The Role of Cohabitation in Recent Family Formation Behaviors……

Stratification in American Family: Single, Cohabiting, or Married at the Birth of a First Child

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Anna M. Cunningham, M.A.

Graduate Program in Sociology

The Ohio State University

2010

Dissertation Committee:

Zhenchao Qian, Advisor

Liana Sayer

Claire Kamp Dush

Anna M. Cunningham

2010

ABSTRACT

The latter half of the twentieth century has ushered in substantial changes in the institution of the American family. The declining significance of marriage and the increasing prevalence of cohabitation in recent decades, both features of even broader economic and social shifts, have reshaped the life course pathways of today’s young adults. Through most of the 20th

Century, across all socio-economic groups, the normative life course pathway in the U.S. prescribed marriage followed by parenthood. However, as young adults delay or forgo marriage, cohabitation and non-marital childbearing have become more prevalent. Although marriage is still viewed as the ideal entrée to parenthood among most young adults, a substantial share of the population forms families outside of marital unions. Currently, nearly 40 percent of all U.S. births occur outside of marriage, and half of these births are to cohabiting couples (Mincieli et al.,

2007). The percentage of those born to single parents or unmarried cohabiting parents raises questions about the present state and future trajectory of American families. Although ample research has addressed the sweeping family changes of the last few decades, cohabitation’s place in U.S. family systems and cycles has been under-explored.

I contribute to this research using the life course perspective to compare individuals across cohorts about their family formation behaviors. I explore the mechanisms that may have contributed to different behaviors within and across generations. Specifically, I examine the relationship contexts of never married women’s first conceptions, and their subsequent union

ii transitions following a first non-marital pregnancy. I then explore the correlations between diverse childhood family experiences and adulthood behaviors.

My results suggest that cohabitation is beginning to play a prominent role in the stratification of the American family. The life course pathway’s of today’s young adults are increasingly divided by race and educational attainment. While college educated White women mostly follow the traditional pathway of marriage followed by childbearing, less educated Whites are increasingly having children while cohabiting. On the other hand, Black women increasingly have children outside of any co-residential arrangement. Indeed, rapid social change has a profound effect on women’s union choice and timing of childbearing. Because the relationship and family trajectories of today’s young adults continue to evolve and life course pathways are growing more complex, the results of this dissertation suggest the need for additional research into the rapidly changing American family.

iii

Dedication

Dedicated to My Three Marie’s

ACKNOWLEDGMENTS

I would like to thank my advisor Zhenchao Qian for the continued support, mentorship and guidance he has offered throughout my ―bumpy‖ graduate career. I have had some unexpected turns in my graduate career and I know I have provided my advisor with a few headaches along the way, but Zhenchao never gave up nor let me give up on the completion of this dissertation. He is an outstanding faculty member of Sociology Department of The Ohio

State University, but he is also an incredible person. It has been an honor to have had him as my advisor, and I am deeply grateful for all he has done for my graduate career.

I am also grateful to Daniel Lichter and Sharon Sassler, who provided me with several research and publication opportunities that have benefited my graduate career immensely. I would like to thank the members of my dissertation committee, Liana Sayer and Clair Kamp-

Dush for the valuable advice and support they have provided throughout my time at Ohio State. I wish to acknowledge the help of the Department of Sociology at Ohio State and the Initiative in

Population Research at Ohio State for providing numerous valuable research opportunities and generous funding.

I would like to think my friend Amanda Kennedy for her support, advice and friendship throughout my graduate student career. I would like to acknowledge Amber Ault for the amazing editorial support she has provided, but also for the invaluable advice and coaching she offered during the final stages of this dissertation. I would like to thank my father for his support.

v I would like to thank my husband John. Over the last few months , while I was dedicating every waking moment to my dissertation, you were working the ―second shift‖ taking care of our new baby girl. Arlie Hochschild would be proud. Thank you my love.

Lastly I would like to thank the three Marie’s in my life. First, my grandmother, Marie

Higgins. She was a pioneer of her time. Born in 1912 in rural Appalachia, she was the first member of her family and was one of the first women of Magoffin County Kentucky to pursue a college education. My grandmother helped to pave the road that I now walk without obstacles.

Second, my mother, Donna Marie Cunningham. You followed in the footsteps of my grandmother. You never took no for answer, and as a child, you instilled in me the strength and persistence I needed to complete my educational endeavors. The last three years of my life have been incredibly trying, but with your unending support I have been able to push forward. You have been my number one cheerleader---as well as agitator---and I thank you for that. You have also provided financial assistance throughout my time as an impoverished graduate student, and that has also been much appreciated. Thank you Mom, I love you.

Finally, my baby girl, Sophia Marie. Thank you for taking long naps so I could write and edit my chapters, and never fussing or complaining--only smiling. Sophia you have only been in my life for 7 months, and yet you are the reason I was able to complete my graduate education.

You have inspired me more than you will ever know and you have taught me what true love really is. Thank you my sweet angel, I love you

vi VITA

August 14, 1978 ………………………….Born – Wheelersburg, Ohio

2000 ……………………………………….B.A. Sociology, Otterbein College

2003………………………………………. M.A. Sociology, The Ohio State University

2008 to present…………………………… Researcher/Data Base Manager Synthesis Inc.

2006 to 2008 ……………………………. .Research Associate, Nationwide Children’s

Hospital

2001-2006…………………………………Graduate Teaching and Research Associate,

The Ohio State University

PUBLICATIONS

Cunningham, Anna M. and Chris Knoester 2007. ―Marital Status, Parenthood, and Psychological Well-being.‖ Sociological Inquiry 77 (2) 264–287

Sassler, Sharon and Anna M. Cunningham 2008. ―How Cohabitors View Childbearing.‖ Sociological Perspectives 51 (1) 3-28.

Sassler, Sharon, Cunningham, Anna M., and Dan Lichter. 2009 ―Intergenerational Patterns of Union Formation and Marital Quality,‖ Journal of Family Issues 30, 757- 786.

Fields of Study

Major Field: Sociology

vii

TABLE OF CONTENTS

Abstract...... ii

Dedication...... iv

Acknowledgements...... v

Vita...... vii

List of Tables ...... ix

List of Figures ...... x

Chapter 1: Introduction ...... 1

Chapter 2: Theoretical Perspectives...... 8

Chapter 3: Data, Measures, and Analytic Strategy ...... 16

Chapter 4: Changing Trends in Non-marital Childbearing and Union Formation: The Role of Cohabitation in Recent Family Formation Behaviors……...... 29

Chapter 5: Childhood Family Structure: Intergenerational Patterns of First Union Formation and Childbearing behaviors………………………………………………...... 77

Chapter 6: Conclusion...... 116

References...... 132

viii LIST OF TABLES

Table 3.1: Sample Characteristics at Wave 1 by Attrition Status and Sex……………………28

Table 4.1: Characteristics of Female Respondents by Cohort...... 67

Table 4.2: Characteristics of Single Non-Cohabiting Women by Conception Status and

Cohort……………………………………………………………………...... 68

Table 4.3: Characteristics of Cohabiting Women by Conception Status and Cohort...... 69

Table 4.4: Odds Ratios for Logistic Regression Analysis Predicting the Likelihood of Conceiving

Among Single Non-Cohabiting Women………………………………………………….…....70

Table 4.5: Odds Ratios for Logistic Regression Analysis Predicting the Likelihood of Conceiving

While Cohabiting Among Women Who Have Entered Into a Cohabitation ...... 71

Table 4.6: Odds Ratios for Logistic Regression Analysis Predicting the Likelihood of Marrying

Within 3 Months of First Child's Birth Among Never Married Pregnant Cohabiting Women…72

Table 4.7: Relative Risk Ratios from Multinomial Logistic Regression Analysis Predicting the

Likelihood of Forming a Union Within 3 months of First Child's Birth Among Never Married

Pregnant Single Women……………………………………………………………………...... 73

Table 5.1: Childhood family structure and background characteristics of focal women…….108

Table 5.2: Focal women's first union formation type by childhood family structure………...109

Table 5.3: Relationship context of focal women's first conception by maternal first conception…………………………………………………………………………………….109

Table 5.4: Odds ratios for focal women's first union formation……………………………...110

ix Table 5.5: Relative risk ratios of focal women experiencing a conception……………………………………………………………………………………111

vii LIST OF FIGURES

Figure 4.1: Educational Attainment of All Women by Cohort……………………………....74

Figure 4.2: Proportion of Women who Cohabited Prior to First Marriage………….……....74

Figure 4.3: Educational Attainment of Women Who Conceived Outside of Any Union by

Cohort…………………………………………………………………………………..……..75

Figure 4.4: Response to a Single Non-cohabiting Conception by Cohort…………………..75

Figure 4.5: Educational Attainment of Cohabiting Women Who Conceived by Cohort…..76

Figure 4.6: Response to a Cohabiting Conception by Cohort…………………………….....76

Figure 5.1: Parental union transitions following a divorce……………………………….....113

Figure 5.2: Focal women's first union formation experience ……………………….……....113

Figure 5.3: Maternal and focal women's nonmarital childbearing behaviors.…………..….114

vii

CHAPTER 1

INTRODUCTION

The latter half of the 20th Century saw substantial changes in the institution of the family in the United States. Delays in age at first marriage, rising divorce rates, increasing cohabitation and greater non-marital childbearing have dramatically reshaped American family patterns. In this project, I focus specifically on two important and related shifts in the family formation process: a) the rise in non-marital childbearing and b) the increasing role of cohabitation in the family formation process. In less than a three-decade span from 1970 to 2000, cohabitation moved from being a relatively rare phenomenon to a relatively normative event in the life course of young adults (Smock 2000). In 2007 there were 6.4 million cohabiting couples, up from 2.9 million in 1996, and 1.5 million in 1980 (U.S. Census Bureau 2007). Recent estimates find that nearly 70 percent of all first co-residential unions now begin with cohabitation rather than marriage (Kennedy and Bumpass 2008).

Coupled with this increase in cohabitation was a substantial rise in fertility outside of marital unions. The proportion of U.S non-marital births reached an all time high of 39 percent of all births in 2006, up from 28 percent in 1990 and 18 percent in 1980 ( Hamilton et al 2007;

Ventura et al; Hamilton et al 1995). Of these non-marital births, over half were to cohabiting women (Mincieli, Manlove, McGarrett, Moore and Ryan 2007). Bumpass and Lu (2000) posit that much of the increase in non-marital childbearing can be attributed to the increase in childbearing among cohabiting parents. While the number of marital births declined, and single

1 non-cohabiting births remained somewhat steady, the number of births to cohabiting women more than doubled (Bumpass and Lu 2000). Today, 1 in every 8 births in the United States is to a cohabiting couple (Manning et al. 2004). So, while most American children are born into two- parent households, the parents in those households are less likely than ever before to be tied by bonds of legal marriage.

Family scholars argue that these rapid demographic changes over the last three decades reflect profound changes in the institution of marriage. A half-century ago, marriage was the dominant institution that defined one’s entrance into adulthood. However, the decline in men’s wages, expanding educational and labor opportunities for women, advances in birth control, increased access to higher education among the working class and women of all racial groups and changing norms and attitudes, the latter half of the twentieth century transformed the institution of marriage from the foundation for family building behaviors into a symbolic marker of prestige and personal achievement (Cherlin 2005; 2009; Nock 2005). Today, marriage has become increasingly less relevant as a context for childbearing and rearing. Consequently, as the relationship between marriage and parenthood weakens, cohabitation has become more prominent in the framework of childbearing.

Although these changes in union and family formation behaviors have been evident across all social groups, the declining role of marriage and increased role of cohabitation and non- marital childbearing have been most prominent among the socio-economically disadvantaged

(Fein et al, 2003). Inversely, marriage has increasingly become a symbol of social status associated both with class and with race. Individuals with higher education, incomes, and stable employment are substantially more likely to be married (Popenoe & Whitehead 2007), while economically disadvantaged Whites and minorities are less likely to enter into marital unions (Fry

& Cohn 2010). Whether one decides to marry is increasingly reflective of one’s social class.

2 While marriage is still highly valued, and the majority of young adults express the desire to marry eventually, disadvantaged women perceive marriage as unattainable due to the social and economic barriers they often encounter (Edin & Reed 2005; Gibson et al, 2005). However, these same barriers have not been applicable to childbearing. Although the postponement of marriage has been experienced across the educational spectrum, the increase in non-marital childbearing has been greatest for women without a college education (Ellwood & Jenks 2004).

Not surprisingly, the family lives of children have been diverging over the last few decades

(McLanahan 2004).

Thus, the increasing role of cohabitation in the family formation process may actually exacerbate a growing disparity in children’s economic resources. As marriage comes to serve as the line of demarcation between the ―haves‖ and the ―have nots,‖ and cohabitation continues to become more prominent in young adults’ family formation decisions, the growing inequalities between married and unmarried families will likely intensify. Understanding the changing role of cohabitation in the family system is imperative, given the detrimental consequences that have been linked to economic deprivation. Importantly, because children raised in unmarried parent households have substantially higher rates of poverty than children raised in married parent households, we need to understand how cohabitation, as an alternative to single parenthood or as a substitution for marriage affects outcomes for children for better or for worse. However, the literature addressing cohabitation's increasing role in the stratification of the American family among recent cohorts of young adults remains limited.

I argue that in order to understand the changing family trajectories of young adults, one must have an understanding of early life course factors. Relationship and family formation behaviors are an intergenerational process, and young adults’ decisions and choices are strongly influenced by the values and behaviors they learned during childhood (Elder 1998; Shanahan

3 2000). Young adults who transitioned into adulthood in the mid to late 1990’s and early 2000’s represent the first cohort of young adults who experienced, as children, the rapid growth in parental divorce, as well as a growing diversity in family type (e.g. cohabitation, step-family, single mother family). Evidence suggests that these rapid family changes during childhood and adolescence have likely reshaped the life course pathways of today’s young adults (Elder et al.

2003). The once standardized pathway of leaving the parental home, completing an education, marrying, and then having children is no longer the normative route for the majority of today’s young adults (Buchmann 1989; Modell & Goodman 1990).

Empirical evidence consistently has demonstrated a link between childhood family experiences and young adult’s relationship and childbearing outcomes (Bulanda and Manning

2008; Cavanagh, Schiller, and Riegle-Crumb 2006; Graefe and Lichter 1999; McLanahan and

Sandefur 1994). However, the influence of childhood experiences is complex and poorly understood because much of the previous research has ignored the increasing diversity in parental union histories. Childhood family instability is often assessed with a simple dichotomous measure of ―divorce‖ versus ―did not divorce‖ or ―intact‖ versus ―non-intact‖ family. Relatively few studies have adequately captured the increasing presence of cohabitation in childhood family experiences on relationship and family trajectories. These diverse family structures have likely played an integral role in the changing union and family formation patterns of today’s adults.

Guided by the life course perspective, I address the changing family formation behaviors over the last couple of decades, and the possible mechanisms that may have contributed to the shifting relationship and family trajectories of today’s young adults. The life course perspective provides me with a theoretical framework that can systematically study family change. It calls on family scholars to assess not only the impact of immediate family environments on individuals, but also the historical contexts within which individuals live when studying family change.

4 Objectives of Study

By utilizing more up-to-date data, I hope to extend the existing literature (Manning 1993;

Raley 2001), by examining the relationship and childbearing patterns across two cohorts of never married pregnant women. Additionally, the incorporation of a more recent cohort of young women permits me to illustrate the growing racial and educational inequalities within the

American family. Finally, I include a more comprehensive measure of childhood family structures in order to assess how families of origin may have influenced these changing behaviors.

As previously mentioned, this study utilizes the life course perspective and its key principles of historical contexts, situating individuals within cohorts, and linking lives within and across generations to address this gap in the literature. First, the life course perspective highlights the importance of cohorts. Young adults who entered into adulthood during the period of rapid demographic change in the last few decades may experience different pathways to union and family formation than individuals who entered into adulthood prior to these substantial changes.

Women who were born at different periods in history have experienced social and economic events that have uniquely shaped their union and family formation behaviors.

In Chapter 4 of this study, I examine the changing relationship contexts of non-marital childbearing across two cohorts of never married women. Women in the early cohort transitioned into adulthood during the 1970’s and early 1980’s when the economic and cultural shifts that swept across the latter half of the 20th Century were only beginning to emerge, while women in the later cohort entered into adulthood during the late 1990’s and early 2000’s, when these changes were well established. By examining changing union and family formation behaviors across these two cohorts of women, I hope to illustrate that cohabitation is increasingly serving as a substitute for marriage- at least among certain groups of women. I further hope to demonstrate

5 how the life course pathways of today’s young adults are closely associated with one’s race and educational attainment. The increasing stratification of the American family warrants a greater understanding of cohabitation’s role in this process.

The life course perspective also highlights the principle of linked lives. That is, experiences and circumstances early in life affect later life outcomes. In Chapter 5, I explore how these changing life trajectories of young adults are influenced by one’s childhood family experiences of divorce, remarriage and/or cohabitation. I hope to illustrate that the increasing presence of cohabitation in young adults’ childhood family experiences has likely influenced their own relationship and family formation decisions.

In sum, the continued growth in cohabitation and non-marital childbearing suggests a need for studies of recent cohorts to assess the extent of cohabitation’s role in the American family (Fein et al. 2003). Although previous studies have examined never married women’s changing marital and cohabitation behaviors following a non-marital pregnancy, no studies, to my knowledge, have assessed these changes in a more recent cohort of young adults. Additionally, research addressing the influence of childhood family structures, including parental cohabitation, on young women’s first union formation and first conception in recent years remains limited.

Summary

In summary, this dissertation is an examination of the emerging demographic trends of the last two decades. By examining early life course events, this study demonstrates the influence of childhood family structures on these changing behaviors. In Chapter 2, I describe the life course perspective, which is the overarching framework for my dissertation research. Data, sample, variable measurement, and methods are presented in Chapter 3. Chapter 4 explores the changing contexts of first union conceptions and subsequent union transitions over the last two decades. In Chapter 5, I examine the intergenerational influence of childhood parental union

6 transitions on young women’s own relationship and family formation behaviors. Chapter 6 consists of the discussions and conclusions from the analysis. The discussion in this chapter summarizes the overall picture of the changing demographic behaviors that have transpired over the last two decades and the influence childhood family structure has had on these changing behaviors.

7 CHAPTER 2

THEORETICAL FRAMEWORK

LIFE COURSE PERSPECTIVE

Over the last few decades the life course pathways of young adults in America have experienced a significant transformation. Previously, individuals experienced a standard sequencing of transition markers that signified the emergence into adulthood (Shanahan 2000;

Modell 1989). The standardized trajectories of departure from the parental home, completion of education, labor force entry, and marriage followed by childbearing were the traditional life course events that defined young adulthood (Buchmann 1989; Modell and Goodman 1990). The rapid demographic changes of divorce, marital delay, cohabitation, and non-marital childbearing that swept across the latter half of the twentieth century drastically altered this pathway. The life course of today’s young adults is becoming more diverse. Though school, work and family are still significant markers of early adulthood, the timing, ordering and duration of these events have substantially changed among young adults in recent years (Modell 1989; Rindfuss 1991;

Shanahan 2000). In this chapter, I discuss the life course perspective which provides an ideal theoretical framework for exploring these changing trajectories and transitions of today’s young adults.

Developed by Elder (1975; 1978), the life course perspective emerged as a framework that could systematically examine change in family processes. Since its origination, this perspective has expanded as an interdisciplinary approach and combines specific elements from

8 psychology, demography, family studies, and sociology (Clausen 1985; Elder, Johnson and

Crosnoe. 2003; Morgan 1996). The life course perspective highlights not only the importance of an individual’s immediate family environment, but also the historical context within which an individual lives. That is, the life course is defined by a series of ordered stages that an individual progresses through over time. Each stage in life provides the groundwork for subsequent stages, and in general, stages cannot be reordered or omitted (Clausen 1985). Understanding the progression of individuals over the life course requires attention to individuals’ past experiences in order to understand the likelihood of future transitions (Elder 1992).

The life course is structured around the trajectories and transitions individuals experience over their life span. Individuals experience several long term trajectories such as work and family, but within each trajectory is the experience of several short term transitions such as graduating college or entering into a marital union. It is the ―interlocking‖ of these trajectories and transitions that structure the individual life course, a change in one may affect all others

(Elder 1992). These transitions and trajectories are experienced at both the individual and the broader social and historical levels. Individuals make a series of sequential choices and actions that help construct the various pathways but their choices are largely constrained and/or permitted by both the historical and familial context in which they live their lives (Elder et al. 2003).

This simplified review of the life course framework emphasizes the importance of past experiences on future outcomes and highlights the importance of historical contexts. The family context in which one grows up, coupled with the historical period in which one lives, will influence relationship and childbearing behaviors and decisions in adulthood. These two key elements of the life course perspective provide an ideal framework for studying the changing relationship and childbearing trajectories of today’s young adults. In the following sections I discuss these two tenets of the life course perspective in further detail.

Historical Context

Young adults who entered into adulthood during the rapid demographic changes of the last few decades may experience a different pathway to union and family formation than individuals who entered into adulthood prior to these substantial changes. Women who were born at different periods in history have experienced social and economic events that have uniquely shaped their fertility and union formation behaviors. In order to describe the influence of historical context on later life events, life course scholars have found the concept of cohort to be very useful (Hutchinson 1986). The study of cohort differences allows an understanding of why a change occurred and what factors may have been involved in creating the impetus behind certain societal transformations (Giele and Elder 1998). The cohort is a central aspect of this study.

A cohort is defined as a group of individuals born at the same historical time who experienced social and culture changes in the same sequence and at the same age (Rosow, 1978;

Ryder, 1965; Settersten and Mayer, 1997). Elder (1994) contends that individuals born in different years encounter different worlds with a different set of circumstances. For example,

Elder (1979) examined the influence of the economic hardships of the Great Depression on family and individual experiences on those born in the early 1920’s compared to those born in the late 1920’s. He found the impact of such dramatic income loss had a greater influence on the younger children, both in childhood and even into adulthood.

Some life course scholars propose that cohorts develop specific strategies to contend with the circumstances they encounter (Hutchinson 1986). This is evident in a study conducted by

Easterlin, Schaeffer, and Macunovich, (1993). The authors find that individuals of the ―baby boomer generation‖ responded to the strong economic competition their cohort faced by increasing the number of women in the work force, postponing marriage, delaying childbearing

10 until later ages, and having fewer children overall. These studies illustrate the concept of a cohort effect. That is, social and cultural changes have divergent influences on the life patterns of successive birth cohorts.

Young adults who transitioned into adulthood in the late 1990’s and early 2000’s represent the first cohort of young adults who experienced, as children, a rapid growth in parental divorce as well as a growing diversity in family type (e.g. cohabitation, step-family, single mother family). For example, in 1960 there were approximately 9 divorces for every 1000 married women. By 1980, this number jumped to nearly 23 divorces per every 1000 married women

(Clarke 1995). Moreover, slightly over 20 births per 1000 unmarried women were reported in

1960, compared to 30 per 1000 unmarried women in 1980 (National Center for Health Statistics

1996). Thus, this cohort of young adults grew up during a historically high and rapid period of family change than earlier or even subsequent cohorts did.

In Chapter four of this study, I examine two cohorts of women who entered into adulthood at very different points in time. Women in the early cohort transitioned into adulthood during the 1970’s and early 1980’s when economic and cultural shifts that swept across the latter half of the twentieth century were only beginning to emerge, while women in the later cohort entered into adulthood during the late 1990’s and early 2000’s when these changes were well established. The later cohort of young women are far more likely than the women in the earlier cohort to have experienced a parental divorce, witnessed a parental cohabitation, and spent time in a single parent home (Sassler et al. 2009). Witnessing these new and alternative family arrangements most likely influenced the union and family formation decisions of women in the later cohort differently than for women in the earlier cohort. Evidence consistently demonstrates that the experience of a parental divorce during childhood significantly increases an adult child’s likelihood of entering into a cohabitation, having a child outside of marriage, and witnessing their

11 own relationships disrupt than children who grew up in a stably married home (Teachman 2002;

2003; 2004; Wolfinger 2001;Wu and Martinson 1993). Yet, as divorce, cohabitation and non- marital childbearing have become more commonplace in the life course of young adults, the negative stigma once associated with these demographic behaviors has substantially weakened over time (Hobcraft 2008).

The influence of these changes is evident in the common practice of cohabitation as first union formation, and in the declining importance of marriage in childbearing decisions among more recent cohorts of young adults (Kennedy and Bumpass 2008). Yet, those born in the earlier cohort where marriage followed by childbirth was the dominant life course pathway, these same demographic shifts did not have the same effect. By utilizing two separate cohorts of young women who transitioned into adulthood at two very different points in time, I hope to extend our current knowledge of the role of cohabitation and non-marital childbearing in the life course of today’s young adults.

Prior experience/Linked Lives

The life course framework also emphasizes the importance of linked lives. In fact,

Shanahan (2000) posits that the increased complexity of life course transitions coincide with ―one of the most significant advances in life course studies‖ (p.675)—the linkage of childhood experiences to outcomes in adulthood. While age and timing are critical in an individual’s transition from one life course stage to the next, previous childhood events and experiences also influence later life transitions (Elder 1998). Family of origin demonstrates this point. Research has consistently established a link between childhood family structure and adult outcomes

(Bulanda and Manning 2008; Cavanagh Schiller and Riegle-Crumb 2006; Graefe and Lichter

1999; McLanahan and Sandefur 1994; Manning and Lamb 2003; Raley and Wildsmith 2004;

Thornton 1991). Parents’ own decisions surrounding relationship and family formation will

12 influence young adults’ subsequent decisions regarding entry into marriage, cohabitation and parenthood (Teachman 2002; 2003; Thornton 1991; Wolfinger 2003; Wu and Martinson 1993;

Wu 1996).

Individuals are typically entrenched in social relationships with family and friends over the life course. The experiences during one’s childhood will affect behaviors and choices during adulthood (Martinson and Wu 1992). Men and women may develop certain propensities towards certain family types, such as cohabitation, due to previous experiences and socialization. For example, the experience of a parental divorce during childhood will often increase the likelihood of an adult offspring’s own experience with divorce and marital discord dissolution (Axinn

Thornton and Xie 2007; McLanahan and Sandefur 1994; Teachman 2002; 2003; Wolfinger 2000;

2001). Those who experienced an intact family during childhood are more likely themselves to enter directly into a marriage versus cohabitation and less likely to divorce (Amato, 1996;

Cherlin, Kiernan, and Lindsay Chase-Lansdale, 1995; McLanahan and Sandefur, 1994;

Teachman 2002).

Whether one enters a cohabiting relationship is also influenced by childhood experiences.

Sassler, Cunningham and Lichter (2009) find that young adults who experienced a parental cohabitation have an elevated odds of cohabiting themselves relative to being married. Similar evidence demonstrates that children who experienced a non-traditional family are more likely to enter into a step-family during adulthood. Goldscheider and Sassler (2006) find young adults who experienced an alternative family structure while growing up were more likely to form a step-family union compared to those whose families remained intact. Clearly, parental choices and decisions have a significant impact on shaping children’s later life course trajectories by guiding individuals toward specific life pathways (Aquilino 1996). In Chapter 5, I argue that the growing diversity in childhood living arrangements among the cohort of young women

13 transitioning into adulthood in the late 1990’s and early 2000s played a critical role in their own union and family formation behaviors. I further contend that the life course pathways of women who experienced a family disruption in childhood are not uniform, and that the experience of alternative family formations during childhood allowed a diversification of pathways into adulthood (Shanahan 2000).

However, as previously mentioned, the increased acceptability of divorce, cohabitation and non-marital childbearing among young adults in recent years has reduced the negative stigma once associated with these demographic behaviors (Hobcraft 2008), and perhaps reducing some of the negative consequences on childhood and adulthood outcomes. The declining stigmatization surrounding these demographic behaviors is reflected in the majority of contemporary young adults’ views regarding union and family formation. In 2001, nearly 60 percent of high school seniors agreed that living together before marriage was a good idea, and more than half believed that a couple who decides to have and raise a child outside of marriage was either experimenting with a worthwhile alternative lifestyle or doing their own thing and not affecting anyone else (Johnston et al. 2002). This is a dramatic rise from approximately 30 percent of high school seniors in 1980 (Thornton and Young-DeMarco 2001). Whether this shift in attitudes regarding family formation behaviors has assuaged some of the negative consequences of parental divorce on adult offspring’s own union and family formation is unclear.

While some find that the negative effects of growing up in a broken home have abated over time

(Wolfinger 1999 ; 2009), others find that the effects of family structure have not attenuated and still persist among today’s young adults (Single-Rushton, Kiernan and Hobcraft 2005).

In summary, the life course perspective understands that the family context in which one grows up is important in determining later life transitions. However, it expands the idea of what is important to take broader societal and historical contexts into account as well. In this study I

14 argue that the rapid demographic and attitudinal changes over the last two decades will have a different impact on union transitions and childbearing behaviors among young adults transitioning into adulthood in the late 1990’s and early 2000’s than those in previous cohorts. I further posit that the experience of these demographic changes during childhood has also significantly contributed to the increasingly diverse life course pathways among today’s young adults.

CHAPTER 3

DATA, MEASURES, AND ANALYTIC STRATEGY

Data

The data source for this dissertation is the National Survey of Families and Households

(NSFH) (Sweet, Bumpass and Call 1988; Sweet and Bumpass 1996; Sweet and Bumpass 2000;

Wright 2003). The first wave of the NSFH was conducted in 1987-88 and involved a multistage probability sample of 13,007 adult respondents. Several portions of the main interview were self- administered to facilitate the collection of sensitive information (Sweet and Bumpass 1996). The original sample contains an oversampling of minorities, single-parent households, families with stepchildren, cohabiting couples, and persons recently married (Sweet and Bumpass 1996). In each household a randomly selected adult was interviewed. The survey included detailed information on the respondents own family formation histories, as well as detailed information on their family environment during childhood. The spouse or cohabiting partner of the main respondent also filled out a self-administered survey. In addition, Wave 1 collected specific information on one randomly selected focal child in which the main respondent had any biological, adopted, step-, or foster children who were under the age of 18. The overall response rate at Wave 1 was 74 percent (Sweet et al. 1988).

In the five year follow-up, the NSFH conducted Wave 2 (1992-1994) and interviewed

10,007 surviving respondents as well current spouses or cohabiting partners of the original Wave

1 sample. Similar to the process in 1987 and 1988, follow up interviews were conducted as a face-to-face, computer-assisted interview with self-administered questionnaires. Wave 2

interviews also included the original spouse or partner, whether or not that person was still living with the primary respondent. A follow-up telephone interview with the focal child

(N=2,505) now aged 10-23 was also conducted. When there was no Wave 1 focal child, and there were any children under age 5 in the household at the time of the Wave 2 survey, a child was randomly selected among the eligible children for a series of questions (Bumpass and Sweet

1996). The response rate for original respondents at the second wave was 82% (Sweet and

Bumpass 1996).

Between 2001 and 2003, a third wave of data was conducted but the scale was much smaller due to budgetary restrictions (Sweet and Bumpass 2002; Wright 2003). A subset of the original sample was re-interviewed including main respondents, spouses and focal children. In this study, I only utilize the focal child data from Wave 3. At wave 3, all interviews were conducted via telephone using CATI technology. Focal children were eligible for a wave 3 interview if they were at least 3 years of age at Wave 1, and had been eligible for a Wave 2 interview (at least 10 years of age at Wave 2). All focal children were 18-34 years of age when interviewed at time 3 (N=1,952). The focal child response rate was 48 percent (Sweet and

Bumpass 2002).

The NSFH provides a number of advantages that make it well suited for my dissertation.

First, the NSFH collected an oversample of cohabiting couples and single parent families.

Extensive questions such as ―Did you and your first (husband/wife) live together before you were married?‖ ―Did you live with anyone else before you were married?‖ were obtained at all three time points for both the main respondent and adult focal child. Furthermore, detailed beginning and end dates of each cohabitation episode as well as marriages were also collected at all three waves. In chapter 4, I focus on the changing role of cohabitation in the family formation process in the last two decades. Given the dramatic rise in the rates of cohabitation as well as the

17 alternative family forms that have emerged within the last few decades, this dataset is best suited for my analysis because of the extensive and detailed marital and cohabitation histories collected, and the nearly two decade span between Wave 1 and Wave 3 of the NSFH.

Second, the NSFH is a large-scale representative survey of families living in the United

States. With the release of the Focal Child Wave 3 data, intergenerational family patterns were made available. The first interviews with the parent and focal child were conducted while the child was on the verge of young adulthood, all focal children were 18 years of age or younger at the initial wave. By Wave 3, all focal children were now young adults aged 18-34. Additionally, data at all three waves were collected independently from parents and children, thus reducing any reporting biases that can often transpire when child data are based solely on parental reports

(Neimi 1968). Furthermore, Wave 2 of the NSFH provided detailed information on any marital or cohabitation changes of the main respondent that transpired between the first and second waves. This allows me to maximize all available information on parental union histories during childhood.

Lastly, the NSFH also collected detailed fertility histories from both the main respondent and the adult focal child. Respondents were asked the questions ―How many children have you given birth to/fathered?‖ and ―(Not counting adopted children,) In what month and year was your

(first, second...) child born?‖ Only information on conceptions that led to live births were collected in the survey.

Nonetheless, there are some challenges with using the NSFH data. In Chapter 4, I examine two cohorts of young women who are between the ages of 18 and 34 at the time of the survey. The first or ―early‖ cohort of women is comprised from the main respondent data from the first wave of the NSFH. The second or ―later‖ cohort of women is based on the focal child data from third wave of the survey. Wave 3 respondents are the focal children of the Wave 1

18 respondents who were parents at the time of their initial interview. Thus, these samples are linked in that respondents at Wave 3 are the children of the main respondents from Wave 1, however, they do not completely overlap. Only 10 percent of women in my later cohort sample have a parent that is represented among the women in my earlier cohort sample.

An additional limitation of the NSFH is the design of the Wave 3 sample selection process. For cases with a completed parental interview at time 3, an attempt was made to interview all focal children who were age 18 and older at time 3, regardless of whether or not an interview was completed at time 2. Therefore, a number of focal children at Wave 3 (N~200) did not have a Wave 2 interview. Thus, I am only able to employ Wave 3 focal child information and unable to utilize focal child data, such as adolescent well-being, obtained at Wave 2 of the NSFH.

Chapter 5 examines the impact of childhood family structure on the union and family formation patterns of adult children. The intergenerational family patterns provided by all three waves of the NSFH are ideal for this chapter, but I have to limit my analyses to Wave 1 and Wave 3 only for any focal child information. In sum, all three Waves of the NSFH data utilized in this dissertation are based on multistage probability samples of the civilian non-institutionalized population of women 18-34 years of age in the United States.

The budgetary constraints and sample selection process previously mentioned, may impact the overall sample I utilize in this dissertation. We know from previous research that respondents who do not participate in the subsequent waves of the NSFH as well as other large nationally representative surveys are more likely to be lower educated, minority, male, and divorced (Dew 2009; Fitzgerald et al. 1998; McNally and Sassler 2003; Smock and Manning

1997). These same traits characterize those most likely to form cohabiting relationships as well as those most likely to conceive outside of marriage (Bumpass and Lu 2000; Kennedy and

Bumpass 2008; Manning 1993). This should be of concern to me because in Chapter 4 the Wave

3 focal children are only selected if a parental interview at Wave 3 was administered. Given the strong intergenerational correlations between education, income, and marital outcomes (Amato

1996; Black et al. 2005; Solon 1992; Teachman 1992), this could be of issue when assessing individual’s most likely to conceive outside of a marital union. In addition, in Chapter 5 I examine the impact of growing up outside of a two parent married home on adult offspring’s own union and childbearing behaviors. Given that rates of sample attrition are greater among those with unstable marital histories (Fitzgerald et al. 1998), this could possibly influence my sample of young adults selected at Wave 3. Additionally, my measures of childhood family structure are assessed from only the primary respondent’s information at Waves 1 and 2 and I do not account for the gender of the primary respondent. However, as mentioned before, research indicates that males are more likely to drop out of the survey compared to females. Again, this may have some impact on my sample of young adults selected at Wave 3. In order to assess the potential impact these issues may have on my study sample, I analyze possible differences between the primary respondents of the Wave 3 survey and the primary respondents who were unavailable for follow- up. Results are presented in Table 3.1.

As displayed in Table 3.1, attrition was greater among younger, less educated, and non-

White respondents. In regards to gender, I do not find a substantial difference between those who left they survey and those who remain in the third wave. Approximately half the original samples of mothers as well as fathers are unavailable for the Wave 3 interview. In terms of family structure, there are relatively equally proportions of primary respondents that remained in the survey (e.g. similar number of married, divorced and/or cohabiting). Since the respondents who left the survey have family structure experiences comparable to the respondents who remain in the study, I do not foresee this having a large impact on my analyses conducted in Chapter 5.

Given the fact that the rates cohabitation and non-marital childbearing are higher among those

20 with lower levels of education and minorities, it is likely if the respondents who had left the survey had actually been retained, my sample size of women who experience a non-marital conception and/or cohabiting relationship would most likely increase. Thus, the results from this study must be interpreted in light of the attrition. The remaining sections of this chapter I describe the analytic sample, measurement variables, and the analytic strategy.

Analytic Sample

The data for all analyses are restricted to women between the ages of 18-34. Chapter 4 analyzes the changing trends and patterns of family formation over the last few decades, and is based on a final sample of 4,165 women—3,118 from Wave 1 and 1,047 from Wave 3 of the

NSFH. Chapter 5 assesses the influence of childhood family background on the relationship and childbearing trajectories of today’s young adults, and is based on a final sample of 933 women.

Variables and Measurement

Dependent Variables

The dependent variables used in this study to explore the changes in women’s union and family formation behaviors over the last two decades include non-marital conceptions, union context of first birth, and first union formation.

Non-marital Conception: Chapter 4 examines the changing trends in non-marital childbearing across cohorts. I create two dichotomous measures of non-marital childbearing. I first explore if a woman conceives outside of any union (including cohabitation). Childless women and women who conceive after marriage are included in this analysis. I then examine if a woman conceives while in a cohabiting union. Only women who have entered into at least one cohabiting union are included in this set of analyses (see Chapter 4 for more detailed explanation this measure).

Union Context of First Birth: Chapter 4 also explores the changing union context of first birth location. I examine the likelihood that a woman who conceives outside of marriage will transition

21 into a marriage, transition into or remain within a cohabitating union, or transition into or remain in a single non-cohabiting state within three months of childbirth. In this dissertation, I use women’s cohabiting and marital behaviors following a non-marital pregnancy as an index of whether cohabitation is becoming its own institutionalized family form (see chapter 4 for a more detailed explanation).

First Union Formation: Chapter 5 examines how childhood family structure contributes to the changing union and childbearing behaviors of young adult women. The third wave of the NSFH obtained detailed marital and cohabitation histories of the focal child respondent. Respondents who had married were asked ―Nowadays, many unmarried couples live together, sometimes they eventually get married and sometimes they don’t. Did you and your first (husband/wife) live together before you married?‖ Did you live with anyone else of the opposite sex before your first marriage?‖ Respondent who had never married were also asked extensive questions on dating and cohabitation histories: ―Nowadays, many unmarried couples live together, sometimes they eventually get married and sometimes they don’t. Have you ever lived with a partner of the opposite sex?

I only examine cohabitations prior to first marriage. In the first set of analyses, I examine the relationship between parental union transitions post-divorce and the likelihood of the focal child’s first union formation. I create a three-category dependent variable to assess focal women’s first union formation patterns. Categories are coded as follows: 1=Entered directly into marriage

(no prior cohabitation); 2=First entered into a cohabitation; 3=Never entered into cohabitation and/or marriage.

Key Independent Variables

The independent variables in my analysis include birth cohort, childhood family structure, race, educational attainment, and controls associated with union and childbearing behaviors.

Birth Cohort: I define two cohorts of women aged 18 to 34 in the analyses. Because the focal children at Wave 3 are only between the ages of 18-34, the same age range is applied to the earlier cohort of women at Wave 1. The first or ―early‖ cohort of women is drawn from the first

Wave of the NSFH. Women in this cohort were born between 1950 and 1969. Over half of the women in this cohort had entered into either a cohabitation or marital union by 1980. The second or ―later‖ cohort of women is drawn from the focal child Wave 3 data of the NSFH. Women in this cohort were born between the years of 1970 and 1983. The majority of women from the later cohort had either entered into a cohabiting or marital union by 2000. The nearly two-decade time span between these two cohorts of women proves ideal when exploring the changing union and family formation behaviors that have transpired over the last few decades.

Childhood Family Structure: In order to maximize the information on parental union histories, I employ both main respondent data from Waves 1 and 2 to assess the family structure of the focal women during childhood. First, I assess if the parental respondent experienced a divorce from the biological parent at either Wave 1 or Wave 2 (between waves). During the first wave interviews, extensive marital histories were collected from the main respondent. This included the number of previous marriages as well as the beginning and end dates of each marital union.

In addition, Wave 1 data included a question that asked the main respondent ―I've recorded.... as your biological children is (each of these children/he/she) the biological child of your current

(husband/wife/partner)?‖ Therefore, I am able to determine parental union transitions at Wave 1.

Wave 2 data of the NSFH contained a set of constructed variables that described the main

23 respondent's union history between Wave 1 and 2 of the NSFH. One measure indicated what the

Wave 1 union status was and what it became by Wave 2. An additional measure described the main respondent's union history since the Wave 1 union.

I employ a more comprehensive measure of childhood family structure by identifying four types of childhood living arrangements following a parental divorce: 1. Remarriage without prior cohabitation 2. Cohabitation prior to a subsequent remarriage, 3. Cohabitation without a subsequent remarriage, 4. Remained single (no cohabitation and/or remarriage).1

Race: In Wave 1 of the NSFH main respondents indicated their race/ethnicity. The question

―Which of the groups on this card best describes you?‖ with nine possible responses.

1-Black 2-White-not of Hispanic origin 3-Mexican American, Chicano, Mexicano 4-Puerto Rican 5-Cuban 6-Other Hispanic 7-American Indian 8-Asian 9-Other This question was not subsequently asked in the Focal Child Wave 3 survey. Therefore, I use parents’ race as a proxy for focal women’s race at Wave 3.

Education at time of conception: I collapse educational attainment into four mutually exclusive categories: less than a high school diploma; high school degree; some college experience; bachelor’s degree or more. I also create a dichotomous measure that assesses if the respondent is currently enrolled in a post-secondary institution.

1 A small number of main respondents were affected by attrition by Wave 2. Approximately 40 of those who were married at Wave 1, and 19 who were divorced and neither remarried nor cohabited at time 1 were lost to follow up. However, statistical tests indicated that there was not a significant difference between the union and family formation outcomes for the adult women whose parents did not have Wave 2 information and those whose did. Therefore, parents who were married at Wave 1 were assigned to the stably married category, and parents were who were divorced at Wave 1 were assigned to the divorce no subsequent co-residential union category.

Measures of Childhood and Family Background

I include a number of childhood and family background measures and potential risk factors for cohabitation and non-marital pregnancies, which may bias the results. These variables are:

Maternal educational attainment: A categorical variable that includes less than a high school diploma, some college experience, and bachelor’s degree or more. Maternal childbearing behaviors: A dichotomous variable that indicates if the focal woman’s mother experienced a non-marital conception. Receipt of public assistance: A dichotomous variable that assesses whether the respondent received a form of public assistance during childhood. Parental divorce: A dichotomous variable that indicates if respondents parent’s marriage disrupted by the age of 14. Family religious affiliation: A categorical variable and an indicator of the respondent’s religious affiliation during childhood. This variable includes women who identify their childhood religious affiliation as Catholic, Protestant, or other Religion. Age at first sex: A continuous variable that measures respondents’ age at first sex. Current age: A continuous variable that measures respondent’s age 18 to 34 years. Cohabitation Experiences: I create three different dichotomous measures that assess previous cohabitation experiences: Ever cohabitated; Multiple Cohabitations; Cohabitation prior to marriage. The measure of cohabitation history utilized is dependent on the model being examined. Analytic Strategy

The primary purpose of this research is to examine the changing patterns of family and union formation among young women, and the influence that childhood family structure may have had on these changing behaviors. To make these assessments I employ both binomial and multinomial logistic regression data analysis techniques. Logistic regression is a statistical model that is utilized when the dependent variable is a dichotomy and the independent variables are of

25 any type. Logistic regression models the odds that an observation will have an indicator equal to

1. The model can be written as follows:

ln[(P/1-P)]= β0+ βiXi+e

Figure 3.1: Logistic Regression Model

Where ln is the natural logarithm, P is the probability of the event occurring, 1-P is the probability of the event not occurring, and Xi are the independent variables discussed above. For ease of interpretation, I convert my results to odds ratios. The odds ratio for the independent variable indicates the relative amount by which the odds of the outcome increase (odds ratio

>1.0) or decrease (odds ratio <1.0) when the value of the independent variable increases by 1 unit. If the odds are the same across groups, then the odds ratio will be 1.0.

In the analyses where the dependent variable has more than two categories, I apply a multinomial logistic regression. Multinomial logistic regression is an extension of a binary regression when the outcome variable of interest consists of two or more categories. Multinomial logistic regression will compare each level of the dependent with the reference category, for each independent variable. I interpret the results using the relative risk ratios, which represent the change in the odds of being in the dependent category versus the comparison category associated with a one unit change on the independent variable.

Conclusion

In this dissertation, I employ the National Survey of Families and Households to investigate the changing union and family formation behaviors over the last two decades. Chapter

4 investigates the likelihood of a non-marital conception and distinguishes between non-marital conceptions that occur within a cohabiting union and conceptions that occur outside of any union.

I then examine women’s subsequent union transitions in response to a non-marital pregnancy.

Chapter 5 explores how the various parental union transitions experienced post-divorce during

26 childhood impact adult offspring’s own first union formation and the likelihood of experiencing a non-marital conception.

Table 3.1: Sample Characteristics at Wave 1 by Attrition Status and Sex NSFH 1 NSFH 1 NSFH Primary Primary 1Primary Respondent Respondent Respondent with focal with focal with focal child child child No Interviewed at NSFH3 NSFH3 interview

Male Female Male Female Male Female

Family Structure at Wave 1 84 52 86 55 84 50 Married to biological parent 6 13 4 10 7 15 Remarried (not biological parent) 2 3 1 3 2 3 Currently cohabiting (Divorced) 8 32 9 32 7 33 Divorced (not currently cohabiting) Educational Attainment at Wave 1 Less than high school education 16 20 21 26 11 13 High school diploma 37 43 39 43 35 44 Some college 23 23 25 21 22 25 College + 24 14 14 10 32 18

Age at Wave 1 37 34 33 31 39 37

Race Black 15 22 18 27 13 18 White 74 66 65 56 80 76 Hispanic 11 12 17 17 7 7

N 1666 3676 719 1843 947 1833

CHAPTER 4

CHANGING TRENDS IN NON-MARITAL CHILDBEARING AND UNION FORMATION:

THE ROLE OF COHABITATION IN RECENT FAMILY FORMATION BEHAVIORS

Introduction

This chapter explores the changing context of first union conceptions and subsequent union transitions over the last two decades. I focus specifically on two important interrelated shifts in the family formation process: a) the dramatic rise in non-marital childbearing and b) the increasing role of cohabitation in the family formation process. The rapid rise in cohabitation, coupled with the rise in conceptions outside of marriage, has considerably altered the landscape of the American family. As young adults have postponed entry into marriage, their rates of entering cohabiting unions have increased dramatically. A recent study finds that nearly 70 percent of all first unions began in cohabitation rather than in marriage (Kennedy and Bumpass

2008). Another consequence of women’s delay into first marriage has been the dramatic rise in fertility outside of marital unions, and increasingly common within a cohabiting union. In fact, much of the recent rise in non-marital childbearing has been driven by the growing proportion of births to cohabiting couples (Bumpass and Lu 2000; Musick 2002; Raley 2001), since the proportion of births to single women has increased only slightly, from 15 percent to 17 percent.

Not surprisingly, as the share of births to cohabiting couples continued to rise, the number of pregnancies ―legitimated‖ by a marital union dramatically declined (Bachu 1999; Lichter, Qian, and Mellott, 2006; Raley, 2001). The continued growth in cohabitation and non-marital childbearing suggests a need for studies of recent cohorts to assess the extent of cohabitation’s

29 role in the American family (Fein et al. 2003). In this study, I analyze the changing fertility and family formation behaviors of never married single and cohabiting women. Guided by the life course perspective, this chapter develops hypotheses for how the role of cohabitation in the family formation process may be changing across cohorts of young women.

Background

The rapid family formation changes over the latter half of the twentieth century reflect the profound changes in the meaning and function of marriage in the life course of young adults.

Post World War II, in the U.S., young adults experienced a somewhat standardized life course pathway. The entry into marriage was nearly a compulsory event that anchored the life course of young adults, and provided the foundation for family building behaviors. Yet, broader economic changes and cultural shifts towards individualism, self fulfillment, and a growing tolerance of more diverse lifestyles, have altered the role of marriage within the life course of today’s young adults (Bianchi and Spain 1996; Casper and Bianchi; Cherlin 2004; Lesthaeghe 1993; Popenoe

1993; Thornton 1989). Once the dominant and only institution that permitted sexual intimacy, co- residence, childbearing and rearing, marriage has now lost its foothold in defining one’s entrance into adulthood. This is evident in the increasing rates of cohabitation, marital delay, and non- marital childbearing among recent cohorts of young adults.

Although these changes in union and family formation behaviors have been evident across all social strata, the declining role of marriage has been most prominent among the socio- economic disadvantaged (Fein et al, 2003). Accordingly, marriage has increasingly become a measure of social stratification based on class and race. Individuals with higher education, incomes, and stable employment are substantially more likely to be married (Popenoe and

Whitehead 2007). While the socio-economic disadvantaged and minorities are increasingly less likely to enter into a marital union (Fry and Cohn 2010). Empirical evidence finds that marriage

30 is still highly valued, and that the vast majority of young adults aspire to marry, but as marriage increasingly becomes a ―luxury‖ status, disadvantaged women perceive it as unattainable due to the social and economic barriers they often encounter (Edin and Reed 2005; Gibson et al, 2005).

Therefore, the class inequalities between married and unmarried families appear to be widening

(McLanahan 2004). In essence, marriage remains central in the life course of young adults, but as the practical importance of marriage declines, and its symbolic importance remains intact, marriage is now the ―capstone‖ of adult life instead of the foundation for family building behaviors (Cherlin 2009, p139).

Consequently, as the link between marriage and parenthood weakens, cohabitation has become more prominent in the framework of childbearing, and the life course of today’s young adults is growing increasingly more diverse and complex. Today, the majority of young adults first cohabit rather than marry directly (Kennedy and Bumpass 2008). Likewise, more than half of all first marriages are preceded by a cohabitation (Bumpass and Lu 2000). The traditional sequence of dating, then marriage followed by parenthood is slowly being replaced by having children and then marrying- at least among certain groups. Whereas in the past, a non-marital pregnancy would often hasten or force young adults to legitimate the birth by entering into a marital union, today the shotgun wedding is becoming an increasingly rare response to a non- marital conception (Bachu 1999). Despite a surfeit of literature that has addressed the sweeping family changes over the last few decades, there is still an ambiguity about cohabitation’s place in the family spectrum. Much of this uncertainty derives from the substantial variability in the meaning and role of cohabitation by race and ethnicity as well as educational attainment.

In this study, I argue that never married women’s responses to an impending birth may serve as an index of whether cohabitation is serving as a substitute for marriage as an acceptable context for childrearing. I assess the changing union context of first conceptions among never

31 married single and cohabiting women, and the subsequent union transitions following their pregnancy. Exploring both single non-cohabiting as well as cohabiting women’s responses to an impending birth permits me the ability to ascertain the extent to which cohabitation is becoming as an accepted context for having and raising children, a function that once marriage only permitted. By incorporating the fertility behaviors of a more recent cohort of women, as well as analyzing both conceptions and births, I hope to extend our current understanding of cohabitation’s place in the family formation system.

Cohabitation and Non-Marital Childbearing

Between 1970 and 2006 the number of cohabiting households increased by more than five- fold, over half of all women have entered into a cohabiting union (Kennedy and Bumpass 2008;

Manning 1995). As couples have increasingly set up households outside of a marital union, they have also been increasingly initiating families outside of a marital context (Hollander 1996). In a decade span, the proportion of children born to cohabiting parents increased from 11 percent in the early 1990’s to 18 percent in 2001 (Mincieli et al 2007). An estimated 40 percent of all children are expected to spend some time in a cohabiting household during their childhood years

(Bumpass and Lu 2000). While the proportion of births to cohabiting mothers nearly doubled, births to single non-cohabiting women remained somewhat stable (Scommegna 2002). Over half of all non-marital births now occur within a cohabitation arrangement (Mincieli et al 2007).

Previously, many of these pregnancies resulted in a marriage before the birth (Manning,

1993), but in recent years fewer single and cohabiting women wed prior to childbirth (Graefe and

Lichter 1999; Lichter Qian and Mellott 2006; Manning, 2001). These changing responses to a non-marital pregnancy highlight the weakening ties between marriage and childbearing. As marriage has become a marker of status that fewer individuals can attain, cohabitation may be serving as an adaptation to these culture shifts (Nock 2006; Smock and Manning 2004).

Literature on cohabitation’s place in the family spectrum suggests that it may be a precursor to marriage, while others argue it is an alternative to marriage or being single (Lillard et al 1995; Manning and Smock 2002; Raley 2001; Sassler 2004; Smock 2000). Yet, recent research suggests that cohabitation and marriage are two separate entities and not substitutable, and these two arrangements serve different purposes and populations (Guzzo 2006; Nock 2006).

The continued growth in cohabiting families suggests cohabitation is becoming its own institutionalized family form - at least among some groups of women. Cohabitation provides disadvantaged women with a socially acceptable route into motherhood, without having to encounter the social and economic barriers associated with marriage (Edin and Reed 2005;

Smock, Manning, and Porter 2005). Although marriage remains a desirable step, the social and financial standards for marriage have been set so high, perhaps too high (Edin and Kefalas 2005), that cohabitation is an attractive option among disadvantage parents. It allows them to share some expenses in a co-parenting context, without having to meet the requirements for marriage

(Reed 2006). Notwithstanding the socio-economic disadvantage’s increased likelihood to form families outside of marriage, there is substantial variation within this subgroup of women. The function and meaning of cohabitation in the family formation process varies considerably by a woman’s educational attainment as well as race and ethnicity, and most often these mechanisms are not mutually exclusive.

Demographic Influences

Education

Concurrent with the recent changes in union formation was the expansion of educational opportunities for young women. In 1980 approximately 14 percent of women had obtained a college education, in 2008 this number was nearly 30 percent (US Census 2009). The number of women enrolled in college institutions now outnumbers men (US Census 2009). Despite the

33 expanding educational opportunities for women, scholars contend that the growing diversity in family formation patterns has been largely driven by women at the bottom of the socio-economic hierarchy. In 1980, roughly 10 percent of mothers in the top third of the educational distribution were unmarried, compared to about 35 percent of mothers in the bottom third. By 2000, the corresponding figures were 7 percent for the most educated mothers and over 40 percent for the least educated (McLanahan 2004). Additional evidence demonstrates that cohabitation is more common, and has grown at a greater pace, among the less educated than among higher educated individuals (Bumpass and Lu 2000;Goldscheider et al. 2001). In the last two decades the proportion of cohabiting women with a high school education increased by over 100 percent, while those with a college education only experienced an increase of approximately 50 percent

(Bumpass and Lu 2000; Chandra, Martinez, Mosher, Abma, and Jones, 2005). And it is women with lower levels of education, who are most likely to bear and rear children while in cohabiting relationships and less likely to marry compared to cohabiting women with higher levels of education (Manning 2001). Indeed, evidence suggests that higher education and economic stability enhances the likelihood of marriage among cohabiting couples ( Manning and Smock

1995). As well educated women are more likely to be matched with well educated men -the prospects of marriage significantly increase (Oppenheimer 2003; Smock and Manning 1997).

Nonetheless, higher educated women are the least likely to conceive while cohabiting or outside of a union, but if they do, they are the most likely to marry in response to a non-marital pregnancy (Landale and Manning 1996; Manning 2001). I would expect that the role of cohabitation in family building will have greater influence among lower educated women. Given the continued growth in educational opportunities for women, I suspect this influence will be greater for the more recent cohort of women. However, much of the educational differences in cohabitation’s role in family building are strongly associated with race. Empirical evidence finds

34 that the likelihood of conceiving outside of marriage is less contingent upon educational attainment for Blacks than it is for both Whites and Hispanics (Upchurch et al, 2002). Thus, the mechanisms of education and race are not mutually exclusive, and often work in tandem with one another in the family building process.

Race

Family formation patterns have consistently varied by racial and ethnic groups.

Historically, Black women have had much lower marriage rates than White and Hispanic women, and Black and Hispanic women have had substantial higher rates of non-marital childbearing than

Whites (Lewis and Ventura 1990;Manning and Landale 1996; Ventura et al, 2000). While cohabitation rates are somewhat similar across all racial and ethnic groups (Casper and Bianchi

2002; Heuveline and Timberlake 2004), the role and function of cohabitation in the family formation process varies substantially (Bumpass and Lu 2000; Upchurch et al, 2001). Both

Hispanic and African American women are significantly more likely than White women to conceive while cohabiting, and substantially less likely to marry in response to their pregnancy

(Manning 2001). Literature suggests that Hispanic culture is more open and approving of cohabitation and it is a more normative context for childbearing compared to Whites and even

Blacks (Landale and Oropesa 2007). However, scholars contend that much of the marriage gap between White and Black women can be explained by a lack of ―marriageable‖ Black men.

Wilson (1987) contends the lower rates of marriage and substantially higher rates of non-marital childbearing among Black women may be attributed to a shortage in the supply of economically attractive ―marriageable men‖ (Edin 2000; Edin and Reed 2005; Wilson 1987).

Nonetheless, recent evidence suggests that these family patterns may be changing.

Although African American and Hispanic women still have substantially higher non-marital birth rates, White women have experienced the greatest increases in non-marital childbearing over the

35 last three decades, and the rates of Whites are now similar to that of Blacks approximately two decades ago (Terry-Humen et al, 2001). Further, ever married rates among White women are now similar to those of Black women roughly 20 years ago (Fry and Cohn 2010). Although marriage rates have declined for White women across all education groups, the decline has been steepest among those with less education (Fry and Cohn 2010). In 2008, 86 percent of college educated White women had ever married, compared to 92 percent in 1980. Among those with less than a high school degree, 81 percent had ever married in 2008, compared to 96 percent in

1980 (Isen 2010). The decline of ever married rates among White women with less than a high school education was more than double than that of college educated women.

As lower educated Whites have been increasingly precluded from the marital institution, they have been turning to cohabitation as a socially accepted context for childrearing. In fact, evidence finds that much of the increase in White non-marital childbearing between 1984 and

1994 has been mostly attributed to the increase in childbearing among cohabiting couples, and the majority of cohabiting pregnancies are concentrated among women with lower levels of education (Manning 2001). While the number of cohabiting births among Whites has been increasing, the number of cohabiting births among Blacks has been declining ( Wu et al, 2001).

This changing racial dynamic in cohabiting families suggest there may be an emerging stratification system even within the context of cohabiting.

In sum, understanding the place of cohabitation in the American family requires attention to an individual’s race and as well as educational attainment. For college educated White women cohabitation has not become an accepted avenue for childbearing, and marriage is still viewed as the appropriate context for having and raising children. Conversely, among lower class White women, marriage is seen increasingly less as a needed step before having children, and cohabitation should become an increasingly common avenue for bearing and raising children.

African American women may see marriage as a desirable step, but given the economic and social barriers they often encounter, their hopes for marriage are rarely achieved. Further, given the small and declining role cohabitation plays in family building behaviors, the majority of

Black women rear and bear children outside of any union (Wu et al., 2001). In 2001, over 60 percent of Hispanic and White non-marital births occurred within a cohabiting household, compared to only 30 percent of Black non-marital births (Manlove et al, 2006).

Literature suggests that cohabitation is less prominent in the family building behaviors for

Whites, and is more of a stage in the marital process (Manning 1993; Manning and Landale

1996), but is an alternative context to marriage or being single for Black and Hispanic women.

However, the increasing role of cohabitation in family building among Whites- or at least disadvantaged Whites- and the declining number of cohabiting births among Blacks, questions if this argument still holds true among more recent cohorts of women. I expect that cohabitation has a stronger influence on the family formation behaviors of White women in the more recent cohort.

Economic and Cultural Arguments

Economic Cost

In the U.S., the role of cohabitation and non-marital childbearing is often understood within an economic framework, which predicts a negative relationship between female educational attainment and non-marital childbearing (Ellwood and Jenks 2004; Willis and Haaga

1996). College educated women have the highest marital rates, lowest divorce rates, and are more likely to delay childbearing until marriage (Popenoe and White 2007), and these trends appear to be growing (Martin 2006). Women at the higher end of the socioeconomic spectrum often find conflict between educational and career opportunities and family opportunities. The event of a non-marital birth can have deleterious consequences on the career trajectories of highly

37 educated women, even within a cohabiting union. The postponement of childbearing and marriage is more advantageous for women with higher socioeconomic status. In fact, women with college degrees who marry after the age of 30 are the only group more likely to bear children after marriage than before (Popenoe and Whitehead 2007). For lower educated women however, the economic impact of non-marital childbearing is not as detrimental on career trajectories, thus, less of a need to delay parenthood. However, this argument largely explains intended childbearing (Musick et al, 2007). Research suggests that single and cohabiting women are not actively planning to have children, and that approximately 70 percent of non-marital conceptions are not planned (Finer and Henshaw, 2006). Evidence suggests that the significant educational disparities in unintended fertility may arise from less effective and reliable contraceptive use among women with lower education, even when their intentions are not to conceive a child

(Forrest 1994; Musick et al. 2007). Musick and colleagues (2007) posit that lower educated women’s lack of education and access to effective contraception may be an important mechanism behind the substantial educational disparities in unintended non-marital childbearing. However, neither of the above arguments directly addresses the substantial racial and ethnic disparities in family building behaviors.

Cultural

Though the increasing number of families formed outside of marriage suggests that lower educated and minority women are eschewing marriage, both quantitative and qualitative research points in the opposite direction. Several studies find that most Americans have high regards for the institution of marriage, including the economic disadvantaged (Bulcroft and Bulcroft, 1993,

Edin and McLanahan 2005; Lichter, Batson and Brown 2004). Yet, as the meaning of marriage has shifted away from a mandatory stage in the life course to a more symbolic representation of individual achievement, poorer individuals face a number of social and economic barriers that

38 impede their hopes for marriage (Cherlin 2004;2009;Gibson-Davis et al, 2005;Wilson and

Neckerman 1986). The financial expectations surrounding marriage further deters those at the lower end of the economic spectrum from entering into a marital union regardless of childbearing desires (Edin, Kefalas and Reed 2004; King and Allen 2009).

This argument is echoed throughout the qualitative literature (Edin and Reed 2005; Edin

2000; Edin Kefalas, and Reed 2004). Gibson-Davis, Edin, and McLanahan (2005) posit that unmarried parents have a list of financial and relationship requirements that must be met in order to marry. In their study among low income cohabiting couples, the authors found that the vast majority of couples indicated a necessity to get their personal finances in order before they could marry. Most respondents expressed the need to establish a middle class lifestyle in order to marry. Of the 60 percent of unmarried parents that felt there was a ―good‖ chance they would marry their newborn child’s other parent, only 15 percent had done so by the child’s third birthday.

Cherlin posits ―People marry now less for the social benefits that marriage provides than for the personal achievement it represents (pg 857). ‖ The emphasis on the financial prerequisites for marriage appears to be greater for Black women than for White (Timmer and Orbuch 2001). But with the declining economic circumstances of men, particularly African American males (Edin

2000;Wilson 1987), coupled with an increased difficulty of finding a partner able or even willing to take on the financial burdens of married parenthood, make their high hopes for marriage remain just that. Thus, childbearing may serve as an alternative marker that signifies adulthood among women at the bottom of the socio-economic hierarchy, and cohabitation provides a route towards ―legitimization‖ for these families (Catanzarite and Ortiz 2002; Edin and Kefalas, 2005;

Kendall et al., 2005).

Trends over time: Cohort Differences

As previously mentioned, over the last few decades the rates of cohabitation dramatically increased. Between 1980 and 1990 the number of cohabiting couple households increased by 80 percent, and another 71 percent the following decade (O’Connell, 2003; U.S. Census Bureau,

2001), and cohabitation has sustained this rapid rate of growth through the present (Kennedy and

Bumpass 2008). As a result, a growing number of children are being born into cohabiting families. Over half of all non-marital births now occur within cohabiting relationships, compared to 40 percent in 1990 and less than 30 percent in 1980 (Mincieli et al. 2007). While the percentage of non-marital births has been skyrocketing, the number of ―shot-gun‖ marriages, or the number of premarital pregnant women marrying before childbirth have substantially declined.

Akerlof, Yellen, and Katz (1996) contend that between 1965 and 1990 approximately 75 percent of the increase in the White out-of-wedlock birth rate, and about 60 percent of the Black increase, can be attributed to the decline in shotgun marriages. The expanding role of cohabitation in union and family forming behaviors suggests a further disconnect between marriage and parenthood. However, research addressing these changing childbearing and union formation patterns among recent cohorts of young adults remains limited.

The life course perspective provides me with an appropriate framework for exploring these changing family patterns across cohorts. First, the life course perspective emphasizes the importance of individual sequencing, in that certain life events will come before others over one’s life course. Individuals make a series of sequential choices and actions that help construct the various pathways, however, their choices are largely constrained and/or permitted by both the historical and familial context in which they live their lives (Elder et al. 2003). Secondly, the life course perspective also highlights the importance of cohort. Young adults who entered into adulthood during the rapid demographic changes of the last few decades may experience a

40 different pathway to union and family formation than individuals who entered into adulthood prior to these substantial changes. Women who were born at different periods in history have experienced social and economic events that have uniquely shaped their union and family formation behaviors.

In this study, the two cohorts of women I analyze entered into adulthood at very different points in time. Women in the early cohort were entering into adulthood and engaging in family formation behaviors when the economic and cultural shifts previously discussed were beginning to emerge, while women in the later cohort entered into adulthood when these changes were well established. As noted by Fein et al. (2003), ―because successive cohorts experience the same phases of life in different social and economic environments, cross-cohort comparisons of effects are needed to absorb fully the implications of period change (p. ix).‖ The later cohort of young women are far more likely than the women in the earlier cohort to have witnessed the disruption of their parents marriage, experienced a parental cohabitation, and to have spent time in a single parent home (Sassler et al., 2009). Experiencing these new choices for family formation most likely influenced the union and family formation decisions of women in the later cohort differently than for the earlier cohort. For instance, we know that the experience of a parental divorce during childhood significantly increases adult children’s likelihood of cohabitation, non- marital childbearing, and even the disruption of their own relationships (Teachman

2002;2003;2004; Wolfinger 2001;Wu and Martinson 1993). Alternatively, the increased acceptability of divorce, cohabitation and non-marital childbearing among recent cohorts of young adults has reduced the negative stigma once associated with these demographic behaviors

(Hobcraft 2008).

The broader socio-historical context in which these young women transitioned into adulthood also differed. The changing ideologies surrounding gender roles and norms that

41 emerged over the latter half of the twentieth century substantially expanded the labor market and educational opportunities for young women. Between 1980 and 2008 the percentage of women over the age of 25 that have obtained a college doubled (U.S. Census 2008). Currently, the gender gap in higher education overall is widening in favor of women (Buchman and DiPrete 2006).

These growing trends in higher education are important as education is strongly related to union and family formation behaviors in the United States. Moreover, this shift in women’s educational opportunities will influence the family building decisions of the two cohorts differently. Given that the risk of a non-marital conception and the rise in cohabitation has been greater for women without a college education, it is likely we may find that lower educated women in the later cohort will have a greater likelihood of conceiving outside of marriage than women in the earlier cohort.

In this chapter I hope to build upon Manning (1993) and Raley’s (2001) research that assessed the weakening tie between childbearing and marriage across cohorts of women. Relying on data from the early 1970’s and 1980’s, Manning (1993) examines women’s marriage and cohabitation behavior after a non-marital conception as an indication of whether cohabitation is serving as an alternative to marriage for childrearing. Her findings suggest that cohabitation is not a substitute for marriage, and is a part of the legitmation process. Yet, this function varies considerably by race. She contends that cohabitation serves as a stepping stone in the marital process among pregnant Whites, but an alternative to remaining single among pregnant Blacks.

More recently, Raley (2001) investigates whether the continued growth in the proportion of births to cohabiting women was merely a result of the increasing number of women entering cohabiting unions, or the beginning of a shift in the meaning of cohabitation. Using data from both the National Survey of Families and Households (NSFH) and National Survey of Family

Growth (NSFG), Raley finds that women’s responses to a non-marital pregnancy are changing, at

42 least among singles. By the early 1990’s, single women were almost as likely to cohabit as to marry in response to a pregnancy. However, she found little change in cohabiting women’s response to pregnancy. Raley concluded that it was too early to ascertain if these changes signified a shift in the meaning of cohabitation. Similar to Manning (1993), Raley’s analyses relied on data from the late 1970’s to early 1990’s. Given that between 1995 and 2002 the percentage of women who ever cohabited increased from 45 percent to 54 percent, and the number of cohabiting parents increased from 11 percent to 18 percent (Kennedy and Bumpass

2008), it is likely the response to a non-marital pregnancy has continued to change. I argue in this chapter that the study of more recent cohorts is needed to see if cohabitation is beginning to replace marriage as an institutionalized family form- at least among certain groups of women.

To ascertain the degree to which cohabitation may be replacing marriage, I examine the changing propensity to conceive outside of marriage among single and cohabiting women, and the subsequent union transitions following a non-marital pregnancy. I examine women who entered into unions in the 1970’s to early 1980’s, when cohabitation was much less normative, and those who entered into unions in the late 1990’s and early 2000’s, when cohabitation had become commonplace. I argue that distinguishing between conceptions that transpired within cohabitation rather than outside a union, as well as examining both conceptions and births improves our current understanding of cohabitation’s role in the family process. I hope by utilizing more up-to-date data, I will be able to extend both Raley’s (2001) and Manning’s (1993) findings, and demonstrate the changing role of cohabitation and non-marital childbearing in young women’s family formation decisions across the last two decades.

In sum, this chapter is guided by the following research questions. How, if at all, has the fertility behavior of never married single and cohabiting women changed across cohorts? How, if at all, do the characteristics of women who bear children within cohabitation compared to women

43 who bear children outside of any union differ? Moreover, have these differences changed over time? How has single and cohabiting women’s response to an impending birth changed across cohorts? Based on the above information, this research is guided by the following hypotheses:

Hypothesis 1- Never married women in the later cohort will be more likely to experience

a first conception outside of a marital union (either in cohabitation or outside of any

union) than never married women in the earlier cohort.

Hypothesis 2- Given the substantial educational differences in non-marital childbearing

and cohabitation, lower educated women will be more likely to conceive outside of

marriage (either in cohabitation or outside any union) than higher educated women.

Lower educated women in the later cohort will be more likely to conceive outside of any

union or within a cohabiting union than lower educated women in the earlier cohort.

Hypothesis 3- Black women will be more likely to conceive outside of any union.

However, Whites will be more likely than Blacks to conceive while cohabiting, but

Hispanics will be more likely than Whites to conceive while in a cohabiting union. Given

the increasing number of births among Hispanic and White cohabitors, Hispanic and

Whites in the later cohort will be more likely to conceive while cohabiting women, than

Hispanic and White women in the earlier cohort.

Hypothesis 4- Cohabiting women in the later cohort will be significantly less likely to

transition into a marriage in response to an impending birth than cohabiting women in the

earlier cohort. Likewise, single women in the later cohort will be significantly less likely

to marry in response to an impending birth, and more likely to cohabit rather than remain

single compared to women in the earlier cohort.

Hypothesis 5- Higher educated women will be most likely to marry in response to a non-

marital pregnancy (among single and cohabiting women). Lower educated women in the

later cohort will be more likely to transition into or remain within a cohabiting union than

lower educated women in the earlier cohort.

Hypothesis 6-Given the increasing role of cohabitation in the family formation decisions

of Hispanic women, Hispanic women will be more likely to cohabit in response to an

impending birth than White women, and this effect should be greater for the later cohort.

However, Whites will be more likely than Blacks to cohabit in response a single non-

cohabiting pregnancy, and this effect should be greater for the later cohort.

Data

For this study, I use data from the first wave (1987-1988) and third wave focal children

(2001-2002) of National Survey of Families and Households (NSFH). The NSFH is a nationally representative sample of 13,007 respondents, including 9,643 main respondents aged 18 and over, plus an oversample of minorities, single parent families, recently married couples and cohabiting couples (Sweet, Bumpass and Call 1988). The Wave 1 response rate was 74 percent. Recently, the National Survey of Families and Households (NSFH) completed data collection on a third wave of data (2001-2002) and conducted extensive interviews of the focal child of the main respondent (Sweet and Bumpass 2002). Detailed marital and cohabitation histories were collected from the focal child, now aged 18-34. Current and previous partner information is also available among the focal child data.

Given the dramatic rise in the rates of cohabitation as well as non-marital childbearing, these samples prove ideal with a nearly 15 year span between the two cohorts. The study sample restricts cohabitation event to women’s cohabitation experience prior to their first marriage in order to reduce bias (Brown 2000; Manning 2004). Post-marital cohabitors are more likely to have already had children, and are less likely to have definite plans to marry their partners and

45 expect to ever remarry (Bumpass, Sweet, and Cherlin 1991). Evidence also finds that post- marital cohabitation nearly doubles the odds of having a non-marital birth (Brown 2000). In addition, non-marital pregnancies are limited to first pregnancies only. Because the focal children sample is between the ages 18 to 34, the same age range is used for the older cohort for comparability and resulting in a final N=4,165. Due to budgetary constraints the Wave 3 sample does suffer from certain limitations. First, if there was a completed parental interview at Wave 3, the NSFH obtained interviews for those focal children who were aged 18 and older at (NSFH3), regardless of whether or not an interview was completed at Wave 2. Therefore, a substantial number of Focal Children at Wave 3 (N~200) did not have Wave 2 information. Secondly, the response rate for focal children at Wave 3 was 48 percent.

Variables

The focus of the main analyses is the likelihood of a non-marital conception and the union context of first births and how this has changed over time. I include additional measures including childhood background and demographic characteristics. There are four dependent variables in these analyses. I first focus on the union context of a non-marital conception, and then I examine any subsequent union transitions made within three months of childbirth.

46 and Piccarreta 2007; Steele 2008)2. Non-marital conceptions are defined as pregnancies to women who had never been married at the time of conception. Because the data do not permit the ability to know if the cohabiting or marital partner is the biological parent of the child, I limit the time to transition into a union within 3 months of childbirth. In the first set of analysis I examine the likelihood of experiencing a single non-cohabiting conception. I employ a dichotomous measure 1=all women who conceived outside of any union 0=all women who did not conceive outside of marriage or cohabitation. Therefore, childless women, women who conceived while in a cohabiting union and women who did not conceive until after marriage are included in this analysis.

The second set of analysis examines conceptions that occurred within a cohabiting union.

Only women who have experienced a cohabiting union prior to first marriage are included in this set of analyses. I create a dichotomous measure 1=never married women who experienced a conception while in a cohabiting union; 0=never married women who did not experience a conception while in a cohabiting union.

The next two sets of analyses examine the changing union context of first births. I first examine the women who experienced a conception while in a cohabiting union and the outcome of this union within three months of childbirth. There are three categories for cohabitation outcome, 0= remained cohabiting, 1 = into a marriage, and 2= ended in a separation. Next, I examine the likelihood that a single non-cohabiting woman who conceived will transition either into a marriage or cohabitation within three months of childbirth as to remaining single 0=remain single, 1= entry into cohabitation, 2= entry into a marriage.

Independent Variables

2 In supplementary analysis I examine conception dates by date of birth minus eight months. However, I find minimal difference in the proportion of births that occurred outside of marriage. 47

Cohort: I defined two cohorts—Early cohort: women in the first wave of the NSFH (collected between 1987-1988); Later Cohort: women in the focal child data sample from the third wave of the NSFH (collected between 2001-2003).

Age: measured as a continuous variable starting at age 18 and ending at age 34.

Education at time of conception: the NSFH collected detailed educational histories. Respondents were asked the highest level of education completed, and information on each spell of schooling

(e.g. full time /part time and level of study). I based my measure of educational attainment on the level of education at the time the female respondent conceived her first child. I collapsed educational attainment into four mutually exclusive categories: less than a high school education; high school degree; some college experience; bachelor’s degree or more. Educational attainment is coded as education at time of the survey for those who do not have children. In addition to the detailed educational histories the NSFH also provides information if the respondent is currently enrolled in a post-secondary institution. I created a dichotomous measure

1=Currently Enrolled 0=Not Enrolled.

Race: coded into three mutually exclusive categories: Black, Hispanic, White/Other.

Maternal Education: the NSFH at Wave 1 asked respondents the highest level of education completed by their mother. However, at Wave 3 this measure is not available. Therefore, because Wave 3 respondents are the children of the main NSFH respondents I linked the focal child back to their parent at Wave 1. I then constructed maternal level of education based on the amount of education reported at the time of the Wave 1 interview. Given that there was substantial number of missing responses for both waves, I substituted paternal education when mother’s level of education was not available.

Parental Divorce: measured if respondent’s parent’s marriage disrupted by the age of 14.

Receipt of Public Assistance: if the respondent received public assistance during childhood.

Cohabitation history: two separate measures of cohabitation history were constructed. For the analysis that examines single non-cohabiting women, I control if the respondent have ever had a cohabitation experience. For the analysis on cohabiting women, I control if the respondent had entered into more than one cohabiting union.

Methods

I employ both logistic and multinomial logistic regression models in this chapter. I first use logistic regression models to estimate the likelihood that a woman will conceive her first child outside of either a marital or cohabiting union. I also use logistic regression to determine the first birth location for women who conceived while cohabiting. Then I use a multinomial logistic regression model to predict the union location of first births among women who conceived outside of a union.

Results

Descriptive trends in union and childbearing behaviors

In order to understand cohabitation and fertility behaviors over the last two decades I first examine the demographic characteristics of women in both cohorts. Descriptive statistics for the total sample of 4,165 women are displayed in Table 4.1. Women in the later cohort are on average two years younger than the older cohort of women. In addition, the later cohort has a smaller proportion of Black and Hispanic women. The shift in educational composition over the last two decades is also demonstrated in Table 4.1. As previously discussed, coinciding with the rapid familial changes in the US was an expansion in educational opportunities for young adults.

In fact, the number of women that earned a college diploma doubled between 1980 and 2000 (US

Census 2007). This fact is replicated in Figure 4.1. The proportion of women who had obtained a college education more than doubled for the later cohort (9% to 26%), while the number of women with a less than a high school education declined by more than half (31% to 13%).

[Table 4.1 and Figure 4.1 about here]

Table 4.1 also shows the increasing number of women entering a post-secondary institution. The later cohort of women has over twice as many women currently enrolled in school. With the exception of maternal education, there is very little difference between the two cohort’s childhood experiences. Approximately one-fourth of both cohorts experienced the disruption of their parent’s marriage, and similar proportions received public assistance during childhood.

Figure 4.2 depicts the substantial rise in the number of cohabitations that preceded first marriage. Sixty percent of women in the later cohort have entered into cohabitation prior to their first marriage, compared to a little over one third of those in the earlier cohort (34 percent).

These findings confirm previous research documenting the rapid rise in cohabitation prior to first marriage (Bumpass and Lu 2000; Kennedy and Bumpass 2008; Smock 2000). Among never married women, there is a two-fold increase in number of women currently cohabiting. However, given the younger age and greater proportion of never married women in the later cohort, there are no substantial differences in the proportion of never married women’s cohabitation histories.

[Figure 4.2 about here]

Table 4.1 also highlights the changing trends in non-marital childbearing. In the fifteen year span between these two cohorts of young women, the number of non-marital conceptions has risen by 10 percent. However, this increase is nearly equally divided between conceptions that occurred in a single non-cohabiting state and those that occurred within a cohabiting union.

Furthermore, the percent of conceptions that occur within a first marital union significantly decline from almost half to only 37 percent. For women in the earlier cohort, the largest share of children are born in married family households. However, among the most recent cohort of women this is no longer the case. Almost half of all births among this cohort of women are in

50 single non-cohabiting households. Table 4.1 confirms that while non-marital fertility has been on the rise over the last two decades, marital fertility has fallen substantially. It is this increasing trend in non-marital fertility that I further explore in the next two tables.

Single Non-Cohabiting

Table 4.2 displays the demographic characteristics of women who conceived outside of any union compared to all women who did not conceive outside of a marital or cohabiting union.

Looking at the demographic differences between these two groups of women can help provide a clearer picture on the type of women who bear children outside of marriage and how this may have changed across time. Consistent with the literature the mean age of mothers that experience a non-marital conception has been increasing over the last two decades (Wu 2002).

[Table 4.2 about here]

The substantial increase in the proportion of women with thirteen to fifteen years of education that experienced a non-marital conception is evident in Table 4.2. Less than 10 percent of women in the earlier cohort who conceived outside of any union have some college experience at the time of conception, however for the later cohort this number is now 18 percent. Yet, there is minimal change in the proportions of women with a college degree at the time of conception.

Similar to Ellwood and Jenks (2004), the growth in non-marital childbearing has been largely confined to women without a college diploma, with a sharp rise in the number of women with 13 to 15 years of education. Figure 4.3 illustrates these differences.

[Figure 4.3 about here]

Table 4.2 demonstrates the growth in non-marital childbearing among White women over the last few decades. In addition, women who conceived outside of any union are from more disadvantaged backgrounds as well as from broken homes, this is true for both cohorts of women.

Finally, the changing response to a non-marital pregnancy is evident in Table 4.2. In less than

51 two decades, the number of women who marry within three months of the child’s birth decreased by more than half (32 % to 14 %). In contrast, the number of never married mothers who transition into cohabitation within three months of the child’s birth has more than tripled (5 % to

17 %). These findings lend credence to my argument that the shotgun marriage is being rapidly replaced by the shotgun cohabitation. In addition, these findings substantiate the need to examine conceptions as well as births. By limiting the analysis to first births only, we limit our understanding of the changing nature of non-marital childbearing. Over 30 percent of women in the earlier cohort, and 14 percent in the later cohort conceive while single non-cohabiting, but transition into a marriage within three months of childbirth Figure 4.4 illustrates these changes.

[Figure 4.4]

Characteristics of Mothers that Conceived within a Cohabitation

In Table 4.3, I examine the demographic characteristics of women who conceived outside of marriage but within a cohabiting union. There does not appear to be any large age differentials between women who conceived in a cohabitation compared to those that did not, nor any substantial shifts over time.

[Table 4.3 about here]

Educational differentials are quite apparent and similar to the previous table. There has been a substantial increase in the proportion of never married women with some college education who conceived while cohabiting in the last two decades (8 % to 22 %). Additionally, there is virtually no change in the proportion of never married college educated women conceiving within a cohabiting household. Similar to Table 4.2, I find that the growth in cohabiting conceptions has been confined to women without a college education. As shown, the

52 growing proportion of conceptions within cohabiting unions is mostly concentrated among those with a high school education and not the least educated (45 % vs. 25 %). However, among the earlier cohort, there is a much smaller difference between women with the least amount of education (51 %) and a high school education who conceive within a cohabitation (40 %).

Figure 4.5 demonstrates these educational differences.

[Figure 4.5]

The gap between cohabiting White women who conceived has narrowed considerably.

This is consistent with the literature that finds births to cohabiting White women dramatically increased in the last two decades (Bumpass and Lu 2000). Yet, it has been the opposite for Black women. Wu et al. (2001) report less than one in five of non-marital births occur to Black cohabiting women, and over time, births within cohabiting unions have constituted a declining fraction of Black non-marital births. Table 4.3 suggests that the proportion of Black women who conceived within a cohabiting union actually declined among the most recent cohort (.16 to .12).

A greater number of women who conceived while cohabiting received some form of public assistance as well as experienced a parental divorce during childhood than women who did not conceive while cohabiting, and this is true for both cohorts of women.

Lastly, I find that the response to impending parenthood among cohabiting women appears to be changing. As seen in Figure 4.6, the majority of women (60 percent) remain cohabiting after the birth of the child. This suggests a decline in the strength of marriage as a prerequisite for childbearing, and a growing importance of cohabitation in the American family process. Similar to Table 4.2, these findings illustrate a substantial part of the story of non-marital childbearing and cohabitation is omitted when only cohabiting births are examined.

[Figure 4.6 about here]

The results from the descriptive analyses suggest that cohabitation is increasingly playing an important role in union and family formation process, however, the extent of this role varies substantial across different groups of women. In the next set of analyses I use logistic regression models in order to predict the extent to which non-marital childbearing and cohabitation has changed in the last two decades.

Union Context of First Non-marital Conceptions

Single Non-Cohabiting Conceptions

I use logistic regression models to assess the likelihood of conceiving outside of any union, marital or cohabitation, and if this has changed over time. I model the odds of a woman conceiving outside of any union versus all women that have not conceived outside of cohabitation or marriage. The results are displayed in Table 4.4. The main effect of cohort can be interpreted as the cohort differences in the likelihood of conceiving outside of any union among women in the later cohort. In support of my hypothesis 1, I find that there has been an increase in the likelihood of conceiving outside of any union compared to not conceiving outside of any union between the two cohorts of women. That is, women in the later cohort are about 60 percent more likely to conceive outside of any union.

[Table 4.4 about here]

I find that even after controlling for demographic and familial characteristics, education differences remain large and significant in predicting the likelihood of conceiving outside of marriage or cohabitation. As hypothesized, women with lower amounts of education are over 5 times more likely to conceive outside of any union. Compared to women with a high school diploma, women with some college education are approximately 60 percent less likely and those with a college education are 90 percent less likely to conceive outside of any union. My findings also confirm that current school enrollment reduces the likelihood of experiencing a non-marital

54 non-union conception by about 15 percent. The racial disparity in non-marital fertility is evident in the fact that I find Black women are almost 5 times as likely to conceive outside of any union relative to White women, this is in support of hypothesis 3. I find no difference between Hispanic and White women’s likelihood of conceiving. In addition, a previous cohabitation significantly increases the odds of conceiving outside of any union.

In order to explore if the above differences have changed over time, I test for potential interaction effects between cohort and age, educational attainment, race and childhood family background. I find only one interaction to attain minimal significance. In support of hypothesis

2, women with less than a high school education are increasingly more likely to conceive outside of any union. In fact, women in the later cohort with less than a high school diploma are almost twice as likely to conceive outside of any union compared to women in the earlier cohort with the same level of education. This finding indicates that the educational disparity between women who conceive outside of marriage and those who do not is still widening.

Cohabitation Conceptions

Table 4.5 estimates the likelihood of conceiving within a cohabiting union among women who have ever cohabited. I find that cohort is not significantly associated with the likelihood of a cohabiting conception, and in contrast to my hypothesis 1. This is different from my expectation that childbearing may have become more common in recent years among cohabiting unions. This finding may reflect that due to the compositional differences between the two cohorts, as evidenced in Table 4.3, the cohort effect has attenuated. This finding is consistent with previous research (Raley 2001) that also found little evidence that fertility is increasing within cohabitation, and that the increasing proportion of births that are to cohabitors is mostly attributed to that fact more people are cohabiting, rather than because a greater share of cohabiting couples are having children

It appears that educational attainment is the strongly associated with the likelihood of pregnancy among cohabitors. As hypothesized, cohabiting women with higher levels of education are significantly less likely to conceive relative to cohabiting women with a high school diploma. Having some college experience relative to a high school education reduces the odds of a cohabiting pregnancy by about 40 percent, and a college education reduces the likelihood by nearly 80 percent. I further find that current school enrollment reduces the likelihood of conceiving among cohabitors by over half. I find that cohabitation may become more central in the union and family building decisions among Whites. African Americans are approximately 30 percent less likely to conceive their first child in a cohabiting union relative to White women.

This finding is in contrast to previous research that finds Black women are more likely to conceive while cohabiting (Manning 2001). In separate analyses (results not shown) I run this model separately for Whites, Blacks, and Hispanics, and find that this changing propensity to conceive within a cohabiting union appears to be concentrated among lower educated Whites.

White women with less than a high school education are 2 times as likely to conceive within a cohabiting union compared to women with a high school diploma. However, I find educational attainment is not significantly associated with Black or Hispanic women’s likelihood of conceiving within a cohabitation.

Additionally, and as hypothesized, I find Hispanic women are 2.5 times as likely to conceive within a cohabitation as Whites. This finding is consistent with the idea that cohabitation is a more culturally accepted avenue for childbearing and rearing among Hispanics, and that the meaning of cohabitation differs considerable for Hispanics than for Whites as well as

Blacks (Landale and Manning 1996; Oropesa 1996).

Women whose mother has some education relative to a high school diploma are approximately 30 percent less likely to experience a cohabiting conception. Furthermore,

56 cohabiting women who received public assistance during childhood are about 40 percent more likely to experience a conception than women who did not receive any public assistance. In order to assess if these factors predicting the likelihood of a cohabiting conception have changed over time, I test for potential interaction effects between cohabiting women’s characteristics and family background and cohort. I find no significant interaction effects.

Union Transitions Following a Non-marital Conception

In this next section I use both binomial and multinomial logistic regressions to estimate union transitions following a non-marital conception. As stated previously, I examine conceptions as well as births because restricting the analyses to births only limits the ability to assess a woman’s response to impending parenthood. A pregnant woman can either marry, cohabit, or break up before the birth of the child, therefore I examine their union transitions after a pregnancy. I first estimate the likelihood a cohabiting woman will marry, remain cohabiting or dissolve the union following a cohabiting conception. I next conduct a multinomial logistic regression to estimate the likelihood that a single non-cohabiting woman will marry, cohabit or remain single following a single non-cohabiting conception.

Union Transitions Among Cohabitors

In order to explore the changing role of cohabitation in women’s responses to an impending birth, I first examine the union transitions following a cohabiting conception. Pregnant women in a cohabiting union may experience either a birth legitimated by a marriage, a birth while cohabiting, or a birth while single. However, due to the fact that less than 8 percent of cohabitors separated prior to the birth of their child, I only examine the likelihood of marrying within three months of childbirth versus remaining in a cohabiting union following a conception.

Just as the descriptive tables indicated, Model 1 of Table 4.6 confirms that cohabitation has become an increasingly important component of unmarried childbearing and responses to a non-marital pregnancy have dramatically changed within the last couple of decades.

[Table 4.6 about here]

Women in the later cohort are 60 percent less likely to enter into a marriage within three months of childbirth. In the span of fifteen years the likelihood of entering marriage has been reduced by over half among the recent cohort of cohabiting women who conceived. This finding is in support of my hypothesis that the ―shotgun‖ marriage is becoming an increasingly rare response to a non-marital pregnancy, and is being replaced by cohabitation. In support of hypothesis 5, I find that women with the least amount of education are significantly less likely to transition into a marital union following a cohabiting conception. Compared to women with a high school education, cohabiting women with less than a high school diploma are about 70 percent less likely to marry within three months of childbirth. In support of hypothesis 6, my findings suggest that an impending birth encourages White cohabitors to marry within three months of childbirth, whereas pregnant Black and Hispanic remain cohabiting following childbirth. In order to see if these race and educational variations have changed across cohorts, I test for a number of interactions. In contrast to my hypotheses, I find no interaction effects.

Union Transitions Among Single Non-Cohabiting Women

In this last set of analyses I examine single non-cohabiting women’s response to impending parenthood. By examining the union transitions following a non-marital pregnancy, I am able to see the changing roles of marriage and cohabitation in the family formation process.

My analyses are limited to never married women that were single non-cohabiting at the time of their first conception. Single women may experience a pregnancy followed by a marriage within

58 three months of the child’s birth, transition into a cohabitation within three months of the child’s birth, or they may remain single by the birth of the child.

Union Transitions

The multivariate results echo my descriptive findings in regards to single women’s response to impending parenthood. In Table 4.7, the first column displays the relative risk of marrying rather than remaining single and the second column displays the relative risk of cohabiting rather than remaining single. As shown in the first column, the cohort effect indicates that the women in the later cohort who conceive while single non-cohabiting are significantly less likely, in fact, nearly 70 percent less likely to enter into a marital union within three months of the child’s birth. This finding supports my hypothesis 4, and confirms the diminishing relationship between marriage and childbearing. Interestingly, I find no significant association with education at time of conception and the likelihood of transitioning into a marriage among single women.

Similar to Table 4.6, I find the response to an impending birth varies substantially by race. A conception outside of any union appears to spur single White women to marry, whereas Black and Hispanic women are significantly more likely to remain single. These findings illustrate the disparity in marriage rates between White and Black women. Single pregnant Black women are

90 percent less likely than pregnant White women to marry within three months of childbirth.

[Table 4.7 about here]

However, the second column reveals that cohabitation is increasingly replacing marriage in response to a non-union conception, particularly for White women. As expected, and in support of my hypothesis 4, women in the later cohort are twice as likely to enter into cohabitation following a non-union conception than the earlier cohort of women. While the proportion of pregnant unmarried women that married at the time of birth has declined, the proportion of these women entering into a cohabitation has increased. However, the risk of

59 transitioning into a cohabiting union varies substantially by race. As hypothesized, single pregnant White women are significantly more likely to move in with their partner in response to a non-union pregnancy than Black women. In fact, pregnant Black women are approximately 60 percent less likely to cohabit and remain single in response to their pregnancy compared to White women. As expected, I find women who had less than a high school education at the time of conception are significantly less likely to transition into a cohabitation compared to women with a high school education at the time of conception.

In supplementary analysis (results not shown) I omit marriage as the reference category.

I find that women who conceive outside of any union are 3 times more likely to remain single than to marry within three months of childbirth and are 6 times more likely to enter into a cohabitation following a conception rather than marry. Although the overall likelihood of experiencing a cohabitation conception has not increased between the two cohorts, the risk of giving birth in a cohabiting union has substantial changed between these two groups of women.

In order to see if the educational and racial differences have changed across cohorts I test for a number of interaction effects, however, I find no significant interactions.

Summary

The findings with regard to the hypotheses are as follows: (1) the likelihood of conceiving outside of any union has increased between the two cohorts; (2) there does not appear to be an increasing propensity to conceive within a cohabiting union; (3) there have been substantial changes in the response to an impending birth. Never married single women are increasingly more likely to choose entering into a cohabitation if pregnant, and never married cohabiting women are significantly less likely to marry in response to a pregnancy; (4) non- marital childbearing is still disproportionately concentrated among those with lower levels of education and minority women.

Discussion

The findings in this chapter demonstrate how the American family is increasingly the basis of social stratification. As marriage now serves as the line of demarcation between the

―haves‖ and the ―have nots‖ (Nock 2006), cohabitation has increasingly become the response among those that lie somewhere in between these two ends of the economic spectrum. Family scholars have long debated the place of cohabitation in the family continuum. For some, cohabitation may serve as a step in the marital process, an alternative to being single, or for others an alternative to marriage. Results from this study suggest cohabitation may be emerging as its own family institution- at least for some. I find that cohabitation is becoming more prominent in family building among lower educated White and Hispanic women, while family formation outside of any union appears to be increasing among women with the fewest economic resources.

While I find there has not been change in the overall likelihood of conceiving within a cohabitation, I do find that the likelihood of conceiving outside of any union has increased over time. Women in the later cohort are more likely to conceive their first child outside of any union compared to women in the earlier cohort. Some of this increase appears to be attributed to an increase in childbearing among those with the least amount of education. Given previous research finds a substantial share of non-marital births are unplanned, my findings lend credence to the argument that perhaps some of these educational disparities in non-marital childbearing arise from the less effective and consistent use of contraception among lower educated women

(Musick et al. 2007).

The growth of non-marital non-union childbearing among the recent cohort of women is disconcerting because it suggests that the economic disparities between children in married families and unmarried families are growing and doing so among those with the least amount of resources. Children living in single non-cohabiting mother families are at the greatest risk for

61 poverty. In fact, the poverty rate for single mother families is double that for cohabiting families, and nearly six times the poverty rate for children living in married families (Acs and Nelson

2002).

Findings from this study highlight the increasing role cohabitation is playing in the family formation behaviors among Whites. However, it appears that this role is primarily concentrated among lower educated Whites. Although White women were substantially less likely than Black women to conceive outside of a union, they were significantly more likely, in fact, 50 percent more likely to conceive within a cohabiting union. This is in contrast to previous works that finds cohabiting pregnancies are more common among Blacks than Whites (Manning

2001). This change in the propensity to conceive between Black and White cohabitors underscores cohabitations evolving role in White women’s family building behaviors, but it also suggest a shift in meaning among Black women. The substantial role of cohabitation in the family formation process among Hispanics was also illustrated. Hispanic women were twice as likely as White women to conceive their first child while in a cohabiting union. Family scholars contend that cohabitation is a more normative context for childbearing and rearing among

Hispanic women. Moreover, a significant number of cohabiting conceptions are actually planned

(Musick 2002 ).

In addition, I also find the response to impending parenthood has dramatically changed over the last couple of decades. The most common response to a non-marital -and most likely unplanned- pregnancy was cohabitation. Both single and cohabiting women in the later cohort were significantly more likely to transition into or remain within a cohabiting union rather than marry or remain single. However, the role and function of cohabitation in these family changes varied considerably by race and educational attainment. Higher educated and White women were more likely to marry in response to impending parenthood compared to women with less

62 education as well as Hispanic and Black women. However, White women were also more likely to transition into a cohabiting union rather than remain single compared to Black women. These findings illustrate the substantial likelihood that an African American child will be born outside of any union.

The results of this study extends Raley’s (2001) work that finds women in the early

1990’s were nearly as likely to cohabit as marry in response to a pregnancy. I find by the decades end, and beginning of the new millennium, single women were substantially more likely to cohabit rather than marry or remain single. Clearly, the ―shotgun‖ marriage is becoming an increasingly rare response to a non-marital pregnancy, and is being rapidly replaced by cohabitation- at least among certain groups of women.

From a methodological standpoint, my findings illustrate the need to distinguish between conceptions and births when assessing the role of non-marital childbearing and cohabitation in the family formation process. As demonstrated in Tables 4.2 and 4.3, a substantial number of women ―legitimated‖ their non-marital conception by transitioning into a marriage within three months of childbirth. Over 30 percent of single women and nearly half of the cohabiting women in the earlier cohort married within three months of childbirth. Similarly, approximately 14 percent of single women and 40 percent of cohabiting women in the later cohort also married within three months of birth. Thus, when researchers limit their analyses to non-marital births, a large percentage of non-marital childbearing is excluded.

Likewise, and perhaps even more importantly, I find that a growing number of single women are cohabiting in response to a pregnancy. Thus, a substantial number of

―cohabiting births‖ actually originated outside of any union. In less than two decades the number of parent cohabitors who began living together in response to a pregnancy rather than as a result of a relationship decision more than tripled ( 5 % to 17%). Again, this finding highlights the

63 growing necessity to examine both conceptions and births when exploring the changing roles of non-marital childbearing and cohabitation. But this growth in ―shotgun‖ cohabitation should also be of concern to both researchers and policy makers alike. Cohabiting parents who chose to remain cohabiting if a pregnancy occurred formed their union around the couple and not the pregnancy. Oftentimes, the transition into a cohabiting union is a fluid and gradual process, and not a deliberate decision (Manning and Smock 2005). In a similar vein, Stanley and colleagues

(2006) refer to such a process as ―sliding not deciding.‖ That is, couples transition into various relationships without fully realizing what is happening. When the event or constraint of an unexpected pregnancy occurs, non-cohabiting couples may slide or transition into a cohabiting relationship without fully comprehending the implications of entering into a co-residential union.

Couples who may have otherwise broken up find themselves in a relationship with greater constraints. Although the number of cohabitations that transition into a marital union has declined over the last few years (Bachu 1999; Lichter, Qian, and Mellott 2006), evidence still finds that the majority of cohabitors began cohabiting with definite plans to marry their partner(

Guzzo 2009; Manning and Smock 2005). Thus, cohabiting unions that were formed around relationship compatibility may be substantially more stable, albeit weaker than marital unions, but more stable than unions formed in response to a non-marital pregnancy. This difference in cohabiting relationships should be of concern given the recent push from public policies that state cohabiting families are logical targets for marriage promotion (Acs and Nelson 2002).

Finally, the substantial educational and racial differences in cohabitation’s role in the family formation process suggest a continued growth in the economic disparities between married and unmarried families, and perhaps even within cohabiting and non-cohabiting families. The results of this study appear to indicate that as marriage has become increasingly stratified among race as well as education, lower educated Whites find themselves increasingly excluded from this

64 institution. Thus, cohabitation is becoming a more prominent framework in their family formation decisions. However, Black women, who have been precluded from the institution of marriage for some time, may be finding themselves increasingly excluded from the realm of cohabitation, and are now being relegated to the bottom of the family spectrum. In fact, Wu and colleagues (2001) found that while births to cohabiting Whites rose dramatically over the last couple of decades, accounting for nearly all of the rise in White non-marital fertility, cohabiting births have actually constituted a declining fraction of non-marital births among Black women.

As illustrated in this study, I find the proportion of cohabiting Black births has declined between these two cohorts of women, and more importantly, Whites are now more likely to conceive in a cohabitation than Blacks, and more likely to transition into a cohabiting union in response to a non-marital pregnancy. In fact, White women were nearly 3 times more likely than Blacks to transition into a cohabiting union rather than remain single in response to a non-marital pregnancy. This should be of concern to policymakers and researchers alike. As marriage continues to stratify along education and race, we may find a similar stratification process emerge within the realm of cohabitation. However, the analyses in this study did not permit the ability to further explore this question. Future research should continue to assess these racial and educational variations within this subpopulation of women.

In sum, the findings from this chapter demonstrate the continued decoupling of marriage and childbearing, and suggest that cohabitation may be emerging as its own family institution, especially among Hispanic women. Though the idea of marriage appears to be prominent in the life course of most women, the reality of this speaks otherwise. A large share of cohabiting parents never do transition into a marriage, and a substantial number break up (Gibson et al,

2005; Reed 2006). As marriage increasingly becomes a marker of prestige and financial achievement, cohabitation may have emerged as an interim context to provide lower educated and

65 minority women some form of a family context, and perhaps temporary alleviates fears and concerns of being a single mother. Because of the high rates of union instability among cohabiting parents, it is difficult to argue that cohabitation is an alternative for marriage (Carlon,

McLanahan and England 2004), particularly since a number of cohabiting women quickly transition into other relationships following the separation from the child’s father (Carlson et al,

2004; Edin and Kefalas 2005). The findings from this chapter add to the complex meaning of cohabitation in the life course of young adults. While cohabitation is becoming more prominent in the family formation decisions of lower educated Whites and Hispanics, women with the least amount of education and Blacks are increasingly forming families outside of any union. This chapter represents a good starting point for understanding the changing trends in union and family formation behaviors over the last couple of decades. However, additional investigation is required into to understanding why such differences have emerged for this later cohort of women.

In Chapter 5 I explore the relationship between childhood family structure and these changing union and family formation behaviors among today’s young adults.

Table 4.1: Characteristics of Female Respondents by Cohort. a b Chi Early Cohort Later Cohort Square/T -Test Background Variables M SD M SD 27.13 4.50 25.46 4.34 Age *** Educational Attainment at Conception Less than high school 0.31 0.37 0.13 0.23 *** High school degree 0.42 0.49 0.29 0.46 *** Some college 0.18 0.36 0.32 0.46 *** College + 0.09 0.44 0.26 0.48 ***

c Current school enrollment 0.10 0.29 0.23 0.42 ***

Race Non-Hispanic Black 0.21 0.41 0.15 0.36 *** Hispanic 0.09 0.29 0.05 0.22 *** Non-Hispanic White/Other 0.70 0.46 0.80 0.40 ***

Childhood Family Background Maternal Education

Less than high school 0.37 0.48 0.12 0.33 *** High school degree 0.39 0.49 0.43 0.49 * Some college 0.14 0.35 0.27 0.44 *** College + 0.10 0.29 0.18 0.39 ***

Receipt of welfare during childhood 0.17 0.37 0.15 0.36 0.25 0.43 0.28 0.45 Parental divorce by Age 14 * Union Status Ever Married 0.68 0.46 0.43 0.50 *** Cohabited prior to first marriage 0.34 0.48 0.60 0.49 *** No cohabitation prior to first marriage 0.66 0.48 0.40 0.49 *** Currently Cohabitating Never Married 0.06 0.23 0.14 0.35 *** Cohabited one time 0.77 0.43 0.70 0.46 Cohabited more than one time 0.23 0.42 0.30 0.46 Currently Not Cohabitating Never Married 0.26 0.44 0.43 0.50 *** Previously cohabited 0.31 0.46 0.25 0.44 * Never cohabited 0.69 0.46 0.75 0.44 * N 3118 1047

Fertility Historyd

Conceived single non-cohabitating 0.42 0.49 0.48 0.50 * Conceived single- cohabitating 0.10 0.29 0.15 0.35 *** Conceived after first marriage 0.48 0.50 0.37 0.48 *** N 2095 457 a women entering first union between 1970 and 1980's b wome entering first union between 1990 and 2000's c attending part or full time at time of survey 67 d includes only women who have experienced a first conception that lead to a live birth Table 4.2: Characteristics of Non-Married Non-Cohabiting Women by Conception Status and Cohort. Early Cohorta Later Cohortb Conceived Outside of Conceived Outside of Did Not Conceive Did Not Conceive Union Union Background Variables M SD M SD M SD M SD Current Age 27.15 4.46 27.09 4.52 26.43 3.96 25.20 4.39 Age at birth 18.47 3.02 ------19.12 2.98 ------Educational Attainment Less than high school 0.62 0.42 0.18 0.34 0.45 0.35 0.05 0.18 High school degree 0.31 0.50 0.46 0.49 0.33 0.50 0.26 0.44 Some college 0.06 0.41 0.23 0.45 0.18 0.46 0.35 0.50 College + 0.01 0.20 0.13 0.40 0.04 0.34 0.32 0.48

Current school enrollmentc 0.06 0.25 0.11 0.31 0.11 0.31 0.27 0.44

Race Non-Hispanic Black 0.43 0.50 0.12 0.33 0.30 0.46 0.11 0.31 Hispanic 0.09 0.28 0.09 0.30 0.08 0.25 0.04 0.21 Non-Hispanic White/Other 0.48 0.50 0.79 0.42 0.63 0.48 0.84 0.37

Childhood Family Background

Maternal Education

Less than high school 0.51 0.50 0.29 0.47 0.24 0.43 0.10 0.29 High school degree 0.34 0.48 0.43 0.49 0.48 0.50 0.42 0.49 Some college 0.10 0.29 0.17 0.36 0.22 0.43 0.27 0.45 College + 0.05 0.21 0.12 0.32 0.07 0.25 0.22 0.41

Receipt of welfare during childhood 0.26 0.44 0.13 0.34 0.26 0.44 0.13 0.33 Parental divorce by age 14 0.30 0.46 0.23 0.42 0.32 0.47 0.27 0.44

Cohabitation History

Ever Cohabited 0.39 0.49 0.36 0.48 0.63 0.48 0.46 0.50 Cohabited more than once 0.05 0.23 0.04 0.22 0.17 0.39 0.10 0.31

Union Status at Birth

Single 0.63 0.48 ------0.69 0.46 ------

Cohabitating 0.05 0.23 ------0.17 0.37 ------

Marriage 0.32 0.46 ------0.14 0.32 ------

N 848 2270 207 840 a women entering first union between 1970 and 1980's b wome entering first union between 1990 and 2000's c attending part or full time at time of survey

Table 4.3: Characteristics of Cohabiting Women by Conception Status and Cohort. Early Cohorta Later Cohortb Conceived within union Did not Conceive Conceived within union Did not Conceive

Background Variables M SD M SD M SD M SD Age 26.73 4.13 27.11 4.32 26.25 3.96 26.30 4.11 20.10 3.26 ------20.30 2.52 ------Age at Birth Educational Attainment Less than high school 0.51 0.48 0.32 0.36 0.25 0.36 0.17 0.24 High school degree 0.40 0.49 0.37 0.49 0.45 0.50 0.28 0.46 Some college 0.08 0.39 0.20 0.45 0.22 0.46 0.29 0.49 College + 0.01 0.18 0.11 0.38 0.08 0.21 0.26 0.42

Current school enrollmentc 0.03 0.17 0.08 0.27 0.03 0.17 0.16 0.37

Race Non-Hispanic Black 0.16 0.37 0.23 0.42 0.12 0.33 0.11 0.31 Hispanic 0.19 0.39 0.05 0.22 0.07 0.26 0.05 0.22 Non-Hispanic White/Other 0.65 0.48 0.72 0.45 0.81 0.40 0.84 0.37

Childhood Family Background Maternal Education Less than high school 0.45 0.50 0.28 0.47 0.23 0.42 0.13 0.34 High school degree 0.42 0.49 0.43 0.49 0.49 0.50 0.41 0.49 Some college 0.07 0.26 0.17 0.36 0.20 0.42 0.31 0.47 College + 0.05 0.22 0.12 0.32 0.08 0.26 0.15 0.35

Receipt of welfare during childhood 0.30 0.46 0.20 0.40 0.31 0.47 0.16 0.37 Parental Divorce by age 14 0.40 0.49 0.29 0.45 0.37 0.49 0.36 0.48

Cohabited More than one time 0.12 0.32 0.14 0.35 0.22 0.42 0.23 0.42 Married Within 3 Months of the Birth 0.47 0.50 ------0.40 0.49 --- N 200 900 67 444 a women entering first union between 1970 and 1980's b wome entering first union between 1990 and 2000's c attending part or full time at time of survey

Table 4.4:Odds Ratios for Logistic Regression Analysis Predicting the Likelihood of Conceiving Among Non- Married Non-Cohabiting Women

MODEL 1 MODEL 2 Age 1.02 ** 1.02 *** Cohort 1.62 *** 1.28

Educational Attainment Less than High School 5.80 *** 3.43 ***

High School Diploma ------Some College 0.38 *** 0.21 *** College 0.09 *** 0.05 ***

Currently Enrolled 0.87 * 0.85 *** Race Black 5.13 *** 5.09 *** Hispanic 0.82 0.81 White ------

Ever Cohabitated 1.02 1.01 Childhood Family Background

Maternal Education

Less than High School 0.90 0.91

High School Diploma ------Some College 0.90 0.91 College 0.77 0.77

Parental Divorce 0.90 0.90 Receipt Welfare 1.13 1.12

Interactions Less than high school*cohort 1.57 + Some College*cohort 1.54 College*Cohort 1.55

LR chi2 1269.76 1273.88 N 4150 4150 *** p<= .001, ** p. <=.01, * p.<=.05, + p.<=10

Table 4.5: Odds Ratios for Logistic Regression Analysis Predicting the Likelihood of Conceiving While Cohabiting Among Women Who Have Entered Into a Cohabitation

MODEL 1 Age 1.00 Cohort 1.00

Educational Attainment Less than High School 1.20 High School Diploma --- Some College 0.58 * College 0.12 ***

Currently Enrolled 0.42 * Race Black 0.63 * Hispanic 2.52 *** White ---

Entered more than one cohabitation 0.82 Childhood Family Background Maternal Education Less than High School 1.01 High School Diploma --- Some College 0.59 * College 0.71

Parental Divorce 1.22 Receipt Welfare 1.41 * LR chi2 133.61 N 1605

*** p<= .001, ** p. <=.01, * p.<=.05, + p.<=10 71

Table 4.6: Odds Ratios for Logistic Regression Analysis Predicting the Likelihood of Marrying Within 3 Months of First Child's Birth Among Never Married Pregnant Cohabiting Women

MODEL 1 Age 0.69 Cohort 0.40 * Educational Attainment Less than a high school degree 0.28 ** High School Degree ---- Some College 0.83 College + 0.20

Currently Enrolled 0.89

Race Black 0.07 *** Hispanic 0.30 ** White ---

Entered into more than one cohabitation 0.55

Childhood Family Background Maternal Education Less than a high school degree 0.84 High School Degree --- Some College 1.06 College + 0.91

Receipt of welfare during childhood 0.88 Parental Divorce by Age 14 1.15

LR chi2 48.66 N 258 *** p<= .001, ** p. <=.01, * p.<=.05, + p.<=10

Table 4.7: Relative Risk Ratios from Multinomial Logistic Regression Analysis Predicting the Likelihood of Forming a Union Within 3 months of First Child's Birth Among Never Married Pregnant Single Women

MODEL 1 Marriage Cohabitation Age 1.07 *** 0.99 Cohort 0.30 *** 1.87 ** Educational Attainment Less than High School 0.86 0.59 * High School Diploma ------Some College 1.08 0.86 College 1.10 1.10

Currently Enrolled 1.36 1.60

Race Black 0.10 *** 0.34 *** Hispanic 0.54 * 0.86 White ------

Entered into more than one cohabitation 0.05 *** 1.18 Childhood Family Background Maternal Education Less than High School 0.93 0.79 High School Diploma ------Some College 0.92 0.81 College 0.95 0.69

Welfare Receipt 0.66 * 1.10

Parental Divorce 0.61 * 1.21

LR chi2 298.91 N 1094 73 *** p<= .001, ** p. <=.01, * p.<=.05, + p.<=10

0.45

0.40

0.35

0.30

0.25

Early Cohort

P roportion 0.20 Later Cohort

0.15

0.10

0.05

0.00 Less than high school High school degree Some college College + Figure 4.1: Educational Attainment at Time of Conception of All Women by Cohort

0.70

0.60

0.50

0.40

Early Cohort Later Cohort

Proportion 0.30

0.20

0.10

0.00 Cohabited Prior to First Marriage No Cohabitation Prior to First Marriage

Figure 4.2: Proportion of Women Who Cohabited Prior to First Marriage by Cohort

1.00 0.01 0.04 0.06

0.90 0.18

0.80 0.31

0.70

0.33 0.60 College + Some college 0.50 High school degree

Proportion Less than high school 0.40

0.30 0.62

0.45 0.20

0.10

0.00 Early Cohort Later Cohort

Figure 4.3: Educational Attainment at Time of Conception of Women Who Conceived Outside of Any Union by Cohort

1.00

0.14 0.90

0.32 0.80 0.17

0.70 0.05

0.60

Transition to marriage 0.50 Transition to cohabiting union

Remaing single Proportion 0.40

0.69 0.63 0.30

0.20

0.10

0.00 Early Cohort Later Cohort Figure 4.4: Response to a Single Non-cohabiting Conception by Cohort

1.00 0.01 0.08 0.08 0.90

0.80 0.22

0.70 0.40

0.60 College + Some college 0.50 0.45 High school degree

Less than high school Proportion 0.40

0.30 0.51

0.20

0.25 0.10

0.00 Early Cohort Later Cohort Figure 4.5: Educational Attainment of Cohabiting Women Who Conceived by Cohort

1.00

0.90

0.80 0.40 0.47

0.70

0.60

Transition to marriage 0.50

Remain in cohabitation Proportion 0.40

0.30 0.60 0.53

0.20

0.10

0.00 Early Cohort Later Cohort Figure 4.6: Response to a Cohabiting Conception by Cohort

CHAPTER 5

CHILDHOOD FAMILY STRUCTURE: INTERGENERATIONAL PATTERNS OF

FIRST UNION FORMATION AND CHILDBEARING BEHAVIORS

The previous chapter demonstrated the rapid changes in union and family formation behaviors over the last two decades. The declining centrality of marriage and the increasing role of cohabitation in the lives of young adults are apparent in recent years. This chapter examines early life course factors that may influence young women’s decisions to enter into cohabiting relationships instead of marriages, as well as their decisions to initiate childbearing outside of marriage. Young adults who transitioned into adulthood in the late 1990’s and early 2000’s represent the first cohort of young adults who experienced, as children, the rapid growth in parental divorce, as well as a growing diversity in family type (e.g. cohabitation, step-family, single mother family). Indeed, approximately 40 percent of today’s young adults experienced the disruption of their parents’ marriages (Bumpass 1984). The majority of these divorced parents transitioned into new unions following the breakup of their marriages (Glick 1989). While a number of parents transitioned directly into a remarriage (Glick 1987), a substantial share of parents began entering into cohabiting unions following their divorce. Though some of these cohabitations resulted in a remarriage, a number of today’s young adults experienced a parental cohabitation post-divorce without a subsequent remarriage (Bumpass et al. 1995). The sharp decline in remarriage rates between 1970 and the early 1980’s has been entirely attributed to the rapid rise in cohabitation (Bumpass and Cherlin 1991). Approximately one fifth of today’s young

77 adults experienced the event of parental cohabitation following the disruption of their parents’ marriage prior to age 16 (Bumpass et al 1991).

Evidence suggests that these rapid family changes during childhood and adolescence may have likely recast the traditional pathway in the life course of today’s young adults (Elder et al.

2003). In the span of only three decades, cohabitation has replaced marriage as the most common first union choice (Bumpass and Lu 2000; Kennedy and Bumpass 2008), and non-marital births now account for nearly 40 percent of all first births (Hamilton et al. 2007). The once standard pathway of leaving the parental home, completing education, marrying then having children is no longer the normative route for the majority of today’s young adults.

Research suggests that individuals’ union formation patterns and childbearing decisions are entrenched in the life course transitions of their parents (Bengtson and Allen 1993; Elder

1984; Hagestad 1984). In particular, empirical evidence has established a link between childhood family experiences and young adult’s relationship and childbearing outcomes (Bulanda and

Manning 2008; Cavanagh, Schiller, and Riegle-Crumb 2006; Graefe and Lichter 1999;

McLanahan and Sandefur 1994). However, the influence of childhood experiences is complex and poorly understood because much of the previous research has ignored the increasing diversity in parental union histories. Childhood family instability is often assessed with a simple dichotomous measure of ―divorce‖ versus ―did not divorce‖ or ―intact‖ versus ―non-intact‖ family. Relatively few studies have adequately captured the increasing presence of cohabitation in childhood family experiences on relationship and family trajectories. When studies have incorporated a more diverse measure of family structure, the role of cohabitation has often been included in the ―single family‖ or ―step-parent family‖ type, thus obscuring the effects of parental cohabitation post- divorce on offspring’s own relationship behaviors. Current literature has yet fully to answer how time spent living with a cohabiting parent influences today’s young adults’

78 first union formation and childbearing behaviors. How, if at all, has cohabitation’s increasing presence in childhood family structures helped shaped young adults’ union and family formation patterns? The role of parental cohabitation during childhood on young adults’ relationship and family trajectories requires additional study.

Life Course Theory

The life course paradigm contends that events and experiences that occur early in life will affect later life circumstances. To understand events and circumstances that transpire in adulthood, one must also have an understanding of an individual’s life in a social and historical context (Elder 1998). Guided by the life course perspective and the principle of linked lives, the main argument of this chapter is that the life trajectories of young adults are entangled with the life course paths of their parents (Bengtson and Allen 1993). Decisions and transitions occurring in one generation have consequences for the lives of the next generation (Elder 1984; Hagestad

1984).

Research has consistently established an association between family structure during childhood and both child and adult life outcomes (Bulanda and Manning 2008; Cavanagh,

Schiller, and Riegle-Crumb 2006; Graefe and Lichter 1999; McLanahan and Sandefur 1994;

Manning and Lamb 2003; Raley and Wildsmith 2004; Thorton 1991). Parents’ union and family formation decisions around marriage, divorce, cohabitation, and non-marital childbearing produce a number of transitions that affect not only the trajectories of their lives, but those of their children as well. Parental divorce, single parenting, step-parenting, and never-married parenthood all have been linked to a number of negative consequences in both childhood and adulthood. Psychological and behavioral problems, lower academic achievement, (Amato 2000;

Furstenberg, Brooks-Gunn, and Morgan 1987; Thomson et al. 1994), a lower probability of high school completion (McLanahan and Sandefur 1994), earlier departure from home,(Aquilino 1991;

Goldscheider and Goldscheider 1998), and increased likelihood of early cohabitation and non- marital childbearing (Sassler and Goldscheider 2004; Thornton 1991; Wu and Martinson 1993;

Wu 1996) have been associated with family disruption during childhood. Clearly, childhood transitions influence later life course trajectories by guiding individuals toward specific life pathways (Aquilino 1996).

However, the life course perspective also highlights the importance of the historical context in which one lives. Life course theory takes period changes, such as the rapid rise in divorce and cohabitation, into account in order to understand the influences of childhood family structure within a broader context of historical and social change (Bengtson and Allen 1993;

Elder 1991; 1994; 2003). The cohort of women in this study experienced the ―divorce revolution‖ as well as an increasing number of alternative family forms as children. From 1960 to 1980, the divorce rate doubled, and nearly half of all marriages were expected to dissolve

(Castro, Martin, and Bumpass 1989). While divorce rates reached a record high during the

1970’s, they have since leveled off (Goldstein 1999; Raley and Bumpass 2003). As a consequence of such a high number of marital disruptions, a number of non-traditional family forms (e.g. cohabitation, step family, single mother) began to emerge. This period of rapid family changes has likely influenced the relationship and family building behaviors of today’s young adults. As divorce, cohabitation and non-marital childbearing have become more commonplace in the life course of young adults, the negative stigma once associated with these demographic behaviors has substantially weakened over time (Hobcraft 2008), perhaps reducing some of the negative consequences on childhood and adulthood outcomes.

In summary, the life course perspective understands that the family context in which one grows up is important in determining later life transitions. However, it argues that the societal and historical contexts in which one grew up must be taken into account as well. In this study, I

80 argue that the rapid demographic and attitudinal changes over the latter half of the twentieth century substantially influenced the life course patterns of young adults transitioning into adulthood in the late 1990’s and early 2000’s. Specifically, the growing complexity in non- traditional family types experienced during childhood has significantly contributed to the increasingly diverse life course pathways among today’s young adults.

Background Information

The changing demography of U.S. families over the last few decades has been well established (Bianchi and Casper 2000; Cherlin 1992; Lilliard, Brien and Waite 1995; Loomis and

Landale 1994; Seltzer 2000). The rise in divorce, the rise in age at first marriage, and the rise in cohabitation have altered the family landscape dramatically. Approximately two-fifths of today’s young adults grew up outside of a home headed by two biological parents married to each other (Bumpass and Lu 1989). Approximately 25 percent spent some time in a step-family household during childhood, and nearly two-thirds of these remarriages began in cohabitation

(Bumpass et al 1995). Coinciding with the rising number of divorces and post-divorce cohabiting family types was a growth in households headed by never-married women. The percentage of births to unmarried women more than doubled between 1965 and 1980 (7.7% to

18.4%) (Ventura et al. 2000). These dramatic changes in childhood living arrangements have most likely influenced the life course transitions of today’s young adults.

Exposure to parental union transitions and family instability carries over into children’s own lives. A considerable amount of scholarly attention has been devoted to the childhood and young adult outcomes of family instability. Concerned over the consequences of parental discord, a surfeit of literature emerged documenting the detrimental effects of a parental divorce on the emotional and psychological well-being during childhood (Amato 2000; Aquilino 1991;

Furstenberg Brooks-Gunn and Morgan, 1987; Goldscheider and Goldscheider 1998 McLanahan

81 and Sandefur 1994). Later research found these negative effects persisted into adulthood (Sassler and Goldscheider 2004; Thornton 1991; Wu and Martinson 1993; Wu 1996). Adults who experienced the disruption of their parents’ marriages still had lower levels of well being, lower educational attainment, more periods of not working or being in school, and lower levels of income (Aquilino 1996; Sigle-Rushton, Hobcraft, andKiernan 2005). Aside from the long-term psychological and socioeconomic effects, growing up outside of a home headed by stably married parents has also been found to a have substantial influence on young adult’s own relationship and childbearing behaviors.

The experience of a parental divorce during childhood appears to shift a child’s first union experience away from marriage and more towards cohabitation (Thornton 1991). Children of divorced parents are more likely to enter into cohabiting arrangements, and do so at younger ages (Sassler and Goldscheider 2004; Teachman 2003; Wolfinger 2003). Aside from union formation behaviors, children who experienced the disruption of their parents’ marriages also tend to differ in the duration and stability of their own unions compared to children who resided with both biological parents continuously. One of the most consistent and well-cited findings throughout the intergenerational transmission literature is the association between parental divorce and adult offspring’s own union dissolution (Axinn, Thorton, and Xie 2007; McLanahan and Sandefur 1994; Teachman 2002; 2003; Wolfinger 2000; 2001). Children who witnessed the disruption of their parents’ marriage are more likely to experience instability in their own relationships. Further, exposure to parental marital conflict and divorce during childhood can influence young adults’ own relationship quality (Teachman 2003). Young adults who witnessed the disruption of their parent’s marriage report greater marital problems and lower satisfaction in their own relationships (Amato 1996; Ross and Mirowsky 1999). The risk of a non-marital conception is also greater for those who experienced the disruption of their parents’ marriage.

Yet, much of the above mentioned research has ignored the growing complexity of childhood family structures, and has failed to use a more detailed measure that includes both marriage and cohabitation.

Currently, knowledge of the relationship between these diverse family contexts, in particular the role of cohabitation, on young adults’ own union formation and childbearing behaviors is limited. Studies that have examined the impact of parental cohabitation on childhood outcomes have primarily focused on the emotional and behavioral well-being of children. Time spent in a household headed by cohabiting parents during childhood is more positively associated with lower academic achievement, lower graduation rates, and emotional and behavioral problems than time spent in households headed by two biological parents, as well as by married stepparents (Brown 2002 2004; Dunifon and Kowalski-Jones 2002; Hao and Xie

2001; Manning and Lamb 2003; Nelson ,Clark, and Acs 2001; Raley et al 2005).

Of late, only a handful of studies have focused attention on the role of parental cohabitation on children’s own union and family formation behaviors during young adulthood and, of those, the role of cohabitation is still unclear. Teachman (2003) does explore the long- term effects of parents’ marriage and cohabiting histories on children’s own relationships. He finds that youth raised in a cohabiting household are more likely to choose cohabitation rather than marriage in young adulthood. However, his sample is based on a cohort of young adults transitioning into young adulthood in the 1970’s to the early 1990’s. Young adults’ relationship and childbearing behaviors have changed substantially since this period. Between 1997 and

2001, almost 68 percent of all first unions began in cohabitation rather than marriage, compared to 60 percent less than a decade earlier 1990-1994 (Kennedy and Bumpass 2008). In addition,

Teachman only examines first union formation and stability, he does not address non-marital childbearing behaviors. Recently, two additional studies (Sassler, Cunningham, and Lichter

2009 and Ryan, Frazetta, and Manlove 2009) have emerged that utilize a more comprehensive measure of childhood family structure. Both studies employ a more recent cohort of young adults entering into relationships in the late 1990’s and early 2000, and find support for utilizing a more nuanced measure of parental relationship histories. Indeed, both demonstrate an increased chance of cohabitation among those whose parents’ cohabitated without remarriage in childhood. In addition, Sassler and colleagues (2009) find that even when compared to children whose parents transitioned into a remarriage, children who experienced a parental cohabitation following a divorce were significantly more likely to enter into a cohabiting relationship themselves.

However, Sassler et al (2009) focus on current union formation behaviors and relationship quality, while Ryan et al (2009) center their interest on early union formation and gender differences. Furthermore, neither study examines young adults’ childbearing behaviors.

Far less attention has been paid to the childbearing behaviors of young adults who experienced parental cohabitation. Bulanda and Manning (2008) find children born to cohabiting parents, when compared to all other family types ( married biological parents, married step parents, and even single parent families), had significantly higher odds of having an early sexual debut and giving birth outside of marriage by age 18 relative to young adults from both married biological parents and married stepparents . However, they limit their analyses to respondents who were adolescents between 13 and 17 years of age. To date, I know of no literature that looks at both first union formation and the childbearing experiences of young adults who transitioned into early adulthood in the late 1990’s and early 2000s.

Theoretical Perspectives

As mentioned previously, the overarching framework for this study is the life course perspective. However, I next detail two mid-level theoretical perspectives for a more thorough understanding of how the union and family formation processes among young adults are

84 influenced by one’s family of origin. A number of perspectives have been utilized to explain the intergenerational link between childhood family structure and young adult outcomes. In this study, I specifically focus on the socialization and instability and change arguments.

Family Socialization

The socialization perspective posits that children acquire attitudes and values and develop certain behaviors that they learn from their parents during childhood. Family scholars posit that the family is an important conduit through which attitudes, behaviors, and values are transferred across generations. Parents’ own attitudes and behaviors regarding union and family formation are strongly transmitted to their children (Axinn and Thornton 1996). As mentioned previously, a large body of research demonstrates the intergenerational transmission of divorce. Children who witnessed the disruption of their parents’ marriages are more likely to see their relationships dissolve (Amato 1996; Cherlin, Kiernan, and Chase-Lansdale 1995; Teachman 2002). Likewise, parents who cohabit have children who are also more likely to cohabit with a partner (Axinn and

Thornton 1996; Cherlin, Kiernan, and Chase-Lansdale, 1995).

The family socialization theory contends that much of the observed intergenerational relationship between childhood family structure and adult offspring’s own union and family formation behaviors is a result of single, divorced, cohabiting and step parents serving as role models and agents of socialization for children (McLanahan, Astone, and Marks 1991; Ryan et al.

2009). Children who experience the disruption of their parents’ marriage have been socialized to have more negative views towards marriage and more positive attitudes toward divorce, cohabitation, and non-marital sex, compared to those whose parents remained married during childhood (Axinn and Thornton 1996; 1993). These attitudes are often a result of both witnessing one’s parents engage in relationship and sexual behavior outside of a marital context, but also the relatively liberal attitudes regarding relationship and family formation behaviors possessed by

85 parents in non-traditional family types. That being said, the family socialization argument also expects differences within non-traditional family types. While exposure to a cohabiting family increases an offspring’s likelihood of entering into a cohabiting relationship, evidence finds that life in a remarried step family may increase the likelihood of marriage in young adulthood (Ryan et al. 2009;Teachman 2002;2003). Parents who choose to reenter into a new marriage following a divorce most likely possess positive attitudes towards marriage.

Having been socialized in a non-traditional family and observing one’s parents engaging in relationships and sex outside of marriage may influence children’s own attitudes, beliefs and behaviors towards relationship and family patterns. Witnessing parental cohabitation during one’s childhood weakens the negative stigma often associated with non-marital unions, as well as the stigma associated with premarital sex and childbearing (Axinn and Thorton 1996; Thorton 1991;

Clarkberg, Stolzenberg, and Waite 1995). Children who grow up in a single parent or cohabiting parent household may not see marriage as necessary component of the union formation process.

Further, the experience of parental cohabitation exposes children to an alternative living arrangement that does not require a marital bond to set up a family household. Young adults, particularly young women, who observed a parental cohabitation, were likely exposed to less traditional sex roles, more liberal attitudes regarding family formation, and a weaker commitment to relationships (Amato and Deboer 2001). Alternatively, the experience of a parent’s remarriage during childhood may model marital behaviors that may improve young adults own attitudes regarding marriage (Axinn and Thornton 1996). Thus, divorced, single, remarried and cohabiting parents have introduced a model of relationship behavior that children are increasingly likely to emulate.

On the other hand, experiencing the negative repercussions of a parental divorce and parents’ negative views towards marriage after a divorce may make children more hesitant to

86 enter into marital unions. If a parent has difficulty in coping or readjusting after a divorce, children may be socialized to be more cautious about marriage or to believe that marriage is an unpleasant experience. Therefore, young adults may be more likely to enter into a cohabiting union in response to their uncertainty or trepidations regarding the marriage formation process.

However, the influence of parental cohabitation on the union formation behaviors of today’s young adults requires additional study. The socialization perspective highlights the need to include parental cohabitation in assessing childhood family structure on young adults’ relationship and childbearing trajectories. Exposure to more liberal attitudes, as well as behaviors regarding relationship and family building behaviors outside of marriage, as well as being socialized with more negative attitudes towards marriage during childhood, suggests children who experienced a parental cohabitation, whether prior to or in lieu of remarriage, would be more likely to first enter into a cohabiting union rather than marry directly. Furthermore, having been socialized with more liberal attitudes towards non-marital sexuality should also increase the risk of non-marital childbearing.

Instability and Change

The instability and change explanation focuses on the stress often caused by a parental marital disruption during childhood. The focus of this perspective highlights the transitions in family structure and not the family type experienced during childhood (Martinson and Wu 1992).

Change in childhood family structure engenders stress for both parent and child (Amato 2000).

The greater number of family structure transitions, the larger the effect on childhood and young adult outcomes (Amato 1993). Proponents of this explanation contend that the stress experienced from the disruption of childhood living arrangements has negative consequences for children’s own union and family formation behaviors (Amato and Keith 1992). The event of a parental divorce can lead to a number of substantial changes in a child’s life. The loss of a parent, a

87 residential move, loss or decline in income, and the introduction of a new parental figure can exacerbate an already stressful situation for a child (Amato 1993; Cherlin Chase-Lansdale and

McRae 1998). The stress associated with multiple transitions during childhood has a cumulative effect and has been found to produce a negative attitude towards marriage and a more positive attitude towards non-marital cohabitation, premarital sex, and non-marital childbearing (Axinn and Thornton 1996; Thorton 1991; Wu 1996). The stress and instability often experienced after a parental disruption may encourage young adults to disengage from a taxing family environment through early sexual debut subsequently increasing their risk for cohabitation as well as childbearing. At the same time, a stressful and unstable home environment may make children more apprehensive about marriage and their own marital success. Individuals who grew up in an unstable family environment may avoid or delay marriage in order to escape the same instability in their own adult life. The instability and change in childhood should increase the likelihood of cohabitation, as either a substitute for marriage or a way to delay an eventual marriage.

In sum, both the family socialization and instability and change perspective highlight the basic principle of linked lives that is key in the life course perspective. Experiences early in an individual's life influence their subsequent behaviors and choices in adulthood (Brofenbrener

1979; Elder 1977; 1978). Childhood family experiences socialize individuals’ attitudes towards certain relationship and childbearing behaviors (Ajzen 1988). The instability and stress generated by family disruptions often produces negative attitudes toward marriage and increases the risk of premarital sex and non-marital union formation.

As the rates of cohabitation now exceed marriage as first union formation and the percentage of births outside of marriage continue to climb, exploring detailed marital and relationship histories of parents may provide a greater insight into who chooses to first enter into a cohabiting union instead of marriage, as well as to initiate childbearing outside of marriage.

Utilizing a more nuanced measure of childhood family structure, as well as up-to-date data, allows a more sophisticated understanding of these relationships. Further, a burgeoning literature in this area finds not all family structures exert equal levels of influence on child and young adult outcomes (Brown 2004; Cavanagh and Huston 2006; Sassler et al. 2009). For example, among step-parent families, children whose parents cohabited prior to remarriage have an even greater likelihood of cohabiting than children whose parents entered into a remarriage directly (Sassler et al. 2009). These findings suggest that even within remarried step-parent families, previous parental union histories can have divergent effects on adult children’s relationship formation patterns. This further highlights the need to include a more comprehensive measure of parental union histories when assessing intergenerational relationships.

In this chapter, I hope to build upon the extant literature (Teachman 2003; Ryan et al.

2009; Sassler et al. 2009) by examining a recent cohort of women transitioning into adulthood in the late 1990’s and early 2000s. This stage in the life course is often characterized by a high degree of change and complexity (Arnett, 2000; Rindfuss, 1991). High school graduation, leaving the parental home, college completion and entry into the work force are all transitions that many young adults experience. Moreover, the development of romantic relationships and even entry into parenthood are two transitions that are also characteristic of entering into young adulthood.

By examining first union formation and childbearing behaviors, I hope to elaborate the role that the diverse childhood family experiences of today’s young adults has had in shaping and changing these life course trajectories.

In sum, this chapter is guided by the following research questions: 1) How have the diverse parental union transitions experienced during childhood shaped adult offspring’s own first union formation and childbearing behaviors? 2) Are certain childhood family experiences more predictive of long term relationship and childbearing trajectories? 3) How do the relationship and

89 childbearing patterns of adults who experienced parental cohabitation differ from those who grew up with both biological parents married, with only a single parent, or in step-parent families?

This research is guided by the following hypotheses:

Hypothesis 1: Having ever lived outside of a married biological two parent family during one’s childhood will be associated with an increased chance of first entering into a cohabitation.

Hypothesis 2: Given that a parental divorce increases early entry into both marriage and cohabitation, a child’s experience of their parents’ divorce will decrease her chances of never having entered into any co-residential union.

Hypothesis 3: Observing parental cohabitation during one’s childhood will increase the likelihood of one’s forming a cohabiting relationship as a first union instead of a marriage. A parental cohabitation in lieu of marriage should have an especially strong link with entering a cohabiting union prior to or in lieu of marriage.

Hypothesis 4: Experiencing a parent’s remarriage, even if preceded by cohabitation, will increase the chance of a person forming a marital union without cohabitation prior to marriage.

Hypothesis 5: I expect growing up outside of a married biological two parent home will be associated with young women’s likelihood of conceiving outside of a marital union (cohabiting or single). Experiencing a parental cohabitation should have an especially strong link with conceiving while in a cohabiting relationship.

Hypothesis 6: Women whose mothers conceived their first child outside of marriage will be more likely to conceive their first child outside of marriage.

Data

In order to examine the intergenerational linkages between parental union transitions during childhood and young adults’ own union and family formation behaviors, intergenerational panel data are needed. The data for this chapter were taken from the National Survey of Families

90 and Households (NSFH), which reports data on multiple generations within one family. The

NSFH is a national probability sample of 13,007 respondents, including 9,643 main respondents aged 18 and over, plus an over-sample of minorities, single parent families, recently married couples, and cohabiting couples (Sweet, Bumpass, and Call 1988). The Wave 1 response rate was 74 percent. The first wave of NSFH was conducted in 1987-1988; a second wave was later conducted in 1992-1994. In the initial survey, a focal child was randomly selected from the eligible children living in the household of the main respondent. These included biological, step, adopted, and foster children under the age of 18. This analysis is only limited to biological and/or adopted children. Stepchildren or children of the respondent’s live-in partner were eliminated from the analysis. Recently, the National Survey of Families and Households (NSFH) completed the collection of a third wave of data (2001-2002), including data from extensive interviews of the focal child of the main respondent (Sweet and Bumpass 2002). These data can be linked to the initial Wave 1 information on parents and focal children (collected in 1987-88), allowing intergenerational family patterns to be studied. In Wave 3, 1,952 interviews were conducted with focal children (Wright 2003). Detailed marital and cohabitation histories were collected from the focal children, now aged 18-34. This provided a wealth of information on young adults’ current and initial union and family formation behaviors. The study sample for this analysis is restricted to only the female biological and/or adopted children of the main respondent. In all,

933 parent-child dyads are examined in this analysis. The response rate for the Focal Child at

Wave 3 was 48 percent (Wright 2003).

Variables

The focus of the main analysis is the impact of parental union transitions during childhood on adult offspring’s own union formation and childbearing behaviors. I include a number of additional measures that can mediate the relationship between childhood family

91 structure and adult relationship outcomes. Childhood social and economic background characteristics and young women’s current characteristics are included in the analyses. There are two dependent variables in these analyses. I first focus on first union formation, and then I examine the relationship context of first conception.

I employ the union history data files for focal children. These data provide a detailed history of the focal child’s current and previous cohabiting and/or marital unions. I create a three- category dependent variable to assess focal women’s first union formation patterns. Categories are coded as follows: 1=Entered directly into marriage (no prior cohabitation); 2=First entered into a cohabitation; 3=Never entered into cohabitation and/or marriage.

Respondents were asked the number of children and the date of birth for each child. For this study however, I limit the analysis to first pregnancy only. Previous research finds that second- and higher-order births are more likely to be intended than first births to unmarried women (Musick 2002), and second and third births are more likely to occur within cohabiting unions than in a single non-cohabiting state (Musick 2000). Therefore, in order to reduce the issue of selectivity in this analysis, conceptions are limited to those that led to first births only. In addition, the NSFH does not include information on conceptions that did not result in a live birth, thus conception dates are constructed by the date of birth minus nine months (see Aassve Billiari and Piccarreta 2007; Steele 2008). Non-marital conceptions are defined as pregnancies to women who had never been married at the time of conception. Because the data do not permit the ability to know if the cohabiting or marital partner is the biological parent of the child, I limit the time to transition into a union within 3 months of childbirth. In this analysis I examine first conception relational contexts. I create four dichotomous variables, defined as follows: 1=Women who conceive outside of any union (cohabiting or marital); 2=Women who conceive within a

92 cohabiting union; 3=Women who conceive within a marital union; 4=Women who did not conceive or whose conceptions did not result in live birth.

Independent Variables

My primary independent variable is parental union transitions during childhood. I assess family structure from both Waves 1 and 2. First, I assess whether the parental respondent had experienced a divorce from the biological parent of the focal child at either Wave 1 or by Wave 2.

If the parental respondent had experienced a marital disruption from the other biological parent, I determined if they entered into any subsequent unions (cohabitation or remarriage) based on the detailed relationship history data. I expand upon previous measures of childhood family structure by identifying four types of childhood living arrangements following a parental divorce: 1.

Remarriage without prior cohabitation; 2. Cohabitation prior to a subsequent remarriage; 3.

Cohabitation without a subsequent remarriage; 4. Remained single (no cohabitation and/or remarriage). 5. Biological parents remained stably married. Due to inadequate sample size

(N=45), parents never married at time of child’s birth were excluded from the overall analysis. I create a summary variable indicating whether the focal child had experienced a parental divorce.

The baseline reference group for parental union status is children whose biological parents remained stably married.

Childhood Social and Economic Background Characteristics

Maternal Education: The NSFH at Wave 1 asks main respondents the highest level of education they have currently completed. I construct maternal level of education based on the amount of education reported at the time of the Wave 1 interview. Given a substantial number of missing responses, I substitute paternal education when mother’s level of education was not available. I collapse educational attainment into four mutually exclusive categories: 1=less than a high school diploma; 2=high school degree; 3=some college experience; 4=bachelor’s degree or more.

Maternal Childbearing Behaviors: Based on the fertility history information collected during

Wave 1, I construct a measure to indicate if the focal woman’s mother experienced a non-marital conception. Consistent with my dependent measure assessing focal women’s fertility history, conceptions are limited to those that led to first births only. In addition, conception dates are constructed by the date of birth minus nine months. Non-marital conceptions are defined as pregnancies to women who had never been married at the time of conception. These are coded as follows: 0=Mother did not conceive first child outside of marriage; 1=Mother conceived first child outside of marriage.

Receipt of Public Assistance: I measure if the respondents’ families received public assistance during their childhoods.

Family Religious Affiliation: I create a measure to assess religious affiliation during childhood. I create three dichotomous variables: 1=Catholic; 2=Protestant; 3=Other. The reference religious group is Protestant.

Focal Women’s Background Characteristics

Age: I measure age as a continuous variable starting at age 18 and ending at age 34.

Education: The NSFH collected detailed educational histories. Respondents were asked the highest level of education completed, and information on each segment of schooling (e.g. full time /part time and level of study). I base my measure of educational attainment on the level of education at the time of the study. I collapse educational attainment into four mutually exclusive categories: 1=less than a high school diploma; 2=high school degree; 3=some college experience;

4=bachelor’s degree or more. In addition to detailed educational histories, the NSFH provides information on the respondents’ current enrollment status in post-secondary institutions. I create a dichotomous measure to capture this: 1=currently enrolled; 0=not enrolled.

Race: I code race into three categories: 1=Black; 2=Hispanic; 3=White/Other.

Age at first sex: I measure age at first sex as a continuous variable. Focal children were asked the age at which they had their first sexual experience. To keep the virgins in the sample, I set their mean age at sex equal to their current age plus one.

Cohabitation Experiences: I control for whether the respondent had ever cohabited and code the categories as follows: 1= cohabited; 0=never entered a cohabiting union. If the respondent is currently cohabiting, I control for whether they have entered into more than one cohabiting union.

Methods

I employ multinomial logistic regression models in this chapter. I first use a multinomial logistic regression model to estimate the likelihood of a woman first entering into a cohabitation, a marriage, or into no co-residential union. Multinomial logistic regression is also applied to estimate the risk of a first conception in various relational circumstances. I differentiate between single non-cohabiting, cohabiting, and marital conceptions. Childless women are the reference group. I initially test bivariate models to assess how a parental marital disruption influences young women’s union formation and childbearing behaviors. I then add a more comprehensive measure of childhood family structure to determine whether these behaviors vary by certain family types. Finally, I present the full model including the mediating factors of childhood and adult background characteristics in the model.

Descriptive Findings

Table 5.1 displays the unweighted descriptive characteristics of the focal female children of the NSFH Wave 1 respondents. The results demonstrate the changing and diverse family patterns this most recent generation of young adults experienced during childhood. Over 40 percent of women experienced the disruption of their parents’ marriages during childhood.

Moreover, the increasingly diverse family patterns that began to emerge in the late 1970’s and early 1980’s are also apparent. Among those parents who dissolved their marriages, a substantial

95 share did not remain single. Many had remarried by the second wave, and the majority of these remarriages were preceded by cohabitation. In fact, a considerable proportion of women whose parents divorced witnessed parental cohabitation during childhood, either in lieu of or prior to a parental remarriage. Figure 5.1 display these results.

[Table 5.1 and Figure 5.1 about here]

Nearly 45 percent of the focal women’s mothers had high school diplomas, while about

25 percent had some college education. Approximately 15 percent of the focal women received some form of public assistance while growing up. The vast majority of the women’s religious affiliation during childhood was Protestant. In addition, the average age of first sex was 17.5 years.

In respect to the focal women’s own characteristics, Table 5.1 reveals that the women were approximately 25 years of age at the time of data collection and Non-Hispanic white. The descriptive results also exhibit the shift in women’s educational attainment over the last two decades. Less than one third of the focal women had only a high school education, compared to nearly 45 percent from their mother’s generation. In addition, over 40 percent of the focal women had at least some college experience, compared to less than 25 percent of the focal women’s mothers.

In regards to the focal women’s own marital experiences, I find that over 40 percent had entered into a first marriage. In addition, the majority of these first marriages, 60 percent, were preceded by cohabitation. Approximately 25 percent of first marriages ended in a divorce. Table

5.1 also shows the increasing presence of cohabitation in the union formation behaviors among today’s young adults. Nearly one half of all focal women have entered into a cohabiting union.

Figure 5.2 displays the growing of role of cohabitation in first union formation experiences.

[Figure 5.2 about here]

The rise in non-marital childbearing within the last two decades is illustrated in Table 5.1.

Twenty-five percent of all women in this sample experienced their first conception outside of marriage, with one quarter of those occurring in a cohabiting relationship. As shown in Figure

5.3, this is a 10 percent increase when compared to their parents’ generation. Approximately 15 percent of focal women’s mother’s conceived their first child outside of a marital union.

[Figure 5.3about here]

In order to illustrate the importance of childhood family structure on adult offspring’s own union and family formation behaviors, I produce simple cross-tabulation tables in this next section. Table 5.2 confirms a significant relationship between parents’ relationship behaviors and adult focal women’s first union formation type. The chi-square test was significant at an alpha level of p<.001. Among women whose parents remained in a stable marriage, nearly 40 percent have yet to enter into any co-residential union, while approximately 22 percent directly marry.

However, among the focal women whose parents cohabited and subsequently remarried, nearly

70 percent first enter into a cohabiting situation, and approximately 16 percent enter into a marriage directly. Similarly, those whose parents cohabited and remained single, over 70 percent first enter into a cohabitation, and only 7 percent directly marry. For women whose parents remained single following their divorce, only about 14 percent directly marry, while similar proportions (approximately 30 %) first cohabit or have yet to enter into a co-residential union.

Finally, among those whose parents remarried directly, between 23 to 26 percent of women directly marry, cohabit, or have yet to enter into a co-residential union. These findings suggest an intergenerational pattern between childhood family structure and adult relationship formation behaviors does exist.

Table 5.3 displays the cross-tabulation of maternal childbearing behaviors and adult focal women’s own childbearing behaviors. The chi-square test was significant at an alpha level of

97 p<.05. Approximately one-fourth of the focal women who conceived their first children outside of marriage (single or within a cohabiting union) had a mother who also conceived her first child outside of marriage. Thus, a 70 percent majority of women who conceived their first children outside of marriage (single or within a cohabiting union) had mothers who conceived their own first children within marriage.

These cross-tabulations provide a straightforward way to get a first impression of the relationship between childhood family background and adult offspring’s relationship and childbearing behaviors. The above tables suggest a strong link between parents’ decisions surrounding union formation and childbearing behaviors and children’s own family formation outcomes. Given the above findings, I now examine the extent to which these various childhood family backgrounds are associated with young women’s current relationship and family process decisions.

Parental Union Transitions and Adult Women’s Union and Childbearing Behaviors

In this section I examine the link between childhood family structure and young women’s first union formation patterns. Applying a multinomial logistic regression, I predict the likelihood of first entering into a cohabiting union or directly into a marital union relative to having never entered a co-residential union (cohabiting or marital). I first enter only the coefficient of parental divorce in Model 1. Next, in Model 2, I distinguish between the types of parental union transitions entered following a divorce. Finally, in Model 3, I enter childhood and adult background characteristics.

First Union Formation

The bivariate results presented in Model 1 of Table 5.4 illustrate how the experience of a parental disruption is associated with greater odds of first entering a cohabiting union rather than remaining single. Women who experience the disruption of their parents’ marriage during their

98 childhood are nearly three times as likely to first form a cohabiting union compared to those whose parents remained married. Interestingly, and in contrast to my expectations, the experience of a parental marital disruption is not significantly associated with the odds of having never entered into a co-residential union.

In order to assess if the type of parental union entered post-divorce influences offspring’s union formation patterns differently, I differentiate between the types of parental union transitions experienced. Indeed, the results of Model 2 underscore the need to include a more diverse measure of parental union histories, particularly parental cohabitation histories, to better understand children’s own relationship behaviors. Although the experience of a parental divorce is associated with higher odds of cohabitation relative to never entering into a co-residential union, the expanded measure of parental union transitions demonstrates not only a divergent effect on offspring’s union formation behaviors, but also a stronger effect than the simple summary measure of parental divorce. As expected, women whose parents cohabited following their divorce, whether in place of marriage or prior to a marital union, are approximately four to five times more likely to first enter a cohabiting relationship than women whose parents remained married. Among the women whose parents divorced but did not subsequently cohabit, the odds of first entering into a cohabiting situation are substantially smaller. In fact, among those whose parents remarried without prior cohabitation, as well as those who remained single following a divorce, the odds of first entering a cohabiting relationship are nearly half that of women whose parents engaged in post-divorce cohabitation. The likelihood of first cohabiting relative to never entering into a co-residential union is substantially smaller if a parent did not cohabit following a divorce. In contrast to my expectations, I do not find an especially strong association with the chance of cohabiting if a parent cohabited in lieu of marriage.

In regards to entering directly into a marriage, the results in Model 2 indicate that a parental remarriage (even if preceded by cohabitation) increases the risk that a woman will enter directly into marriage instead of staying single. However, these findings are on the verge of significance. Moreover, there appears to be little difference in the likelihood of a marriage if the parental remarriage was preceded by cohabitation.

Lastly, in Model 3 I enter childhood and background characteristics that may mediate the effects of childhood family structure on the likelihood of first union formation. As hypothesized, even after family and adult background characteristics are accounted for, the effect of family structure on first union formation persists, but only for those who experienced parental cohabitation. In fact, among the women whose parents entered into a cohabiting union post- divorce, the likelihood of first entering into their own cohabitation is still three times higher than those whose parents remained married. Interestingly, the influence of childhood family type among those whose parents remained single or remarried directly has dissipated once the intervening variables are entered into the model. These findings suggest that the intergenerational transmission of relationship behaviors may be weakening when there was no exposure to a parental cohabitation post-divorce. However, there remains a strong intergenerational component of cohabitation. Additionally, and in contrast to my hypotheses, I find the relationship between childhood family structure and the likelihood of first entering into a marital union is no longer significant once the control variables are entered into the model.

Women who may come from lower socio-economic backgrounds are more likely to first cohabit. Those whose mothers have less than a high school education are nearly two times as likely to first form a cohabitation rather than remain single. Catholics are significantly more likely to cohabit and less likely to marry than to be in no relationship at all, compared to

Protestants. Those of other religious affiliations are more likely to have never entered into a

100 union rather than cohabit or marry compared to Protestants. Not surprisingly, those who delay their first sexual encounter till later ages are significantly less likely to cohabit relative to having never entered into a co-residential union. As expected, women are significantly more likely to enter into a cohabitation as well as marriage as they age, rather than never entering into a union.

Blacks are significantly more likely to remain single than Whites. In fact, Black women are approximately 70 percent less likely than White women to first enter a cohabitation and 90 percent less likely to marry relative to never having entered into a co-residential union. Lastly, women who have a college education as well as currently attending school are significantly more likely to be single than to cohabit or marry.

Non-marital Childbearing

In the next analysis I examine the influence of childhood family structure on young women’s family formation behaviors. Research addressing the influence of parental cohabitation on young women’s childbearing behaviors in recent years remains limited, if not absent. I apply a multinomial logistic regression to analyze this relationship. I first examine the baseline relationship between parental divorce and maternal childbearing behaviors on the likelihood of a first conception occurring in a single non-cohabiting state, cohabiting union, or within a marital union, as compared to childless women. Model 1 displays only the coefficients of a parental divorce and maternal childbearing behaviors. Unlike previous models, I include whether a focal woman’s mother conceived her first child outside of marriage in Model 1. In Model 2, I distinguish between the various types of parental union transitions following a divorce, and in

Model 3, I control for childhood and adult background characteristics.

Table 5.5 displays the results of the relationship context of focal women’s first conceptions. The bivariate estimates presented in Model 1 reveal that the experience of a parental divorce significantly increases the odds of a non-marital conception. Indeed, women whose

101 parents divorced during childhood are twice as likely to conceive while single and non-cohabiting as well as twice as likely to conceive within a cohabiting union relative to childless women. A parental divorce also elevates the odds of a marital conception as compared to no conception. As expected, if a woman’s mother conceived her first child outside of a marital union, she is also more likely to conceive her first child outside of a marital union. A maternal non-marital conception doubles the odds of a single non-cohabiting conception relative to no conception.

Maternal non-marital childbearing also increases the odds of a marital conception.

Model 2 displays the coefficients of the various parental union transitions and maternal childbearing. The influence of parental union transitions elevates the risk of conceiving outside of a marital union. Furthermore, witnessing a parental cohabitation is associated with an increased risk of conceiving while in a cohabiting union. However, these findings are on the verge of significance (p=.07). A maternal non-marital conception increases the odds of conceiving outside of marriage.

Interestingly, and in contrast to my hypotheses, I find the effects of parental marital and childbearing experiences dissipate after childhood and adult background characteristics are entered into Model 3. These findings suggest that mechanisms of childhood socio-economic background and women’s race and educational attainment may largely explain differences in the relationship context of a first conception, rather than family structure itself. Perhaps as divorce, remarriage, and cohabitation have become more commonplace, the intergenerational correlations between parent and offspring’s childbearing behaviors have weakened. On the other hand, the lack of significance may reflect my relatively small sample of women who conceived outside of marriage. For example, less than 100 single women and only 45 cohabiting women whose parents divorced experienced a non-marital conception. Thus, these findings may be due to a lack of statistical power.

102

Receipt of public assistance and religious affiliation are the only childhood social and economic background measures significantly associated with a non-marital conception. Women from economically disadvantaged backgrounds are more than twice as likely to conceive outside of any union and four times as likely to conceive within a cohabiting union. Catholics are more likely to conceive within a cohabiting union, and less likely to conceive in a marital union compared to Protestants.

In regards to focal women’s background characteristics, I find the odds of conceiving outside of any union increase with age. Additionally, and in support of the previous literature, women who delay age of first sexual encounter are significantly less likely to experience any conception. Black women are over 90 percent less likely to conceive their first child within a marital union than are White women. Hispanic women are twice as likely to conceive outside of any union compared with Whites. Women with higher levels of educational attainment are significantly less likely to conceive outside of marriage. Lastly, current school enrollment and ever having entered into a cohabiting union reduces the odds of conceiving.

Summary

The findings with regard to the hypotheses are as follows: (1) there remains a strong intergenerational link between parent and filial generations union formation behaviors however, only if a parental cohabitation was experienced during childhood; (2) As divorce and remarriage have become institutionalized in American society, and more commonplace, intergenerational correlations between parental and filial union formation behaviors may be weakening; (3) there is an intergenerational correlation between childhood family structure and non-marital childbearing, however, the relationship attenuates after mediating variables are entered into the model.

Discussion

103

The primary aim of this chapter was to assess how the relationship and childbearing behaviors of young women were shaped by their childhood experiences of parental divorce, remarriage and cohabitation. The results from my analyses illustrate an association between parental relationship behaviors and young adult’s union formation patterns, but only among certain childhood family experiences. That is, not all parental union experiences exert the same influence on young adults’ relationship and childbearing trajectories. I find the link between parental union behaviors and young women’s first union choice was only apparent among those who experienced a parental cohabitation during their formative years. Even after a number of intervening mechanisms were taken into account, a parental cohabitation elevated the chance of first forming a cohabiting union two to three times more than women whose parents remained married. For those whose parents remained single or directly remarried following their divorce, the likelihood of first cohabiting had dissipated after background variables were entered into the model.

The most persuasive explanation for these findings is most likely the socialization process that transpires within non-traditional family types. Young women who resided in a non- traditional home during childhood learned either first hand or through role modeling, relatively liberal attitudes regarding sexuality and non-marital relationships, and thus were socialized to possess similar attitudes regarding union and family formation behaviors. Divorce, single, remarried and cohabiting parents have introduced a model of relationship behavior that children are likely to replicate in their union formation decisions. The experience of a parental cohabitation may lead children to view marriage as an insignificant component of the relationship as well as family building process because their primary role model is unmarried themselves (Wu and Martinson 1993). The socialization process argues that the lives of parents and children are

104 inextricably linked and the behaviors and actions of parents have a direct impact on shaping the actions of their children’s own union and family formation behaviors via role modeling.

On the other hand, the increased risk of cohabitation could also be due to the economic hardships often encountered when growing up outside of a married two biological parent family.

Indeed, the poverty rate for children residing with a cohabiting parent is twice as high as that for children living with married biological parents (Acs and Nelson 2002). As demonstrated in

Chapter 4, as well as in previous studies (Bumpass and Lu 2000), cohabitation is most common among those with lower levels of education and financial resources. Furthermore, Model 3 of

Table 5.4 in this current chapter illustrated that the odds of first forming a cohabiting union were nearly two times higher for women whose mothers had less than a high school education. Given the social and financial prerequisites deemed necessary to marry, those who grew up with fewer economic resources may be more likely to first cohabit rather than marry directly.

Another plausible explanation for the intergenerational patterns between parental and filial generations found in this chapter could be the increasing commonality of divorce during childhood. As illustrated in Table 5.1, 40 percent of women in my sample experienced the disruption of their parents’ marriage during childhood, compared to only 16 percent who experienced a parental cohabitation post-divorce. As divorce has become more commonplace and less stigmatized, the negative outcomes often associated with a family disruption may have weakened. The lack of significance of parents’ remarrying directly and remaining single post- divorce in predicting union formation patterns suggests this could be the case. Parental cohabitation, however, was still primarily concentrated among a more highly selective at-risk population, thus demonstrating the persisting effect of a parental cohabitation.

I argue it is most likely a combination of all of the above explanations. The increasingly diverse family experiences of today’s young adults exposed individuals to relatively liberal

105 attitudes regarding sexuality and non-marital relationships, and therefore young adults are more likely to follow in their parents’ footsteps with their own union formation behaviors. Yet, cohabitation has consistently been concentrated among women at the lower end of the socio- economic spectrum. Thus, parental and filial economic resources most likely play some role in these findings. Lastly, the lack of significance of a parents’ remarrying directly or remaining single post-divorce in predicting union formation patterns suggests that the negative stigma once associated with a parental divorce has substantially weakened over time. In 1980, over half of all

U.S. marriages were expected to disrupt (Goldstein 1999), as divorce was becoming an increasingly normalized process of the American family system. Therefore, as divorce has become institutionalized in American society, and less selective of certain populations, the intergenerational component has likely abated.

Interestingly, women whose parents remained single and did not enter a co-residential union post-divorce (cohabitation or marriage) were still significantly more likely to cohabit than to stay single. This seems to contradict the literature that contends it is the instability associated with the frequency (not type) of parental transitions post-divorce that should be assessed when exploring young adults’ relationship behaviors. Having experienced no parental transitions following a parental divorce was associated with slightly higher odds of cohabiting than was having experienced a parental remarriage without prior cohabitation (one transition). This finding suggest that the event of marriage (even a remarriage) models behaviors and attitudes that socialize children to have more a positive view and desire for their own marriage (Axinn and

Thornton 1996).

In sum, the life course principal of linked lives was evident in this chapter. Childhood family experiences have clearly contributed to the growing complexity and diversity of the life course pathway’s among today’s young adults. The union and childbearing decisions made in the

106 parental generation had a direct impact on the adult children’s own relationship behaviors, although I found minimal support for childhood family background on adult childbearing behaviors once the intervening mechanisms were entered into the model. This chapter underscores the necessity to differentiate between the various types of childhood family structures. A more detailed measure of parental union transitions revealed that the type of union a parent enters post-divorce does not exert the same influence on adult children’s relationship and family behaviors. While the influence of a parental remarriage without cohabitation, as well as remaining single was minimal, a parental cohabitation, either prior to or in substitution of a remarriage, was highly predictive of first union choice.

These findings have important implications for researchers and policy makers alike investigating the impact of childhood family experiences on young adults’ first union formation and childbearing behaviors. When parental cohabitation is ignored or combined with another non-traditional family type, a substantial part of the picture is omitted. In fact, my analyses demonstrated that parental cohabitation plays an integral role in the union formation and childbearing decisions of today’s young adults. Further, the results from this chapter suggest that the increasing normalization of divorce has perhaps attenuated some of the outcomes once associated with a parental disruption. This finding extends prior work that contends the intergenerational influence of divorce has not abated and its influence still persists in the union formation patterns of young adults. Future work should assess the degree to which parental cohabitation and not divorce itself plays in the intergenerational correlation of parental and filial relationship outcomes. The findings from this chapter demonstrate the necessity to illuminate the growing diversity in childhood family structures in order to understand the changing life course pathways of today’s young adults.

107

Table 5.1: Childhood family structure and background characteristics of focal women Variable Description Mean Std. Dev. Parental marital experience Stable Marriage Still married to parnter by Wave 2 0.57 0.50 Divorced Focal Child's parents divorced by Wave 2 0.43 0.50 Parental union transitions following divorce Remarried no cohabitation Entered into new marriage by Wave 2 0.07 0.27 Cohabitation to remarriage Cohabited and then married by Wave 2 0.10 0.30 Cohabitation remained single Cohabited but did not remarry by Wave 2 0.06 0.24 Remained single Parent did not remarry nor cohabit by Wave 2 0.20 0.40 Maternal non-marital childbearing Conceived outside of marriage 0.15 0.35 Childhood family characteristics

Maternal Education Less than high school 0.12 0.33 High school degree 0.44 0.50 Some college experience 0.24 0.43 College 0.20 0.40

Receipt of public assistance 0.15 0.34 Childhood religious affliation Catholic 0.25 0.43 Protestant 0.67 0.47 Other 0.08 0.27 Focal childs age of first sex 17.55 2.98

Focal child attributes Age 25.47 4.34 Race Black 0.12 0.33 Hispanic 0.05 0.23 Non-Hispanic White 0.82 0.38

Educational attainment Less than high school 0.05 0.21 High school degree 0.29 0.46 Some college experience 0.43 0.49 College or more 0.23 0.42 Currently enrolled in school 0.24 0.43

First Union Formation Entered directly into marriage 0.19 0.50 Cohabited prior to marriage 0.24 0.50 Cohabited Never Married 0.25 0.43 Never Cohabited Never married 0.32 0.49 Marital Disruption First married ended in divorce 0.11 0.50

Childbearing Behaviors Has child 0.41 0.49 single non cohabitating conception 0.19 0.45 cohabitating conception 0.06 0.43 Marriage conception 0.16 0.44

N 108 933

Table 5.2: Focal women's first union formation type by childhood family structure

First union formation type Direct Cohabitation Cohabit No co- Total N marriage prior to first never residential marriage married union Parental union status during childhood, %

Remarried no cohabitation 26.0 26.0 23.3 24.7 100 73 Cohabitation to remarriage 16.5 34.1 35.2 14.3 100 91 Cohabitation remained single 7.0 40.4 33.3 19.3 100 57 Remained single 13.7 30.6 29.5 26.2 100 183 Stably married 21.9 18.2 20.4 39.5 100 529 Note: Sample total N=933 χ2 =68.29, df=12; p <= .001

Table 5.3:Relationship context of focal women's first conception by maternal first conception

Relationship context of focal women's first No Single Cohabiting Marital conception conception conception conception conception

Maternal first conception, % Did not conceive outside of marriage 87.7 79.0 85.0 81.3 Conceived first child outside of marriage 12.3 21.0 15.0 18.8 Total 100 100 100 100 N 543 186 60 144 Note: Sample total N=933 χ2 =8.71, df=3; p <= .05

109

Table 5.4: Odds ratios for focal women's first union formation Model 1 Model 2 Model 3

Cohabit Married Cohabit Married Cohabit Married

Versus no co-residential union Versus no co-residential union Versus no co-residential union

Parental Divorce 2.85 *** 1.26 Parental Union Transitions

Remarried no cohabitation 2.05 * 1.90 + 1.27 1.03

Cohabitation to remarriage 4.96 *** 2.08 + 3.27 *** 1.43 Cohabitation remained single 3.91 *** 0.66 2.94 ** 0.48 Remained single 2.35 *** 0.94 1.61 + 0.58 Remaind stably married ------Maternal non-marital childbearing

Conceived outside of marriage 1.21 0.83 Childhood family characteristics Maternal Education Less than high school 1.95 * 1.50 High school degree ------Some college experience 1.46 1.15 College 0.76 0.54 +

Receipt of public assistance 1.61 1.35

Childhood religious affliation Catholic 1.01 * 0.32 *** Other 0.89 * 0.35 * Protestant ------

Age of first sex 0.73 *** 0.79 ***

Focal woman attributes

Age 1.33 *** 1.47 *** Race Black 0.29 * 0.12 *** Hispanic 0.75 2.24 White ------Educational attainment Less than a high school degree 1.25 1.10 High School Degree ------Some College 0.93 0.86 College + 0.56 + 0.61

Currently enrolled 0.39 *** 0.34 *** LR chi2 52.69 69.51 518.30 N 933 933 933 *** p<= .001, ** p. <=.01, * p.<=.05, + p.<=.10

110

Table 5.5: Relative risk ratios of focal women experiencing a conception Model 1 Model 2 Model 3 Single Cohabitating Married Single Cohabitating Married Single Cohabitating Married versus remaining childless versus remaining childless versus remaining childless Divorced 2.09 *** 1.91 * 1.43 + Maternal nonmarital conception 2.32 *** 1.01 1.87 * Parental Union Transitions Remarried no cohabitation 2.35 ** 1.80 1.87 + 1.40 0.97 1.31 Cohabitation to remarriage 1.73 + 2.15 + 1.68 1.32 0.92 2.15 + Cohabitation remained single 1.64 2.42 + 1.00 0.78 0.59 0.78 Remained single 2.33 *** 1.65 1.30 1.07 0.57 0.89

111 Remaind stably married ------Maternal non-marital childbearing

Conceived outside of marriage 2.33 *** 1.06 1.90 * 1.47 0.78 1.54 Childhood family characteristics Maternal Education Less than high school 1.77 + 1.54 1.38 High school degree Some college experience 0.94 0.60 1.00 College 0.67 1.45 0.77

Receipt of public assistance 2.78 *** 4.84 *** 1.80 Childhood religious affliation Catholic 0.72 2.39 * 0.57 + Other 1.34 0.53 0.57 Protestant --- Age of first sex 0.83 *** 0.73 *** 0.82 *** LR chi2 28.14 32.87 562.42 N 933 933.00 933 *** p<= .001, ** p. <=.01, * p.<=.05, + p.<=.10

111

Table 5.5: continued Model 1 Model 2 Model 3 Single Cohabitating Married Single Cohabitating Married Single Cohabitating Married versus remaining childless versus remaining childless versus remaining childless Focal woman attributes Age 1.35 *** 1.29 *** 1.57 ***

1 Race 12 Black 1.69 1.31 0.04 **

Hispanic 2.23 + 1.27 2.41 White --- Educational attainment Less than a high school degree 2.22 + 2.38 0.77 High School Degree Some College 0.54 * 0.27 *** 0.70 College + 0.07 *** 0.03 *** 0.22 ***

Currently enrolled 0.34 *** 0.20 * 0.28 ** Ever cohabited 0.27 *** 1.39 0.16 ***

LR chi2 28.14 32.87 562.42 N 933 933.00 933 *** p<= .001, ** p. <=.01, * p.<=.05, + p.<=.10

112

0.25

0.20

0.15

Remarried no cohabitation

0.10 Cohabitation to remarriage

Cohabitation remained single Proportion

Remained Single (no cohabitation and/or marriage) 0.05

0.00

Figure 5.1: Parental union transitions following a divorce

0.60

0.50

0.40

0.30 Proportion

0.20

0.10

0.00 Entered directly into marriage Entered into cohabitation Never cohabited never married Figure 5.2: Focal women's first union formation experience

113

0.30

0.25

0.20

0.15 Maternal Focal Woman

Proportion 0.10

0.05

0.00 Nonmarital Conception Figure 5.3: Maternal and focal women's nonmarital childbearing behaviors

114

CHAPTER 6

CONCLUSION

Introduction

American family structures are experiencing rapid change. The declining significance of marriage and the increasing prevalence of cohabitation in recent decades, both features of even broader economic and social shifts, have reshaped the life course pathways of today’s young adults. Through most of the 20th Century, across all socio-economic groups, the normative life course pathway in the U.S. prescribed marriage followed by parenthood. However, as young adults delay or forgo marriage, cohabitation and non-marital childbearing have become more prevalent in their life course pathways. Although marriage is still viewed as the ideal entrée to parenthood among most young adults, the real practice of a substantial share of the population is to form families outside of marital unions. Nearly 40 percent of all US births now occur outside of marriage. Half of these births are to cohabiting couples, suggesting that 80 percent of

American children are still born into two-parent households, (Mincieli al., 2007). The percentage of those born to single parents or cohabiting parents who are not married raises interesting questions about the present state and future trajectory of US families. Although ample research has addressed the sweeping family changes of the last few decades, the subtleties of cohabitation’s place in U.S. family systems and cycles have been under-explored.

I contribute to this research using the life course perspective with the key principles of historical contexts, situating individuals within cohorts, and linking lives within and across generations to address this gap in the literature. By assessing family formation behaviors over recent decades, I explore the mechanisms that may have contributed to the changing relationship

115 and family trajectories of today’s young adults. Specifically, I examine the relationship contexts of never married women’s first conceptions, and their subsequent union transitions following a first full-term pregnancy. I then explore the correlations between the growing diversity in childhood family experiences and these changing adulthood behaviors.

The rapid social changes that increased divorce, marital delay, cohabitation, and non- marital childbearing during the latter half of the 20th Century have influenced the life course trajectories of young adults. Though school, work, and family are still significant markers of early adulthood, the timing, ordering, and duration of these events have substantially changed

(Modell 1989; Rindfuss 1991; Shanahan 2000). The increasing prevalence and social acceptability of cohabitation has played a greater role in the family formation behaviors of women in recent decades. Young adults are becoming increasingly receptive to diversity in their attitudes toward marriage, sexuality, and childbearing (Thornton and Young- Demarco 1999); the research presented here demonstrates how these attitudes and decisions are strongly influenced by the values and behaviors they learned in childhood. Even though the impact of rapid social change on individual behavior is undeniable, this research demonstrates that young women’s relationship and childbearing trajectories can still be strongly anchored within local family contexts.

I use data from the National Survey of Families and Households to analyze these changing behaviors and their connections to family histories. The data include detailed marital and fertility histories of women, an oversample of cohabiting couples and single parent families, and multi-generational data. This data set enables me to explore changing relationship and childbearing behaviors across two cohorts of young women. Furthermore, it permits me the ability to examine the intergenerational linkages between parental union transitions during the

116 child-raising years, and the subsequent family formation behaviors of the young women offspring in the next generation

I asked three primary research questions through this project. First, how has the fertility behavior of never married single and cohabiting women changed across cohorts? Second, how has single and cohabiting women’s relational response to pregnancy changed across cohorts?

Third, how have the diverse parental union transitions young women experienced or observed during childhood shaped their own first union formation and childbearing behaviors?

How has the fertility behavior of never married women changed across cohorts?

Rapid family changes over the latter half of the twentieth century reflect profound changes in the meanings and functions of marriage in the life course trajectories of young adults.

Economic and cultural changes over the last few decades have shifted the role of marriage within the life course of today’s young adults (Bianchi & Spain 1996; Casper & Bianchi; Cherlin 2004;

Lesthaeghe 1993; Popenoe 1993; Thornton 1989). Following World War II, in the U.S., young adults experienced a somewhat standardized life course pathway. The entry into marriage was nearly a compulsory event that anchored the life course of young adults, and it provided the foundation for family building behaviors. However, expanding labor and educational opportunities for women, decline in men’s wages, advancement in birth control coupled with changing attitudes and norms (Cherlin 2005; Thornton and Young-Demarco 2001), have resulted in new patterns of delayed marriage, increased cohabitation, and non-marital childbearing that have challenged this traditional pathway.

While these changing relationship and family trajectories have been experienced across all social groups, race and educational status mitigate these larger patterns. A broader growing inequality between the wealthy and the rest of the population in the US, and its reflection in marital statistics, has led scholars to argue that marriage is increasingly reflective of as well as the

117 basis for social stratification, with the economic disparities between unmarried and married couples widening. ―So vast is the difference, one is tempted to replace the traditional notion of social class with the more descriptive term marriage class (Nock 2006 p.324).‖ Currently, whether one enters into a marital union is largely dictated by their social class background.

The findings from Chapter 4 confirm just this. Exploring the impact of historical change on cohorts, I first assess the changing union context of first conceptions across two cohorts of never married single and cohabiting women. Life course scholars contend that individuals born in different years encounter different worlds with a different set of circumstances (Elder 1998;

2003). Young women who entered into adulthood during the rapid demographic changes of the last few decades may experience a different pathway to union and family formation than individuals who entered into adulthood prior to these substantial changes, or who were part of the first wave of adults to participate in this shift. The two cohorts examined in this chapter include women who entered into adulthood at very different points in time. Women in the early cohort transitioned into adulthood during the 1970’s and early 1980’s, when the economic and cultural shifts that swept across the latter half of the 20th Century were only beginning to emerge, while women in the later cohort entered into adulthood during the late 1990’s and early 2000’s when these changes were well established.

In support of my hypothesis, I find that women in the later cohort were significantly more likely to conceive their first child outside of any union compared to women in the earlier cohort.

Additional analyses indicated that the likelihood of conceiving outside of a union has changed among those without a high school education, with these women being over-represented among those who conceive outside of marriage. My findings illustrate a persisting and growing disparity in parental economic resources (McLanahan 2004). In contrast to my expectations I found that the overall propensity to conceive within a cohabitation has not increased between the two

118 cohorts of women. I suspect this may be due to compositional differences between the two cohorts.

How has single and cohabiting women’s response to an impending birth changed across cohorts?

In addition to the increasing propensity to conceive outside of a union, I find changes have also taken place in women’s responses to impending parenthood. Examining never married pregnant women’s responses to an impending birth extends the current literature on non-marital fertility and family formation. First, I show a substantial portion of non-marital childbearing is over looked when only births are examined. Women may chose to transition into or remain within a cohabiting union, to marry, or to dissolve their relationship by the time the child is born. As shown in Chapter 4, a considerable number of women in this sample transitioned into marriages within three months of childbirth. Over 30 percent of single women and nearly half of the cohabiting women in the earlier cohort married within three months of childbirth. Similarly, approximately 14 percent of single women and 40 percent of cohabiting women in the later cohort also married within three months of birth. Additionally, the number of parents who began living together in response to a pregnancy rather than as a result of a non-pregnancy-related relationship decision more than tripled ( 5 % to 17%). Thus, examining pregnant women’s relationship behavior following a non-marital conception provides a more in-depth picture of non-marital childbearing.

Secondly, and in support of my hypotheses, as well as the life course perspective, my analyses in Chapter 4 indicate that the growing disconnection of marriage from childbearing is also apparent among women in the most recent cohort. Women in the recent cohort are more likely to cohabit in response to a non-marital pregnancy rather than marry or stay single. This extends prior research that finds single women were nearly as likely to cohabit as to marry in

119 response to a non-marital pregnancy (Raley 2001). I find within a decade’s time, single women are now more likely to cohabit rather than stay single or marry in response to an impending birth.

The ―shotgun cohabitation‖ has largely replaced the ―shotgun marriage‖ of years past. This changing response most likely reflects the substantial changes in the meaning and function of marriage in the life course of young adults. While marriage is still highly valued, and the majority of young adults express a desire to eventually marry, the social and financial prerequisites deemed necessary for marriage impede some women from entering into a marital union.

However, these same barriers do not apply to childbearing. My findings suggest that cohabitation is becoming a substitution for marriage among certain groups of women.

Consistent with a pattern of economic stratification in marriage rates, the data analyzed for this dissertation indicate that single women’s family formation decisions significantly differ by race and educational attainment. Although college educated White women are the least likely to conceive outside of marriage, when they do, they are the most likely to marry in response to the pregnancy. African American women are the most likely to remain single and non-cohabiting within three months of childbirth, even when compared to Hispanic women.

My findings suggest that cohabitation is beginning to play a more prominent role in the stratification of the American family. As marriage continues to increase among the more educated and affluent, less educated Whites find themselves increasingly excluded from this institution. Thus, cohabitation is becoming a more important option in their family formation choices. At the same time, it appears to be declining in the family formation choices among

Black women. Previously, White women were the least likely to conceive outside of a marital union, even within a cohabiting union (Manning 2001). My results suggest this may no longer be the case. Although White women are less likely than Black women to conceive outside of a union, they are 50 percent more likely than Black women to conceive while cohabiting.

120

Furthermore, my descriptive results indicate the proportion of births to Black women who are cohabiting have declined between these two cohorts of women. This supports the conclusions of

Wu and colleagues (2001) who find that cohabiting births have actually constituted a declining fraction of non-marital births among Black women. As cohabitation increasingly serves as a substitution for marriage among less- educated Whites, Black women, who have been precluded from the institution of marriage for some time, may also be finding themselves increasingly excluded from the realm of cohabitation, and are now being relegated to the bottom of the family spectrum. The same factors that have contributed to the decline in marriage among Black women, such as a shortage of desirable Black men, may also be contributing the decline in domestic partnerships as well. We know that in the last couple of decades the rate of employment for young Black males has actually declined (Holzer 2009; Holzer et al., 2005). Thus, an increasing number of black women are opting out of cohabitation and choosing intergenerational households in their family formation decisions (Oberlander 2007).

An additional factor that may be contributing to the continuing decline in marriage, is the substantial share of single mothers, particularly African American mothers, who receive welfare benefits (Moffitt and Gottschalk 2001). Some family scholars have posited that the American welfare state has contributed significantly to the declining marriage rates among the poor. Due to the significant marriage ―tax‖ implemented on single women who chose to marry the child’s father, many women may choose to remain unmarried in fear of the potential loss of their welfare benefits. However, in a study conducted by Moffitt (2002), he finds that although there is a negative association with welfare benefits and the likelihood of marriage, the effect is small and does not explain the lower marriage rates among the poor.

In sum, the life course pathways of today’s young adults follow patterns differentially associated with race and educational attainment. While college educated White women have

121 deviated little from the 20th Century traditional pathway of marriage followed by childbearing, less educated Whites are increasingly forming families within cohabiting unions. However,

Black women are increasingly forming families outside of any co-residential arrangement with an adult male. The married/unmarried dichotomy that is often applied to describe the growing disparities in American families may no longer be adequate. Findings from this research suggest that there may be an emerging stratification system even within the context of cohabiting, and that families are increasingly segmented into married, cohabiting, and the single no co-residential union.

How have the diverse parental union transitions experienced during childhood shaped adult offspring’s own first union formation and childbearing behaviors?

The final goal of this dissertation is to explore the intergenerational patterns of union formation and childbearing. A considerable amount of scholarly attention has been devoted to the childhood and young adult outcomes of family instability. Concerned over the consequences of parental divorce, a large body of research has emerged documenting the intergenerational correlations between parental marital behaviors and childhood and young adult outcomes (Amato

2000; Aquilino 1991; Furstenberg Brooks-Gunn and Morgan, 1987; Goldscheider &

Goldscheider 1998 McLanahan and Sandefur 1994). Scholars have argued that the relationship and childbearing trajectories of young adults are strongly influenced by their parents’ own life course pathways. Parents’ own decisions surrounding relationship and family formation influence young adults’ subsequent decisions regarding entry into marriage, cohabitation and parenthood (Teachman 2002; 2003; Thornton 1991; Wolfinger 2003; Wu and Martinson 1993;

Wu 1996).

Current research focused on young adults’ union and family formation behaviors often neglects this growing diversity in childhood family experiences. Family instability is often

122 assessed with a simple dichotomous measure that ignores the complexity of parental relationship histories, particularly the growing role of cohabitation in parents’ post- divorce union relationships. In Chapter 5 of this dissertation, I apply the life course principle of linked lives, and explore how this growing diversity in childhood family experiences influences young adults’ own union formation and childbearing decisions. My results demonstrate the necessity to utilize a more nuanced measure of childhood family structure when assessing young adult outcomes. I provide a more complete picture of how parental union histories during childhood impact offspring’s own relationship and childbearing behaviors in adulthood. Studies that fail to differentiate between the increasingly diverse childhood family structures experienced by today’s young adults substantially limit their ability to assess the intergenerational transmission of relationship and family formation behaviors. When parental cohabitation is ignored or combined with another family type, a considerable part of the story is overlooked.

Indeed, the findings from this study indicate children’s later life trajectories are inextricably linked to the behaviors and actions of their parents. In support of the life course perspective, results from this study demonstrate that the first union formation patterns of young adults are significantly influenced by their parents’ relationship histories. However, not all childhood family experiences exert the same influence on young adults’ relationship trajectories.

The experience of parental cohabitation during a person’s childhood increases the person’s likelihood of first forming a cohabiting union by two to three times, even after a number of intervening mechanisms are entered into the model. Yet, for those whose parents remained single or directly remarried following their divorce, the influence on women’s first union formation attenuates, once background characteristics are taken into account. Consistent with the principle of linked lives, family structure in the parental generation affects the family formation decisions of the next generation.

123

In order to provide insight into the possible mechanisms behind the long-term effects of childhood family structure, two mid-level explanations --- family socialization and instability and change --- are presented in Chapter 5. The family socialization argument contends that children who grew up outside a traditional two parent married home have been socialized to have more positive attitudes toward non-marital sex, childbearing and cohabitation (Axinn & Thornton,

1996; McLanahan & Sandefur, 1994). In contrast, the instability and change perspective contends that the number of transition and changes in family structure and the instability and stress associated with these changes generate more positive attitudes toward non-marital sex, cohabitation and childbearing and more negative attitudes towards marriage (Axinn and Thornton

1996). My analysis supports the socialization argument. Emphasizing the attitudes, values, and parental role modeling learned during childhood, socialization theories contend that children will emulate their parents’ behaviors in their own adult relationships. Young women who resided in a cohabiting home during childhood learned either first hand or through role modeling attitudes regarding sexuality and non-marital relationships that are positively disposed toward cohabitation, and thus were socialized to possess similar attitudes regarding union and family formation behaviors. A parental cohabitation may lead children to view marriage as an insignificant component of the relationship or family building process because their primary role model is unmarried themselves (Wu & Martinson 1993).

In essence, young adults’ lives are strongly anchored within the family context.

Although marital delay, divorce, cohabitation, and non-marital childbearing are increasingly prevalent in the lives of young adults, their decisions and choices regarding family formation are still strongly linked to their experiences during childhood.

Implications and directions for future research

124

Why should the growing racial and educational disparities in family formation and childbearing behaviors elicit concern? These changes in union and childbearing behaviors have implications for the emotional and economic well-being for children as well as adults. In Chapter

4, I found that women with the fewest years of education in the recent cohort are twice more likely to conceive a child outside of a union than similarly educated women in the earlier cohort.

Additionally, I found African American women are significantly more likely to conceive outside of a partnership, and less likely to marry or even cohabit in response to this pregnancy, than both

White and Hispanic women. These findings are disconcerting because children living in single non-cohabiting mother families are at the greatest risk for poverty. In fact, the poverty rate for single mother families is double that for cohabiting couples, and nearly six times the poverty rate for children living with married couples (Acs & Nelson 2002). The problem is particularly serious for African American children. Regardless of race, almost 60 percent of children under the age of 6 living in single mother households are in poverty (White & Rogers 2000), and this persistently high rate of childhood poverty has been attributed reciprocally to the rapid growth in non-marital childbearing (Hoynes et al. 2006).

Second, Chapter 4 demonstrates that cohabitation is becoming more prominent among less educated Whites who have children. Yet, a recent study revealed that children born to cohabiting parents have five times the risk of experiencing a parental disruption than do children of married parents, and this risk is greatest for White children. Furthermore, we know that the rates of union instability are greater among those with lower levels of education (Lichter et al.

2006). Many of these cohabiting families only became ―a family‖ in response to an unexpected pregnancy. Couples who may have been in a casual relationship with no intentions of living together may have been pushed into a cohabiting relationship. Thus, this increasing number of already unstable unions could have detrimental effects on children’s economic and emotional

125 well-being. Moreover, the current initiatives to promote marriage, particularly among cohabiting parents (Acs & Nelson, 2004), could further exacerbate these problems. Encouraging these cohabiting families to marry could likely increase the rate of divorce that is already higher among those at bottom of socio-economic ladder. Furthermore, pushing a marriage that may very well end in a divorce would only worsen the financial hardships of these women and children. Future research should examine the stability of cohabiting unions formed in response to an unexpected pregnancy versus the impact of childbirth on unions already formed on another basis, such as couple compatibility.

The growing economic inequalities within the American family also have important implications for society at large. As marriage becomes increasingly limited to higher educated

Whites, and individuals continue to marry those most similar in terms of race, education, and income, the most advantaged continue to gain. While the less-educated, economically disadvantaged, and racially marginalized continue to lose ground. This increasing stratification of the American family ―systematically undermines attempts to create equal opportunities across lines of class and race (Max 2007 p.582)‖.

Finally, Chapter 5 affirms the linked lives life course theory argument that despite rapid economic and cultural changes, including expanding educational and labor force opportunities for women, children continue to follow in their parents footsteps in their relationship and family formation choices. However, as previously mentioned, not all parental relationship structures exert the same influence on young women’s family formation behaviors. Only those who experienced a parental cohabitation during childhood had an elevated risk of cohabiting themselves. This project is one of the few studies that incorporate a more comprehensive measure of childhood family structure in order to assess both first union formation and childbearing behaviors among recent cohorts of young adults. Future studies should continue to explore

126 whether parental cohabitation, as opposed to divorce per se, significantly affects children’s and young adults’ life course trajectories.

In sum, the life course perspective provides family scholars with a theoretical framework that can systematically study family change. This perspective highlights not only the significance of immediate family environments on individuals, but also the historical contexts within which families are situated. The findings from this study support just that. While the rapid economic and cultural changes over the last few decades have dramatically altered the life course pathways of young adults, childhood family experiences still have a persistent and long-term effect well into adulthood. My findings indicate that researchers should apply life course perspectives to studies of the changing family formation behaviors of young adults and the influence childhood family structure may have on these changing family patterns.

Limitations

While the NSFH provides the best available data for conducting the analyses for this research, it has limitations. For example, one limitation of the NSFH data is the design of the

Wave 3 sample selection process. For cases with a completed parental interview at time 3, an attempt was made to interview all focal children who were age 18 and older at time 3, regardless of whether an interview was completed at time 2. Therefore, a number of focal children at Wave

3 (N~200) did not have a Wave 2 interview and I am unable to employ any Wave 2 measures. I am also limited in my analyses by sample size. I have only a small number of women in the third wave of the NSFH who experienced a cohabiting conception that lead to a birth. The lack of significance in certain analyses assessing non-marital childbearing could be due to the lack of statistical power that can often result from such small samples.

In Chapter 4, the women in the later cohort are the focal children of the Wave 1 respondents who were parents at the time of their initial interview. Thus, these samples are

127 linked in that respondents at Wave 3 are the children of the main respondents from Wave 1, although they do not completely overlap. Therefore, conclusions drawn on the changing union and family formation behaviors of young adults in the later cohort may not be observed in other or future cohorts.

Additionally, I limit my analyses in this project to female respondents only. It is possible that union formation and childbearing decisions may differ for men and women. The mechanisms behind an individual’s decision to remain single, enter into a cohabitation, or transition into a martial union following a non-marital pregnancy may function differently for women than they do for men. However, I am limited in the inclusion of male partner characteristics by the NSFH study design. At Wave 3, there is not a measure that specifies if the current partner is the father of the specified child. In Chapter 5 I assess the impact of childhood family experiences on young adult’s own union and childbearing behaviors. However, some evidence suggests one’ childhood family structure may influence later life decisions of men and women differently (McLanahan &

Sandefur, 1994). Axinn and Thornton (1993) found that maternal attitudes regarding cohabitation had a stronger influence on daughters' union formation behaviors than for sons.

In addition, I do not fully address the within gender dynamics. That is, marriage, cohabitation, and parenthood provide individuals with cultural scripts that help guide their union and family formation behaviors. However, these scripts significantly vary by one’s social class background. While the vast majority of all women are entering into cohabitations, only lower educated women perceive it has an appropriate framework to bear and rear children. For college educated women, parenthood and cohabitation are two mutually exclusive realms that rarely intersect. Future work should continue to explore the gender dynamics in relationship and childbearing behaviors.

128

This study also cannot make strong causal arguments about the relationship between parental and adult offspring behaviors based on the analyses utilized in Chapter 5. I have not controlled for a number of sources of unobserved heterogeneity. Most important among these is the fact that, the focal women in Wave 3 are the children of the original Wave 1 respondents. In addition, the diversity of childhood family experiences is limited by my sample. For example, due to a small sample size I had to omit never married parent households. Future work should continue to examine this gender dynamic in relationship and childbearing behaviors.

Chapter 4 is also limited by the analytic methods utilized. In this dissertation I only employ logistic and multinomial logistic regression. However, event history analysis would most likely be better suited for addressing the hypotheses proposed in Chapter 4. Employing a competing risk event history model that assesses a young woman’s risk of conceiving outside of any union versus a cohabiting union would most likely provide a more comprehensive understanding of non-marital childbearing in the life course of young adults.

Finally, in Chapter 5 my study has also centered on the adult children born to a specific cohort of parents. Thus, I can not determine if the intergenerational patterns witnessed in this cohort of young adults would be observed in current or future cohorts. The results presented in this dissertation should be tempered by the limitations of my analyses.

Conclusion

This dissertation explores the changing union and family formation patterns that have transpired over the last couple of decades. Guided by the life course perspective as a framework for understanding how familial and historical contexts influence individuals’ life trajectories, I focus on the changing union context of non-marital childbearing, never married women’s changing responses in to an impending birth, and how one’s family of origin has helped shape the changing life course pathways of today’s young adults. My findings indicate that non-marital

129 childbearing is increasing among those with the fewest economic resources. The economic disparities between married and unmarried couples are growing. Furthermore, the role of cohabitation appears to be increasing in the family formation decisions of less educated White women, but declining in those of Black women. Lastly, I find that one’s childhood family structure still plays an integral role in young adults’ relationship and family trajectories.

However, this influence on relationship outcomes appears to be concentrated among those who were exposed to a parental cohabitation. Because the relationship and family trajectories of today’s young adults continue to evolve and life course pathways are growing more complex, the results of this dissertation suggest the need for additional research into the rapidly changing

American family.

130

REFERENCES

Aassve, Arnstein., Simon Burgess, Carol Propper, and Matt Dickson. 2006. "Employment, family union and childbearing decisions in Great Britain." Journal of the Royal Statistical Society Series a-Statistics in Society 169:781-804.

Aassve Arnstein , A., Francesco. C. Billari, and Raffaella Piccarreta. 2007. "Strings of adulthood: A sequence analysis of young British women's work-family trajectories." Pp. 369-388.

Akerlof, George. A., Janice L. Yellen, and Micheal. L. Katz. 1996. "An analysis of out-of- wedlock childbearing in the United States." Quarterly Journal of Economics 111:277- 317.

Allen, Vern L. Bengston and Katherine R. 1993. "―The Life Course Perspective Applied to Families over Time.‖ " Pp. 452-475 in Sourcebook of Family Theories and Methods: A Contextual Approach, edited by W. D. Boss, R. LaRossa, W. Schumm, and S. Steinmetz Boston, MA: : Allyn and Bacon.

Amara, Bachu. 1999. "Trends in Premarital Childbearing: 1930 to 1994." Current Population Reports U.S. Census Bureau, Washington, DC:P23-197.

Amato, Paul. R. 1996. "Explaining the intergenerational transmission of divorce." Journal of Marriage and the Family 58:628-640.

Amato, Paul R. 2000. "The consequences of divorce for adults and children." Journal of Marriage and the Family 62:1269-1287.

—. 2005. "The impact of family formation change on the cognitive, social, and emotional well- being of the next generation." Future of Children 15:75-96.

Amato, Paul R. and Allan. Booth. 1996. "A prospective study of divorce and parent-child relationships." Journal of Marriage and the Family 58:356-365.

Amato, Paul R. and Danella D. DeBoer. 2001. "The transmission of marital instability across generations: Relationship skills or commitment to marriage?" Journal of Marriage and the Family 63:1038-1051.

Andrew J. Cherlin, P. Lindsay Chase-Lansdale, and Christine McRae. 1998. "Effects of Parental Divorce on Mental Health throughout the Life Course " American Sociological Review 63:239-49.

131

Aquilino, William. S. 1994. "Impact of childhood family disruption on young adults relationships with parents." Journal of Marriage and the Family 56:295-313.

—. 1996. "The life course of children born to unmarried mothers: Childhood living arrangements and young adult outcomes." Journal of Marriage and the Family 58:293-310.

Avellar, Sarah. and Pamela. J. Smock. 2005. "The economic consequences of the dissolution of cohabiting unions." Journal of Marriage and the Family 67:315-327.

Baizan, Pau, Arnstein Aassve, and Francesco. C. Billari. 2004. "The interrelations between cohabitation, marriage and first birth in Germany and Sweden." Population and Environment 25:531-561.

Bernhardt, Eva. and Franics. Goldscheider. 2002. "Children and union formation in Sweden." European Sociological Review 18:289-299.

Bernhardt, Eva. M. and Francis. K. Goldscheider. 2001. "Men, resources, and family living: The determinants of union and parental status in the United States and Sweden." Journal of Marriage and the Family 63:793-803.

Berrington, Anne. 1996. "Dynamics in marriage and cohabitation: An inter-temporal, life course analysis of first union formation and dissolution - Manting,D." European Journal of Population-Revue Europeenne De Demographie 12:192-193.

—. 2001. "Entry into parenthood and the outcome of cohabiting partnerships in Britain." Journal of Marriage and the Family 63:80-96.

Berrington, Anne. and I. Diamond. 1999. "Marital dissolution among the 1958 British birth cohort: The role of cohabitation." Population Studies-a Journal of Demography 53:19- 38.

—. 2000. "Marriage or cohabitation: a competing risks analysis of first-partnership formation among the 1958 British birth cohort." Journal of the Royal Statistical Society Series a- Statistics in Society 163:127-151.

Bianchi, Suzanne M., and Lynne M. Casper. . 2000. "―American Families.‖" Population Bulletin Washington DC: Population Reference Bureau. 55.

Bianchi, Suzanne M. and Daphne Spain. . 1986. American Women in Transition. . New York: : Russell Sage.

Blossfeld, Hans-Peter . P. and Johannes. Huinink. 1991. "HUMAN-CAPITAL INVESTMENTS OR NORMS OF ROLE TRANSITION - HOW WOMENS SCHOOLING AND CAREER AFFECT THE PROCESS OF FAMILY FORMATION." American Journal of Sociology 97:143-168.

Brannen, Julia and Anne Nilsen. 2006. "From fatherhood to fathering: Transmission and change among British fathers in four-generation families." Sociology-the Journal of the British

132

Sociological Association 40:335-352.

Brien, Micheal J., Lee. A. Lillard, and Steve Stern. 2006. "Cohabitation, marriage, and divorce in a model of match quality." International Economic Review 47:451-494.

Brown, Susan L. 2000. "Fertility Following Marital Dissolution: The Role of Cohabitation." Journal of Family Issues 21:501-524.

Brown, Susan. L. 2004. "Family structure and child well-being: The significance of parental cohabitation." Journal of Marriage and the Family 66:351-367.

Buchmann, Claudia and Thomas A. DiPrete. 2006. "―The Growing Female Advantage in College Completion: The Role of Family Background and Academic Achievement.‖" American Sociological Review 71:515-41.

Bulanda, Ronald E. and Wendy. D. Manning. 2003. "Parental cohabitation experiences and adolescent behavioral outcomes." Pp. 593-618 in 115th Annual Meeting of the American- Economic-Association. Washington, Dc.

Bulanda, Ronald E. and Wendy D. Manning. 2008. "Parental cohabitation experiences and adolescent behavioral outcomes." Population Research and Policy Review 27:593-618.

Bulcroft, Richard A. and Kris A. Bulcroft. 1993. "Race differences in attitudinal and motivational factors in the decision to marry." Journal of Marriage and the Family 55:338-355.

Bumpass, Larry. and Hsien-hen Lu. 2000. "Trends in cohabitation and implications for children's family contexts in the United States." Population Studies-a Journal of Demography 54:29-41.

Bumpass, Larry L. 2004. "Social change and the American family." Pp. 213-219 in Understanding and Optimizing Human Development:: from Cells to Patients to Populations, vol. 1038, Annals of the New York Academy of Sciences.

Bumpass, Larry L., Kelly Raley, and James A. Sweet. 1995. "The changing character of stepfamilies- Impications of cohabitation and nonmarital childbearing." Demography 32:425-436.

Bumpass, Larry L. and Raley K. Raley. 1995. "Redefining single parent famlies: cohabitation and chaning family reality." Demography 32:97-109.

Bumpass, Larry L. and James A. Sweet. 1989. "Childrens experience in single-parent families: Implications of cohabitaiton and marital transitions." Family Planning Perspectives 21:256-260.

—. 1989. "National Estimates of Cohabitaiton." Demography 26:615-625.

Bumpass, Larry L., James A. Sweet, and Andrew Cherlin. 1991. "The role of cohabitation in the declining rates of marriage." Journal of Marriage and the Family 53:913-927.

133

Carlson, Marcia, Sarah McLanahan, and Paula England. 2004. "Union formation in fragile families." Demography 41:237-261.

Casper, Lynn M. and Paul N. Cohen. 2000. "How does POSSLQ measure up? Historical estimates of cohabitation." Demography 37:237-245.

Catanzarite, Lisa and Vilma Ortiz. 2002. "Too few good men? Available partners and single motherhood among Latinas, African Americans, and whites." Hispanic Journal of Behavioral Sciences 24:278-295.

Cavanagh, Shannon. E, Catherine Riegle-Crumb, and Robert Crosnoe. 2007. "Puberty and the education of girls." Social Psychology Quarterly 70:186-198.

Cavanagh, Shannon E. 2008. "Family Structure History and Adolescent Adjustment." Journal of Family Issues 29:944-980.

Cavanagh, Shannon E., S. R. Crissey, and R. Kelly Raley. 2008. "Family structure history and adolescent romance." Journal of Marriage and the Family 70:698-714.

Cavanagh, Shannon E. and Aletha. C. Huston. 2006. "Family instability and children's early problem behavior." Social Forces 85:551-581.

Cavanagh, Shannon E. and Aletha C. Huston. 2008. "The Timing of Family Instability and Children's Social Development." Journal of Marriage and the Family 70:1258-1270.

Chandra Anjani, Gladys M. Martinez, William D. Mosher,Joyce C. Abma, and Jo Jones. 2005. "Fertility, family planning, and reproductive health of U.S. women: Data from the 2002 National Survey of Family Growth." Washington, DC: National Center for Health Statistics.

Cherlin, Andrew. 2008. "The future of marriage." Population and Development Review 34:175- 177.

Cherlin, Andrew J. 2004. "The Deinstitutionalization of American marriage." Journal of Marriage and the Family 66:848-861.

—. 2005. "American marriage in the early twenty-first century." Future of Children 15:33-55.

Cherlin, Andrew J. and Frank F. Furstenberg. 1994. "Stepfamilies in the United STates: A reconsideration." Annual Review of Sociology 20:359-381.

Cherlin, Andrew J., Kathleen. E. Kiernan, and P. Lindsay Chaselansdale. 1995. "Parental divorce in childhood and demographic outcomes in young adulthood." Demography 32:299-318.

Clarkberg, Marin, Ross M. Stolzenberg, and Linda J. Waite,. 1995. "Attitudes, values, and entrance into cohabitational versus marital unions." Social Forces 74:609-632.

134

Condon, John T., Phillip Boyce, and Carolyn J. Corkindale. 2004. "The First-Time Fathers Study: a prospective study of the mental health and wellbeing of men during the transition to parenthood." Australian and New Zealand Journal of Psychiatry 38:56-64.

Cunningham, Micheal and Arland Thornton. 2005. "The influence of union transitions on white adults' attitudes toward cohabitation." Journal of Marriage and the Family 67:710-720.

Cunningham, Micheal. and Arland. Thornton. 2006. "The influence of parents' marital quality on adult children's attitudes toward marriage and its alternatives: Main and moderating effects." Demography 43:659-672.

David J. Fein, Nancy R. Burstein, Greta G. Fein and Laura D. Lindberg. 2003. "The Determinants o fMarriage and Cohabitation among Disadvantaged Americans: Research Findings and Needs." Administration for Children and Families.

Dush, Claire. M. K. and Paul R. Amato. 2005. "Consequences of relationship status and quality for subjective well-being." Journal of Social and Personal Relationships 22:607-627.

Dush, Claire M. K., Catherine L. Cohan, and Paul R. Amato. 2003. "The relationship between cohabitation and marital quality and stability: Change across cohorts?" Journal of Marriage and the Family 65:539-549.

Edin, Kathyrn. 2000. "What do low-income single mothers say about marriage?" Social Problems 47:112-133.

Edin, Kathyrn, M. J. Kefalas, and Joanna. M. Reed. 2004. "A peek inside the black box: What marriage means for poor unmarried parents." Journal of Marriage and the Family 66:1007-1014.

Edin, Kathryn and Joanna M. Reed. 2005. "Why don't they just get married? Barriers to marriage among the disadvantaged." Future of Children 15:117-137.

Eggebeen, David J. 2002. "The changing course of fatherhood - Men's experiences with children in demographic perspective." Journal of Family Issues 23:486-506.

Elder, GlennH., J. K. Liker, and C. E. Cross. 1984. "Parent child-behaviors in the Great Depression: Life course and intergenerational influences." Life-Span Development and Behavior 6:109-158.

Elder, Glenn H. 1975. "Age differentiation and Life course." Annual Review of Sociology 1:165- 190.

—. 1977. "Family History and Life Course." Journal of Family History 2:279-304.

—. 1984. "Families, Kin, and the Life Course: A Sociological Perspective." Review of Child Development Research 7:80-136.

—. 1985. Life course dynamics, trajectories and transitions 1968-1980. Ithaca: Cornell

135

University Press.

—. 1987. "Famlies and Lives: Some Developments in life course studies." Journal of Family History 12:179-199.

—. 1992. "Life Course." in Encyclopedia of Sociology, edited by B. M. L. Borgatta E.F. New York: Macmillan.

—. 1994. "Time, Human Agency, and Social Change: Perspectives on the Life Course." Social Psychology Quarterly 57:4-15.

—. 1998. "The life course as developmental theory." Child Development 69:1-12.

Elder. GLenn H., Monica K. Johnson and Robert Crosnoe. 2003. "The emergence and development of life course theory." Pp. 3-19 in Handbook of the life course., edited by J. T. S. Mortimer, M.J. New York: Kluwer Academic Pres.

Ellwood, David T. and Christopher Jenks. "The Spread of Single-Parent Families in the United States Since 1960." Pp. 26-65 in The Future of the Family, edited by T. M. S. a. L. R. D. P. Moynihan. New York: Russell Sage Foundation.

Fields, Jason and Lynne M. Casper. . 2001. "merican Families and Living Arrangements." Current Population Report P20-537.

Foster, E. Micheal and Saul. D. Hoffman. 1996. "Nonmarital childbearing in the 1980s: assessing the importance of women 25 and older." Family Planning Perspectives 28:117-119.

Fry, Richard and D'Vera Cohn. 2010. "Women, Men and the New Economics of Marriage." Pp. 1-37 Pew Research Center; Social & Demographic Trends Project.

Furstenberg, Frank. F., Jr., Jeanne Brooks-Gunn, J., & Morgan S. Phillip. 1987. " Adolescent mothers and their children in later life." Family Planning Perspectives 19:142-51.

Furstenberg, Frank F. 1987. "The divorce revolution: The unexpcted social and economic consquences for rwomen and children in American " Pp. 556-558 in Contemporary Sociology-a Journal of Reviews, vol. 16, edited by L. Weitzman.

Furstenberg, Frank F., Saul D. Hoffman, and L. Shrestha. 1995. "The effect of divorce on intergenterational transfers: New evidence " Demography 32:319-333.

Gibson-Davis, Christine M., Kathryn. Edin, and Sarah McLanahan. 2005. "High hopes but even higher expectations: The retreat from marriage among low-income couples." Journal of Marriage and the Family 67:1301-1312.

136

Giele, Janet A. and Glen H. Elder, Jr. 1998. Methods of Life Course Research: Qualitative and Quantitative Approaches. Thousand Oaks: Sage.

Glick, Jennifer E., Stacey D. Ruf, Michael J. White, and Frances Goldscheider. 2006. "Educational Engagement and Early Family Formation: Differences by Ethnicity and Generation." Social Forces 84:1391-1415.

Glick, Paul C. 1984. "Marriage, Divorce, and Living Arrangements: Prospective Changes." Journal of Family Issues 5:7-26.

Glick, Paul C. and S. L. Lin. 1987. "Remarriage after divorce: Recent changes and demographic variations." Sociological Perspectives 30:162-179.

Golan, M., A. Weizman, A. Apter, and M. Fainaru. 1998. "Parents as the exclusive agents of change in the treatment of childhood obesity." American Journal of Clinical Nutrition 67:1130-1135.

Goldscheider, Francis and Sharon Sassler. 2006. "Creating stepfamilies: Integrating children into the study of union formation." Journal of Marriage and the Family 68:275-291.

Goldscheider, Frances K. and Calvin Goldscheider. 1998. "The Effects of Childhood Family Structure on Leaving and Returning Home." Journal of Marriage & Family 60:745-756.

Goldstein, Joshua R. 1999. "The leveling of divorce in the United States." Demography 36:409- 414.

Graefe, Deborah Roempke and Daniel T. Lichter. 1999. "Life Course Transitions of American Children: Parental Cohabitation, Marriage, and Single Motherhood." Demography 36:205-217.

Graefe, Deborah. R. and Daniel T. Lichter. 2002. "Marriage among unwed mothers: Whites, blacks and Hispanics compared." Perspectives on Sexual and Reproductive Health 34:286-293.

Gupta, Sanjiv, Pamela. J. Smock, and Wendy D. Manning. 2004. "Moving out: Transition to nonresidence among resident fathers in the United States, 1968-1997." Journal of Marriage and the Family 66:627-638.

Guzzo, Katherine. B. 2006. "How do marriage market conditions affect entrance into cohabitation vs. marriage?" Social Science Research 35:332-355.

—. 2006. "The relationship between life course events and union formation." Social Science Research 35:384-408.

Hagestad, Gunhild O. 1984. "Parent-child relations in adulthood: The impact of divorce in middle age. P." in Parenthood: Psychodynamic Perspectives. , edited by S. W. R.Cohen, & B. Cohler New York: : Guilford Press.

137

Hamilton, Brady E., Joyce A. Martin, and Stephanie J. Ventura. 2007. "Births: Preliminary Data for 2006." National Vital Statistics Reports 56.

Harknett, Kristen and Sarah S. McLanahan. 2004. "Racial and ethnic differences in marriage after the birth of a child." American Sociological Review 69:790-811.

Heuveline, Paul and Jeffrey M. Timberlake. 2004. "The role of cohabitation in family formation: The United States in comparative perspective." Journal of Marriage and the Family 66:1214-1230.

Hobcraft, John. 2008. "The timing and partnership context of becoming a parent: Cohort and gender commonalities and differences in childhood antecedents." Demographic Research 19:1281-1322.

Hogan, Dennis P. 1978. "Variable order of events in the life course." American Sociological Review 43:573-586.

Hollander, Dore. 1996. "Nonmarital Childbearing in the United States: A Government Report." Family Planning Perspectives 28.

Holzer, Harry J. 2009. "The Labor Market and Young Black Men: Updating Moynihan's Perspective." Annals of the American Academy of Political and Social Science 621:47-69.

Holzer, Harry J., Paul Offner, and Elaine Sorensen. 2005. "What explains the continuing decline in labor force activity among young black men?" Labor History 46:37-55.

Jefferies, Julia, Anne Berrington, and Ian Diamond. 2000. "Childbearing following marital dissolution in britain." European Journal of Population-Revue Europeenne De Demographie 16:193-210.

Joshi, Pamela, James M. Quane, and Andrew J. Cherlin. 2009. "Contemporary Work and Family Issues Affecting Marriage and Cohabitation Among Low-Income Single Mothers." Family Relations 58:647-661.

Juby, H. and C. Le Bourdais. 1998. "The changing context of fatherhood in Canada: A life course analysis." Population Studies-a Journal of Demography 52:163-175.

Kaufman, Gayle and Frances Goldscheider. 2007. "Do men need a spouse more than women? Perceptions of importance of marriage for men and women " Sociological Quarterly 48:29-46.

Kennedy, Sheila and Larry L. Bumpass. 2008. "Cohabitation and children's living arrangements: New estimates from the United States." Demographic Research 19:1663-1692.

Kiernan, Katherine. 2004. "Redrawing the boundaries of marriage." Journal of Marriage and the Family 66:980-987.

—. 2008. "Marriage and cohabitation." European Journal of Population-Revue Europeenne De

138

Demographie 24:341-342.

King, A. E. O. and T. T. Allen. 2009. "Personal Characteristics of the Ideal African American Marriage Partner A Survey of Adult Black Men and Women." Journal of Black Studies 39:570-588.

Kreider, Rose and Diane Elliot. 2009. "America’s Families and Living Arrangements: 2007." Current Population Report Series P20-561.

Lamb, Micheal. E. 2000. "The history of research on father involvement: An overview." Marriage and Family Review 29:23-42.

Lampard, R. and K. Peggs. 1999. "Repartnering: the relevance of parenthood and gender to cohabitation and remarriage among the formerly married." British Journal of Sociology 50:443-465.

Landale, Nancy. S. and R. S. Oropesa. 2007. "Hispanic families: Stability and change." Annual Review of Sociology 33:381-405.

LaRossa, R. 2004. "The culture of fatherhood in the fifties: A closer look." Journal of Family History 29:47-70.

Lesthaeghe, R. 1998. "On theory development: Applications to the study of family formation." Population and Development Review 24:1-+.

Lesthaeghe, R. J. and L. Neidert. 2006. "The second demographic transition in the United States: Exception or textbook example?" Population and Development Review 32:669-+.

Lichter, D. T., C. D. Batson, and J. B. Brown. 2004. "Welfare reform and marriage promotion: The marital expectations and desires of single and cohabiting mothers." Social Service Review 78:2-25.

Lichter, D. T. and D. R. Graefe. 2001. "Finding a mate? The marital and cohabitation histories of unwed mothers." Pp. 317-343 in Out of Wedlock: Causes and Consequences of Nonmarital Fertility, edited by L. Wu and B. Wolfe.

Lichter, D. T., D. R. Graefe, and J. B. Brown. 2003. "Is marriage a panacea? Union formation among economically disadvantaged unwed mothers." Social Problems 50:60-86.

Lichter, D. T., Z. C. Qian, and L. M. Mellott. 2006. "Marriage or dissolution? Union transitions among poor cohabiting women." Demography 43:223-240.

Lillard, L. A., M. J. Brien, and L. J. Waite. 1995. "Premarital cohabitaiton and subsquent marital dissolution: A matter of self selection." Demography 32:437-457.

Manlove, Jennifer, Elizabeth Terry-Humen and Erum Ikramullah. 2006. " Young teens and older sexual partners: Correlates and consequences for males and females." Perspectives on Sexual and Reproductive Health 38:197-207.

139

Manlove, J., E. Terry-Humen, and E. Ikramullah. 2006. "Young teenagers and older sexual partners: Correlates and consequences for males and females." Perspectives on Sexual and Reproductive Health 38:197-207.

Manning, Wendy D. 1993. "MARRIAGE AND COHABITATION FOLLOWING PREMARITAL CONCEPTION." Journal of Marriage and the Family 55:839-850.

—. 2001. "Childbearing in cohabiting unions: Racial and ethnic differences." Family Planning Perspectives 33:217-223.

Manning, Wendy. D. 2001. "Childbearing in cohabiting unions: Racial and ethnic differences." Family Planning Perspectives 33:217-223.

Manning, Wendy D. 2004. "Children and the stability of cohabiting couples." Journal of Marriage and the Family 66:674-689.

Manning, Wendy D., M. A. Longmore, and P. C. Giordano. 2007. "The changing institution of marriage: Adolescents' expectations to cohabit and to marry." Journal of Marriage and the Family 69:559-575.

Manning, Wendy D. and Pamela J. Smock. 1997. "Children's living arrangements in unmarried- mother families." Journal of Family Issues 18:526-544.

Manning, Wendy. D. and Pamela J. Smock. 2002. "First comes cohabitation and then comes marriage? A research note." Journal of Family Issues 23:1065-1087.

Manning, Wendy D. and Pamela J. Smock. 2005. "Measuring and modeling cohabitation: New perspectives from qualitative data." Journal of Marriage and the Family 67:989-1002.

Manning, Wendy D., Pamela J. Smock, and D. Majumdar. 2004. "The relative stability of cohabiting and marital unions for children." Population Research and Policy Review 23:135-159.

Manning, Wendy D., S. D. Stewart, and Pamela J. Smock. 2003. "The complexity of fathers' parenting responsibilities and involvement with nonresident children." Journal of Family Issues 24:645-667.

Manning, Wendy. D and Kathleen A. Lamb. 2003. "Adolescent well-being in cohabiting, married, and single-parent families." Journal of Marriage and the Family 65:876-893.

Manning, Wendy D. and Pamela J. Smock. 1997. "Children's living arrangements in unmarried- mother families." Journal of Family Issues 18:526-544.

Martin, Steven P. 2006. "Trends in marital dissolution by women's education in the United States." Demographic Research 15:537-559.

Martin, T. C. and Larry L. Bumpass. 1989. "Recent trends in marital disruption." Demography

140

26:37-51.

Mathews TJ, and Ventura SJ. 1997. "Birth and fertility rates by educational attainment: United States, 1994." Monthly vital statistics report 45.

McLanahan, S. 2004. "Diverging destinies: How children are faring under the second demographic transition." Demography 41:607-627.

—. 2007. "Should government promote marriage?" Journal of Policy Analysis and Management 26:951-951.

McLanahan, Sara and Gary Sandefur. 1994. Growing Up with a Single Parent. . Cambridge: Harvard University Press.

McLanahan, Sarah & Sandefur, Gary 1994. Growing up with a single parent: What hurts, what helps. . Cambridge, MA: Harvard University Press.

Mincieli, Lisa, Jennifer Manlove, Molly McGarrett, Kristin Moore and Suzanne Ryan. May 2007. "The Relationship Context of Births Outside of Marriage: The Rise of Cohabitation." Child Trends Research Brief.

Mulder, C. H. 2003. "The effects of singlehood and cohabitation on the transition to parenthood in the Netherlands." Journal of Family Issues 24:291-313.

Musick, Kelly. 2002. "Planned and Unplanned Childbearing among Unmarried Women." Journal of Marriage and the Family 64:915-929.

—. 2007. "Cohabitation, nonmarital childbearing, and the marriage process." Demographic Research 16.

Nock, Steven. 2006. "Illustrations of family scholarship: Introduction to the special issue." Social Science Research 35:322–331.

Nock, Steven L. 2005. "Marriage as a public issue." Future of Children 15:13-32.

Norton, A. J. and Paul. C. Glick. 1986. "One parent fmailies: A social and economic profile." Family Relations 35:9-17.

Oconnell, M. and C. C. Rogers. 1982. "Differential fertilty in the United States:1976-1980." Family Planning Perspectives 14:281-&.

O'Connell, Tavia Simmons and Martin. 2000. "Married-Couple and Unmarried-Partner Households:2000." Census 2000 Special Reports (U.S. Census Bureau, February 2003).

Oropesa, R. S. and N. S. Landale. 2004. "The future of marriage and Hispanics." Journal of Marriage and the Family 66:901-920.

Osborne, C., W. D. Manning, and P. J. Smock. 2007. "Married and cohabiting parents'

141

relationship stability: A focus on race and ethnicity." Journal of Marriage and the Family 69:1345-1366.

Parnell, A. M., G. Swicegood, and G. Stevens. 1994. "Nonmarital pregancies and marriage in the United States." Social Forces 73:263-287.

Patrick, Heuveline and M. Timberlake Jeffrey. 2004. "The role of cohabitation in family formation: The United States in comparative perspective." Journal of Marriage and Family 66:1214-1230.

Phillips, J. A. and M. M. Sweeney. 2005. "Premarital cohabitation and marital disruption among White, Black, and Mexican American women." Journal of Marriage and the Family 67:296-314.

Popenoe, David. 1993. "American family decline1960-1990 - A Review and Appraisal." Journal of Marriage and the Family 55:527-542.

—. 1993. "American Family Decline, 1960-1990: A Review and Appraisal." Journal of Marriage and Family 55:527-542.

Popenoe, David and Barbara Whitehead. 2007. "The state of our unions 2007." Piscataway, NJ: Rutgers University. : National Marriage Project.

Qian, Z. C., D. T. Lichter, and L. M. Mellott. 2005. "Out-of-wedlock childbearing, marital prospects and mate selection." Social Forces 84:473-491.

Raley, R. Kelly. 2001. "Increasing fertility in cohabiting unions: Evidence for the second demographic transition in the United States?" Demography 38:59-66.

Raley, R. Kelly and Elisabeth Wildsmith. 2004. "Cohabitation and children's family instability." Journal of Marriage and the Family 66:210-219.

Reed, Joanna M. 2006. "Not crossing the "extra line": How cohabitors with children view their unions." Journal of Marriage and the Family 68:1117-1131.

Rindfuss, Richard R., S. P. Morgan, and K. Offutt. 1996. "Education and the changing age pattern of American fertility: 1963-1989." Demography 33:277-290.

Ross, C. E. and J. Mirowsky. 1999. "Parental divorce, life-course disruption, and adult depression." Journal of Marriage and the Family 61:1034-1045.

Ryan, S., K. Franzetta, E. Schelar, and J. Manlove. 2009. "Family Structure History: Links to Relationship Formation Behaviors in Young Adulthood." Journal of Marriage and the Family 71:935-953.

Ryder, Norman B. 1965. "The cohort as a concept in the study of social change." American Sociological Review 30:843-861.

142

Sassler, S. 2004. "The process of entering into cohabiting unions." Journal of Marriage and the Family 66:491-505.

Sassler, S. and A. Cunningham. 2008. "How cohabitors view childbearing." Sociological Perspectives 51:3-28.

Sassler, S., A. Cunningham, and D. T. Lichter. 2009. "Intergenerational Patterns of Union Formation and Relationship Quality." Journal of Family Issues 30:757-786.

Sassler, S., A. Miller, and S. M. Favinger. 2009. "Planned Parenthood? Fertility Intentions and Experiences Among Cohabiting Couples." Journal of Family Issues 30:206-232.

Scommegna, Paolo. 2002. " ―Increased Cohabitation Changing Children’s Family Settings,‖ " in Population Today, National Institute of Child Health and Human Development (NICHD), National Institutes of Health, , vol. September and October.

Segal, K., T. Holmes, C. B. Hollenbeck, and A. Chasson. 2004. "The effectiveness of the shapedown program, a family-based weight control program for children and adolescents." Journal of the American Dietetic Association 104:32.

Seltzer, Judith A. 2000. "Families Formed Outside of Marriage." Journal of Marriage and Family 62:1247-1268.

Seltzer, J. A. 2004. "Cohabitation in the United States and Britain: Demography, kinship, and the future." Journal of Marriage and the Family 66:921-928.

Shanahan, Michael J. 2000. "Pathways to adulthood in changing societies: Variability and Mechanisms in Life Course Perspective." Annual Review of Sociology 26:667.

Shanahan, M. J., R. A. Miech, and G. H. Elder. 1998. "Changing pathways to attainment in men's lives: Historical patterns of school, work, and social class"." Social Forces 77:231-256.

Sigle-Rushton, W., J. Hobcraft, and K. Kiernan. 2005. "Parental divorce and subsequent disadvantage: A cross-cohort comparison." Demography 42:427-446.

Skinner, K. B., S. J. Bahr, D. R. Crane, and V. R. Call. 2002. "Cohabitation, marriage, and remarriage - A comparison of relationship quality over time." Journal of Family Issues 23:74-90.

Smock, P. J. 2000. "Cohabitation in the United States: An appraisal of research themes, findings, and implications." Annual Review of Sociology 26:1-20.

—. 2004. "The wax and wane of marriage: Prospects for marriage in the 21st century." Journal of Marriage and the Family 66:966-973.

Smock, P. J. and W. D. Manning. 1997. "Cohabiting partners' economic circumstances and marriage." Demography 34:331-341.

143

Smock, Pamela J. and Wendy D. Manning. 2004. "Living Together Unmarried in the United States: Demographic Perspectives and Implications for Family Policy." Law & Policy 26:87-117.

Smock, P. J., W. D. Manning, and M. Porter. 2005. ""Everything's there except money": How money shapes decisions to marry among cohabitors." Journal of Marriage and the Family 67:680-696.

South, Scott J. 1999. "Historical Changes and Life Course Variation in the Determinants of Premarital Childbearing." Journal of Marriage & Family 61:752-763.

Spruijt, E. and V. Duindam. 2002. "Was there an increase in caring fatherhood in the 1990s? Two Dutch longitudinal studies." Social Behavior and Personality 30:683-695.

Steele, F. and S. Curtis. 2003. "Appropriate methods for analyzing the effect of method choice on contraceptive discontinuation." Demography 40:1-22.

Steele, F., H. Joshi, C. Kallis, and H. Goldstein. 2006. "Changing compatibility of cohabitation and childbearing between young British women born in 1958 and 1970." Population Studies-a Journal of Demography 60:137-152.

Steele, F., C. Kallis, H. Goldstein, and H. Joshi. 2005. "The relationship between childbearing and transitions from marriage and cohabitation in Britain." Demography 42:647-673.

Steele, F., C. Kallis, and H. Joshi. 2006. "The formation and outcomes of cohabiting and marital partnerships in early adulthood: the role of previous partnership experience." Journal of the Royal Statistical Society Series a-Statistics in Society 169:757-779.

Stewart, Susan. D., Wendy. D. Manning, and Pamela. J. Smock. 2003. "Union formation among men in the US: Does having prior children matter?" Journal of Marriage and the Family 65:90-104.

Sweeney, Megan M. 2007. "Stepfather Families and the Emotional Well-Being of Adolescents." Journal of Health & Social Behavior 48:33-49.

Sweet, James. A. and Larry. Bumpass. 1984. "Progress report on census mongraph on families and households: With speicall attention to the increase in cohabitaiton." Population Index 50:423-423.

Sweet, James A. and Larry. L. Bumpass. 1992. "YOUNG-ADULTS VIEWS OF MARRIAGE, COHABITATION, AND FAMILY." Pp. 143-170 in Changing American Family - Sociological and Demographic Perspectives, edited by S. J. South and S. E. Tolnay.

Sweet, James A.; Bumpass, Larry L. and Call, V. 1988. "The design and content of the National Survey of Families and Households." NSFH Working Paper 1 Madison, WI: Center for Demography and Ecology, University of Wisconsin-Madison.

Teachman, Jay. 2003. "Childhood Living Arrangements and the Formation of Coresidential

144

Unions." Journal of Marriage and Family 65:507-524.

Teachman, Jay. 2003. "Premarital sex, premarital cohabitation, and the risk of subsequent marital dissolution among women." Journal of Marriage and the Family 65:444-455.

Teachman, Jay D. 2002. "Childhood living arrangements and the intergenerational transmission of divorce." Journal of Marriage and the Family 64:717-729.

—. 2002. "Stability across cohorts in divorce risk factors." Demography 39:331-351.

Teachman, Jay. D. and Kevin. A. Polonko. 1990. "Cohabitation and marital stability in the United States." Social Forces 69:207-220.

Teachman, Jay. D., Lucky M. Tedrow, and Kyle. D. Crowder. 2000. "The changing demography of America's families." Journal of Marriage and the Family 62:1234-1246.

Teachman, Jay. D., Jeffrey. Thomas, and Kathleen. Paasch. 1991. "Legal status and the stability of coresidential unions." Demography 28:571-586.

Terry-Humen, Elizabeth, Jennifer Manlove, and Kristen A. Moore. . April 2001. "Births Outside of Marriage: Perceptions vs. Reality. ." Child Trends Research Brief.

Thomson, Elizabeth., T. L. Hanson, and Sarah. S. McLanahan. 1994. "Family structure and child well-being: Economic resources vs. parental behaviors." Social Forces 73:221-242.

Thornton, Arland. 1991. "Influence of the marital history of parents on the marital and cohabitational experienes of children." American Journal of Sociology 96:868-894.

Thornton, Arland, William G. Axinn and Yu Xie. 2007. Marriage and Cohabitation. Chicago: The University of Chicago Press.

Thornton, Arland., William. G. Axinn, and Jay D. Teachman. 1995. "The influence of school enrollment and accumulation on cohabitationa nd marriage in early adulthood. ." American Sociological Review 60:762-774.

Thornton, Arland and LindaYoung-DeMarco. 2001. "Four decades of trends in attitudes toward family issues in the United States: The 1960s through the 1990s." Journal of Marriage and the Family 63:1009-1037.

Timmer, Susan. G. and Terri. L. Orbuch. 2001. "The links between premarital parenthood, meanings of marriage, and marital outcomes." Family Relations 50:178-185.

Turcotte, Pierre and Frances Goldscheider. 1998. "The Evolution of Factors Influencing First Union Formation in Canada." Canadian Studies in Population and Development Review 25:145-174.

145

United States Bureau of the Census, 2009. "Years of School Completed by People 25 Years and Over, by Age and Sex: Selected Years 1940 to 2009 " Current Population Survey

United States Bureau of the Census, 2009. Educational Attainment in the United States.

United States Bureau of the Census 2007. ―Annual Social and Economic Supplement, selected years, 1970 to 2007.‖ Current Population Survey.

Upchurch, Dawn. M., Lillard. A. Lillard, and Constantijn. W. A. Panis. 2002. "Nonmarital childbearing: Influences of education, marriage, and fertility." Demography 39:311-329.

Ventura, Stephanie J., Christine A. Bachrach, Laura Hill, Kelleen Kaye, Pamela Holcomb, and Elisa Koff. . 1995. "The Demography of Nonmarital Childbearing." in Report to Congress on Out-of-Wedlock Childbearing. Washington, DC: U.S. Government Printing Office. Department of Health and Human Services. Department of Health and Human Services Department of Health and Human Services:3–133.

VENTURA, Stephanie.J., Willilam.D. MOSHER, Sally C. CURTIN, Joyce.C. ABMA, Stanley. HENSHAW. 2000. "Trends in Pregnancies and Pregnancy Rates by Outcome: Estimates for the United States." National Center for Health Statistics. Vital Health Statistics 21.

White, Lynn and Stacy J. Rogers. 2000. "Economic Circumstances and Family Outcomes: A Review of the 1990s." Journal of Marriage & Family 62:1035-1051.

Willis, Rober. J. and John. G. Haaga. 1996. "Economic approaches to understanding nonmarital fertility." Population and Development Review 22:67-86.

Wilson, William Julius. 1987. . The Truly Disadvantaged: The City, the Underclass, and Public Policy. Chicago: University of Chicago Press.

Wilson, William J. and Kathryn Neckerman 1987. The truly disadvantaged. Chicago: University of Chicago Press.

Wolfinger, Nicholas. H. 2000. "Beyond the intergenerational transmission of divorce - Do people replicate the patterns of marital instability they grew up with?" Journal of Family Issues 21:1061-1086.

—. 2001. "The effects of family structure of origin on offspring cohabitation duration." Sociological Inquiry 71:293-313.

—. 2003. "Family structure homogamy: The effects of parental divorce on partner selection and marital stability." Social Science Research 32:80-97.

—. Forthcoming. "More Evidence for Trends in the Intergenerational Transmission of Divorce: A Completed Cohort Approach using Data from the General Social Survey." Demography.

Wolfinger, Nicholas H. 2003. "Parental Divorce and Offspring Marriage: Early or Late?" Social Forces 82:337-353.

146

Woodward, Liana. J., David. M. Fergusson, and L. John. Horwood. 2006. "Gender differences in the transition to early parenthood." Development and Psychopathology 18:275-294.

Wu, Lawrence. L., Larry. L. Bumpass, and Kelly. Musick. 2001. "Historical and life course trajectories of nonmarital childbearing." Pp. 3-48 in Out of Wedlock: Causes and Consequences of Nonmarital Fertility, edited by L. Wu and B. Wolfe.

Wu, Lawrence. L. and Brian. C. Martinson. 1993. "Family structure and the risk of a premarital birth." American Sociological Review 58:210-232.

Wu, Lawrence. L. and Kelly. Musick. 2008. "Stability of Marital and Cohabiting Unions Following a First Birth." Population Research and Policy Review 27:713-727.

Wu, Lawrence. L. and Barbara. Wolfe. 2001. "Out of wedlock - Causes and consequences of nonmarital fertility - Introduction." Pp. XIII-XXXII in Out of Wedlock: Causes and Consequences of Nonmarital Fertility, edited by L. Wu and B. Wolfe.

147