research Instruments in Social Gerontology (Vol. 2). Mangen. D. and Peterson, W. (eds). Minneapolis, MN: University of Minnesota Press. Berardo, F.M. (1980). Decade Preview: Some Trends and Directions for Family Research and Theory in the 1980's. Journal of Marriage and the Family. 42.4:723-728. Berardo. F.M. (1981). Family Research Theory': Emergent Topics in the 197Q's and the Prospects for the 1980's. Journal of Marriage and the Family. 43.2:25 1 -254. Biblarz, T.J., Bengtson. V.L.. and Bucur. A. (199 6). Social Mobility Across Three Generations. Journal of Marriage and the Family. 58:1 88-200. Binstock, R.H., and George, L.K. (1 996). Handbook of Aging and the Social Sciences. Sand Diego: Academic Press. Binstock, R.H., and George, L.K. (eds.). (2001). Handbook of Aging and the Social Sciences. 5''' Ed. San Diego. C A: Academic Press. Boss, P.G.. Doherty, W.J, LaRossa, R., Schumm. W.R.. and Steinmetz, S.K. (eds.) (1993). Sourcebook of Family Theories ami Methods. A Contextual Approach. New York: Plenum Press. Bray, J., and Kelly, J. (1998). Stepfamilies: Love. Marriage and Parenting in the First Decade. New York: Broadway Books.

PAGE 140

126 Brayfield, A. (1995). A Bargain at Any Price? Childcare Costs and Women's Employment. Social Science Research. 24:188-214. Brewster, K.L., and Padavic, I. (2000). Change in Gender-Ideology, 1977-1996: The Contributions of Intracohort Change and Population Turnover. Journal of Marriage and the Family. (62:477-487). Browne, C.V. (1998). Women. Feminism, and Aging. New York: Springer Publishing. Calasanti, T.M. (1996). Gender and Life Satisfaction in Retirement: An Assessment of the Male Model. Journal of Gerontology. 5 IB: SI 8-29. Carr, D. (2004). My Daughter Has a Career I Just Raised Babies": The Psychological Consequences of Women's Intergenerational Social Comparisons. Social Psychology Quarterly. Carr, D. (2002). The Psychological Consequences of Work-Family Tradeoffs for Three Cohorts of Men and Women. Social Psychology Quarterly. 65:103-24. Carr, D. (1997). The Fulfillment of Career Dreams at Midlife: Does it Matter for Women's Mental Health? Journal of Health and Social Behavior. 38:33 1-344. Carstensen, L.L. (1992). Social and Emotinal Patterns in Adulthood: Support for Socioemotional Selectivity Theory. Psychology of Aging. 7:331-338. Cherlin, A.F. (1988). The Changing European Family. Journal of Family Issues. 9:291-297. Connidis, I.A. (2001). Family Ties and Aging. Thousand Oaks, CA: Sage Publications. Coontz, S. (1992). The Way We Never Were: American Families and the Nostalgia Trap. New York: Basic Books. Coontz, S. (1997). The Way We Really Are: Coming to Terms with America 's Changing Families. New York: Basic Books. Dannefer, D. (1984). Adult Development and Social Theory: A Pragmatic Reappraisal. American Sociological Review. 49,1 : 100-1 16. DeGenova, M.K. (ed). (1997). Families in Cultural Context: Strengths and Challenges in Diversity. Mountain View, CA: Mayfield Publishing Co. Dilworth-Anderson, P., and Burton, L.M. (1996). Rethinking Family Development: Critical Conceptual Issues in the Study of Diverse Groups. Journal of Social and Personal Relationships. 1 3,3 :325-334. Duck, S. (ed.). (1997). Handbook of Personal Relationships. 2""^ Ed New York, NY: Wiley.

PAGE 141

127 Duvall, EM. (1977). Family Development. Philidelphia. Lippincott. Duvall. E.M., and Hill, R.L. (1948). Report of the Committee on the Dynamics of Family Interaction. Washington, D.C.: National Conference on Family Life. Elder, G.H. Jr. (1998). The Life Course and Human Development. \n Handbook of Child Psychology: Vol. 1, Theoretical Models of Human Development (5'^ ed). Lemer, R.M. (ed). New York: Wiley. Estes, C. (1991). The New Political Economy of Aging: Introduction and Critique. In M. Minkler and C. Estes (eds.). Critical Perspectives on Aging. Amity ville, NY: Baywood Press. Felder, D.G. (1999). A Century of Women: The Most Influential Events in the 20"' Century Women 's History. Secaucus, NJ: Carol Publishing Group. Festinger. Leon. (1954). A Theory of Human Relations. Human Relations. 7:114-140. Fingerman. K.L. (2001). Aging Mothers and Their Adult Daughters: A Study inMixed Emotions. New York: Springer Publishing. Fingerman, K.L. (1996). Sources of Tension in the Aging Mother and Adult Daughter Relationship. Psychology and Aging. 11:591 -606. Fingerman, K.L., and Bermann, E. (2000). Applications of Family Systems Theory to the Study of Adulthood. International Journal of Aging and Human Development. 51,1:5-29. Gerson, K. (1985). Hard Choices: How Women Decide About Work. Career, and Motherhood. Berkeley: University of California Press. Gerson, K. (1986). Briefcase. Baby, or Both? Whether Women are Raised to be Mothers or Pursue Careers, Their Adult Lives Often Take Unexpected Paths. Psychology Today. 20:30-36. Giamisso. R, Stallings, M.,andBengtson, V.L. (1995). The 'Intergenerational Stake' Hypothesis Revisited: Parent-Child Differences in Perceptions of Relationships 20 Years Later. In V.L. Bengtson, K.W. Schaie, and L.M. Burton (eds). Aduh Intergenerational Relations: Effects of Social Change. New York: Springer Publishing. Glass, J, Bengtson, V.L, and Dunham, C.C. (1986). Attitude Similarity in ThreeGeneration Families: Socialization, Status Inheritance, or Reciprocal Influence? American Sociological Review. 5 1 : 685-698. Glass, J., and Polisar, D. (1987). A Method and Metric for Assessing Similarities Among Dyads. Journal of Marriage and the Family. 49:663-668.

PAGE 142

128 Glenn, E.N., Nakano, E., Chang, G., and Forcey, L.R. (Eds). (1994). Mothering: Ideology, Experience, and Agency. New York: Routledge. Glick, P., and Parke, R. (1965). New Approaches in Studying the Life Cycle of the Family. Demography, 2:187-202. Greil, A.L. (1991), Not Yet Pregnant: Infertile Couples in America. New Brunswick: Rutgers University Press. Gubrium, J.F.. and Holstein, J.A. (1999). Constructionist Perspectives on Aging. In V.L. Bengtson and K. W. Schaie (eds.). Handbook of Theories of Aging. New York: Springer Publishing. Hagestad, G.O., and Dannefer, D. (2001). Concepts and Theories of Aging. In R.H. Binstock and L.K. George (eds). Handbook of Aging and the Social Sciences. San Diego, CA: Hartcourt Brace Hagestad, G.O.. and Nuegarten, B. (1985). Age and the Life Course. In In R.H. Binstock and L.K. George (eds). Handbook of Aging and the Social Sciences. San Diego, CA: Hartcourt Brace. Hakmiller, K.L. (1 962). Social Comparison Processes Under Differential Conditions of Ego Threat. Hattery, A. (2001). Women, Work, and Family: Balancing arui Weaving. Thousand Oaks, CA: Sage Publications. Havighurst, A. (1963). Successful Aging. In R.H. Williams, C.Tibbitts, and W. Donohue (eds.). Process of Aging, Social and Psychological Perspectives, Vol. 1. New York: Atherton. Hays, S. (1996). The Cultural Contradiction of Motherhood. New Haven, CT: Yale University Press. Hesse-Biber, S., and Carter, G.L. (2000). Working Women in America: Split Dreams. New York: Oxford University Press. Hill, R..andMattessich, P. (1979). Family Development Theory and Life Span Development. In P. Baltes and O. Brim (eds). Life Span Development and Behavior. Vol. 3. New York: Academic Press. Hill-Collins, P. (1994). Shifting the Center: Race. Class, and Feminist Theorizing About Motherhood. \n Mothering: Ideology, Experience, and Agency. Glenn, E.N., Nakano, E., Chang, G., and Forcey, L.R. (Eds). New York: Routledge. Hoeming, E.M., and Schaeffer, D. (Eds.). (1984). Inter generational Relationships. New York: C.J. Hogrefe, Inc.

PAGE 143

129 Hochschild, A.R. (1989). The Second Shift. New York: Avon. Hochschild, A.R. (1997). The Time Bind: When Work Becomes Home and Home Becomes Work. New York: Metropolitan Books. Klein, E. (1984). Gender PoUtics. Cambndge, MA: Harvard University Press. Kurdek, L.A. (1993). Predicting Marital Dissolution: A 5-Year Prospective Longitudinal Study of Newly wed Couples. Journal of Personality and Social Psychology. 64,2:221-242. Langley, W.E. (ed.). (1994). Women's Rights In the United States: A Documentary History. Westport, CN: Greenwood Press. Lawton, L, Silverstein, M., andBengtson, V.L. (1994). Affection, Social Contact and Geographic Distance Between Adult Children and Their Parents. Journal of Marriage and the Family. 56:57-68. Lemer, R.M. (ed.). (1998). Handbook of Child Psychology: Vol 1, Theoretical Models of Human Development (5'^ ed.). New York: Wiley. Levinson, D.J., and Levinson, J.D. (1996). Seasons of a Woman's Life. New York, NY: Knopf Locke, D. (1992). Increasing Multicultural Understanding. Newbury Park, CA: Sage. Mandell, N., and Duffy, A. (Eds). (2000). Canadian Families: Diversity. Conflict, and Change. 2"'' ltd. Toronto: Harcourt Brace. Mangen, D.J., Bengtson, V.L., and Landr>', PH. (eds.). (1988). Measurement of Inter generational Relations. Newbury Park, CA: Sage Publications. Mangen, D., and Peterson, W. (eds.). (1982). Handbook of Research Instruments in Social Gerontology (Vol. 2). Minneapolis, MN: University of Minnesota Press. Manheim, K. (1952). The Problem on Generations. Essays in the Sociology of Knowledge. London: Routledge and Kegan Paul. Margolis, Maxine. (2000). True to Her Nature: Changing Advice to American Women. Waveland Press. Martin, J, Hamilton, B, Ventura, S.,Menacker,F., and Park, M. (2002). Births: Final Data for 2000. National Vital Statistics Reports. 50.5. Martin-Matthews, A. (2000). Change and Diversity' in Aging Families and Intergenerational Relations. In Canadian Families: Diversity. Conflict, and Change. 2"'' Ed. Mandell, N. and Duffy, A. (Eds). Toronto: Harcourt Brace.

PAGE 144

130 Mattessich, P., and Hill, R. (1987). Life Cycle and Family Development. In M.B. Sussman and S.K. Steinmentz, Handbook of Marriage and Family. New York: Plenum Press. McChesney, K.Y., andBengtson, V.L. (1988). Solidarit\\ Integration, and Cohesion in Families: Concepts and Theones. In Measurement of Intergenerational Relations. Mangen, D.J., Bengtosn, V.L., and Landry. PH. (eds). Newbury Park, CA: Sage Publications. McMahon, M. (1995). Engendering Motherhood: Identity and Self-Transformation in Women 's Lives. New York: Guilford Press. McPherson, B.D. (1998). Aging as a Social Process. 3'''^ Ed. Toronto: Harcourt Brace. Milardo, R.M. (2001). Understanding Families Into the New Millennium: A Decade in Review. Minneapolis, MN: National Council on Family Relations. Miller, R.B., and Glass, J. (1989). Parent-Child Attitude Similarity Across the Life Course. Journal of Marriage and the Family. 5 1 : 99 1 -997. Mills, T.L., Wakeman, M.A., and Fea, C. (2001). Adult Grandchildren's Perceptions of Emotional Closeness and Consensus With Their Maternal and Paternal Grandparents. Journal of Family Issues. 22,4: 427-455. Minkler, M., and Estes, C. (eds.). (1991). Critical Perspectives on Aging. Amityville, NY: Baywood Press. Moen, P. (2001). The Gendered Life Course. In Handbook of Aging and the Social Sciences. 5'^ Ed. Binstock, R.H. and George, L.K. (Eds). San Diego, CA: Academic Press. Moen, P. (1992). Women 's Two Roles: A Contemporary Dilemma. New York: Aubum House. Moore, K.A., and Sawhill, I.V. (1984). Implications of Women's Employment for Home and Family Life. In Work and Family: Changing Roles for Men and Women. Voydanoff, P. (ed.). Palo Alto. CA: Mayfield. Neugarten, B.L. (1968). The Awareness of Middle Age. InNeugarten, B.L. (Ed). Middle Age and Aging. Chicago: University of Chicago Press. Papemow, PL. (1 993). Becoming a Stepfamily: Patterns of Development in Remarried Families. San Francisco: Jossey-Bass. Pillemer, K., and Suitor, J.J. (1998). Baby Boom Families: Relations with Aging Parents. Generations. 22,1:65-70.

PAGE 145

131 Popenoe, D. (1999). Paternal Androgyny. In C. Albers (ed.). Sociology of Families: Readings. Thousand Oaks, CA: Pine Forge Press. Reskin, B., and Padavic, I. (1994). Women and Men at Work. Thousand Oaks, CA: Pine Forge Press. Rexroat, C, and Shehan, C. (1984). Expected Versus Actual Work Roles of Women. American Sociological Review. 49:349-358. Riley, M.W. (1971). Social Gerontology and the Age Stratification of Society. American Sociological Review. 52: 1-14. Rile>'. M.W. (1987). On the Significance of Age in Sociology. American Sociological Review. 52,1:1-14. Ritzer, G. (1996). Modern Sociological Theory. New York: McGrawHill. Roberts, R.E.L, and Bengtson, V.L. (1993). Relationships with Parents, Self-Esteem, and Psychological Weil-Being in Young Adulthood. Social Psychology Quarterly. 56,4:263-277. Rodgers, R.H, and White, J.M. (1993). Family Devleopment Theory. In P.O. Boss, W.J. Doherty, R. LaRossa, W.R. Schumm. and S.K. Steinmetz (eds). Sourcebook of Family Theories and Methods, A Contextual Approach. New York: Plenum Press. Stewart, A.J., and Vandewater, E.A. (1999). If I Had it to Do Over Again": Midlife Review, Midcourse Corrections, and Women's Well-Being in Midlife. Journal of Personality and Social Psychology. 76,2:270-28 1 Roper Organization. (1995). The Virginia Slims American Women's Opinion Poll. Storrs. CT: The Roper Center. Rossi, A.S. (ed.). (1985). Gender and the Life Course. Hawthorne, NY: Aldine Publishing Co Rossi, AS., and Rossi, PH. (1990). Of Human Bonding: Parent-Child Relations Across the Life-Course. New York: Aldine de Gruyter. Rosow, I. (1974). Socialization to Old Age. Berkeley, CA: University' of California Press. Ryder, N.B. (1965). The Cohort as a Concept in the Study of Social Change. American Sociological Review. 30,6: 843-86 1

PAGE 146

132 Sabatelli, R.M., and Shehan, C.L. (1993). Exchange and Resource Theories. In P.G. Boss, W J Doherty, R. LaRossa, W.R. Schumm, and S.K. Steinmetz (eds.). Sourcebook of Family Theories and Methods. A Contextual Approach. New York: Plenum Press. Sanchez, Y.M. (1997). Families of Mexican Origin. In M.K. DeGenova(ed.), Fa/n//ie5 in Cultural Context: Strengths and Challenges in Diversity. Mountain View, CA: Mayfield Publishing Co. Schwartz, M.A., and Scott, B.M. (2000). Marriages and Families, Diversity and Change. New Jersey: Prentice Hall. Segura, DA. (1994). Working at Motherhood: Chicana and Mexican Immigrant Mothers and Employment. \x\ Mothering: Ideology. Experience, and Agency. Glenn, E.N., Nakano, E., Chang, G., and Forcey, L.R. (Eds). New York: Routledge. Silverstein, M, and Bengtson, V.L. (1997). Intergenerational Solidarity and the Structure of Adult Child-Parent Relationships in American Families. American Journal of Sociology. 103,2:429-460. Silverstein, M., Conroy, S.J., Wang, H., Giarrusso, R., and Bengtson, V.L. (2002). Reciprocity m Parent-Child Relations Over the Adult Life Course. Journal of Gerontology: Social Sciences. 57B,1:S3-S13. Stetson, D. M. (1997). Women's Rights in the U.S.A.: Policy Debates and Gender Roles (2nd edition). Garland Publishing. Suls, J., and Mullen, B. (1982). From Cradle to the Grave: Comparison and Self Evaluation Across the Life Span. Psychological Perspectives on the Self Hillsdale, NJ: Lawrence Earlbaum. Suls, J, and Wheeler, L. (eds.) (2000a). Handbookof Social Comparison: Theory and Research. New York, NY: Kluwer Academi/Plenum Publishers. Suls, J. and Wheeler, L. (2000b). A Selective History of Classic and New-Social Companson Theory. In Handbook of Social Comparison: Theory and Research. New York, NY: Kluwer Academi/Plenum Publishers. Sussman, M.B., and Steinmetz, S.K. (eds). (1987). Handbook of Marriage and Family. New York: Plenum Press. Thornton, A, Orbuch, T, and Axinn, W. (1995). Parent-Child Relationships Dunng the Transition to Adulthood. Journal of Family Issues. 16:538-564.

PAGE 147

133 Thornton, A., and Young-DeMarco, L. (2001). Four Decades of Trends in Attitudes Toward Family Issues in the United States. The 1960's Through the 1990's. Journal of Marriage and the Family. 63 : 1 0091 03 7 Tubman, J., and Lemer, R. (1994). Continuity and Discontinuity in the Affective Experiences of Parents and Children: Evidence from the New York Longitudindal Study. American Journal of Orthopsychiatry. 64: 112-125. U.S Bureau of the Census. (1998). Statistical Abstract of the United States. Washington, DC: US Government Pnnting Office: 409. Table 655. U.S. Department of Education, National Center for Eduation Statistics. (1998). Digest of Education Statistics. Washington, DC: U.S. Government Pnnting Office. U.S. Department of Labor. (2000). Employment and Earnings. Washington, DC: U.S. Government Printing Office. Voy danoff. P (Ed. ). ( 1 984). Work and Family: Changing Roles for Men and Women. Palo Alto, CA: Mayfield. Weldon, M. (1997). Father Only Families are Increasing. Chicago Tribune, July6: sec. 13.3. White, L.K. (1990). Determinants of Divorce: A Review of Research in the Eighties. Journal of Marriage and the Family. 52 : 904-9 1 2. Whiteford, L.M.,andGozalez, L, (1995). Stigma: The Hidden Burden of hifertility. Social Science and Medicine; 40, January :27-36. Wilkinson. D.Y. (1997). American Families of African Descent. In M.K. DeGenova (Ed.). Families in Cultural Context: Strengths and Challenges in Diversity. Mountain View, CA: Mayfield Publishing Co. William, R.H.. Tibbitts, C, and Donohue, W. (1963). Process of Aging, Social and Psychological Perspectives, Vol. 1. New York: Atherton. Wilmouth, J.M. (2002). Synthesis Across Disciplines: Examples from Research on Family Relationships in Later Life. The Gerontologist. 42,2:281-284.

PAGE 148

BIOGRAPHICAL SKETCH Melanie Ann Wakeman was bom in 1970, in Boynton Beach, Florida. After living in Florida, Wales, and Spain, Melanie attended and finished high school in San Diego, California. From there, she attended California State University Fullerton where she received a bachelor's degree in sociology in 1997, and a master's degree in sociology in 1999. In 2001, Melanie was awarded a predoctoral fellowship at the University of Florida Institute on Aging. This fellowship allowed her to concentrate on her research efforts, focusing on the intersections of aging, gender, and family. After completing her doctorate in sociology from the University of Florida, Melanie will begin a postdoctoral position at the University of Southern California, Andrus Gerontology Center. 134

PAGE 149

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Cbrtstance L. Sl[}6ha/i, Chair Professor of Sociology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Amy Mehj;a^an Pienta, Cochair Assistant Professor of Sociology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Leonard Beeghley Professor of Sociology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. f^^h^i^c (^^^-^-x^ Barbara A. Zseitibik Associate Professor of Sociology

PAGE 150

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. oi' Tanya KoropeckyjiCox Assistant Professor of Sociology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Angel Kwolek-Folland Professor of History This dissertation was submitted to the Graduate Faculty of the Department of Sociology in the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. May 2005 Dean, Graduate School

PAGE 151

. Wl H-'~! UNIVERSITY OF FLO^'D* 3 1 262 08285 333 3


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EDL021V3J_NKN7K6 INGEST_TIME 2017-07-11T22:16:36Z PACKAGE UF00100965_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES