RELATIONSHIP DISSOLUTION IN COMPLEX FAMILY STRUCTURES: THE ROLE OF MULTIPARTNERED FERTILITY

Pajarita Charles

A dissertation submitted to the faculty of the University of North Carolina at Chapel Hill in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the

School of Social Work

Chapel Hill 2009

Approved by:

Chair: Dennis K. Orthner, Ph.D. Shenyang Guo, Ph.D. Anne Jones, Ph.D. Susan Parish, Ph.D. Kathleen M. Harris, Ph.D.

ii

ABSTRACT

PAJARITA CHARLES: Relationship Dissolution in Complex Family Structures:

The Role of Multipartnered Fertility (Under the direction of Dennis K. Orthner, Ph.D.)

Multipartnered fertility (MPF) is a growing phenomenon in many families and

occurs in as many as 74% of couples in certain socio-economic groups (Meyer, Cancian,

& Cook, 2005). MPF describes the occurrence of parents having children with more than

one partner. As these are often couples that social workers meet in various service

environments, understanding their needs is an important consideration for social work

practice, policy, and research.

Previous evidence demonstrates that couples with MPF are at an increased risk for

unstable relationships (Teachman, 2008a), yet we know little about the timing of

relationship dissolution and the differential role that MPF plays in union instability. The

primary research question for this study is: “To what extent does multipartnered fertility

influence whether and when a couple divorces or separates if they are married,

cohabiting, or dating?”

The data (N = 3,022) come from three waves of the Fragile Families and Child

Wellbeing Study, a multistage stratified probability sample of hospital births in 20 large

U.S. cities. Kaplan-Meier estimates are used to describe the event by illustrating the

length of time couples remain in their relationship and to test group differences. Survival

iii

analysis using discrete-time models is used to estimate the effects of MPF and covariates

on relationship dissolution.

The survivor function suggests a decreasing rate of remaining in the relationship

over the study period, especially in father-only and father/mother MPF cases.

Furthermore, the hazard function indicates a fast rate of dissolution in the early period

following birth, especially among these two groups. The discrete-time models show that

father-only and father/mother MPF cases are significantly more likely to end their

relationship than couples without MPF, after other factors are accounted for. Moreover,

unmarried couples, previously incarcerated fathers, younger mothers, and unsupported

mothers are more likely to separate.

Multipartnered fertility among both mothers and fathers may play a critical role in

the outcome of couple relationships. The findings from this study suggest that programs

and policies to strengthen unmarried couples need to take MPF into consideration, and

should carefully consider the timing of interventions to ensure that they are provided at

the appropriate time.

iv

ACKNOWLEDGEMENTS I am grateful to many people who accompanied me on this journey. I am

especially indebted to Dr. Dennis Orthner, my chair, advisor, mentor, and friend. Dennis

played an instrumental role in supporting me through this entire process and I thank him

sincerely for his wisdom, guidance, dependability, and thoughtfulness. I will take with

me all that he has offered, both professionally and personally, and will always remember

Dennis for his unwavering commitment to improving the lives of disadvantaged children

and families.

I am also particularly grateful to Dr. Shenyang Guo, who taught me much of what

I know about statistics. His guidance and feedback was critical at many stages of my

dissertation and I thank him for his willingness to correspond over email since most of

my dissertating was done from California! Another special thanks goes to Dr. Anne

Jones, a committee member, and my mentor and friend. Her contribution to my thinking

about how best to support low-income, unmarried couples has been very important to me

over the years of our work together. I am also grateful to Dr. Susan Parish whose

feedback on issues related to my topic was especially helpful. And finally, I thank Dr.

Kathleen Mullan Harris whose substantive and methodological expertise was invaluable

at many points in my research.

I also want to thank Dr. Michal Grinstein-Weiss, my mentor and good friend.

Michal supported me in many ways over the past several years and often selflessly

v

offered me the extra assistance and encouragement I needed to make it to the next stage. I

will always remember that “Where there’s a Will, there’s a Way.”

I have always felt very connected to UNC and know that without the support of

many people at the School of Social Work my experience would have been very

different. I want to express my gratitude to Dean Jack Richman, Dr. Kathleen Rounds,

and all the staff at CITU for being there in countless ways. And to all my friends in the

doctoral program, especially Mary, thank you for traveling the road alongside me.

Many special people in California were supportive of me throughout the doctoral

program and I want to thank them. The PS Moms brought humor and grace to my life in

the midst of stress and strain. The Rivera family provided help in ways that no one else

could. And finally, thanks to Pierre, Ilana, and Colby, who offered reliability and security

when I needed it most.

Most importantly, I want to thank my family. I owe tremendous gratitude to my

mother who has been my friend, confidant, and biggest fan. Her unwavering support at

every step in this process made me always believe that I could do it, no matter the odds. I

will always remember the phone calls and emails with all her words of encouragement

and strength to get me through the day.

Finally, I owe the most to my husband and best friend, Simeon, who inspired me

to do this from the very beginning. He really made this possible, never doubting for a

moment that I could finish. His firm belief in me as a researcher, wife, and mother helped

me overcome what seemed to be the impossible! I will always thank him for being there

in countless ways. And lastly, I thank Sorin and Caelan, our two boys. They were as

vi

much a part of this process as anyone else and taught me what really matters, and that

anything, really anything at all, is possible.

vii

TABLE OF CONTENTS LIST OF TABLES…..……………………………………………………………………ix

LIST OF FIGURES.……………………………………………………………………....x

Chapter

I. INTRODUCTION……..……………………………………………………..1

II. INSTABILITY IN NONMARITAL UNIONS.……………………………...4

Nonmarital Unions & Childbearing….……………………………………….4

Union Instability…..…………………………………………………………..5

Implications of Union Instability.…………………………………………….6

Policy Context……………………………………………………………….10

Multipartnered Fertility……………………………………………………...12

Correlates of Relationship Instability………………………………………..14

Theoretical Framework: Transition to Parenthood…………………………..20

III. RESEARCH METHODS……………………………………………………25

Research Aims and Hypotheses……………………………………………..25

Data………………………………………………………………………….26

Sample……………………………………………………………………….28

Censoring……...……………………………………………………………..30

Dependent Variable, Origin of Time, and Study Window…………………..32

Independent Variables……………………………………………………….33

Data Management……………………………………………………………37

viii

Data Analysis………………………………………………………………...37

Model Evaluation and Diagnostic Procedures……………………………….42

IV. RESULTS……………………………………………………………………45

Univariate and Bivariate Results…………………………………………….45

Survivor and Hazard Functions……………………………………………...49

Multivariate Model Results………………………………………………….53

Model Predicted Survivor Curves…………………………………………...61

V. DISCUSSION AND CONCLUSION.………………………………………63

Summary of Major Findings…………………………………………………64

Strengths and Limitations……………………………………………………73

Implications…………………………………………………………………..76

Conclusion.………………………………………………………………......82

APPENDICES…………………………………………………………………………...84

REFERENCES.………………………………………………………………………….87

ix

LIST OF TABLES

Table Page

1. Descriptive Statistics for Couples by Relationship Status…………..……….46

2. 90th Percentile of the Survival Function of Selected Variables...…………....48

3. Odds Ratios from Discrete-Time Event History Model Predicting Relationship Dissolution vs. Staying Together……………………………....55

4. Odds Ratios from Discrete-Time Event History Model Using

Alternative Reference Group for MPF………………………………………59

x

LIST OF FIGURES

Figure Page

1. Five-domain structural model of marital and family adaptation .…………...21

2. Modified five-domain structural model of couple and family adaptation…………………………………………………………………….23

3. Survivor function of relationship dissolution ..……………………………...49

4. Hazard function of relationship dissolution ………………………………..50

5. Survivor function by multipartnered fertility history.…………………….....51

6. Survivor function by relationship status…………………………………….53

7. Hazard function of relationship dissolution by multipartnered fertility status ..……………………………………………...54

8. Model predicted survivor curves…………………………………………….62

CHAPTER 1

INTRODUCTION

Among the most concerning of recent changes to the traditional two-parent

nuclear family model is relationship instability. Couple instability has garnered attention

because of its negative effects on children, especially among low-income families.

Despite what is known about the consequences and causes of relationship instability

among married couples, less is known about the mechanisms behind union dissolution

among unmarried couples, who are likely to differ in important ways from couples who

marry.

Although most parents are romantically involved at the time of their child’s birth,

unmarried couples tend to end their relationships quickly (Bumpass & Lu, 2000). Over

half (54%) of cohabiting couples, for example, end their relationship either through

separation or marriage within one year (Binstock & Thornton, 2003). The risk of

relationship dissolution is particularly high during the early stages of parenting, when

couples face stressors that are often difficult to manage (Center for Research on Child

Wellbeing, 2003; C. P. Cowan & Cowan, 2000; Twenge, Campbell, & Foster, 2003),

leaving children more likely to grow up without both parents.

The proportion of children who live with a lone parent is more than double that of

35 years ago, and nearly half of all U.S. children will spend some part of their life in a

single-parent household (Andersson, 2002; Bumpass & Lu, 2000). This disturbing

2

consequence of the rapid changes in family structure has become significant to policy

makers and researchers because single parenting has been found to be associated with

detrimental effects on children on a range of economic, educational, behavioral, and

psychological outcomes (Amato & Booth, 1997; McLanahan, 1997; McLanahan &

Sandefur, 1994; Thomas & Sawhill, 2005).

Single parenting, however, does not account for all family arrangements outside

the two-biological-married-parent structure. Multipartnered fertility is a growing

phenomenon that describes the occurrence of parents having children with more than one

partner. It occurs in as few as 8% of couple relationships and in as many as 74% in

certain socio-economic/racial-ethnic groups (Guzzo & Furstenberg, 2007a; Meyer et al.,

2005). Previous research on stepfamilies has shown that couples with children from

previous partners are at an increased risk for unstable relationships (Teachman, 2008a).

Furthermore, evidence suggests that fathers with children in multiple families can be a

source of stress for many couples and a risk factor for relationship dissolution (Lichter et

al., 2006). Despite this evidence, we know little about the timing of union dissolution and

the differential effects of multipartnered fertility when it occurs with one or both parents

compared to neither parent.

Although the effects of couple dissolution are well documented, evidence-based

practice and treatment methods to address it have not been well developed. Despite this

apparent lag in research, social policy initiatives to stem relationship instability abound.

The Temporary Assistance for Needy Families Reauthorization Bill, passed in February

2006 as part of the Deficit Reduction Act of 2005, appropriated up to $750 million to

support healthy marriage and responsible fatherhood programs. In 2006, there were over

3

250 Healthy Marriage grants awarded to service providers in states across the country

working to strengthen and sustain marriages and relationships (Administration for

Children and Families, 2006)1. Many of these programs operate in settings serviced by

social workers. Since social workers often work with at-risk couples with multipartnered

fertility in these and a variety of other social service environments, understanding their

needs is an important consideration for social work practice, policy, and research.

Several social and economic explanations exist for the recent demographic

changes observed in the traditional two-parent family model. Some posit that the

weakening link or “decoupling” of marriage and parenthood, as well as a shift in norms

and an increased acceptance of alternatives in family structure, play a part (Cherlin,

Cross-Barnet, Burton, & Garrett-Peters, 2008; Edin & Kefalas, 2005). Changes in gender

role expectations of men and women, as well as women’s desire to increase their self-

efficacy through childbearing (Edin & Kefalas, 2005) are other possible reasons for the

shifts in family patterns. Economic explanations focusing on the expansion of economic

opportunities for women (Becker, 1991), incentive effects due to welfare (Bitler,

Gelbach, Hoynes, & Zavodny, 2004), and the decrease in men’s earning potential

(Oppenheimer, 1994), also exist . Despite the numerous hypotheses about changes to

family formation patterns, no single theory explains a satisfactory portion of the problem.

In fact, finding an accurate explanation has proven to be one of the most difficult and

frustrating problems for social scientists in years (Ellwood & Jencks, 2001). To help

address this challenge, this paper seeks to improve our understanding of the relationship

between multipartnered fertility and union instability among a sample of parents who

recently had a child together. 1 This is likely an undercount of grantees because it does not include later contracts that were awarded.

CHAPTER 2

INSTABILITY IN NONMARITAL UNIONS

Nonmarital Unions & Childbearing

Nonmarital childbearing has grown dramatically during the last 50 years,

increasing from 5% in 1960 to almost 40% in 2007 (Martin et al., 2007; Ventura, 2009).

The number of infants born to unmarried women rose to 1.7 million in 2007, the highest

number ever recorded for which comparable statistical data are available. Proportions of

births to unmarried women vary widely by race and ethnicity. For instance, 70% of all

births to non-Hispanic black women were to mothers who were unmarried in 2005,

compared to 48% and 25% for Hispanic and white women respectively (Martin et al.,

2007).

Contrary to popular belief, many unmarried women who bear children do so with

a partner who is actively engaged with the mother during and, at least for some time, after

pregnancy. While the findings are somewhat mixed, cohabitation arrangements appear to

account for a large portion of what are conventionally thought of as non-marital, single-

mother births. According to Bumpass & Lu (2000), completely unattached mothers (i.e.,

women in no type of sustained partner relationship at all) are actually quite rare, with

evidence suggesting that the rise in childbirth among unmarried mothers is for the most

part due to a rise in births to cohabiting, two-parent couples.

5

Approximately 4.6 million couples in the U.S. are in a cohabiting relationship, a

stark contrast to the .4 million households with cohabiting couples in 1960 (Seltzer,

2004). With half of couples cohabiting before marriage, this co-residential romantic

relationship has become the normative experience for many Americans (Cherlin, 2005).

“It has in fact become so prevalent that the majority of marriages and remarriages now

begin as cohabiting relationships, and most younger men and women cohabitate at some

point in their lives” (Smock, 2000, p. 1). The dramatic increase in cohabitation over the

last several decades has garnered significant attention because of numerous social

implications associated with it. Cohabitation has been associated with an increase in

marital unhappiness and a weakening of people’s commitment to the norm of marriage

(Amato, Johnson, Booth, & Stacey, 2003; Axinn & Thornton, 1992). It has also been

linked to unstable relationships (Bumpass & Lu, 2000) and poor child outcomes (Bulanda

& Manning, 2008; Teachman, 2008b).

Union Instability

Despite the increase in rates of cohabitation, stability within such unions is

becoming less common (Lichter et al., 2006; Manning, Smock, & Majumdar, 2004).

Cohabiting couples are more likely to end their relationships compared to married

couples and are more likely to divorce once they have gotten married; the exception to

this is when women cohabit exclusively with their husbands (Teachman, 2003). The

proportion of cohabiting couples that separated between 1980 and 1994, regardless of

eventual marriage to each other, increased from 45% to 54% (Bumpass & Lu, 2000).

Cohabiting relationships tend to be short-lived, with about half such relationships lasting

one year or less. Within five years, about 40% of cohabiting couples separate and another

6

50% end the cohabitation through marriage (Binstock & Thornton, 2003; Bumpass & Lu,

2000; Smock, 2000). As cohabitation becomes more socially acceptable and more people

enter into cohabiting relationships for convenience, it is expected that commitment levels

within relationships will drop leading to even higher rates of dissolution (Bumpass & Lu,

2000).

Differences in relationship outcomes among cohabiting couples who are poor

compared to their nonpoor counterparts suggest these two groups may face different

relationship stability risks. While transitions out of cohabitation are prevalent for both

groups, in a study by Lichter and colleagues (2006) over half (52%) of the relationships

ended in dissolution for disadvantaged women compared to 42% for nonpoor women.

Among those who were poor, only one-third ended in marriage compared to 51% for the

nonpoor group.

Despite the instability of most nonmarital unions, the majority of couples in

cohabiting and dating relationships report very favorable attitudes toward marriage and

most cohabiters expect to marry eventually. Approximately three-quarters of cohabiting

women, for example, report that they intend to marry their partner in the future (Manning

& Smock, 2002; McLanahan, 2009). This supports the view that unmarried couples,

specifically those cohabiting, see themselves as taking steps toward marriage, not as

seeking an alternative to it (Guzzo, 2009; Oppenheimer, 2003).

Implications of Union Instability

Economic Consequences

Relationships that involve children and that eventually dissolve are a concern to

policymakers and researchers because of the risk for lower economic well-being among

7

single parents and children in these family types. Thomas and Sawhill (2005) compared

median income across family types after adjusting for tax credits and liabilities, food

stamp benefits, child care costs, and family size and found that married-parent families

are generally better off financially than either single-parent or cohabiting families by as

much as 45% and 35%, respectively. Single-parent families, especially single-mother

families, have the highest poverty rate (34%) of all other family configurations, including

married (7.6%), and cohabiting (21.5%) families (Thomas & Sawhill, 2005). Despite

these inequalities, previous research has found that low-income single mothers have some

economic advantages that resemble those of married, two-parent families. For instance,

Orthner et al. (2004) found that single parents’ ability to save and to pay bills on time

approached that of married, dual parents primarily because of advantages from

government assistance programs.

Research on income variation among unmarried mothers shows noteworthy

differences between cohabiting, romantically involved, and completely unattached

mothers (i.e., mothers who are not cohabiting or romantically involved with the father of

their baby). Jackson, Tienda, and Huang (2001) found that single mothers in no

partnership at all face the most precarious economic circumstances among all relationship

types. Married mothers reported $55,000 in household earnings compared to

approximately $24,000 for cohabiting women and $20,000 for unattached mothers.

Furthermore, only 40% of unattached mothers claimed that they received cash assistance

from the father of their baby compared to 95% of cohabiting mothers and 83% of

romantically involved mothers.

8

Couple stability also appears to affect wealth and asset accumulation. Cohabiting

individuals, and especially those in short-term relationships, are generally at a

disadvantage when it comes to long-term wealth accumulation and rarely generate levels

of wealth similar to those of married couples (Wilmoth & Koso, 2002). For example,

married individuals in a study of savings performances among low-income participants in

an Individual Development Account savings program had higher levels of home

ownership and car ownership than unmarried participants by as much as 20% (Grinstein-

Weiss, Zhan, & Sherraden, 2006).

Although the economic impact of union instability on women is generally well

understood, the effects of leaving a relationship for men are somewhat less clear and

seem to depend on whether the transition is out of a cohabiting relationship or a marriage.

For instance, Avellar and Smock (2005) found that following divorce, married men

typically experience increases in their personal earnings and decreases in their poverty

levels. This is contrary to the experience of cohabiting men whose poverty levels tend to

increase, a finding potentially attributable to cohabiting men’s likelihood to have fewer

skills and less education than those who marry.

Research on the economic effects of relationship stability and family formation is

hampered by the question of whether the differences in income among family types are

driven by family structure per se or by selection (Parke, 2003). It is possible, for example,

that couples with the most resources choose marriage while those with fewer resources

opt for cohabitation, and those with the least resources end up as single parents.

Researchers have, however, attempted to control for these possible selection biases (using

fixed and random effects models, for example) and have found that couple arrangements

9

and marriage are still associated with economic benefits for mothers and children

(Lerman, 2002; Thomas & Sawhill, 2005).

Child Well-Being

Instability in unmarried couple relationships has negative implications beyond

those of economic effects. The consequences for children are of primary concern early

and later in life because of evidence that points to a strong link between family structure

and behavioral, educational, psychological, employment, and physical health outcomes

(McLanahan, 1997). Children raised by two biological married parents tend to fare better

than children from other family structures, particularly single-parent families

(McLanahan & Sandefur, 1994). Children raised in single-parent families are more likely

to be at high risk for lower socioeconomic achievement, poor educational performance,

poorer psychological well-being, and lower social integration (Amato & Booth, 1997;

Pagani, Boulerice, & Tremblay, 1997; Teachman, 2008b; Wen, 2008). They are also at

risk for lower rates of high school and college completion, teenage childbearing, idleness

in young adulthood (McLanahan & Sandefur, 1994) and higher rates of poverty and

lower earnings later in life (Corcoran & Adams, 1997) .

Previous research has found that once important family characteristics such as

income, race, and socioeconomic status are taken into account, children from families

with only one parent are more likely to experience problems both during childhood and in

adulthood (Duncan & Brooks-Gunn, 1997). More recent research, however, suggests that

the negative effects of parental living arrangements and father absence have been

exaggerated in public discourse and are largely attenuated once income and education are

taken into account (DeBell, 2008; Foster & Kalil, 2007). This suggests that the

10

determinants of child outcomes stem from multiple factors, not only family living

arrangements. Further, it may indicate that family structure plays less of a role as a main

effect and is more of a proxy for other processes that affect child outcomes (e.g.,

fatherhood involvement) (Foster & Kalil, 2007).

Although children in cohabiting and married families generally appear different

on several socio-emotional and educational outcomes, the evidence remains inconsistent.

For example, Dunifon & Kowaleski-Jones (2002) found that while white children with

cohabiting parents had lower math scores in school, black children with cohabiting

parents had higher test scores. Another study showed no effects on behavior but

significant negative effects on math and reading scores for children who remained in a

cohabiting family compared to those whose mothers eventually married (Acs, 2005). In

the same study, children in families who transitioned from cohabitation to marriage,

however, had higher math and reading scores even before their mothers married,

suggesting that a high quality union may explain more of the benefits of having married

parents than marriage itself. Similar findings were made with the Fragile Families Study

showing that children from cohabiting families demonstrate more problem behaviors

(e.g., depression, anxiety, withdrawal, and aggressiveness) than children from married

families, although the differences were found to be largely attributable to background

characteristics of the parents who chose marriage (Center for Research on Child

Wellbeing, 2005b).

Policy Context

Heated debate over the implementation of welfare reform legislation has centered

on the complex issue of whether, and if so how, the government should assist

11

disadvantaged couples in sustaining their relationships and potentially getting married,

given the range of negative outcomes associated with nonmarital unions and

childbearing. On one side are marriage supporters who argue that marriage would

dramatically decrease the child poverty rate (Rector, Johnson, Fagan, & Noyes, 2003).

On the other side are critics of marriage promotion who maintain that women and

children who are victims of domestic violence will be unnecessarily pressured into

staying in violent relationships under the guise of “healthy marriage” goals (Catlett &

Artis, 2004). Others fall somewhere in the middle and contend that while some couples

are likely to benefit from government-sponsored pro-marriage programs, many of them

will need additional support to form stable families, including mental health services,

employment services, and assistance after reentry from incarceration (McLanahan, 2009).

As reported above, the evidence on children’s outcomes from unmarried versus

married parents remains mixed. Regardless of the true effects of marriage on poverty and

child well-being, it is known that most unmarried couples aspire to marry and believe that

marriage is good for themselves and for their children. Despite their high hopes of

marrying, some groups face significant barriers to doing so and are at significant risk of

being alone or in serial relationships. Many factors have been identified with union

instability and yet programs and policies to help couples stabilize and strengthen their

relationship still remain in their infancy, particularly those that serve low-income,

minority populations (Reardon-Anderson, Stagner, Macomber, & Murray, 2005).

Additional evidence is needed to guide policymakers and program planners as to how

best to support couples at risk of dissolution. This study serves to inform this knowledge

12

gap through examination of the role that multipartnered fertility, along with other factors,

plays in the stability of low-income couple relationships.

Multipartnered Fertility

Multipartnered fertility (MPF) refers to adults having children with more than one

partner, either outside the current relationship or in a former relationship (M. Carlson &

Furstenberg, 2006). Despite this growing phenomenon, information on the prevalence of

multipartnered fertility is limited because of the difficulty in capturing complex family

relationships, especially among poor parents who typically experience frequent

transitions. Accurate fertility histories of men are almost non-existent, and thus much of

what is known comes from mother reports like those in the Fragile Families Study. While

the Fragile Families Study is useful, it is nonetheless limited to the population of

nonmarital births in large U.S. cities. Some other sources of information exist, however,

that are useful in creating a general picture of multipartnered fertility in the U.S.

A Wisconsin study of welfare recipients in the late 1990s found that among the

mothers and fathers interviewed, three-fourths had a child with a previous partner (Meyer

et al., 2005). Data from the 2002 National Survey of Family Growth provide a useful

snapshot of multipartnered fertility among a nationally representative sample of men

aged 15-44. Guzzo and Furstenberg (2007a) found that nearly 8% of all American men

reported having had children with more than one partner. Despite this rather low

percentage, a different story was revealed when the researchers examined only fathers. In

this case, 17% of the men reported having had a child with at least one other mother.

Using the National Longitudinal Study of Adolescent Health (Add Health), Guzzo and

Furstenberg (2007b) found that among women aged 19-25 with a nonmarital first birth,

13

14% subsequently went on to have a second child with a new partner. Moreover, 41% of

women with two or more children reported having had their children with multiple

partners.

Research shows that multipartnered fertility is problematic because of how it

affects family formation decisions and stability. For example, multipartnered fertility acts

as a risk factor for separation among married couples and is a disincentive to sustain a

relationship among unmarried couples, particularly when they are economically

disadvantaged (Lichter et al., 2006). Moreover, when a father has a child from a prior

relationship, it acts as a risk factor for dissolution among both married and cohabiting

couples (Osborne, Manning, & Smock, 2004). Additional evidence suggests that

unmarried parents are less likely to progress into a cohabiting or marital relationship after

having had a baby when the father or mother has a child from a previous relationship (M.

Carlson, McLanahan, & England, 2004; Harknett & McLanahan, 2004; Upchurch,

Lilliard, & Panis, 2001).

The negative effects of multiple partnerships on union formation appear to differ

by gender. For women, it may be explained by a hesitation to move the relationship to a

more committed level when the partner is paying child support to another woman or is

potentially inclined to sustain old romantic ties (Edin & Kefalas, 2005). Multipartnered

fertility may be equally unattractive for a man because the choice to marry or to move in

with a woman who already has a child often means bearing the economic cost of another

man’s child (Koo, Suchindran, & Griffith, 1984; Lichter & Graefe, 2001).

Multipartnered fertility has implications for mothers, fathers, and children beyond

those related to family structure. For example, Harknett and Knab (2007) found that

14

multipartnered fertility among both mothers and fathers was associated with the

perception of reduced support from social networks in the form of housing, financial, and

childcare assistance. Other research shows that fathers’ multipartnered fertility is

associated with deleterious effects on both his prior and new family because of

ineffective parenting and an inability to cooperate with either partner, i.e., the former

mother and the new mother (M. Carlson & Furstenberg, 2007).

Despite the good intentions of lone fathers and mothers to find a stable partner

and stepparent for their child, the complex family configurations that result from

multipartnered fertility often further reduce the chances of forming an enduring

relationship and family system. This has significant effects on family formation

processes, and makes it particularly unlikely for children born to unmarried

disadvantaged parents to rise out of poverty (McLanahan, 2009).

Correlates of Relationship Instability

Prior research has shown that multipartnered fertility is one of numerous factors

affecting relationship outcomes among unmarried couples. Other factors include

demographic, socioeconomic, and relationship characteristics.

Previous research has consistently shown that family-of-origin factors, including

family structure, play an important role in predicting future family formation outcomes

(Amato & Cheadle, 2005; Teachman, 2002). Parental arrangements have been shown to

act as a template for offspring (Sassler, Cunningham, & Lichter, 2009) and evidence

suggests that growing up without both biological parents can lead to a lower chance of

getting married and developing high quality relationships later in life (South, 2001).

15

Other studies have noted the importance of race and ethnicity on the stability of

relationships. Racial and ethnic variation among cohabiting couples who eventually

separate suggests that blacks are more likely to dissolve their unions than are Hispanics

and whites (Manning et al., 2004; Osborne et al., 2004). Black couples are also more

likely to divorce than white couples during the first 14 years of marriage (Orbuch, Veroff,

Hassan, & Horrocks, 2002) and are less likely to marry than whites at all socioeconomic

levels (Lichter, Kephart, McLaughlin, & Landry, 1992). Asian and Latina women have

lower odds of dissolving their relationships than white women (Lewin, 2005).

These findings are qualified, however, by a closer look at the details of couple

relationships. Brown (2000) found that black couples were less likely to formalize their

relationships through marriage than were whites, but were also more likely to maintain a

cohabiting, stable relationship than white couples who transitioned into marriage more

quickly. This may be an indication that cohabitation and other nonmarital relationship

ties (e.g., serious dating), serve as an alternative to marriage for black couples more often

than for whites who view it as a temporary relationship state (Manning & Landale, 1996).

Age acts as a protective factor against relationship and marital dissolution (Lewin,

2005; Lichter et al., 2006; Osborne et al., 2004; Upchurch et al., 2001). The older one is

at the start of a cohabiting union for example, the less the risk of separation (Wu &

Balakrishnan, 1995). Couples who marry early, for example, have less time to search for

their best match and subsequently obtain less information about their partner. In turn, this

increases the risk of future union dissolution (Becker, 1991; Teachman, Tedrow, & Hall,

2006).

16

Although religious involvement has been found to act as a protective mechanism

against marital disruption and to increase the odds of marriage compared to staying single

or cohabiting (Lehrer & Chiswick, 1993; Stewart, Manning, & Smock, 2003), more

recent research on the role of religion in understanding union dissolution in unmarried

relationships appears mixed. For example, several studies on union stability show that

mothers’ regular attendance at church or having a specific religious affiliation had no

effect on rates of separation in cohabiting and dating relationships (Manning et al., 2004;

Osborne, 2005; Wu & Balakrishnan, 1995). Osborne and colleagues (2004) found the

opposite effect, with weekly attendance at religious services decreasing the odds of

separation for cohabiting couples.

These contradictory findings may in part be explained by the lack of detailed

information in data sets on the religiosity of both parents and on details of specific

religious affiliation that can bear on family formation decisions (Lehrer, 2004).

Information about the religious attendance of fathers is particularly important given

recent findings that joint attendance and fathers’ attendance alone (unlike mothers’

attendance alone) are predictive of higher relationship quality in unmarried and married

couples (Wilcox & Wolfinger, 2006). The findings may also be related more generally to

the differences observed in the role that religion plays in marital versus nonmarital

relationships. It is possible, for instance, that religion plays a smaller role in union

outcomes among unmarried couples because men and women who select into cohabiting

and dating relationships are less religious to start with and hold less traditional views

about marriage and family (Thornton, Axinn, & Hill, 1992).

17

A history of incarceration has also been shown to be associated with unstable

romantic relationships. Ex-offenders may be undesirable as marriage partners for various

social and economic reasons and are less likely to be engaged in normative spousal and

parenting roles (Lopoo & Western, 2005). Evidence suggests that a history of

incarceration is strongly linked to being in a non-marital (and often unstable) co-

residential relationship either with or without children (London & Parker, 2009).

Socioeconomic Factors

Men’s earnings play an important role in predicting relationship stability. Low

earnings ($10,000-$24,999) for example, are negatively related to maintaining a dating

relationship (relative to breaking up) and higher earnings ($25,000 and up) are positively

associated with marriage (M. Carlson et al., 2004). Research on the effects of income and

employment stability on relationship outcomes demonstrates similar findings: the odds of

union dissolution are higher in cases when the man has low earnings and high

unemployment (Lewin, 2005).

This economic effect among men is supported by qualitative research that

suggests women have little patience for men who are economically ill-equipped to

support their family (Charles et al., 2006; Edin & Kefalas, 2005; Gibson, Edin, &

McLanahan, 2003). This is particularly the case among black couples who highly value

the traditional role of the male’s breadwinner ability (Cutrona et al., 2003). The negative

effect of low earnings does not appear, however, to operate the same among Hispanic

couples. Mexican-American marriages are more similar in rates to whites and exceed

those of blacks despite the relatively low levels of education and earnings that many

Mexican-Americans experience (Rosenfeld, 2002).

18

Not all evidence, however, suggests that men’s income has the only effect on

union stability. Research by Osborne (2005) suggests that women’s earnings actually

play a more important role than men’s earnings in predicting marriage and separation

among unmarried couples. Specifically, she finds that women in a dating relationship

who are higher earners have ten times the odds of marrying compared to women with no

earnings. Even some earnings (less than $10,000) compared to zero earnings decrease the

odds of union dissolution for both cohabiting and dating mothers by half. In Osborne’s

study, no association was found between men’s earnings and separation or transition to

marriage.

Education for both men and women also plays an important role in relationship

stability. College education among women is associated with marriage among cohabiters

but has been found to have no effect among dating couples (Osborne, 2005). Additional

evidence suggests that college education for fathers significantly reduces the odds of

union dissolution in married and cohabiting couples and that having the same education

as the father acts as a protective effect against dissolution among mothers (Osborne et al.,

2004). Despite the protective mechanism that educational homogamy plays in reducing

union disruption, there has been a decreasing trend in marriage among couples with less

education, a phenomenon indicative of a growing social divide between those with more

and less education (Qian & Preston, 1993; Schwartz & Mare, 2005).

Homeownership, a measure of wealth, is another factor potentially associated

with family structure and relationship stability. Prior studies have found that

homeownership and assets were related to lower levels of risk for divorce in married

couples (Bracher, Santow, Morgan, & Trussell, 1993; Dew, 2008). Because of the scant

19

research on this topic that includes unmarried couples, the way in which homeownership

operates in unmarried samples is unknown. However, the assumption in the current study

is that the processes that seem to exist for married couples might hold true for unmarried

couples as well.

Relationship Characteristics

Prior research has found that the quality of a couple’s relationship is linked to

union outcomes: as anticipated, better relationship quality and lower conflict is associated

with higher odds of marriage and lower odds of separation (Osborne, 2005). The odds of

separation among cohabiting couples, for example, are higher when there is unhappiness

in the relationship, disagreement, infrequent partner interaction, and poor conflict

resolution skills (Brown, 2000). Analysis of data from the Fragile Families Study

indicates that gender distrust and supportiveness are both significant predictors of

transitions into cohabitation or marriage one year after a couple has a baby (M. Carlson et

al., 2004). Qualitative research using a subset of respondents from the Fragile Families

Study has been especially helpful in understanding issues related to trust. Hill (2007)

found that distrust and sexual jealousy from incidents of infidelity were not only quite

common but often signaled an imminent end to the relationship.

A father’s physical violence toward the mother, frequent conflict, and paternal

substance abuse have been found to be associated with a higher risk of separation

(DeMaris, 2000). Leaving an abusive relationship because of domestic violence is a fairly

consistent finding among samples of married and cohabiting women (Amato &

Hohmann-Marriott, 2007; Zlotnick, Johnson, & Kohn, 2006). Not all aggression results

in poor relationship outcomes, however. For example, DeMaris (2000) found no

20

significant impact on dissolution from verbal conflict or when the violence was

perpetrated by the woman.

The length of time a couple knows each other prior to pregnancy is also predicted

to influence the outcome of the relationship. Prior research has found that relationships of

short duration are associated with unplanned pregnancies (Bouchard, 2005). Historically,

unmarried couples who got pregnant would rush to marry to avoid the shame associated

with having pre-marital sex (England & Edin, 2007). The frequency with which this

happens today, however, is quite low and it is anticipated that short relationships, in

combination with pregnancy (especially unplanned pregnancies), are more likely to be

vulnerable to dissolution.

Finally, attitudes about marriage could play a role in the relationship outcomes of

romantic couples since marriage is a social institution commonly associated with higher

levels of relationship commitment (Riggio & Weiser, 2008). If a man or woman

expresses strong feelings about the importance of marriage and the role it plays in

children’s lives, he or she might be less inclined to separate from her partner (even her

unmarried partner) in hopes of reaping some of these rewards.

Theoretical Framework: Transition to Parenthood

Previous research has consistently demonstrated that the transition to parenthood

for first-time parents is associated with negative effects on a couple’s marital satisfaction

(C. P. Cowan & Cowan, 2000; Hawkins, Fawcett, Carroll, & Gilliland, 2006; Twenge et

al., 2003). The mechanisms underlying this process are not fully understood however,

thus Cowan and Cowan (1988) argued for better comprehension of the development of

relationship change that occurs when a baby is added to the family system. To address

21

this problem, they developed the five-domain structural model of marital and family

adaptation to improve the understanding of the process of change that occurs in a

couple’s relationship after the birth of their baby.

The basis of Cowan and Cowan’s model stems in part from Bronfenbrenner’s

ecological approach (1979), which suggests the need for a multilevel system of analysis

in order to account for the full context in which individuals function and change. As seen

in Figure 1, the five domains in the model include elements from different levels of

family organization -- the individual, dyad, triad, three generations, and surrounding

social systems.

The focus is on the connection between the elements in each level. The elements

include (a) the characteristics of each individual in the family, with emphasis on self-

concept and self-esteem (individual), (b) the quality of the husband-wife or partner

relationship, with special emphasis on their division of labor and patterns of

Individual

Dyad

Triad

3 generations

Social systems

Figure 1: Five-domain structural model of marital and family adaptation

22

communication (dyad), (c) the relationship between each parent and the child (triad), (d)

the connection between patterns in the new family system and the two families of origin

(three generations), and (e) the balance between parents’ external sources of stress and

support, with special emphasis on social networks and jobs or careers (social system).

The model is particularly useful because of its inclusion of different elements within the

family system and the attention given to the interaction across these domains. “Our model

suggests that what happens in each of the domains combines to influence satisfaction or

dissatisfaction and adaptation or distress for the individuals, the marriage, or the family

as a whole” (C. P. Cowan & Cowan, 2000, p. 123). The dynamic processes that occur

between and within the internal and external elements of the model can help explain

some aspects of the mechanism driving the effects of multipartnered fertility.

As depicted in Figure 2, if the model is modified to include “affiliate” individuals

and families, i.e., partners and children from prior relationships (plus new partners and

children that ex-partners currently have), in the outermost domain (social systems), then

multipartnered fertility becomes a key element in the system of the focal family. Take for

example, the case of a family in which the father has a child from a previous relationship

and has limited education and employment skills. Poor job prospects and depressed

wages will likely limit his ability to generate adequate income to support his current

family and to pay child support to his previous family. The combined negative impact of

the social system (work opportunities and multipartnered fertility) on both families, and

in particular on the partner relationship, suggests that re-partnering puts some couples at a

significant disadvantage for long-term stability.

23

Despite the fact that the model was originally intended for married couples with

first-time births, it serves as a useful framework for conceptualizing how multipartnered

fertility influences relationship stability in other couples as well, including unmarried

low-income minority couples with previous children. An application of the model in this

way has in fact been put to use in a recent intervention study testing the effectiveness of a

couple-strengthening program for low-income couples (C. P. Cowan, Cowan, Pruett, &

Pruett, 2007). Preliminary results indicate significant positive outcomes for participants

compared to control group couples.

Under the condition of a modified framework, it becomes clearer how the

relationship of the focal couple is subjected to significant stress when there are pre-

existing parental, romantic, and various obligatory ties to other families and individuals.

This is especially relevant when the couple has just had a baby and the father has

Individual

Dyad

Triad

3 generations

Social systems

Figure 2: Modified five-domain structural model of couple and family adaptation

Father’s MPF Family: prior mother + child +

(her new male partner + any children they have)

Mother’s MPF Partner: prior father + (his new female partner + any children they have)

24

multipartnered fertility. Under these circumstances, the couple is likely to face significant

challenges given the resource demands (e.g., time, money, emotion, physical presence)

that exist across households. This can impart stress to all the children and partners

involved, leaving the focal couple with the newborn baby at higher risk for union

dissolution. The overall research question and hypotheses for this study were generated

with this framework in mind, as well as an understanding of the correlates of relationship

dissolution.

CHAPTER 3

RESEARCH METHODS

Research Aims and Hypotheses

The purpose of this study is to explore the timing of relationship dissolution and

the factors associated with its occurrence among couples who just had a baby together.

The primary research question is: “To what extent does multipartnered fertility influence

whether and when a couple divorces or separates when they are married, cohabiting, or

dating?” Several stages of analysis address the research question. First, the extent to

which couples end their relationship in the period following the birth of their baby by

initial relationship status is examined. Second, the pattern of leaving the relationship over

time and tests of group differences according to multipartnered fertility history are

explored. Third, survival analysis is employed to evaluate the association between union

break up and multipartnered fertility by the father, mother, or both parents compared to

neither one of them. Lastly, predicted probabilities are used to examine the pattern of

relationship stability according to multipartnered fertility history and current relationship

status with model-predicted survivor curves.

Within the framework of the primary research question, there are five hypotheses

that will be tested in this study:

1. Couples with multipartnered fertility have a faster rate of separation than couples

without multipartnered fertility.

26

2. Cohabiting and dating couples have a faster rate of separation than married

couples.

3. Couples with multipartnered fertility from the father face greater risk for

dissolution relative to couples with multipartnered fertility from the mother.

4. Economically disadvantaged couples are at a higher risk for relationship

dissolution than couples who are more economically stable.

5. Couples with higher levels of relationship supportiveness and lower levels of

violence and conflict face a reduced risk for dissolution relative to couples with

poor relationship quality.

Hypothesis 1 will be tested using Kaplan-Meier estimates to obtain the survivor

function of the sample according to respondents’ multipartnered fertility status.

Hypothesis 2 will also be tested using Kaplan-Meier estimates to examine the survivor

function of respondents according to their relationship status at birth. After visual

inspection of the survivor curves, both hypotheses will be further verified using the

Generalized (Breslow Wilcoxon) Test to test for significant differences of the median

survivor function. If this is not possible because more than 50% of the couples remain in

their relationship by the end of the study window, then differences between groups at

some other percentile (e.g., 75th, 85th, 90th ) will be tested instead. The last three

hypotheses will be tested using discrete-time survival models examining the effect of

multipartnered fertility and covariates on the risk of dissolution.

Data

This study uses data from the first three waves of the Fragile Families and Child

Wellbeing Study to explore the timing of relationship instability among parents with a

27

newborn child. The Fragile Families Study is a longitudinal stratified random sample of

hospital births in 20 large U.S. cities. The data were collected using a stratified multistage

clustered sample design where respondents were not selected independently or with equal

probability. Unmarried mothers were over-sampled to allow for a greater focus on births

to vulnerable populations. The first stage of selection was the city, the second stage

hospitals, and the third stage births. The sampling frame consisted of 77 cities with

populations of more than 200,000 people. These cities were then grouped into 9 different

strata according to their policy environments (i.e., generosity of welfare and child support

enforcement) and local labor market conditions (Reichman, Teitler, Garfinkel, &

McLanahan, 2001). One city was selected from the first 8 strata and 8 cities were selected

from the 9th strata, totaling 16 cities in the national sample. Four additional cities were

selected for special reasons and cannot be included in analyses weighting back to the

national sample. From within the national sample, one or more hospitals with high rates

of nonmarital births were identified; a total of 75 hospitals were selected. Finally, births

in these hospitals were selected until a certain sample size and sampling rate was met

(about 75% for unmarried mothers and 25% for married mothers). When weighted, the

data are representative of nonmarital births (and nearly representative of marital births) in

U.S. cities with populations over 200,000 in 1999 (B. L. Carlson, 2008).

Between 1998 and 2000, approximately 3,500 unmarried mothers and 1,500

married mothers were interviewed in the hospital immediately following their child’s

birth. For the majority of births, both mothers and fathers were interviewed within three

days of delivery. The Fragile Families study consists of interviews at birth in addition to

28

several follow-up waves. Interviews following the baseline survey were conducted in

person and over the telephone.

The present study utilizes the baseline interview (i.e., wave 1), and two

consecutive waves of data collected when the child was approximately one and three-

years old (i.e., wave 2 and wave 3). The wave 2 and wave 3 interviews were conducted

between 1999-2002 and 2001-2003 respectively (Center for Research on Child

Wellbeing, 2008). The surveys cover topics in the following eight areas: child health and

well-being, parent-child and mother-father relationships, demographic characteristics,

marriage attitudes, family background, health, religion, and socioeconomic

characteristics.

At baseline, response rates for eligible mothers and fathers approached in the

hospital were 87% for unmarried mothers, 82% for married mothers, 75% for unmarried

fathers, and 89% for married fathers. The baseline dataset included 4,898 completed

mother and 3,830 completed father interviews. Across the three waves of the study, 86%

of fathers were interviewed at least one time; 82% of mothers and 55% of fathers were

interviewed at all three waves (Center for Research on Child Wellbeing, 2005a).

Sample

From the initial 4,898 interviews, this analysis is limited to the 4,245 mothers who

were married or romantically involved at the baseline interview. In order to focus on

couples in a romantic relationship, three categories of observations were omitted from the

baseline sample: mothers who reported being friends with their partner or not talking to

him, mothers whose child’s father was unknown, or relationships for which union status

was missing (N=653). Of these 4,245 cases, 396 mothers and fathers with missing

29

interviews at wave 3 were dropped, resulting in a sample of 3,849. The sample was

further reduced to 3,819 when 30 observations with event dates before the start of the

study were excluded2. Finally, due to missing values on the dependent and independent

variables the analytic sample was reduced to 3,022 mothers. The sample includes 813

(27%) married mothers, 1,329 (44%) cohabiting mothers, and 880 (29%) dating

mothers3, (i.e., romantically involved with, but living apart from, the baby’s father).

Missing values on the independent variables ranged from 0% to 7.04%. Bivariate

tests (chi-square for categorical variables and Adjusted Wald Test for continuous

variables) using the weighted data were conducted to determine whether excluded

couples differed significantly from couples included in the analytic sample4. Fortunately,

the included cases did not differ from the excluded cases on most variables except in

three instances. The first was for fertility history (7.04% missing) where approximately

20% of fathers with multipartnered fertility were excluded compared to 15% when both

parents had multipartnered fertility, 12% when mothers had multipartnered fertility, and

6% when neither had multipartnered fertility F(2.41, 77.12) = 6.75, p < .01. Second,

mothers who reported substance abuse (.21% missing) were more likely to be missing

than mothers who reported not having a substance abuse problem (48% vs. 15%) F(1,

145) = 5.56, p < .05. Third, mothers who were dropped from the sample had on average a

shorter relationship prior to pregnancy than mothers not dropped from the sample (.79%

missing) (5.4 years vs. 7.4 years) F(1, 32) = 12.96, p < .01. Some caution should be used

2 These were omitted since event dates that precede the onset of the relationship would result in a “negative duration” of time to event. 3 In other studies using Fragile Families data, researchers usually refer to this group as “visiting.” 4 When a test of independence is conducted in Stata using weighted data, the test is based on the usual Pearson χ2 statistic for two-way tables. However, to account for the survey design, the statistic is converted into an F statistic with a correction.

30

when interpreting results from this study because of these differences. Specifically, the

study sample’s lower proportion of fathers with multipartnered fertility, mothers with a

substance abuse problem, and couples with shorter relationship duration prior to

pregnancy all need to be taken into account because of the limitation this may cause with

regard to making generalizations about the study’s findings.

Censoring

There are two types of censoring problems that occur in event history analysis –

left and right censoring. Censoring occurs when the time to event is unknown for some

portion of the sample. Guang Guo (1993) defines left-censoring in the following manner:

“a left-censored subject is known to have experienced the event, but the exact failure time

is unknown,” (p. 220); in survival analysis, the event of interest occurs before the start of

the observation period. Right-censoring occurs when the event of interest happens to a

subject after the end of the observation period or because of a loss to follow-up during

the study window, e.g., the subject drops out of the study for some reason (i.e., death,

refusal to participate, or failure to locate the subject at follow-up interviews).

Censored cases (N=1,655) are defined in this study as those respondents whose

exact time to event (relationship dissolution) is unknown because it did not occur during

the study window. These are right-hand censored cases because we only know that the

event has not occurred by the end of the study window and may or may not occur after

the end of the observation period. All censored cases with a valid mother interview at

wave 3 are assigned the interview date at wave 3 as the censoring date. If the mother’s

wave 3 interview is missing, I supplement with the father’s interview date if available.

31

Random censored cases are defined as those study subjects whose starting and

ending points are known but the ending date occurs before the end of the observation

period for a reason other than the event of interest. There are two types of random

censoring in this study; the first is when fathers die before the end of the study. For these

cases, the month and year of the death was used as the censoring date because the

censoring is noninformative and thus does not introduce serious bias into the model

estimates (S. Guo, Forthcoming, p. 24). These cases are retained in the analytic data set

(N = 27). The other form of random censoring is due to attrition, i.e., the mother and

father have interviews at wave 1 and wave 2 (or just at wave 1) and then drop out of the

study for an unknown reason before having a follow-up interview at wave 3 (N = 396).

More specifically, there were 206 cases with missing interviews at waves 2 and 3, and

190 cases with missing interviews at wave 3 only. These cases are treated as missing and

are dropped from the analytic sample because they run the risk of being informative; that

is, there may be some systematic pattern to study subjects who drop out early that is

related to relationship dissolution.

There is some portion of the Fragile Families sample that is left-censored; that is,

they experienced a relationship change before coming under observation. For example,

women in some of the couples reported not being in any relationship at all with the father

of the baby at the time of the baseline survey (N = 653). These cases are considered left-

censored because the relationship change (in this case separation) had already occurred.

As recommended by Paul Allison (1984), these cases can be excluded without

introducing significant bias and were omitted from the analytic data set.

32

Dependent Variable, Origin of Time, and Study Window

The event of interest in this study is whether the couple divorces or separates if

married, cohabiting, or dating. The dependent variable was created with two pieces of

information: the length of time or duration of the relationship and a dichotomous measure

indicating whether the event occurred or the case was censored (1=divorced/separated,

0=censored). Censored cases are those respondents who sustained their relationship until

the end of the study period.

The origin of time is determined by the date of the baseline interview for the

mother. The interview date was selected as the origin of time for two reasons: 1) the real

origin of time is unknown, and 2) the focus of the study is change in the relationship

following the transition to parenthood. Since interviews typically occurred in the hospital

within 24-48 hours of the birth of the baby, I use the interview as the origin of time and

examine change in the relationship from that point forward. An alternative to using the

interview date as the origin of time would be the use of a measure in the data set that

indicates the length of time the couple knew each other prior to the pregnancy. However,

this is used as a covariate instead of as a proxy for the relationship duration since it is not

possible to verify that this time is the actual length of the romantic relationship.

The unit of time to relationship change is analyzed as a discrete-time measure

calculated in the metric of months. Thus, the “duration” variable was created using the

number of months from entry into the study to the point of relationship dissolution or

censoring. The study window goes from the baseline interview for each study subject to

wave 3 of the study. The study window is 50 months long. For descriptive purposes (e.g.,

survivor functions), only the first 47 months are utilized because the last event is

33

observed in month 47; all 50 months are used in the discrete-time models. Because time

dummies for each month were necessary in the discrete-time models, the months were

grouped into thirteen intervals of three months each, to avoid the use of 49 dummy

variables (plus a reference group) in the discrete-time model. The choice of using 3-

month intervals, plus one longer interval, was based on careful examination of the data

and the distribution of the number of months contributed to the data set by respondents.

Three months are included in each of the first 12 intervals (i.e., months 1-3 = interval 1,

months 4-6 = interval 2, and so on,…months 37-50 = interval 13). The l3th interval

includes months 37-50 because of the smaller number of observations in those months

than in earlier months. This coarser way of categorizing the data serves to build a more

parsimonious model while retaining the full range of requisite months. The first interval,

months 1-3, is used as the reference group in the discrete-time models.

Independent Variables

The independent variables were selected on the basis of theoretical relevance, as

well as previous research and include fertility history, demographic factors,

socioeconomic characteristics, relationship status and quality.

Fertility History

Several questions about the parents’ childbearing outside the relationship are used

to determine multipartnered fertility. I use both mothers’ and fathers’ reports about

fertility status and use mothers’ reports about her partner’s fertility history in cases where

it is missing in order to preserve sample size since there are fewer fathers in the study. A

multiple category variable is used to capture four types of multipartnered fertility cases:

1=no MPF (reference), 2=mother-only MPF, 3=father-only MPF, and 4=both MPF.

34

Demographic Factors

For all father variables, the father’s information is used wherever possible and

supplemented with mothers’ reports about the father in order to maintain as large a

sample as possible. The mother’s race is specified as three categories: 1=White non-

Hispanic (reference), 2=Black non-Hispanic, 3=Hispanic/other race. A dummy variable

is used to indicate whether the father’s race differs from the mother’s race. Age of the

mother at the time of the baseline interview is modeled as a continuous variable. A

dummy variable is used to indicate if the father’s age differs from the mother’s age by 10

years or more because of the high correlation between mothers and fathers ages. Mothers’

religious involvement is measured using a dichotomous variable indicating frequent or

infrequent attendance at religious services. Frequent involvement was defined as

attending religious services at least several times a month and infrequent as attending

services several times per year or less. A dummy variable indicating whether the mother

has a substance abuse problem is based on the question, “In the past year, has drinking or

using drugs ever interfered with your work on a job or with your personal relationships?”

For fathers, a dummy variable is used to indicate whether the mother reports that he was

ever in jail. Family background is represented by a dichotomous variable indicating

whether the mother lived with her biological parents at age 15.

Socioeconomic Characteristics

Maternal education was specified as four different categories: 1=less than high

school, 2=high school or GED, 3=some college, or 4=college graduate (reference). Three

dummy variables, with college graduate as the reference group, were used in the models.

Due to high correlations between maternal and paternal education variables, only those

35

from the mother are included. Work status is measured with a dummy variable indicating

1 if the mother worked in the year before the birth and 0 otherwise. The father’s version

of this variable differs slightly; work status is defined by whether the father worked in the

week prior to the birth, as opposed to the year before the birth. The welfare status of the

mother is captured using a dummy variable if the respondent received income from

public assistance, food stamps, or welfare in the previous year. Finally, a dichotomous

variable indicates whether the mother or father owns versus rents the home they live in.

Relationship Status and Quality

A multiple category variable indicating whether the couple is married, cohabiting,

or dating is used to capture the relationship status of the couple at the time of birth:

1=married (reference), 2=cohabiting, 3=dating. A continuous variable measured in years

is used to capture the time that the couple knew one another prior to the birth of the baby.

Relationship quality is based on a series of variables widely used by other researchers

(M. Carlson et al., 2004; Osborne, 2005) to measure both negative and positive aspects of

the relationship at the time of birth: emotional supportiveness, conflict, violence, distrust

of the opposite sex, and attitudes about marriage. The measure of emotional support is a

scale that was based on the mean of maternal reports of how frequently (1=often,

2=sometimes, 3=never) their partners: 1) were fair and willing to compromise (reversed),

2) expressed love and affection (reversed), 3) insulted and criticized them, and 4)

encouraged and helped them (reversed) (α = .67). Responses were recoded so that a high

value indicates a high level of support. These are similar to items included in Straus’

(1979) “Reasoning” and “Verbal Aggression” scales in the Conflict Tactics Scale.

36

Conflict is represented as the mean of a six-item scale regarding the frequency

with which mothers disagreed with fathers in the last month before the birth about

money, spending time together, sex, the pregnancy, drugs or alcohol, and being faithful

(1=often, 2=sometimes, 3=never) (α = .64). Responses were all reverse coded so that a

high value indicates more frequent conflict.

Physical violence is based on how frequently both parents report being hit or

slapped during their relationship prior to the birth of the baby (1=often, 2=sometimes,

3=never). Because even sometimes hitting or slapping a partner is significant, the

variable is dichotomized to indicate any reported (i.e., often or sometimes) physical

violence. Distrust of the opposite sex is measured using two items that women report on:

1) “Men cannot be trusted to be faithful,” and 2) “In a dating relationship, a man is

largely out to take advantage of a woman.” Responses ranged from 1=strongly disagree

to 4=strongly agree, with higher values reflective of a distrustful perspective. This

measure was dichotomized so that responses of 3 or 4 (i.e., agree or strongly agree)

indicated distrust of the opposite sex.

The measure of attitudes about marriage is based on two items that ask mothers

about their level of agreement with the following statements: 1) “It is better for a couple

to get married than just live together” and 2) “It is better for children if their parents are

married.” Responses ranged from a low of 1=strongly disagree to a high of 4=strongly

agree, with higher values (3 or 4) indicative of a positive attitude toward marriage. This

variable was also dichotomized so that responses of 3 or 4 reflected a pro-marriage

attitude.

37

Data Management

Using information about the study window, the duration of the relationship,

censoring, and the event, the original dataset which was in person-level format (one row

of data per observation, N = 3,022) was converted into person-time format (multiple rows

of data for each respondent, N = 106,293). For example, a couple who came under

observation at wave 1 (month 1) and sustained their relationship until wave 3 (month 36)

would contribute 36 rows of data. A couple whose relationship ended after one and a half

years (or 18 months) would contribute 18 rows of data. The difference in these two cases

is that the second couple experienced the event while first couple was censored.

The dependent variable (1=event, 0=censored) is structured such that it attributes

the event to the respondent only in the period in which the event actually occurred. Using

the 18-month case example, the dependent variable would be coded 0 for the first 17

months and then coded 1 (event) in the 18th month (or in the 18th row). For the couple

whose relationship lasted 36 months (i.e., censored because they sustained their

relationship for the entire study window), the dependent variable would be coded as 0 on

all 36 rows of data.

Data Analysis

Descriptive Analyses

Univariate and bivariate analyses were conducted to describe the study sample.

Variables reflecting fertility history, demographics, socioeconomic status, and

relationship quality and characteristics are displayed in Table 1 using weighted

percentages to reflect births to unmarried women in large U.S. cities. Because public use

data files are utilized in this study, replicate weights provided in the data are used in place

38

of stratum and primary sample unit variables that would otherwise be used to estimate

variances. Thus weights are used to account for attrition and for the fact that respondents

were not selected independently or with equal probability. Adjustments were made for

clustering using the svy commands in Stata.

Bivariate analyses on selected variables of importance are presented in Table 2 (in

Chapter 4) using the 90th percentile of the survivor function based on Kaplan-Meier

estimates. The 90th percentile was selected because more than 50% of the couples

remained in their relationship by the end of the study window thus making the median

survival time meaningless. The Breslow (Generalized Wilcoxon) Test was used to assess

for statistically significant differences across groups. The stci procedure in Stata 10 was

used to obtain the 90th percentiles and the sts test with the wilcoxon option was used to

test for group differences.

Kaplan-Meier estimates were used to describe and explore the events of interest

by obtaining plots of the hazard functions, survivor functions, and by conducting tests of

group differences. The following explanation of hazard and survival functions provides a

formal description of these two important functions.

The study times for the subjects in the Fragile Families sample form a

distribution, known as the survival distribution (S. Guo, Forthcoming). Various functions

characterize survival distributions that serve as the basis for survival analysis. These

include: the hazard function h(t), probability density function (PDF) f(t), cumulative

distribution function (CDF) F(f), and survivor function S(t) (Allison, 1995). If T denotes

the time to the event (i.e., time when the couple “fails” or separates), these four functions

describe the probability distribution for T. Given one of these four functions, the other

39

three can be determined (Cleves, Gould, & Gutierrez, 2004a), and thus for convenience

purposes, the following description of the survival distribution focuses on the survivor

S(t) and hazard h(t) functions only.

The hazard function, also referred to as a rate, is perhaps the most central concept

in survival analysis and is defined as the instantaneous rate of failure or the instantaneous

risk that an event will occur at time t. The hazard is a quantity that takes the form of the

number of events per interval of time (similar to “miles per hour”), which is why it is

referred to as a rate (Allison, 1995). It is expressed as the instantaneous probability that

the event will occur in a specific and small interval of time, conditional upon the subject

having survived up to the beginning of the interval, divided by the time interval:

t

tTttTtth

t ∆≥∆+<≤

=→∆

}|Pr{lim)(

0

The numerator of the hazard function is the conditional probability of the event

occurring at time t + ∆t. In other words, assuming that the event has not yet occurred at

time t, what is the probability of the event occurring in the time interval of t + ∆t (S. Guo,

Forthcoming)? The hazard function can vary from zero (meaning no risk at all) to infinity

(meaning the certainty of failure at that instant) (Cleves et al., 2004a). The hazard rate

can exceed 1, although cannot be less than 0. Over the study window, the hazard function

can increase, decrease, remain steady, or change shape. The risk of event occurrence

follows the shape of the hazard function: when the risk is zero, the hazard is zero; as risk

rises with time, so does the hazard; as risk decreases, the hazard drops; if the risk is

constant, then the hazard for that particularly instant is the same.

The survivor function, S(t), measures the probability of surviving beyond time t:

)(1}Pr{)( tFtTtS −=≥=

40

The survivor function is the reverse of the cumulative distribution function, i.e., F(t) =

Pr(T < t). It provides the probability that there is no event prior to time t. The function is

equal to one at t = 0 and decreases toward zero as t goes to infinity (Cleves et al., 2004a).

An illustration of the overall length of time couples “survive” in their relationship

is presented in Figure 3 (in Chapter 4), as well as survivor curves according to

multipartnered fertility history and relationship status at birth (Figure 5 and Figure 6

respectively in Chapter 4). The sts graph procedure was used to obtain these plots.

The estimated hazard function for the overall sample and by multipartnered

fertility status is also presented. Note that the hazard functions were not estimated using

Kaplan-Meier estimates, but rather the life-table method and a kernel smoothing

procedure conducted within Stata (S. Guo, Forthcoming; StataCorp, 2007). The sts

graph, hazard procedure was used to obtain the hazard curves.

Event History Analyses

Event history analysis, and specifically, discrete-time models (Allison, 1982)

were then used to estimate the effects of multipartnered fertility history and a series of

covariates on the relationship outcomes of married, cohabiting, and dating couples.

Discrete-time survival analysis was selected to help answer the question of why

relationships change at different times for different couples and specifically to answer:

“What is the relationship between the risk of event occurrence in each time period and a

set of predictors?”

The dependent variable of interest is the discrete-time hazard of divorce or

separation for married, cohabiting, and dating couples. Couples that do not experience a

change in the relationship by the end of the study window are considered censored. The

41

logistic procedure in Stata 10 was used for the binomial regression estimations. To

address the research question, all changes to the relationship are treated the same.

Hazards are calculated to express the relative risk of divorce/separation for each

independent variable. For example, the hazard for welfare use is the relative risk of

divorce/separation for mothers on welfare versus mothers not on welfare, controlling for

all other covariates.

The discrete-time models use a binary logit model with pooled event histories,

and the probability of event occurrence as a proxy for the hazard rate, to estimate the

effect of predictor variables on the probability of relationship dissolution (Allison, 1982;

S. Guo, Forthcoming). Study participants contribute person-months to the data until they

experience the event or are censored. The model takes the following form:

iitit

it ZXP

P''

1log 21 ββα ++=

−

where P is the probability of dissolution given that couple i has not separated prior to

month t. αt is a set of t-1 dummy variables used to control for time dependence. Xi is a

vector of multipartnered fertility history variables, and Zi is a vector of time-invariant

control variables.

A total of four models were analyzed. The key predictor variable, multipartnered

fertility, is included in all the models. Model 1 also includes relationship status and the

time dummies. Model 2 adds a set of demographic control variables, and Model 3 adds

variables related to socioeconomic status. The fourth and final model adds relationship

characteristics. The selection of variables in Model 4 was based on several steps of model

building, including careful examination of individual predictors and their contribution to

42

the model, strength of theoretical significance, and percentage of missing. The preferred

model (Model 4) is the result of this process.

Finally, model-predicted survivor curves based on predicted probabilities are

presented to summarize trends over time for values from multipartnered fertility history

and relationship status at birth controlling for all other covariates at their mean. The

model-predicted survivor curves are based on a slightly modified version of Model 4

from the discrete-time survival analysis. The primary predictor variable, multipartnered

fertility, was re-categorized into a dummy variable (1=any multipartnered fertility,

0=none) in order to limit the number of curves that would be produced in the plot. In this

way, only six curves are presented: married/no MPF, married/MPF, cohabitation/no

MPF, cohabitation/MPF, dating/no MPF, dating/MPF.

The use of discrete-time survival analysis is particularly advantageous for a

number of reasons. First, it is suitable for the analysis of data collected in settings where

continuous data are not available because of cost and logistical restrictions. Second,

discrete-time survival analysis does not require special software; estimations can be made

using traditional software packages and relatively simple analysis techniques such as

logistic regression. Finally, it is a relatively intuitive approach and lends itself to research

with an application focus (Allison, 1982; Willett & Singer, 1993).

Model Evaluation and Diagnostic Procedures

Interaction Terms

Three sets of interaction terms were tested to evaluate improvement in model fit.

The terms were selected based on substantive relevance: relationship status X

multipartnered fertility, relationship status X race, welfare X multipartnered fertility. All

43

the interaction terms were non-significant at the p < .05 level. The interaction terms were

tested one at a time in the preferred discrete-time model. Once it was determined that the

term was non-significant, it was taken out and replaced with the next term, and so on

until each of the three interaction terms was tested. Since inclusion of interaction terms in

a model that is not significant at traditional levels of statistical significance tends to

increase the standard errors without changing the point estimates, the terms were omitted

(Hosmer & Lemeshow, 2000).

Multicollinearity

Once the final selection of independent variables was complete, a test for

multicollinearity was conducted by examining the variance inflation factor (VIF). The vif

procedure in Stata was used for this assessment. Because VIF was tested using ordinary

least squares regression, the nature of censoring on the dependent variable was ignored in

this procedure. None of the VIF values exceeded 10, suggesting that multicollinearity

was unproblematic (Kutner, Nachtsheim, & Neter, 2004). The VIF values ranged from a

low of 1.03 (mothers’ substance abuse) to a high of 3.64 (less than a high school diploma

for mothers) with an average VIF of 1.56.

Model Assessment

Three measures of model fit were used to evaluate the proposed models. As seen

at the bottom of Table 3 (in Chapter 4), likelihood ratio tests were used for overall model

evaluation and model chi-square and pseudo-R2 were used for goodness-of-fit. Generally,

a higher pseudo-R2 suggests better model fit. Pseudo-R2’s, however, should only be used

when predicting the same outcome on the same dataset with the same predictor variables

(UCLA Academic Technology Services, 2008). Unfortunately, there are no available

44

corrective measures for these indices for use with survey data; thus, they should be

interpreted with some caution.

Influential Observations

A graph of Cook’s D was obtained for all the independent variables in order to

assess the influence on parameter estimates of each individual observation. As seen in

Appendix A, the plot indicates that only one or two cases are apart from the rest of the

observations and thus their influence is likely to be minimal, so they were retained in the

sample.

45

CHAPTER 4

RESULTS

Univariate and Bivariate Results

Descriptive statistics for the entire sample and separately by initial relationship

status are reported in Table 1. Dating couples were much more likely than cohabiting and

married couples to end their relationship, 70.4% versus 35.6% and 13.3% respectively.

Cohabiting and dating couples have considerably higher rates of multipartnered fertility

than married couples overall. Married mothers are more often white, whereas dating and

cohabiting mothers are more often black and Hispanic/other race, respectively. In

approximately 15% of the sample the father’s race differs from the mother’s race.

Cohabiting mothers are the least religious among all couple types.

The mean age of mothers is 27 years old. The mean age of dating mothers (22.6)

is lower than it is for cohabiting (24.6) and married (29.5) mothers. A small percentage

(8.7%) of fathers differ in age from the mother by 10 or more years. Mothers who date

have higher rates of substance abuse than cohabiting and married mothers. Overall, more

than half the mothers (55%) lived with their own parents at age 15.

Approximately 23% of the mothers have less than a high school diploma whereas

30% have a high school diploma or GED. Seventy-three percent of all mothers worked in

the year prior to the baby’s birth while 90% of fathers worked in the week before the

birth. Approximately 46% of dating mothers and 41% of cohabiting mothers received

46

Table 1. Descriptive Statistics for Couples by Rel ationship Status

Total N=3,022 Percent /

Mean (SD)

Married N=813

Percent / Mean (SD)

Cohabiting N=1,329 Percent /

Mean (SD)

Dating N=880

Percent / Mean (SD)

Dependent VariableRelationship DissolutionSeparation 24.9 13.3 35.6 70.4No change (censored) 75.2 86.7 64.4 29.7Independent VariablesMultipartner Fertility HistoryNo MPF children 66.5 78.2 42.7 43.0MPF (mother) 10.1 6.6 18.6 14.8MPF (father) 13.4 10.4 20.2 18.4MPF (both) 10.1 4.9 18.6 23.8DemographicsMother's race

White 42.7 51.9 28.6 16.3Black 20.0 10.2 31.3 55.3Hispanic/other 37.2 37.9 40.1 28.4

Father's race differs from mother 14.7 13.6 17.2 16.2Mother's religiousness 40.0 43.4 29.0 40.0Mother's age 27.6(5.9) 29.5(3.4) 24.6(7.6) 22.6(8.6)Father is 10 years older/younger than mother 8.7 6.9 11.2 14.2Mother's substance abuse problem 1.1 0.3 1.9 4.0Mother lived with both parents at age 15 55.3 63.5 43.2 31.0Father ever in jail (mother's report) 16.1 7.7 31.6 35.6Socioeconomic StatusMother's education

<High school 23.4 15.9 36.4 43.0High school/GED 30.4 24.6 44.0 38.8Some college 19.8 21.1 18.1 15.9College or more 26.4 38.5 1.5 2.3

Mother worked in year before birth 73.0 74.7 73.4 62.7Father worked in week before birth 90.4 95.8 84.0 71.6Mother on welfare 22.0 11.9 40.6 45.7Mother or father owns home 47.2 57.2 21.8 35.6Relationship CharacteristicsYears known each other 7.4(5.6) 9.0(3.4) 4.3(6.0) 3.7(5.4)Mother feels supported (1-3) 2.7(.3) 2.7(.2) 2.7(.4) 2.6 (.5)Mother reports conflict (1-3) 1.4(.3) 1.3(.2) 1.4(.5) 1.5(.6)Mother reports violence 1.6 1.8 1.2 1.5Mother distrusts men 20.6 17.3 24.3 32.8Mother has pro-marriage attitude 85.0 90.8 73.8 71.9Note: N's are unweighted; percentages and means are weighted.

welfare assistance, compared to only 12% of married mothers. A considerable proportion

of respondents also owned their own home (47.2%).

47

Table 1 shows that parents who live together or date have known each other much

shorter periods of time than married parents (4.3% and 3.7% vs. 9%). Overall, mothers

report high levels of support from the father (2.7 on a scale of 1-3) and low levels of

conflict (1.4 on a scale of 1-3). Less than 2% of mothers reported physical violence

perpetrated by the father. Dating mothers have higher rates of distrust toward men (33%)

than cohabiting (24%) and married mothers (17%). Finally, all couple types tend to have

rather high rates of pro-marriage attitudes ranging from a low of 72% among dating

mothers to a high of 91% among married mothers.

Table 2 presents the 90th percentiles of the survivor function for selected key

variables, including multipartnered fertility, relationship status at birth, race, education,

welfare, and violence. The 90th percentiles for multipartnered fertility are 8 months for

couples without MPF, 4 months for mothers with MPF, and 3 months each for fathers

with MPF and for couples that both have MPF. The 90th percentile can be interpreted as

the length of time it takes for 10% of the couples to divorce or separate. For example, it

takes longer (8 months) for couples without MPF children to end their relationship than

for couples in which both the mother and father have MPF (3 months). This finding

supports Hypothesis 1 by showing that couples with multipartnered fertility have faster

rates of separation than those without multipartnered fertility. The Breslow Test shows

that these differences in the survivor functions for MPF are significant, χ2 (3, N = 1165) =

158.79, p < .001. Despite these initial findings, the bivariate tests do not control for other

covariates as will be done in the multivariate modeling procedure.

48

Additional findings from the 90th percentile of the survivor function indicate that

married couples sustain their relationship much longer than both cohabiting and dating

mothers, χ2 (2, N = 1165) = 569.05, p < .001. This supports Hypothesis 2, which

predicted that non-married couples would have a faster rate of dissolution than married

couples.

VariableMultipartner Fertility HistoryNo MPF children 9 ***MPF (mother) 4 ***MPF (father) 3 ***MPF (both) 3 ***Relationship Status at Child's BirthMarried 33 ***Cohabiting 6 ***Dating 3 ***DemographicsMother's race

White 12 ***Black 3 ***Hispanic/other 5 ***

Socioeconomic StatusMother's education

<High school 3 ***High school/GED 3 ***Some college 6 ***College or more 30 ***

Mother on welfare 3 ***Relationship CharacteristicsMother reports violence 3 ******p < .001, Breslow (Generalized Wilcoxon) Test

Table 2. 90 th Percentile of the Survival Function of Selected Variables

Number of Months

Additional results show that white mothers compared to minority mothers sustain their

relationships longer (12 months versus 3 and 5 months for black and Hispanic/other race

respectively) χ2 (2, N = 1165) = 181.37, p < .001. Mothers with a college education tend

49

75th Percentile of the Survivor Function (Month 16)

0.00

0.25

0.50

0.75

1.00

Sur

vivo

r F

unct

ion

0 10 20 30 40 50Months

* Based on unweighted data

Survivor Function of Relationship Status*

to sustain their relationships far longer than those with less education (31 months for

college goers versus 3 or 4 months for every other category) χ2 (3, N = 1165) = 150.38, p

< .001. Relationships end earlier among mothers on welfare compared to those not

needing assistance (3 months versus 5 months), χ2 (1, N = 1165) = 88.3, p < .001. Finally,

when violence is reported by the mother, the relationship tends to end sooner (3 months

versus 4 months) χ2 (1, N = 1165) = 15.3, p < .001.

Survivor and Hazard Functions

In Figure 3, we see the survivor function for the entire sample of couples. The

survivor curve shows that at the start of the study window, or time 0, all couples are in

their relationship with no events of dissolution. Over time, the estimates indicate a

Figure 3. Survivor function of relationship dissolution

50

steadily decreasing rate of remaining in a relationship after having a child. In month 3,

the estimated probability that a couple will remain in the relationship drops from .98 to

.90, the largest decline observed within a one-month span.

As indicated earlier, the median survival time is not observable, meaning that

more than 50% of the couples sustained their relationship by the end of the study

window. In month 16, however, we observe the 75th percentile of the survivor function,

indicating that three-fourths of the sample have “survived” or remained in the

relationship up to that time.

Figure 4 presents the hazard plot of relationship dissolution for all couples in the

study sample. The hazard function provides information about the speed or rate of change

for the event of interest.

Figure 4. Hazard function of relationship dissolution

.01

.012

.014

.016

Haz

ard

Fun

ctio

n

0 10 20 30 40

Months* Based on unweighted data

Smoothed Hazard Estimate of Relationship Dissolution*

51

Notice the high spike at the beginning of the window; this indicates a fast speed of

change at that time. This spike corresponds to the large drop observed in the survivor

function around month three. Then as the risk of dissolution declines over time, the

hazard drops. Starting at approximately month 24 or two years after birth, there is a slight

increase in the hazard for six months, after which it declines again. While the hazard plot

tells a similar story to the survivor function, it also suggests that the speed with which

relationships change is not constant over the study period.

In Figures 5 and 6, we see the survivor function again but this time according to

multipartnered fertility history and relationship status, respectively. Figure 5 presents

survivor curves to provide a sense of the sustainability of relationships according to the

study’s primary predictor variable: multipartnered fertility history.

Figure 5. Survivor function by multipartnered fertility history

0.00

0.25

0.50

0.75

1.00

Sur

vivo

r F

unct

ion

0 10 20 30 40 50Months

No MPF Mother MPFFather MPF Both MPF

Survivor Function by Multipartner Fertility

52

The curves are very similar in the first three months after birth followed by a decrease for

couples when they both have MPF and when only the father has MPF. The survivor

function for couples without MPF (the top curve) suggests that they maintain their

relationship more than the other couple types. On the whole, those with children from

previous relationships are most at risk for separation compared to those couples who

share only biological children together, χ2 (3, N = 1165) = 158.79, p < .001.

In Figure 6, the curve for married couples (top) reflects the slowest rate of change

while the curves for cohabiting (middle) and dating (bottom) indicate faster speeds of

change. Up to month three, the survivor curves appear similar with little evidence of

difference by relationship status.

At month three, however, there is a significant drop in the survivor curve for

dating couples. The median survival time for this group is month 20, meaning that 50%

of the cases separated by month 20 and 50% sustained their relationship beyond this

point. The plot clearly shows that the proportion of couples who successfully sustained

their relationship the most are married, followed by cohabiting couples, and finally dating

couples. The 90th percentile of the survivor function across these three groups differs

significantly, χ2 (2, N = 1165) = 569.05, p < .001. Visual inspection of the survivor curves

in Figures 3 and 4 provide additional support for the first two hypotheses; that is, MPF

couples appear to be at greater risk for dissolution relative to non-MPF couples

(Hypothesis 1) and further, nonmarried couples are at greater risk for separation than

married couples (Hypothesis 2).

53

0.00

0.25

0.50

0.75

1.00

Sur

vivo

r F

unct

ion

0 10 20 30 40 50

Months

MarriedCohabitingDating

Survivor Function by Relationship Status

Figure 6. Survivor function by relationship status

Finally, in Figure 7 the hazard function is revisited, this time by multipartnered

fertility status. Of particular importance is the high hazard rate early on in the study

window when both parents have multipartnered fertility and when only the father has

multipartnered fertility. These compare to the hazard curve for mother-only

multipartnered fertility cases, which has a low and fairly constant hazard until

approximately 23 months when it starts to rise. The hazard is the lowest and most

constant for cases in which neither parent has multipartnered fertility.

Multivariate Model Results

Event history models are used to better understand the risk of dissolution using

multipartnered fertility and a variety of socioeconomic and demographic covariates. I

54

used binomial logistic regression to estimate the probability of breaking up relative to

sustaining the relationship. Table 3 presents the odds ratios of dissolution relative to

staying in the relationship for all couple types. The models are based on the weighted

data since analyses not shown here indicated substantial differences between the

weighted and unweighted results.

Figure 7. Hazard function of relationship dissolution by multipartnered fertility history

The analyses show that a couple’s history of multipartnered fertility was consistently

associated with the risk of relationship dissolution. Model 1 estimates the effects of

multipartnered (father, mother, and both versus neither) on relationship dissolution

controlling for relationship status. The first model shows that fathers with multipartnered

.01

.015

.02

.025

.03

Haz

ard

Fun

ctio

n

0 10 20 30 40Months

No MPF Mother MPFFather MPF Both MPF

Smoothed Hazard Estimate of Relationship Dissolution by MPF

55

VariablesMultipartner Fertility HistoryNo MPF children (ref)MPF (mother) 1.287 1.581 1.724 1.687MPF (father) 2.542 ** 2.720 *** 3.084 *** 3.108 ***MPF (both) 1.548 † 1.876 ** 2.104 ** 2.554 **Relationship Status at Child's BirthMarried (ref)Cohabiting 2.665 ** 1.870 * 1.909 † 2.286 *Dating 8.383 *** 5.101 *** 5.298 *** 5.970 ***DemographicsMother's race

White (ref)Black 1.321 1.353 1.178Hispanic/Other race 0.779 0.719 0.713

Father's race differs from mother 0.910 0.926 1.070Mother's religiousness 1.071 1.060 1.145Mother's age 0.957 † 0.957 † 0.935 *Father is 10 years older/younger than mother 0.739 0.674 0.695Mother's substance abuse problem 1.046 1.016 0.849Mother lived with both parents at age 15 1.466 1.507 1.536Father ever in jail (mother's report) 1.680 ** 1.690 * 1.544 †

Socioeconomic StatusMother's education

<High school 0.984 0.881High school/GED 0.719 0.616Some college 0.711 0.629College or more (ref)

Mother worked in year before birth 0.697 0.705Father worked in week before birth 1.160 1.149Mother on welfare 0.854 0.766Mother or father owns home 0.943 0.952Relationship CharacteristicsYears known each other 1.031Mother feels supported (1-3) 0.327 **Mother reports conflict (1-3) 1.347Mother reports violence 0.623Mother distrusts men 0.859Mother has pro-marriage attitude 1.118Time IntervalsInterval 1 (months 1-3) (ref)Interval 2 (months 4-6) 0.397 † 0.394 † 0.399 † 0.402 †

Interval 3 (months 7-9) 0.426 ** 0.427 ** 0.435 * 0.446 *Interval 4 (months 10-12) 0.617 0.622 0.638 0.664Interval 5 (months 13-15) 0.376 * 0.382 * 0.394 * 0.404 †

Interval 6 (months 16-18) 0.200 *** 0.203 *** 0.209 *** 0.215 ***Interval 7 (months 19-21) 0.360 † 0.369 † 0.379 0.394Interval 8 (months 22-24) 0.487 † 0.508 † 0.523 † 0.547Interval 9 (months 25-27) 0.457 0.476 0.490 0.510Interval 10 (months 28-30) 0.235 ** 0.246 ** 0.254 ** 0.270 *Interval 11 (months 31-33) 0.408 0.430 0.444 0.476Interval 12 (months 34-36) 0.111 *** 0.115 *** 0.119 *** 0.128 ***Interval 13 (months 37-50) 0.211 * 0.222 * 0.227 * 0.231 *Model Chi-Square (df) 928.89(17) *** 1050.51(26) *** 1064.02(33) *** 1158.64(39) ***

Pseudo-R 2 0.095 0.107 0.108 0.118

Likelihood Ratio χ2 (df) 121.62(9)*** 13.50(7)†

102.57(9)***Note: Estimates based on weighted data. Fit statistics based on unweighted data. †p < .10, *p < .05, **p < .01, ***p < .001

Table 3. Odds Ratios from Discrete-Time Event Hist ory Model Predicting Relationship Dissolution vs. S taying Together (N = 36,401 Person-Periods)

Model 1 Model 2 Model 3 Model 4

56

fertility had more than two and a half times the risk of dissolution compared to cases

without multipartnered fertility. Furthermore, although it only reached significance at the

10% level, couples who both had multipartnered fertility were 55% more likely to end

their relationship than couples without children from a previous relationship. Model 1

also shows that a couple’s relationship status at the time of birth is strongly associated

with dissolution: cohabiting couples were more than two and a half times as likely to

break up and dating couples eight times as likely to break up as their married

counterparts.

Model 2 takes into account demographic features including race, religiosity, age,

substance abuse of the mother, whether the mother lived her biological parents at age 15,

and incarceration history of the father. Results show that fathers with multipartnered

fertility and couples who both have multipartnered fertility are still at an increased risk

for dissolution (2.7 times and 1.9 times respectively) compared to their counterparts

without multipartnered fertility. These odds ratios are just slightly larger than in Model 1,

and couples who both have multipartnered fertility now reach significance at the p < .01

level.

Relationship status continues to be associated with dissolution in Model 2 but

with lower odds (5.1 for dating couples and 1.9 for cohabiters) once controlling for

demographic characteristics. Two additional characteristics show evidence of being

associated with dissolution: mother’s age and father’s incarceration history. Model 2

shows that as age increases the risk of dissolution decreases by 4.3%, and fathers with a

history of incarceration are more than one and a half times as likely to end their

relationship as their counterparts without any incarceration history.

57

In Model 3, socioeconomic characteristics are added to the model and include

mother’s education, work history for both parents, mother’s welfare status, and whether

the couple owns or rents their home. Similar to the effect observed from adding

additional controls in Model 2, the odds of dissolution for fathers with multipartnered

fertility and for couples who both have multipartnered fertility continues to rise. The risk

of dissolution for fathers with multipartnered fertility is now more than three times that of

couples without multipartnered fertility, and more than two times that of couples who

both have multipartnered fertility children. Relationship status maintains significance

(although only at the 10% level now for cohabiting couples) and the odds ratios remain

approximately the same (2 times the risk of dissolution for cohabiters and more than 5

times the risk for dating couples compared to married couples). Mother’s age and father’s

incarceration history go unchanged, and none of the socioeconomic characteristics appear

to be associated with dissolution. Despite the fact that these variables do not make a

statistical contribution to the model, they are kept in Model 3 and Model 4 as important

control variables. The nonsignificant socioeconomic variables suggest that Hypothesis 4

should be rejected; economically disadvantaged couples do not appear to be at higher risk

for dissolution relative to economically more stable couples all else being equal.

Finally, in Model 4 a set of variables that control for relationship characteristics

are added. Continuing with the same pattern observed in Models 1-3, Model 4 increases

the difference in the risk of dissolution between fathers and couples with multipartnered

and nonmultipartnered fertility families. Fathers with children from previous

relationships have over three times the risk of separation compared to their counterparts

without multipartnered fertility, and couples who both have multipartnered fertility now

58

have over two and a half times the risk of dissolution. Similarly, Model 4 increases the

difference in risk of dissolution between unmarried and married couples. The odds were

2.28 for cohabiters and 5.97 for dating couples.

Finally, relationship characteristics, with just one exception, explain very little of

the risk of dissolution. The only significant variable was mother’s feelings of support

from the father, and as expected, every one-unit increase in feelings of support decreases

the odds of dissolution by 67%. This finding lends only partial support to Hypothesis 5:

couples with higher levels of relationship supportiveness and lower levels of violence and

conflict have lower odds of dissolution than couples with poor relationship quality. As it

turned out, violence and conflict were not associated with dissolution in this model with

this sample.

In order to carry out the test for Hypothesis 3 (i.e., couples with multipartnered

fertility from the father face greater risk of dissolution than couples with multipartnered

fertility from the mother), Model 4 was re-analyzed using a different reference group for

the multipartnered fertility variable. Model 4.1 in Table 4 includes all of the same

variables as Model 4 but uses “father’s multipartnered fertility” as the reference group

instead of “no multipartnered fertility” in order to make a direct comparison between

mothers and fathers with multipartnered fertility. Results show that couples in which both

parents have multipartnered fertility, as well as couples where only the mother has

multipartnered fertility, are not at higher risk for relationship dissolution than couples

where only the father has multipartnered fertility, when controlling for all other

covariates. This finding leads to a rejection of Hypothesis 3; there are, in fact, no

apparent differences in the risk for dissolution between mothers and fathers with

59

VariablesMultipartner Fertility HistoryNo MPF children 0.322 ***MPF (mother) 0.543MPF (father) (ref)MPF (both) 0.822Relationship Status at Child's BirthMarried (ref)Cohabiting 2.286 *Dating 5.970 ***DemographicsMother's race

White (ref)Black 1.178Hispanic/Other race 0.713

Father's race differs from mother 1.070Mother's religiousness 1.145Mother's age 0.935 *Father is 10 years older/younger than mother 0.695Mother's substance abuse problem 0.849Mother lived with both parents at age 15 1.536Father ever in jail (mother's report) 1.544 †

Socioeconomic StatusMother's education

<High school 0.881High school/GED 0.616Some college 0.629College or more (ref)

Mother worked in year before birth 0.705Father worked in week before birth 1.149Mother on welfare 0.766Mother or father owns home 0.952Relationship CharacteristicsYears known each other 1.031Mother feels supported (1-3) 0.327 **Mother reports conflict (1-3) 1.347Mother reports violence 0.623Mother distrusts men 0.859Mother has pro-marriage attitude 1.118Time IntervalsInterval 1 (months 1-3) (ref)Interval 2 (months 4-6) 0.402 †

Interval 3 (months 7-9) 0.446 *Interval 4 (months 10-12) 0.664Interval 5 (months 13-15) 0.404 †

Interval 6 (months 16-18) 0.215 ***Interval 7 (months 19-21) 0.394Interval 8 (months 22-24) 0.547Interval 9 (months 25-27) 0.510Interval 10 (months 28-30) 0.270 *Interval 11 (months 31-33) 0.476Interval 12 (months 34-36) 0.128 ***Interval 13 (months 37-50) 0.231 *Model Chi-Square (df) 1158.64(39) ***

Pseudo-R 20.118

Note: Estimates based on weighted data. Fit statistics based on unweighted data. †p < .10, *p < .05, **p < .01, ***p < .001

Table 4. Odds Ratios from Discrete-Time Event Hist ory Model Using Alternative Reference Group for MPF (N = 3 6,401 Person-Periods)

Model 4.1

60

multipartnered fertility. As expected, however, couples without multipartnered fertility

are 68% less likely to experience dissolution compared to fathers with multipartnered

fertility.

In each of the four models, controls for time are included (Intervals 1-13). These

dummy variables describe the shape of the baseline logit hazard function and indicate

whether risk for relationship dissolution increases, decreases, or remains steady over time

(Singer & Willett, 2003). In general, the odds ratios show that the risk of relationship

dissolution decreases over time. There is a slight spike early on in Interval 4 and then the

risk flattens out until the last two intervals (12 and 13) when the risk decreases.

Sensitivity Analyses

As discussed in the Methods section, there are 396 cases that were dropped from

the analytic data set due to attrition (random censoring) at wave 2 or wave 3. Specifically,

there were 206 cases with missing interviews at waves 2 and 3 and 190 cases with

missing interviews at wave 3 only. All of these cases were initially dropped because they

run the risk of being informative; that is there could be a systematic pattern to

respondents who drop out early that is related to the event of relationship dissolution.

Because there is no formal way to test if these dropped cases are informative, a

sensitivity analysis was conducted on the cases with missing at wave 3 only (N = 190).

(The 206 cases with missing at waves 2 and 3 were still excluded.) In this way, the 190

cases are treated as censored because they did not experience the event and were no

longer under observation by the end of the study period. The final model (Model 4) from

the original set of discrete-time survival analyses was re-analyzed with these cases

included. Results of the sensitivity analysis are presented in Appendix B, Table 5

61

(alongside the original Model 4 results) and indicate no substantive differences in p

values and odds ratios, as would be expected since none of these cases experienced the

event.

A second sensitivity analysis was conducted to assess for possible bias that might

have been caused by including father information on certain variables when most of the

data on the characteristics of the couples are derived from mother-provided information.

These five variables included: multipartnered fertility, whether the mother and father

differed in race, whether there was an age difference of more than ten years between the

mother and father, whether the father worked, and whether the couple owned or rented a

home. The discrete-time models were modified to exclude these variables.

The results of this modified model are presented in Appendix C, Table 6

(alongside the original Model 4 results). There are small differences between the original

Model 4 results which included the father information and the modified model results

without this information but no substantive changes in the interpretations of variables that

contribute to the model. Thus, exclusion of this father information thus does not appear

warranted.

Model Predicted Survivor Curves

Next, model predicted survivor curves were generated to demonstrate graphically

the relative risk of dissolution among couples that might be particularly disadvantaged;

that is, those who are both unmarried and have multipartnered fertility. Using the

parameter estimates from Model 4 for multipartnered fertility and relationship status,

62

while controlling for all other covariates at their mean, survivor curves for each

combination of multipartnered fertility and relationship status were produced5.

As seen in Figure 6, the risk of break-up among dating couples with

multipartnered fertility far outweighs the risk of other combinations. In particular,

married couples without multipartnered fertility appear to face the least risk among all

groups, followed by cohabiting couples with multipartnered fertility, married couples

without multipartnered fertility, cohabiting with multipartnered fertility, and finally,

dating without multipartnered fertility.

Figure 6. Model predicted survivor curves

5 Twelve curves would have been produced had the original form of MPF been used (four categories: neither-MPF, father-MPF, mother-MPF, both-MPF), along with relationship status (three categories: married, cohabiting, and dating). Instead, Model 4 from the multivariate results was re-analyzed using a collapsed (dummy) version of multipartnered fertility, i.e., 1 = any multipartnered fertility, 0 = no multipartnered fertility) thus producing only six curves.

Married/No MPF

Cohabiting/No MPF

Married/MPF

Cohabiting/MPF

Dating/No MPF

Dating/MPF

.2

.4

.6

.8

1

Pro

port

ion

Rem

aini

ng in

Rel

atio

nshi

p

0 5 10 15Intervals (1-13)

Impact of Multipartnered Fertility and Relationship Status on Dissolution

63

CHAPTER 5

DISCUSSION AND CONCLUSION

This dissertation examined the timing and effects of multipartnered fertility on the

relationship stability of married, cohabiting, and dating couples after the birth of a baby.

While previous research has explored various aspects of the consequences and causes of

relationship instability, the role that multipartnered fertility plays in destabilizing

romantic unions is still an understudied area. Moreover, we know very little about the

timing of relationship breakups in the context of the transition to parenthood, and

virtually nothing about the timing of dissolution among dating couples bearing children.

This study makes a contribution to the research literature in three ways: 1) it included not

only married and cohabiting couples as is commonly done in relationship research, but

dating couples as well; 2) it examined the timing of dissolution during the first three

years after birth; and 3) it examined the role of multipartnered fertility in explaining

relationship dissolution. To do this, the study utilized three waves of data from the

Fragile Families and Child Wellbeing Study. Kaplan-Meier estimates and discrete-time

event history models were used to explore the research questions of interest. In this

chapter, the major findings of the study are summarized, a discussion of the study’s

strengths and limitations is presented, and, finally, implications for research, practice, and

policy are discussed.

64

Summary of Major Findings

The Timing and Rate of Relationship Dissolution

Given that a limited number of studies have examined the timing of relationship

dissolution among unmarried couples after childbirth, the current study began by

examining the pattern of leaving the relationship during the first several years after

having had a baby6. The estimates suggest a steadily decreasing rate of remaining in the

relationship over the study period. In other words, during the first several years after

having had a baby, the proportion of couples sustaining their relationship gradually

decreases. By the end of the one-year mark, 22% of couples have ended their

relationship, leaving 78% of the unions intact7. By the end of the study (47 months long),

43% of the couples have separated. This study found that romantic unions are not only

vulnerable to dissolution over time, but are especially at risk of separation in the period

immediately following birth, i.e., within the first several months. The high point in the

hazard function (Figure 4) reflects a fast speed of change during this period.

As hypothesized, couples with multipartnered fertility are more at risk for

separation than couples who share only biological children. This was specifically the case

when the father and both the mother and father had children from previous relationships.

An elevated risk of dissolution among fathers with children in other households seems

quite plausible and is similar to what has been found in previous research (M. Carlson et

al., 2004; Guzzo, 2009). The complexity that comes with sharing scarce resources with

6 This study focused on literature using U.S. samples because family structure and family formation patterns in Western European and non-European English-speaking countries tend to differ from demographic changes in the United States due to cultural and social differences (Cherlin, 2009). 7 Based on unweighted Kaplan-Meier estimates.

65

multiple families can cause considerable tension throughout the family system and

especially within the mother-father dyad.

Multipartnered fertility burdens the father financially, making it difficult to meet

child support obligations (Meyer et al., 2005). It also potentially undermines a father’s

investment in his current role as a partner and parent, which can act to exacerbate the

problems in what is an already tenuous relationship (M. Carlson & Furstenberg, 2007). It

intensifies mothers’ sense of insecurity and jealousy because of possible romantic ties to

prior partners (Edin & Kefalas, 2005). Finally, it negatively affects children’s well-being.

For the time that the father is present, his commitment to the children may be low and his

attention splintered. He may also distract the mother and inadvertently take her time and

energy away from child rearing (Cherlin, 2009). And finally, if the relationship ends (as

this study indicates it might), the child must adjust to yet another transition when the

father exits.

As hypothesized, the study also found that dating and cohabiting couples are at

higher risk for separation than their married counterparts. This mirrors previous research

that has found that the odds of family stability are considerably lower for cohabiting

mothers than married mothers (Manning et al., 2004). The findings from this study differ

from those of previous research, however, in that very few studies have included dating

couples (for exceptions see M. Carlson et al., 2004; Hsueh, Morrison, & Doss, 2009).

This is perhaps a reflection of their lower prevalence in the childbirth population or a

limitation in availability of quality data on this group. Despite the possible data

limitations, this is an important group to study. In some subgroups of the dating

population, such as young, low-income African Americans, the fluid boundaries

66

characteristic of dating relationships often result in serial relationships and pregnancy

with little long-term commitment from one or both partners, making this group especially

vulnerable (Edin & Kefalas, 2005). If children were not involved, the effects of such

courtships might be negligible. Unfortunately, this is rarely the case, and children raised

without both parents are generally at a significant disadvantage (Amato & Booth, 1997;

McLanahan & Sandefur, 1994; Wen, 2008). Dating couples with children warrant

attention in the research, policy, and practice domains.

Discrete-Time Hazard Models

Findings from multivariate discrete-time hazard models suggest that an

association between multipartnered fertility and relationship instability holds when other

factors are taken into account: multipartnered fertility from the father and from both

parents is associated with a higher risk of dissolution compared to couples without any

children from previous relationships, all things being equal. Dating and cohabiting

couples are also more vulnerable to separation than are married couples. Finally, younger

mothers are more likely to separate, as are couples in which the father has a history of

incarceration.

These findings are similar to those of a recent study by Guzzo (2009), in which

she examined the effects of marital intentions and other covariates on the relationship

outcomes of cohabiting couples. Although she analyzed models separately for men and

women, she still found that when fathers had children from a prior relationship it

significantly increased the odds of dissolution. Nevertheless, not all studies have found

similar effects. Osborne et al.’s study (2007) found no effects of multipartnered fertility

on separation regardless of the origin (mother or father). However, similar to this study,

67

Osborne found that cohabiting couples were at an increased risk for relationship

dissolution relative to married couples. The difference in our findings may be driven by

the fact that the present study’s sample included dating couples, who while having similar

rates of multipartnered fertility compared to cohabiting couples, have considerably higher

rates of dissolution than cohabiting or married couples. It is feasible then that dating

couples with multipartnered fertility are increasing the likelihood of dissolution compared

to other relationship types.

Another key question of interest in this study was whether there are differences in

relationship instability between fathers and mothers who have multipartnered fertility. In

other words, is the risk of dissolution different when the father has had children with

other mothers than when the mother has had children with other fathers? If couples face a

higher risk of instability when they have children from previous romances generally, then

knowing the answer to this question could have important implications for program

developers and practitioners when structuring interventions for couples regarding what

and who to target. The study hypothesized that fathers compared to mothers with

multipartnered fertility (as opposed to fathers compared to couples without

multipartnered fertility) pose a greater risk of separation. Such a finding would suggest

that programs need to focus on fathers rather than mothers with children from prior

relationships. As shown in Table 4, however, this is not the case. Although the odds are

lower for mother-multipartnered fertility cases compared to father cases, the difference is

not statistically significant and hence does not support the study’s hypothesis regarding

fathers versus mothers.

68

One possible explanation for the rejection of Hypothesis 3 is that the relationship

between multipartnered fertility and dissolution is similar enough in mothers and fathers

such that any actual difference cannot be observed. This does not in any way, however,

minimize or run counter to this study’s findings that father-only MPF and joint-MPF

cases (i.e., both mother and father) have a significant positive effect on dissolution. This

is a different research question and suggests that multipartnered fertility is a risk factor

for dissolution compared to not having multipartnered fertility; instead, the rejection of

Hypothesis 3 simply suggests that father-only MPF does not present a higher risk factor

for dissolution compared to mother-only MPF.

Unexpectedly, this study found that socioeconomic variables did very little to

explain relationship dissolution, thus rejecting the fourth hypothesis of the study:

economically disadvantaged mothers do not appear to be at higher risk for dissolution

than more advantaged mothers. This aligns with the findings from Orthner et al.’s (2004)

cross-sectional research on low-income families who were determined to be as effective

at building relationship strengths as middle income families and developed effective

strategies to sustain their relationships and to meet their needs. While the study did not

examine relationship dissolution specifically, it suggests that poverty does not necessarily

drive all negative family outcomes, including relationship dissolution, and that low-

income couples have great potential to build lasting relationships and cohesive families.

Another explanation for the non-significant findings is the specific selection of

socioeconomic control variables. Previous quantitative and qualitative research has

suggested that men’s earnings play a crucial role in predicting relationship stability (M.

Carlson et al., 2004; Charles et al., 2006; Edin & Kefalas, 2005; Lewin, 2005). However,

69

the current study did not include measures of income or education for men because of the

large proportion of missing values on these variables. While this could possibly account

for the study’s findings, other research has found that women’s earnings play a more

important role in explaining separation among dating couples compared to men’s

earnings, thus suggesting that the selection of variables was appropriate. For example,

Osborne (2005) found no effect on stability based on income from fathers but significant

effects from mothers’ incomes: higher levels of maternal income were associated with

lower risks of dissolution.

The finding that welfare does not play a role in dissolution may also partially be

explained by the definition of the variable in the Fragile Families study: a respondent was

considered a welfare recipient if they received “public assistance, welfare, or food

stamps.” This broad interpretation of welfare use, as opposed to a measure that uses a

more stringent definition based on TANF (cash assistance) alone, has more than likely

captured couples with fewer risk factors than couples might otherwise have if the

measure had been more narrowly defined. In other words, because of more lenient

eligibility criteria for Food Stamps there may be couples included under welfare who are

offsetting the negative effects of poverty that we would otherwise expect to find.

Another challenge in interpreting the lack of findings related to welfare

participation is that the sample as a whole is of low-income. Thus, the non-welfare

households are also at some risk of living at the economic margin, making them

potentially more similar to the welfare using households compared to a sample that may

have included more higher income families.

70

The study’s finding that welfare is not associated with relationship dissolution

may also be a reflection of previous research by Moffit (2000) on welfare and female

headship. Using longitudinal data, he found little evidence of an association between

welfare benefit levels and female headship, especially among black women, suggesting

that welfare may not discourage the formation of two-parent families as previously

thought. This combination of findings suggests that future research should attempt to

clarify the relationship between men and women’s socioeconomic status and relationship

instability and should utilize other economic measures besides individual-level welfare,

e.g., average state welfare payments, income.

Another noteworthy finding of the study is that mothers who reported feelings of

support by the father had a lower risk of separation than their counterparts who did not

share in this experience. This partially supports the study’s final hypothesis that mothers

with high levels of support and low levels of conflict and violence would be less likely to

exit the relationship. Results indicate that while it is true that supportive relationships

have higher chances of enduring, conflict and violence did not help explain the

mechanism underlying relationship dissolution.

This coincides with research by Osborne et al. (2007), who found similar

associations between support and violence and relationship dissolution. The lack of

significance between violence and relationship stability is surprising but may be

explained by the limited strategy used to capture this aspect of a relationship: the measure

is based on a single question that asks whether the father hits or slaps the mother. The

true nature of physically violent relationships can take many forms and so the actual

effect of violence on the union may be biased downward. Furthermore, because mothers

71

are likely to underreport violence the parameter estimates may be biased thus reducing

the ability to accurately assess its relationship with union stability. The influence of

violence on relationship dissolution may have been different had a household-level of

domestic violence that included fathers’ experiences been included instead. Previous

research using Fragile Families data indicates that fathers tend to report much higher

levels of violence than mothers (8.2% compared to 1.7%) (Charles & Perreira, 2007).

Once the missing values on this measure are imputed, it could be helpful in capturing

potentially unobserved characteristics of the relationship.

Model-Predicted Survivor Curves

Finally, the study attempted to show what would happen in the population if the

pattern of dissolution were to hold based on the model of independent and control

variables. Model-predicted survivor curves for a combination of multipartnered fertility

and baseline relationship status were generated holding all other covariates at their mean.

The results indicate that dating couples with multipartnered fertility from the mother

and/or the father are at a considerable disadvantage compared to all other types of

couples, i.e., married/no MPF, married/MPF, cohabiting/MPF, cohabiting/no MPF,

dating/no MPF. These model-predicted survivor curves depict how an average couple

who is dating with any multipartnered fertility (mother, father, or both parents) is at a

much higher risk of dissolution in each time period among the three groups of couples

with and without multipartnered fertility. The predicted survivor curves suggest that

dating parents with children from other relationships need especially strong support

services and skills in order to sustain their relationship.

72

Transition to Parenthood

The theoretical framework for this study utilized a modified version of Cowan

and Cowan’s five-domain structural model of marital and family adaption to the birth of a

baby. Using five elements within the family system (i.e., individual, couple, triad

[mother, father, child], three generations [child, parents, and grandparents or other

extended family], and the external social system), the purpose of the model is to enhance

our understanding of the mechanisms at play between and within the five domains. The

current study modified Cowan and Cowan’s (2000) original model by specifying an

interplay between “affiliate” individuals and families, e.g., partners and children from

previous relationships, and the focal family. In this regard, multipartnered fertility

becomes a key element in the system. As supported by the findings in this study,

multipartnered fertility affects the family system, and specifically the couple dyad. Future

research should expand our knowledge of the role that multipartnered fertility plays in

affecting other domains in the model, e.g., child outcomes, social supports, kin networks,

employment demands.

Despite the model’s potential utility in offering a framework for understanding

the impact that multipartnered fertility may have on relationship stability, it has

limitations that warrant further reflection. First, the model does not explicitly indicate the

direction of effects that may occur across the five domains. For example, as hypothesized

in this study, the parents’ work patterns and other measures of socioeconomic status (fifth

and outermost level of the model) may influence the couple dyad (second level). On the

other hand, the individual (first and innermost level) is likely to influence the overall

parenting strategy and triadic relationship with the child (at third level). Identifying the

73

direction of hypothesized relationships like these between the five domains in the family

system would be useful for further theory building and for testing this model specifically.

Second, the model fails to include any dimension of time. As suggested by this

study, relationship stability among parents with a newborn baby is at risk of not being

constant in the months and years after childbirth. A framework that includes some aspect

of time with regard to relationship dissolution would be useful in improving our

appreciation of how parenthood and childbirth, major life course events, influence

whether and when a union dissolves.

Strengths and Limitations

This study contributes to extant research on union stability and multipartnered

fertility history in a number of areas. The study provides important insight into the timing

of relationship dissolution following the birth of a baby; it captures a fuller picture of

instability than previous studies by including dating couples who bear children in

addition to married and cohabiting couples; and it examines how a growing phenomenon

– multipartnered fertility – affects union outcomes. Additionally, using model-predicted

survivor curves, the study presents key findings in an efficient and clear method that is

readily understood by most audiences. Finally, the findings are applicable to a nationally

representative sample of non-marital births in large U.S. cities.

Despites the study’s strengths, there are several limitations that warrant

discussion. Several time-varying variables should be included in the discrete-time event

history models in order to take full advantage of the longitudinal nature of the data and to

accurately capture the effects of certain couple characteristics. These time-varying

variables include: work experience and welfare status, as well as the relationship

74

characteristic variables, e.g., support, violence, distrust. The primary purpose of including

time-varying covariates in a model is to connect the independent variable of interest at a

specific time to the study time (S. Guo, Forthcoming) because the risk of hazard changes

when the values of the variable change (Cleves, Gould, & Gutierrez, 2004b). Excluding a

time-varying version of employment, for example, limits the ability to control for an

anticipated event, such as unemployment, which may be negatively associated with

relationship dissolution.

Future research should also take advantage of the availability of multiple

imputation methods (Rose & Fraser, 2008) to impute missing data values instead of

relying on listwise deletion (also known as case deletion or complete case analysis)

(Schafer & Graham, 2002). Although the proportion of missing data was relatively small

(no variable was missing more than 7% of its values), the ability to impute missing data

would eliminate the need to delete observations which for several variables (i.e.,

multipartnered fertility, substance abuse, relationship time prior to pregnancy) that were

shown to differ on the missing versus non-missing cases. These differences suggest that

the missing data are “nonignorable” (Allison, 2002) or “missing not at random” (MNAR)

(Schafer & Graham, 2002) and represent a threat to the validity of generalizing the

study’s findings. Alternative methods of handling missing data were considered,

including dummy variable adjustment and mean substitution; however, both methods

were described by Allison (2002) as even more biased and problematic than using

listwise deletion. Thus, listwise deletion was ultimately selected as the least egregious of

the available options. Multiple imputation, however, could go a long way in improving

75

this study’s analysis and increasing its applicability and will be considered in future

research.

Due to an empty interval problem with the married sample, i.e., no married couple

experienced the event in the 13th and final interval, separate analyses of the model for

each group (married, cohabiting, and dating) were not presented. Although the model for

the married group converged, the 13th interval was dropped and the resulting estimates

and standard errors appeared suspect (i.e., most of the significant estimates were in the

opposite direction of what was expected and standard errors on the MPF variables were

unusually large). However, a solution to this problem proposed by Allison (1995), in

which the coefficient for the empty interval is constrained to be the same as that of the

adjacent interval, should solve this problem in future analyses of this data.

A number of important variables were included in the multivariate models to

capture characteristics of the relationship, e.g., support, conflict, distrust. The study lacks,

however, potentially important variables that measure expectations that partners typically

have in a relationship about the role that each person should play, e.g., child care,

household tasks. Future waves of the Fragile Families study (or other studies) should

include survey questions that address these aspects of the relationship. Additionally, this

study’s findings are limited to urban couples. Future research should be extended to

include couples from rural environments as well.

A final limitation concerning the analytical method is worth noting. There is a

potential disadvantage to using a discrete-time model because of the time metric used,

e.g., month, quarter, year. As Guo (Forthcoming) points out, these relatively coarse

measures of the change rate may result in a loss of information. When using quarters for

76

the time metric, for example, a couple whose relationship lasted 1 day is treated the same

as a couple whose relationship lasted 90 days. This does not, however, seem to present a

serious problem when using longitudinal data from multi-wave panel studies.

Implications

Research Implications

A key premise of this study was that transitions to parenthood present

considerable challenges to most parents, and especially low-income parents who already

face a host of disadvantages. To capture the effects of the transition to parenthood on

couple outcomes requires that measures of relationship quality and other relevant

variables be obtained before and after birth. The ideal data set would start at the initiation

of a relationship and follow the couple through courtship, pregnancy and marriage (if

applicable) or sustained co-parenting, and beyond for many years at regular intervals

(Fein, Burstein, Fein, & Lindberg, 2003). The data would also include measures of other

existing relationships in the immediate social system (i.e., those that resulted in

multipartnered fertility). Data of this nature would allow for a fuller assessment of factors

that affect the trajectories of couples’ relationships and the outcomes of their children as

well.

This study included several measures of relationship quality known to affect

union outcomes; however, future research should consider how to improve the validity of

such indicators. Violence is one such example. The reliability of accurately capturing

violent behavior in this study is questionable because the measure was based on only one

question (“How frequently did your partner hit or slap you?”). More precise measures of

this behavior are especially needed because of its known effect on relationships around

77

the time of birth and because violence is more prevalent among low-income and

unmarried couples (Saltzman, Johnson, Gilbert, & Goodwin, 2003). The National

Institute for Justice recommends that violence be measured using five different

categories: physical abuse, sexual violence, threats of sexual or physical violence,

stalking, and psychological abuse (Centers for Disease Control and Prevention, 2000).

Although inclusion of extraneous variables is never warranted, expanding the current

mechanism for capturing the effects of violence on relationship stability should be

considered. Following the work of Straus (1979), conflict as it is manifested across a

broad continuum of aggression should continue to be used. In this study, several items

that resembled questions from the “Reasoning” and “Verbal Aggression” scales of the

Conflict Tactics Scale (Straus, 1979) were utilized and should also be considered in

future research.

This study focused on one aspect of relationship stability among parents with

newborn children – union dissolution. However, future research should consider other

possible exits from the relationship, especially for dating parents, since this has not been

widely explored. For instance, it would be useful to have an understanding of the timing

of transitions into other relationship states, such as marriage and cohabitation, among

parents who begin as “daters.” Moreover, it could be beneficial to examine multiple or

repeated events, i.e., when a parent exits from one relationship and enters into another.

Serial relationships are bound to have significant effects on children; thus, extending our

knowledge about the role of “serial fatherhood figures” in children’s outcomes at

different developmental stages could be informative for policy makers as well as

practitioners.

78

Finally, the current study cannot make causal claims about multipartnered fertility

and its effects on relationship outcomes. As with any data produced from observational

studies as opposed to experimental studies, it is possible that the association between

multipartnered fertility and dissolution is attributable to unobserved characteristics of

parents who tend to have serial relationships with children. In other words, the effect of

multipartnered fertility on relationship status is likely confounded with other factors, that

is, those that led the parents to form previous partnerships (that included childbearing) to

begin with.

Including a rich set of control variables that are confounders is one way of helping

to reduce this bias. If possible, however, research efforts in the future should include

other mechanisms to reduce these selection effects in order to move closer to establishing

the causal relationship between multipartnered fertility and union instability. One method

for achieving this would be to use a fixed effects estimation approach as conducted in

previous studies on parental relationships (M. Carlson & Furstenberg, 2007; Foster &

Kalil, 2007). Fixed-effects models would be used to compare the same couple at different

points in time, holding constant all couple characteristics that are stable over time (or are

time invariant). In non-mathematical terms, fixed-effects models can be thought of as

regression models that include a dummy variable for each individual in the sample, thus

controlling for their constant characteristics even when those characteristics have not

been explicitly measured (c.f. Grogan-Kaylor, 2004).

Practice and Policy Implications

One of the more important findings of this study was that relationships among

couples who recently had a child together are at significant risk of dissolution very soon

79

after the birth of the baby. It appears that dating couples and couples with multipartnered

fertility from the father may be driving this effect. Several recently developed

interventions to strengthen low-income couples’ relationships were designed with this

very problem in mind. For example, the Strong-Couples and Strong-Children program

(Jones, 2008), and the Building Strong Families program, both projects of the federal

Healthy Marriage Initiative, were designed to provide relationship skills and support

services to low-income unwed expectant or new parents around the time of the “magic

moment” (Dion et al., 2003). This is the period around the birth when couples are

typically emotionally close from the pregnancy and birth and are potentially more open to

interventions than at other times.

Waiting until the “magic moment” around the time of birth, however, may be too

late. This raises an important timing question about when interventions should be

delivered. Policy makers and program managers may need to target couples earlier at the

“pre-contemplation” stage of parenting before couples actually get pregnant (but are

romantically involved). A modification of this kind in relationship strengthening

programs could prompt yet another difficult question: How should policies or programs

be designed so that they target couples “far enough” along on the relationship timeline

that it is advantageous to offer them an intervention but not so late that the couple is in

imminent danger of separating. Program managers from existing interventions, such as

Strong Couples-Strong Children, may be able to inform this policy and practice

conundrum.

The second important finding from this study that has considerable policy and

practice implications is the role that multipartnered fertility plays in relationship

80

instability and, subsequently, in decisions to marry (or not) among low-income African-

American couples. Upon initial release of the Fragile Families data, policy makers were

encouraged to find that while a considerable number of new parents with low-incomes

were unmarried, they were still romantically involved at the time of birth and had high

expectations for marriage with their partner. Approximately 74% of mothers and 90% of

fathers reported that they had a 50-50 chance or better of marrying their partner

(McLanahan et al., 2003). Furthermore, the majority of the mothers and fathers expressed

a strong belief in marriage, indicating that marriage is good for themselves and their

children.

With such high expectations for marriage among couples with nonmarital births,

policy makers chose to commit considerable amounts of funding to marriage promotion

and relationship stabilization programs using funds from the Temporary Assistance to

Needy Families program. The hope was that these efforts would result in reduced poverty

levels and improved child outcomes among the most vulnerable families, especially those

consisting of unmarried mothers and children.

Despite the high hopes of both policy makers and the couples themselves for

marriage, the effects of multipartnered fertility on relationship stability, and subsequently

on marriage, was not fully considered. Having a child from a previous relationship lowers

the probability that the current union will remain intact and that the couple will marry at

all (Mincy, 2002). Both fathers and mothers are hesitant to make a serious, long-term,

marital commitment to partners with children from other parents.

Mothers are reluctant to marry fathers because of financial obligations he has to

other households, and because of fears about ongoing sexual ties to the previous

81

partner(s) (Monte, 2007). Fathers avoid transitioning to marriage because they are

hesitant to take on the financial responsibility of another man’s child (Lichter & Graefe,

2001) in addition to the child support obligations they may already have to other families.

Further, welfare policies still penalize families by reducing or eliminating their benefits

once couples have married (Orthner et al., 2004). Under these conditions, social policy

efforts to promote marriage as a poverty reduction strategy will more than likely be

ineffective.

Instead, policy makers should consider alternative and complementary measures

that have the potential to serve the same purpose, that is, to reduce poverty levels among

unmarried couples and improve the well-being of their children. Research suggests that

poor labor market opportunities and incarceration are associated with decreased child

support payments among unmarried fathers (Magnuson & Gibson-Davis, 2007).

Strengthening workforce development efforts to help fathers find and sustain stable

employment could help them more easily provide for all of their children (Mincy, 2002).

Specific clinical strategies that help parents co-parent across households are also

warranted. Evidence suggests that fathers’ apparent disengagement from their children

who live in multiple households has as much to do with their inability to provide

financial support as it does with their incapacity to coparent effectively (M. Carlson &

Furstenberg, 2007). Interventions to strengthen couple relationships with specific

attention to fostering positive father involvement are crucial to the success of any family-

focused social policy efforts (C. P. Cowan et al., 2007).

82

Conclusion

Many couples in America today will partner and re-partner multiple times before

making the final commitment to marriage (Cherlin, 2009). These cycles of re-partnering

raise an important policy question: Would it be more effective to help couples avoid

pregnancy altogether while they complete or reach a more mature point in their courtship

process than to help them sustain their relationship once they have a child? Is waiting

until they are pregnant or have a child too late? Cherlin (2009) addresses this social

policy question by going a step further. He suggests that couples not only be encouraged

to hold off on childbearing, but to slow down the entire courtship process to begin with.

I suggest that we advise lone parents to take their time in finding partners and to be confident that a relationship will last before they bring a stepparent into their home. We should urge them to choose carefully and deliberately so that their subsequent partnerships, if any, have the greatest chance of enduring. (pp. 196-197).

Since there are no indications that nonmarital childbearing is currently slowing

down, efforts to strengthen parental relationships outside the institution of marriage

should continue. This is certainly the case among unmarried couples with children. The

key question raised in this study was whether having children with multiple partners

increased the risk of relationship dissolution. The answer is yes, and knowing this should

potentially enable practitioners and policy makers to improve the designs of existing and

forthcoming programs aimed at reducing poverty through marriage and relationship

strengthening efforts.

The results of this study suggest that multipartnered fertility, especially when the

father has children from previous partners, has the potential to play a critical role in the

outcome of couple relationships. Findings suggest that programs and policies need to

83

help fathers navigate the complexities that come with children born in serial relationships

if marriage promotion efforts are to succeed in increasing family unity. This could

possibly be achieved by offering father-specific groups that focus on co-parenting,

increasing men’s ability to pay child support across households, and structuring family-

centered services so that the needs of multiple families with shared biological ties can be

met. In this way, mothers and fathers with shared children in multiple families can be

supported in their effort to raise strong children and live productive lives.

84

APPENDIX A

Cook’s D Plot

111111111111333333333333

5

777777777778888888888888999999999999

1011

1212

12

13131313

13

141414141414141414

14

1616161616161616161616161818181818181818181818181919191919191919191919212121

21

22222222222222222222222225252525252525

25

262626262626262626262626262727272727272727272727272828282828

28

2929292929292929

29

30303030303030303030303030

31

32323232

32

33333333333333333333333336363636363636363636363636383838383838383838383838393939393939393939393939

40

41

45

464646464646464646464646464848484848484848484848

50

52525252525252525252525252545454545454545454545455555555555555555555555555565656565656565656565656575757

57

5858585858585858585858

58

60

626262626262626262626262626363636363636363636363636464

64

66666666666666666666666668686868686868686868686868696969696969696969696969707070707070707070707070717171717171717171717171717272727272727272

72

7575757575757575757575757676767676767676767676767678787878787878

78

7979797979797979797979

81

828282828282828282828282828383838383838383838383838384848484

84

84848484

84

86868686868686868686

86

88

898989898989898989898989919191

91

9292929292929292929292929393939393939393939393

95

969696969696969696969696979797

97

100100100100100100100100100100100100100101101101101101101101101101101101101

102105

106

106

107107107107107107107107107107107107109109109109109109109109109109109109110110110110110110110110110110110110

111

112112112112112112112112112112112112114114114114114114114114114114

114

115

116116116116116116116116116116116116116117117117117117117117117117117117117118118118118118118118118118118118119119119119119119119119119119119119120120120120120120120120120120120120122122122

122

125125125125125125125125125125125125125126

126

127127127127127127127127127127127128128128128128

128

129129129129129129129129129129129

130130

130

130130130130130130130130130130133133133133133133133133133133133133133135135135135135135135135135135135135137137137137137137137137137137137137138138138138138138138138138138138138138140140140140

140

141

141

143

143

144144144144144144144144144144144145145145145145145145145145145145145146146146146146146146146

146

147

147

149149149149149149149149149149149149

149

150

150

153153153153153153153153153153153154154154154154154154154154154154154154

156

158158158158158158158158158158158158158159159159159159159159

159

159159159160160

160

162162162162162162162162162162162162163163163163163163163163163163163163164164164164164164164164164164164164165165165165165165165165165165165166166166166166166166166166166166166166168168168168168168168168168168168169169169169169169169169169

169

170170170170170170170170170170170170171171171171171171171171171171171171174174174174174174174174174174174177177177177177177177177177177177178178178178178178178178178178178178178180180180180180180180180180180180180180182182182182182182182182182182182182182183183183183183183183183183183183183184184184184184184184184184184184184184186186186186186186186186186186186186186187187187187187187187187187187187187189189189189189189189189189189189189189192192192192192192192192192192192192193193193193193193193193193193

193

194194194194194194194194194194194194195195195195195195195195195195195197197197197197197

197

201

202202202202202202202202202202202202

203

207207207207207207207207207207207208208208208208208208208208208208208209209209209209209209209209209209209210210210210210210210210210210210210211211211211211211211211211211211213213213213213213213213213213213214214214214214214214214214214214214217217217217217217217217217217217217218218218218218218218218

218

220220220220220220220220220220220220221221221221221221221221221221221221

222

223223223223223223223223223223223223223224224224224224224224224224224224224225225225225

225228

230230230230

230

231231231231231231231231231231231231231232232232232232232232232232232232232233233233233233233233233233233233

236

238238238238238

238

239239239239

239

240240240

240

241241241241241241241241241241241241242242242242

242

243

249249249249249249249249249249249249249250250250250250250250250250250250250253253253253253253253253253253253253254254254254254254254254254254254255255255255255255255255255255255256256256256256256256256256256256256262262262262262262262262262262262264264264264264264264264264264264264264267267267267267267267267267267267267268268

268

269269269269269269269269269269269271271271271

271

272272272272272272272272272272272273273273273273273273273273273273273274274274274274274274274274274274274275275275275275275275275275275275275276276276276276276276276276276276280280280280280280280280280280280280282282282

282

283283283283283283283283283283283283284284284284284284284284284284284284285285285285285285285285285285285285285286286286286286286286286286286286286287287287287287287287287287

287

289289289289289289289289289289289289

289

290290290

290

291291291291291291291291291291291291291292292292292292292292292292292292293293293293293293293293293293293293293295295295295295295295295295295295295296296296296296296296296

296

297297297297297297297297297297297298298298298298298298298298298298298300300300300300300300300300300300303303303303303303303303303303303303306306306306306306306306306306306306307

307

308

308

310

310

311311311311311311311311311311311311313313313313313313313313

313

313313314314314314314314314314314314314316316316316316316316316316316316316316317317317317317317317317

317

318318318318318318318318318318318318319319319319319319319319319319319319

323

324324324324324324324324324324324326326326326326326326326326326326

327

329329329329329329329329329329329329

331

332332332332332332332332332332332332333333333333333333333333

333

334334334334334334334334334334334334334335335335335335335335335335335335336336336

336

338

338

339339339339339339339339339339339339340340340340340340340340340340340

340

343

343

344344344344344344344344344344344344

345

347347347347347347347347347347347347347

348348

348

352352352

352

353353353353353353353353353353353353354354354354354354354354354

354

355355355355355355355355355355355358358358358358358358358358358358358358360360360360360360360360360360360360361361361361361361361361361361361361362362362362362362362362362362362362363363

363

365365365365365365365365365365365365366366366366366366366366366366366366367367367367367367367367367367367367368

368

369369369369369369369369369369369370370370370370370370370370370

370

371371371371371371371371371

371

373373373373373373373373373373373373374374374374374374374374374374374376376376376

376

377377377377377377377377377377377377378378378378378378378378378378378378378380380380380380380380380380380380380382382382382382382382382382382382382383383383383

383

384384384384384384384384384384384384385385385385385385385385385385385385385386386386

386387

388388388388388

388

390390390390390390390390390390390391391391391391391391391391391391391392392392392392392392

392

393393

393

394394394394394394394394394394394394395395395395395395395395395395395397

397

398398398398398398398398398398398398398399399399399399399399399399399399399400400400

400

401

404404404404404404404404404404404405405405405405405405405405405405405406406406406406406406

406

407407407407407407407407407407407409409409409409409409409409409409413413413413413413413413413413413413413415415415415415415415415415415415415416416416416416416416416416416416416417417417417417417417417417417

417

417418418418418418418

418

419419419419419419419419419419419419420420420420420420420420420420420420420421421421421421421421421421421421421423423423423423423423423423423423423424424424424424424424424424424424425425425425425425425425425425425425425426426426426426426426426426426426426426427427427427427427427427427427427427428428428428428428428428428

428

431431431431431431431

431

433433433433433

433

434434434434434434434434434434434434435435435435435435435435435435435

436

438438438438438438438438438438438438440440

440

441441441441441441441441441441441441442442442442442442442442442442442444444444444444444444444444444444444447447447447447447447447447447447447447448448448448448448448448448448448

449

449449449449449449449449449449450450450450450450450450450450450450

451

453453453453453453453

453

455455455455455455455455455455455456456456456456456456456456456456456456457

457

458458458

458459

460460460460460460460460460460460

461

464464464464464464464464464464464465465465465465465465465465

465

468468468468468468468468468468468468470470470470470470470470470470470472472472

472

473473473

473

474474474474474

474

475475475475475475475475475475475475476

476477

478478478478478478478478478478478478479479479479479479479479

479

480480480480480480480480480480480480481481481481481481481481481481481482482482482482482482482482482482482485485485485485485

485

486486486486486486486486486486486486486487487487487487487487487487487487487488488488488488488488488488488488488489489489489489489489489489489489491491491491491491491491491491491491493493493493493493493493493493493494494494494494494494494494

494

496496496496496496496496

496

498498498498498498498

498

499499499499499499499499499499499499504504504504504504504504504504504504505505505505505505505505505505505505505508508508508508508508508508508508508509509509509509509509509509509509511511511511511511511511511511511

512

513513513513513

513

515515515515515515515515515515515515516516516516516516516516516516516519519519519519519519519519519519519520520520520520520520520

520

522

522

523523523523523523523523523523523524524524524524524524524524524524524526526526526526526526526526526526526529529529529529529529529529529529529529531531531531531531531531531531

531

532532532532532532532532532

532

533533533533533533533533533533533

537

539539539539539539539539539539

539

540540

540

541541541541541541541541541541541542542542

542

543543543

543

544544544544544544544544544544544544

545547

548548548548548548548548548548548548550550550550

550

551

551

552552552552552552552

552

553553553553553553553553553553553553554554554554554554554554554554554554560560560560560560

560

560560560560560560

561

562562562562562562562562562562562562563563563563563563563563563563563563564564564564

564566

567567567567567567567

567

569569569569569569569569569569569569569570570570

570

571571571571571571571571571571571571

572

574574574

574

575575

575

576577

577

579579579579579579579579579579579581581581581581581581581581581581581582582582582582582582582582582582582582583583583583583583583583583583583583584584584584584584584584584584584584585585585585585585585585585585585

585

586586586586586586586586586586586586590590590590590590590590590590590590591591591591591591591591591591591591593593593593593593593593593593593593595595595595595595595595595595

595

598598598598598598598598598598598600600600600600600600600600600600600602602602602602602602602602602602602602605605605605605605605605605605605605606606606606606606606606606606606606608608608608608608608608608608608608609609609609609609609609609609609609610610610610610610610610610610610611611611611611611611611611611611612612612612612612612

612

613613613613613

613

615615615615615615615615615615

615

615616616616616616616616616616616616616616618618618618618618618618

618

619

620620620620620620620620620620620620621621621621621621621621621621621621621624624624624624624624624624624624624624627627627627627627627627627627627627628628628628628628628628628628628631631631631631631631631631631631631632632632632632632632632632632632632633633633633633633633633633633633634634634634634634634634634634634634635635635635635635635635635635635636636636636636636636636636636636636637637637637637

637

638638638638638638638638638638638638639639639639639639639639

639

640640640640640640640640640640640640

641

642642642642642642642642642642642642644644644644644644644644644644644646646646646646646646646646646646646647647647647647647647647647647647647648648648648648648648648

648

649

649

650650650650650650650650650650650650652652652652652652652652652652652652653653653653653653

653

654654654654654654

654

655

657657657657657657657657657657657659659659659659659659659659659659659659661661661661661661661661661661661662662662662662662662662662662662662663663663663663

663

664

666666666666666666666666666666666666669669669669669669669669669

669

670

670

671671

671

672672672672672672672672672672672676676676676676676676676676676676677677677677677677677677677677677677679

679

681681681681681681681681681681681681682682682

682

683683683683683683683683683683683683

688

690690690

690

691691691691691691691691691691691691692692692692692692692692692692692692693693693693693693693693693693693693695695695

695

696696696696696696696696696696696696697697697697697697697697697697697697698698698698698698698698698698698698701701701701701701701701701701701702702702702702702702702702702702702703703703703703703703703703703703703703704704704704704704704704704704704704704705705705705705705705705705705705

706

707707707707707707707707707707707

708

710710710710710710710710710710710710713713713713713713713713713713713713713714714714714714714

714

715715715715715715715715715715715715717717717717717717717717717717717717

719

720720720720720720

720

721721721721721721721721721721721722722722722722722722722

722

725725725725725725725725725725725

726

727727727727727727727727727727727727731731731731731731731731731731731733733733733733733733733733733733734734734734734734734734734734734734734736736736736736736736

736

738

739739739739739739739739739739739740740

740

742742742

742

743743743743

743

744744744744744744744744744744744744745745745745745745745745

745

745745746746746746746746746746746746746746

747

748748748748748748748748748748748748748749749749749749749749749749

749

750750750750750750750750750750750750751751751751751751751751

751

752752752752752752

752

753

753

755755755755755755755755755755755755758758758758758758758758758758758758760760760760760760760760760760760760761761761761761761761761761761761762762762762762762762762762762762762762763763763763763763763763763763763763763764764764764764764764764764764764764

765

766766766766766766766766766766766

766

768768768768768768768768768768768768768769769769769769769769769769769769769770770770770770770770770770770770770770771771771771771771771771771771771771

771

773773773773773773773773773773773773773774774774774774774774774774774774774779779779779

779

780780780780780780780780780780780

780

781

783783783783783783783783783783783783783784784784784784784784784784784784786786786786786786786786786786

786

787787787787787787787787787787787790790790790790790790790790790790790791791791791791791791791791791791791792792792792792792792792792792792794794794794794794

794

796796796796796796796796796796796796797797

797

798798798798798798798798798798798800800800800800800800800800800800800800801801

801

802802802

802

803803803803803803803803803803803803803

804807

809809809809809809809809809809809810810810810810810810810810810810810812

812

813813813813813813813813813813813813814814814814814

814

815

816816816816816816816816816816816816816818818818818818818818818818818818818819

819

821

821

822

822822

823823823823823823823823823823823823823825825825825825825825825825825825825

828

830830830830830830830830830830830830831831831831831831831831831831831831834834834

834

835835835835835835835835835835835835835836836836836836836836836836836836836836837837837837837837837837837

837

840840840840840840840

840

841841841841841841841841841841841

841

843843843843843843843843843843843843844844844844844844844844844844844844844847847

847

849849849849849849849849849849849850850850850850850850850850850850852852852852852852852852852852852852853853853853853853853853853853853853853855855855855855855855855855855855855855856856856856856856856856856856856857857857857857857857857857857857859859859859

859

860860860860860860860860860860860860864864864864864864864864864864864864864865865865865865865865865865865865866866866866866866866

866

867867867867867867867867867867867868868868868868868868868868868868868868869869869869869869869869869869869869870870870870870870870870

870

871871

871

872872872872872872872872872872872874874

874

875875875

875

878878878878878878878878878878878878878879879879

879

880880880880880880880880

880

880880880881881881881881881881881881881881881881882882882882882882882882882882882882883883883883883883883883883883883883885885885885885885885885885885885885885886886

886

887

891891891891891891891891891891

891

893893893893893893893893893893893893894894894894894894894894894894894894894897897897897897897897897897897897897898

898

899899899

899

900900900900900900900900900900900

902

905905905905905905905905905905905905906906906906906906906906906906906906908908908908908908908908908908908908909909909909909909909909909

909

913913913913913913913913913913913913913915915915915915915915915915915915915917917917917917917917917917917917917918918918918918918918918918918918918918922

922

923923923923

923

925925925925925925925925925925925926926926926926926926926926926926926927927927927927927927927927927927927929929929929929929929929929929930930930930930930930930930930930930931931931931931931931931931931931932932932932932932932932932932932933933933933933933933933933933933933937937937937937937937937937937937937937938938938938938938938938938938938938938939939939939939939939939939939939939941941941941941941941941941941941943943943943943943943943943943943943944944944944944944944944944944944945945945945945945945945945945945947947947947947947947947947947947947

948

950950950950950950950950950950950950950951951951951951951951951951951951951952952952952952952952952952952952952955955955955955955955955955955955955955956956956

956

957957957957957957957957957957957957958

958

959959959959959959959959959959959959959960960960960960960960960960960960960960962962962962962962962962962962962

963

964964964964964

964

965965965965965965965965965965965965965966966966966966966966966966966966968968968968968968968968968968968968970970970970970970970970970970970972972972972972972972972972972972972973973973973973973973

973

975975975

975

976976976976976976976976976976976978978978978978978978978978978978980980980980980980980980980980980980981981981981981981981981

981

982982982982982982

982

983983983983983983983983983983983983983984984984984984984984984984984984985

985

988988988988988

988

991

992992992992992992992992992992992992993993993993993993993993993993993993994

994

997997997997997997997

997

1000100010001000100010001000

1000

100110011001100110011001100110011001100110011001100210021002100210021002100210021002100210021002100310031003100310031003100310031003100310031003100410041004100410041004100410041004100410041004100510051005100510051005100510051005100510051005100610061006100610061006100610061006100610061006100710071007100710071007100710071007100710071007

1008

100910091009100910091009100910091009100910091011101110111011101110111011101110111011101110111012101210121012101210121012

1012

10131013

1013

101410141014101410141014101410141014101410141015101510151015101510151015101510151015101510151016101610161016101610161016101610161016101610161017101710171017101710171017101710171017101710191019101910191019101910191019

1019

1021

1021

1022102210221022102210221022102210221022102210221024102410241024102410241024102410241024102410241024102510251025102510251025102510251025102510251025

1028

102910291029102910291029102910291029102910291030103010301030103010301030

1030

1031103110311031

1031

10331033103310331033103310331033103310331033103410341034

1034

10361036103610361036103610361036103610361036103610361038103810381038103810381038103810381038103810391039103910391039103910391039103910391039103910401040

1040

104210421042104210421042104210421042104210421043104310431043104310431043104310431043104310431046

1046

1047104710471047

1047

10481048104810481048104810481048104810481048

1048

1050105010501050

1050

105110511051105110511051105110511051105110511052

1052

105310531053105310531053105310531053105310531055

1055

1056

10571057105710571057105710571057105710571057105710571059105910591059105910591059105910591059105910591061106110611061106110611061106110611061106110621062106210621062106210621062106210621062106210631063106310631063106310631063106310631063106310631065106510651065106510651065106510651065106510651065106610661066106610661066106610661066106610661066

1067

10681068106810681068106810681068106810681068106810691069106910691069106910691069106910691069106910701070107010701070107010701070107010701070107010711071107110711071107110711071107110711071107110721072

1072

10741074107410741074107410741074107410741074107410741075107510751075107510751075107510751075107510811081108110811081108110811081108110811081108310831083108310831083108310831083108310831083108310841084108410841084108410841084108410841084108410841086108610861086108610861086108610861086108610861086

1087

10881088108810881088108810881088108810881088108910891089108910891089108910891089108910891089108910901090109010901090109010901090109010901090109010911091

1091

1092

109310931093109310931093

1093

10941094

1094

10951095109510951095

1095

10961096109610961096109610961096109610961096109610971097109710971097109710971097109710971097109711001100110011001100110011001100110011001100110011021102110211021102110211021102

1102

1104

110511051105110511051105110511051105110511051105

1106

110711071107110711071107110711071107110711071109110911091109110911091109110911091109110911091110111011101110

1110

11141114111411141114111411141114

1114

1115111511151115111511151115111511151115111511151116111611161116111611161116111611161116111611171117111711171117111711171117111711171117111711181118

1118

111911191119111911191119111911191119111911191119111911201120112011201120112011201120112011201120112011211121112111211121112111211121112111211121112111211123112311231123112311231123112311231123

1123

11231124112411241124112411241124112411241124

1124

1125112511251125112511251125112511251125112511261126112611261126112611261126112611261126112611261128112811281128112811281128112811281128

1128

1129112911291129112911291129112911291129112911291130113011301130113011301130113011301130113011301132

1132

11331133113311331133113311331133

1133

11341134113411341134113411341134113411341134113411361136113611361136113611361136

1136

1137113711371137113711371137113711371137113711371139113911391139113911391139113911391139113911391140114011401140114011401140114011401140114011411141114111411141114111411141114111411141114111411142114211421142114211421142114211421142114211421142114311431143114311431143114311431143114311431143114311441144114411441144114411441144114411441144114411451145114511451145114511451145114511451145114511451146114611461146114611461146114611461146114611461147114711471147114711471147114711471147114711471148114811481148

1148

114911491149114911491149114911491149114911491149115011501150115011501150115011501150115011501150115111511151115111511151115111511151115111511151

1152

115311531153115311531153115311531153115311531153115311551155115511551155115511551155

1155

11571157115711571157115711571157115711571157116011601160116011601160116011601160116011601160116011611161116111611161116111611161116111611161116111611162116211621162116211621162116211621162116211631163116311631163116311631163116311631163116311631164116411641164116411641164116411641164116411641164116511651165

1165

1166116611661166116611661166116611661166

1166

1168116811681168116811681168

1168

11701170117011701170117011701170

1170

1171

11711174

117611761176117611761176117611761176117611761176117611801180

1180

11831183118311831183118311831183118311831183118311831184118411841184118411841184

1184

1187118711871187

11871189

1191119111911191119111911191119111911191119111911191119311931193119311931193119311931193119311931194119411941194119411941194119411941194119411941195119511951195119511951195119511951195119511951196119611961196119611961196119611961196119611961197119711971197119711971197119711971197119711981198

1198

119911991199119911991199119911991199119911991199120012001200120012001200

1200

12011201120112011201120112011201120112011201120112021202120212021202120212021202120212021202

1203

120512051205120512051205120512051205120512051205120612061206120612061206120612061206120612061206120712071207120712071207120712071207120712071207120812081208120812081208120812081208120812081210121012101210121012101210121012101210121012101210121112111211121112111211

1211

1213

1213

121312131213121312131213121312131213121312141214121412141214121412141214121412141214121412151215121512151215121512151215121512151215121612161216121612161216121612161216121612161216121712171217121712171217

1217

121912191219

1219

1220

12211221122112211221122112211221122112211221122112221222122212221222122212221222122212221222122212221223122312231223122312231223122312231223122312241224122412241224122412241224122412241224122512251225122512251225122512251225122512251226122612261226122612261226122612261226122612271227122712271227122712271227122712271227122712281228122812281228122812281228122812281228122812291229122912291229

1229

1231

123212321232123212321232

1232

123312331233123312331233123312331233123312331233123412341234123412341234123412341234123412341237123712371237123712371237123712371237123712391239123912391239123912391239123912391239

1241

12421242124212421242124212421242124212421242124212431243124312431243124312431243

1243

12441244

1244

12451245124512451245124512451245124512451245124512461246124612461246124612461246124612461246

1248

12491249124912491249

1249

12501250125012501250125012501250

1250

12511251125112511251125112511251

1251

125112511251125212521252125212521252125212521252125212521252125312531253125312531253125312531253125312531253

1254

1260126012601260126012601260126012601260126012621262

1262

1263126312631263126312631263126312631263126312651265126512651265126512651265126512651265126512661266126612661266126612661266126612661266126612691269126912691269126912691269126912691269

1269

12701270127012701270127012701270127012701270127012711271

1271

12731273127312731273127312731273127312731273127412741274

1274

1275

1275

1275

12771277127712771277127712771277127712771277127712771278127812781278127812781278127812781278127812781278128212821282

1282

12831283128312831283128312831283128312831283128312851285128512851285128512851285128512851285128612861286128612861286128612861286128612861286

1287

1288

128912891289

1289

1289

1290

1290

12911291129112911291129112911291

1291

1291129112931293129312931293129312931293

1293

1293129312931293129412941294129412941294129412941294129412941294

1295

13001300130013001300130013001300

1300

1302130213021302130213021302130213021302130213021302130313031303

1303

130413041304130413041304130413041304130413041304130513051305130513051305130513051305130513051305130613061306130613061306

1306

13071307130713071307130713071307130713071307131013101310131013101310131013101310131013101310131313131313

1313

1314131413141314131413141314131413141314131413141314131513151315131513151315131513151315131513151316

1316

1319

132013201320132013201320132013201320132013201320

1322

1323

1324132413241324132413241324132413241324132413241325132513251325132513251325132513251325132513251325132613261326132613261326132613261326132613261326132713271327

1327

13291329132913291329132913291329132913291329132913291331133113311331133113311331133113311331

1331

1333133313331333133313331333133313331333133313351335133513351335133513351335133513351335133613361336133613361336

1336

1337

13411341

1341

1344134413441344134413441344

1344

13451345134513451345134513451345134513451345134513451346134613461346134613461346

1346

134713471347134713471347134713471347134713471348134813481348134813481348134813481348134813481348134913491349134913491349134913491349134913491349134913511351135113511351135113511351135113511351135313531353135313531353135313531353135313531353

1354

13551355135513551355135513551355135513551355135513551357135713571357135713571357135713571357

1357

1358

1359135913591359135913591359

1359

13601360136013601360136013601360136013601360136113611361136113611361136113611361136113611361136313631363

1363

13641364136413641364136413641364136413641364136413641365136513651365136513651365136513651365136513661366

1366

13671367136713671367136713671367136713671367136713691369136913691369136913691369136913691369136913721372137213721372137213721372

1372

137213721372137313731373

13731375

1376137613761376137613761376

1376

13771377137713771377137713771377

1377

1377137713781378137813781378137813781378137813781378137813801380138013801380138013801380138013801380138013801381138113811381138113811381138113811381138113811381138213821382138213821382138213821382138213821382138313831383138313831383138313831383138313831383138613861386

1386

13881388138813881388138813881388138813881388

13891390

1391139113911391139113911391139113911391139113911392139213921392139213921392139213921392139213931393139313931393139313931393139313931393139313941394139413941394139413941394

1394

13961396139613961396139613961396139613961396139613981398139813981398139813981398

1398

139913991399139913991399139913991399139913991399139914001400140014001400

1400

140214021402140214021402140214021402140214021402140414041404140414041404140414041404140414041404

1405

140614061406140614061406140614061406

1406

14081408

1408

14091409140914091409140914091409140914091409141014101410141014101410141014101410141014101410141314131413141314131413141314131413

1413

141514151415141514151415141514151415141514151416141614161416141614161416141614161416141614171417141714171417141714171417141714171417

1419

14211421142114211421142114211421142114211421142114211422142214221422

1422

142414241424142414241424142414241424142414241424142414251425

1425

142714271427142714271427

1427

14281428142814281428142814281428142814281428142814311431143114311431143114311431143114311431

1432

14341434143414341434143414341434143414341434143414341435143514351435143514351435143514351435143514351436143614361436

1436

14361436143614361436143614361439143914391439143914391439143914391439143914391440144014401440144014401440144014401440144014401441144114411441144114411441

1441

144114411441144214421442144214421442144214421442144214421442144214431443144314431443144314431443144314431443144314451445144514451445144514451445144514451445144514451447144714471447144714471447

1447

14551455

1455

1457

145814581458145814581458145814581458145814581458146014601460146014601460146014601460146014601460146114611461146114611461146114611461

1461

14621462146214621462146214621462146214621462146214621463146314631463146314631463146314631463146314631464146414641464146414641464146414641464146414641465146514651465146514651465

1465

14661466146614661466146614661466146614661466146714671467146714671467146714671467146714671467

1468

147014701470147014701470147014701470147014701470147114711471147114711471147114711471147114711471147314731473147314731473147314731473147314731473147414741474147414741474147414741474147414741474147514751475147514751475147514751475147514751476147614761476147614761476147614761476147614771477147714771477147714771477147714771477147714791479147914791479147914791479147914791479147914791481148114811481148114811481148114811481148114821482148214821482148214821482148214821482148314831483148314831483148314831483148314831483148514851485148514851485148514851485148514851485

1486

14881488

1488

14891489148914891489

1489

1490149014901490

1490

1493149314931493149314931493149314931493

1493

14931494149414941494149414941494149414941494149414941496149614961496149614961496149614961496149614961497149714971497149714971497149714971497149714971498149814981498149814981498149814981498149814981498149914991499149914991499149914991499149914991499149915001500150015001500150015001500150015001500150115011501150115011501150115011501

1501

1502

1502

150315031503150315031503150315031503150315031503150615061506150615061506150615061506150615061506150715071507150715071507150715071507150715071507

1509

151015101510

1510

151115111511151115111511

1511

15121512151215121512151215121512151215121512151315131513151315131513151315131513151315131513151415141514151415141514151415141514151415141514151415161516151615161516

1516

15181518151815181518151815181518151815181518

1520

1522152215221522152215221522152215221522152215231523

1523

15241524152415241524152415241524

1524

152415241524152515251525152515251525

1525

15271527152715271527

1527

15281528152815281528152815281528

1528

15291529

1529

153015301530153015301531153115311531153115311531

1531

1532153215321532153215321532153215321532153215321533153315331533153315331533153315331533153315331534153415341534153415341534153415341534153415341535153515351535153515351535153515351535153515351535153615361536153615361536153615361536153615361537153715371537153715371537153715371537153715371538153815381538153815381538153815381538153815381539153915391539153915391539153915391539153915391539

1541

1542154215421542154215421542154215421542154215451545154515451545154515451545154515451545154515461546154615461546154615461546154615461546155015501550155015501550155015501550155015501550155015521552

1552

155315531553155315531553155315531553155315531557

1557

155815581558155815581558155815581558155815581558155915591559155915591559155915591559155915591559155915601560156015601560156015601560156015601560156015631563156315631563156315631563

1563

156515651565156515651565156515651565156515651565156515661566156615661566156615661566156615661566

1567

1569156915691569156915691569156915691569

1569

156915711571157115711571

1571

15721572157215721572157215721572157215721572157215741574157415741574157415741574157415741574157415751575

1575

157715771577157715771577157715771577157715771577157815781578157815781578157815781578157815781578157915791579

1579

15811581

1581

1583158315831583158315831583

1583

1584

1585158515851585158515851585158515851585158515871587158715871587158715871587

1587

158815881588158815881588158815881588158815881588158915891589158915891589158915891589158915891590159015901590159015901590159015901590159015901592

1592

15931593159315931593159315931593

1593

1597159715971597

1597

15981598159815981598159815981598159815981598159815991599159915991599159915991599159915991599159915991600

1600

1601160116011601160116011601160116011601160116011602160216021602160216021602160216021602160216021603160316031603160316031603160316031603

1603

1604

1607160716071607160716071607160716071607160716071608160816081608160816081608160816081608160816091609160916091609160916091609

1609

161116111611

1611

161316131613161316131613161316131613161316131613161316141614161416141614161416141614161416141614161416141615161516151615161516151615161516151615161516151616161616161616161616161616161616161616161616161616161816181618161816181618

1618

1619

1619

1620

16211621162116211621

16211623

1624162416241624162416241624162416241624

1624

1624162616261626162616261626162616261626162616261626162616271627162716271627162716271627

1627

162716271629162916291629162916291629162916291629

1629

16291633163316331633163316331633163316331633163316331635163516351635163516351635163516351635163516361636163616361636163616361636163616361636163716371637

1637

16381638163816381638163816381638

1638

1638163816381638163916391639163916391639163916391639163916391639164016401640

1640

16411641164116411641164116411641164116411641

1643

1644164416441644164416441644164416441644164416441644

1645

1647164716471647164716471647

1647

16481648164816481648164816481648164816481648164916491649164916491649164916491649164916491649165016501650165016501650165016501650165016501650165116511651165116511651165116511651165116511651165216521652165216521652165216521652165216521652165416541654165416541654165416541654165416541655165516551655165516551655165516551655165516551655165616561656165616561656165616561656

1656

166016601660166016601660

1660

16631663166316631663166316631663166316631663166416641664166416641664166416641664166416641668166816681668166816681668166816681668166816681670167016701670167016701670167016701670167016701670

1671

167316731673167316731673

1673

1677167716771677167716771677167716771677167716781678167816781678167816781678

1678

1679

1680168016801680168016801680168016801680168016801681168116811681

1681

168216821682168216821682

1682

16831683

1683

1685168516851685168516851685168516851685168516851685168616861686168616861686168616861686168616861686

1688

168916891689168916891689168916891689168916891690

1690

1692169216921692169216921692169216921692169216921693169316931693169316931693169316931693169316931693169416941694

1694

1695169516951695169516951695169516951695169516951695

1697

16981698169816981698169816981698169816981698169916991699169916991699169916991699169916991700170017001700170017001700170017001700170017001703170317031703170317031703170317031703170317031703

1706

170717071707170717071707170717071707170717071710171017101710171017101710171017101710171017101712

1712

1712171217121712171217121712171217121712

1713

17131713171317131713171317131713171317131713

1714

1716171617161716171617161716171617161716171617161717171717171717171717171717171717171717171717171717

1718

17201720172017201720172017201720172017201720172017211721172117211721172117211721

1721

17221722172217221722172217221722172217221722172217221723172317231723172317231723172317231723172317231723172417241724172417241724172417241724172417241724172417261726172617261726172617261726172617261726172617261727172717271727172717271727172717271727

1727

17281728172817281728

1728

1729

173017301730173017301730173017301730173017301730173117311731

1731

1732173217321732173217321732173217321732173217321734173417341734173417341734

1734

1735173517351735

1735

173717371737173717371737173717371737173717371738173817381738173817381738173817381738173817381739173917391739173917391739173917391739173917391740174017401740174017401740174017401740174017401741174117411741174117411741174117411741174117411742174217421742174217421742174217421742174217421742174317431743174317431743174317431743174317431744174417441744

1744

174517451745174517451745174517451745174517451746174617461746174617461746174617461746174617471747174717471747174717471747174717471747174717471748174817481748174817481748174817481748174817481749174917491749

1749

17511751175117511751175117511751175117511751175317531753

17531754

175517551755175517551755175517551755175517551755

17581759175917591759175917591759175917591759

1759

1760176017601760176017601760176017601760176017601760176117611761176117611761176117611761

1761

1762176317631763

1763

1764176417641764176417641764176417641764176417641766176617661766176617661766176617661766176617671767176717671767176717671767

1767

176817681768176817681768176817681768176817681769176917691769176917691769

1769

177017701770177017701770177017701770177017701770177117711771177117711771177117711771177117711771177217721772177217721772177217721772177217721772177417741774177417741774177417741774177417741774177617761776177617761776177617761776177617761778177817781778177817781778

1778

177917791779177917791779177917791779177917791779

1780

17811781178117811781178117811781

1781

1781178117811781178317831783178317831783178317831783178317831784178417841784

1784

178617861786

1786

17871787

1787

17891789178917891789178917891789178917891789179017901790179017901790179017901790179017901790

1791

179217921792179217921792179217921792179217921792179317931793179317931793179317931793179317931793179417941794179417941794179417941794179417941795179517951795179517951795179517951795179517981798179817981798179817981798179817981798

1801

18051805180518051805180518051805180518051805180618061806180618061806180618061806180618061806180818081808180818081808180818081808180818081809180918091809180918091809180918091809180918091809181018101810181018101810181018101810181018101810181018111811181118111811181118111811181118111811

1812

1814181418141814181418141814

1814

18151815181518151815181518151815181518151815181518161816181618161816181618161816181618161816

1820

18211821

1821

18231823182318231823182318231823

1823

18241824182418241824182418241824182418241824182518251825182518251825182518251825182518251825182618261826182618261826182618261826182618261827182718271827182718271827182718271827182718271828182818281828182818281828182818281828182818281828182918291829182918291829182918291829182918291829183018301830183018301830183018301830183018301830183118311831183118311831183118311831183118311832

1832

1834183418341834

1834

1834183418341834183418341834183518351835183518351835183518351835183518351836183618361836183618361836

1836

183718371837183718371837183718371837183718371837

18381840

18411841184118411841184118411841184118411841184218421842184218421842184218421842184218421842184418441844184418441844184418441844184418441844184618461846184618461846184618461846184618461847184718471847184718471847184718471847184718471848

1848

1849

185018501850185018501850185018501850185018501850185118511851185118511851185118511851185118511851

1852

18531853185318531853185318531853185318531853185518551855185518551855185518551855185518551855185618561856185618561856185618561856185618561856185618571857

1857

185918591859

1859

1861

1862

1862

18641864186418641864

1864

1866186618661866186618661866186618661866186618661866

1868

18701870

1870

187318731873187318731873187318731873187318731874187418741874187418741874187418741874

1874

18751875187518751875187518751875187518751875

1875

1879187918791879187918791879187918791879187918791880188018801880188018801880188018801880

1880

1881

1884188418841884188418841884

1884

18851885188518851885188518851885188518851885188518861886188618861886

1886

188718871887188718871887188718871887188718871887188818881888188818881888188818881888188818881888189018901890189018901890189018901890189018901890189118911891189118911891189118911891189118911891

1891

189218921892

1892

189618961896189618961896189618961896189618961898189818981898189818981898189818981898189818981899189918991899189918991899189918991899189919001900190019001900190019001900190019001900190019011901190119011901190119011901

1901

1902190219021902190219021902190219021902190219031903190319031903190319031903190319031903190319041904190419041904190419041904190419041904190419051905190519051905190519051905190519051905

1906

191019101910191019101910191019101910191019101910191019141914191419141914191419141914191419141914191519151915191519151915191519151915191519151915

1916

1917191719171917191719171917191719171917191719171918191819181918191819181918191819181918191819181919191919191919191919191919191919191919191919191920192019201920192019201920192019201920192019211921192119211921192119211921192119211921192119221922192219221922192219221922192219221922192219241924192419241924192419241924192419241924192519251925192519251925192519251925192519251925192519261926192619261926192619261926192619261926192719271927

1927

19281928

1928

192919291929192919291929192919291929192919291929193219321932

1932

1934

193519351935193519351935193519351935193519351935

19361937

1938193819381938193819381938193819381938193819391939193919391939193919391939193919391939194019401940194019401940194019401940194019401940194119411941194119411941194119411941194119411942194219421942194219421942194219421942194219421942194319431943194319431943

1943

19441944194419441944194419441944194419441944194419441945194519451945194519451945

1945

194619461946

19461947

19481948194819481948194819481948194819481948194819491949194919491949194919491949194919491949194919491951195119511951195119511951195119511951195119511952195219521952195219521952195219521952195219521953195319531953

1953

1954195419541954195419541954195419541954

1954

1956195619561956

1956

19571957195719571957195719571957195719571957195719581958195819581958195819581958195819581958

1958

1959195919591959195919591959195919591959195919601960196019601960196019601960196019601960196219621962196219621962196219621962196219621964196419641964

1964

19651965196519651965196519651965196519651965

1965

196619661966196619661966196619661966

1966

19671967196719671967196719671967196719671967196719681968

1968

1970197019701970197019701970

1970

19701970197019711971197119711971

1971

1973197319731973

1973

197419741974197419741974197419741974197419741974197719771977197719771977197719771977197719771978197819781978197819781978197819781978197819781979197919791979197919791979197919791979197919791979198119811981198119811981198119811981198119811981198119821982198219821982198219821982198219821982198219831983198319831983198319831983198319831983198319841984198419841984198419841984198419841984198419841985198519851985198519851985198519851985198519881988

1988

198919891989198919891989198919891989

1989

19901990199019901990199019901990199019901990199019911991199119911991199119911991199119911991199119951995199519951995199519951995

1995

19991999199919991999199919991999199919991999199920002000200020002000200020002000200020002000

2000

20032003200320032003200320032003200320032003200320032005200520052005

2005

200720072007200720072007200720072007200720072007200720082008200820082008200820082008

2008

2009200920092009

2009

201320132013201320132013201320132013201320132013

2015

2016

2016

20182018201820182018201820182018201820182018201920192019201920192019201920192019201920192019202120212021202120212021202120212021202120212021202120222022202220222022202220222022202220222022202220232023202320232023202320232023202320232023202420242024202420242024202420242024202420242024202720272027202720272027202720272027202720272027202920292029202920292029202920292029202920292029203120312031203120312031203120312031

2031

20332033203320332033203320332033

2033

2035203520352035203520352035203520352035203520352036203620362036203620362036203620362036203620362038203820382038203820382038203820382038203820382038

2040

2040

20412041204120412041204120412041204120412041

2042

204320432043204320432043204320432043204320432044204420442044204420442044204420442044

2044

20452045

2045

204620462046

2046

20472047204720472047204720472047204720472047

2048

2050205020502050205020502050205020502050205020512051205120512051205120512051205120512051205120512052205220522052205220522052

2052

20532053

2053

2055

2056

20592059205920592059205920592059205920592059205920602060206020602060206020602060206020602060

2061

206220622062206220622062206220622062206220622062206320632063206320632063206320632063206320632063

2064

2067206720672067206720672067206720672067206720692069206920692069206920692069206920692069

2070

20712071207120712071207120712071207120712071207120722072207220722072207220722072207220722072207220732073207320732073

2073

2076207620762076207620762076

2076

2077

2077

20782078207820782078207820782078207820782078208020802080208020802080208020802080208020802080208020842084208420842084

2084

2085208520852085

2085

208620862086208620862086208620862086208620862087208720872087

2087

20882088208820882088

2088

20902090209020902090209020902090209020902090

2090

209120912091209120912091209120912091209120912092209220922092209220922092209220922092209220922093209320932093209320932093209320932093209320932096

2096

2097209720972097209720972097209720972097209720972098209820982098209820982098209820982098209820982098209920992099209920992099209920992099209920992099210021002100210021002100210021002100210021002101210121012101210121012101210121012101210121022102210221022102210221022102

2102

21032103210321032103210321032103210321032103210321042104210421042104210421042104210421042104210621062106210621062106210621062106210621062107210721072107210721072107210721072107210721072107210821082108210821082108210821082108210821082108

2109

211121112111

2111

21122112211221122112211221122112

2112

21152115211521152115211521152115211521152115211521162116211621162116211621162116211621162116211621162117211721172117

2117

21172117211721172117211721172118211821182118211821182118211821182118211821182119211921192119211921192119

2119

2120

21212121212121212121212121212121

2121

21212121212321232123

2123

2124212421242124212421242124212421242124212421242124

2125

212721272127212721272127212721272127

2127

212921292129212921292129212921292129212921292129212921312131213121312131213121312131213121312131213121312133213321332133213321332133

2133

2134

2136213621362136

2136

213721372137213721372137213721372137213721372138213821382138213821382138

2138

213921392139213921392139213921392139213921392139214021402140

2140

214121412141

2141

21422142214221422142214221422142214221422142

2143

2144

2144

2146214621462146214621462146214621462146214621462147214721472147214721472147214721472147214721482148214821482148214821482148214821482148214821492149214921492149214921492149214921492149214921512151215121512151215121512151215121512151

2151

2152215221522152215221522152215221522152215221522154215421542154215421542154215421542154215421542155215521552155215521552155215521552155

2155

2156

215721572157

2157

2159215921592159215921592159215921592159215921592162216221622162216221622162216221622162216221622164216421642164216421642164216421642164

2164

21652165

2165

2166216621662166216621662166216621662166216621662166216721672167

2167

2168216821682168

2168

21692169216921692169216921692169216921692169

2169

2170217021702170217021702170217021702170217021702171217121712171217121712171217121712171217121712172

2172

2174217421742174217421742174217421742174217421742175

2175

217721772177

2177

2178217821782178217821782178217821782178217821782179217921792179217921792179217921792179217921792180218021802180218021802180218021802180218021802183218321832183

2183

2183218321832183218321832184218421842184218421842184218421842184218421842186

2186

21872187218721872187218721872187218721872187

2190

2192219221922192219221922192219221922192219221922194219421942194219421942194219421942194219421942197219721972197219721972197219721972197

2197

219921992199219921992199219921992199219921992201220122012201220122012201220122012201220122012201

2203

220422042204220422042204220422042204220422042204220622062206220622062206220622062206220622062206220722072207220722072207220722072207220722072207220822082208220822082208220822082208220822082209220922092209220922092209220922092209220922112211221122112211221122112211221122112211

2211

221222122212221222122212221222122212221222122212221322132213

2213

221622162216

2216

221722172217

2217

22182218221822182218221822182218221822182218221922192219221922192219221922192219221922192219221922212221222122212221222122212221222122212221222222222222222222222222222222222222222222222222222422242224222422242224222422242224222422242226222622262226

2226

22272227222722272227222722272227222722272227222722282228222822282228222822282228222822282228222822322232223222322232223222322232223222322232223222332233223322332233223322332233223322332233

2234

2237223722372237223722372237223722372237223722372237

2240

22422242224222422242224222422242224222422242224322432243224322432243224322432243224322432243

2244

22452245224522452245224522452245224522452245224622462246224622462246224622462246224622462246224722472247224722472247224722472247224722472251225122512251225122512251225122512251225122512251

2252

22552255225522552255225522552255225522552255225522562256225622562256225622562256225622562256

2257

22582258225822582258

2258

2260

2262226222622262226222622262226222622262226222622262226422642264226422642264226422642264226422642265226522652265226522652265226522652265226522652266226622662266226622662266226622662266226622662267226722672267226722672267226722672267226722672268226822682268226822682268226822682268226822682268

2269

227022702270227022702270227022702270227022702270227122712271227122712271227122712271227122712271227222722272227222722272227222722272227222722272227222742274227422742274227422742274227422742274227522752275227522752275227522752275

2275

2277227722772277227722772277227722772277227722772277

22782281

22872287228722872287228722872287228722872287228722882288228822882288228822882288228822882288228822892289228922892289228922892289228922892289228922902290229022902290

2290

2293229322932293229322932293229322932293229322932296229622962296229622962296229622962296229622962296229722972297229722972297229722972297229722972297229822982298229822982298229822982298229822982298229922992299229922992299

2299

23002300230023002300230023002300230023002300230623062306230623062306230623062306230623062306230723102310231023102310231023102310231023102310231023112311231123112311231123112311231123112311231123142314231423142314231423142314231423142314231423152315231523152315231523152315231523152315231623162316231623162316231623162316231623162317

2317

23182318

2318

23232323232323232323232323232323232323232323232323242324232423242324232423242324232423242324232423252325232523252325232523252325232523252325232523262326

2326

23282328232823282328232823282328232823282328232823282329

2329

2330233023302330

2330

23322332233223322332233223322332233223322332

2332

2334233423342334233423342334

2334

233523352335

2335

233623362336233623362336233623362336233623362338233823382338233823382338233823382338

2338

2339

2339

234023402340

2340

23412341234123412341234123412341234123412341234123432343234323432343234323432343234323432343234323432344234423442344234423442344234423442344

2344

2346

2348234823482348

2348

235023502350235023502350235023502350235023502350

2351

23512351

2351

235223522352235223522352235223522352235223522352235223552355235523552355235523552355235523552355235623562356235623562356235623562356235623562356235623582358235823582358235823582358235823582358

2358

2360236023602360236023602360236023602360236023602361236123612361236123612361236123612361

2361

23622362236223622362236223622362236223622362236323632363

2363

236523652365236523652365236523652365236523652365236523662366236623662366236623662366236623662366

2367

23682368236823682368236823682368236823682368236923692369236923692369236923692369

2369

23702370237023702370237023702370237023702370237123712371

2371

2372

23732373237323732373237323732373237323732373237423742374237423742374237423742374237423742374237523752375237523752375237523752375237523752375237523762376237623762376237623762376237623762376

2377

23802380

2380

23812381

2381

238223822382

2382

238323832383238323832383

2383

2384238423842384

2384

23862386238623862386238623862386238623862386238623862388238823882388238823882388238823882388238823882389238923892389238923892389238923892389238923892389239023902390239023902390239023902390239023902392239223922392239223922392239223922392239223922393239323932393239323932393239323932393239323942394239423942394

2394

2395239523952395239523952395239523952395

2395

23952396239623962396239623962396239623962396239623962397239723972397239723972397239723972397239723972399239923992399239923992399239923992399

2399

240024002400240024002400240024002400240024002400240124012401240124012401240124012401240124012403240324032403240324032403240324032403240324032404240424042404240424042404240424042404240424052405240524052405240524052405

2405

24052405240524052406240624062406240624062406240624062406240624062407240724072407240724072407240724072407240724072408240824082408240824082408240824082408240824082411241124112411241124112411241124112411241124112412

2412

241324132413241324132413241324132413241324132413

2413

24142414241424142414241424142414241424142414241424152415241524152415

2415

24162416241624162416241624162416241624162416241624172417241724172417241724172417241724172417241724172419241924192419241924192419241924192419241924192420242024202420242024202420242024202420242024242424242424242424242424242424242424242424

2424

2425242524252425

2425

24262426242624262426242624262426242624262426242624272427

2427

2428242824282428242824282428242824282428

2428

242924292429242924292429242924292429242924292429243324332433243324332433243324332433243324332433243424342434243424342434243424342434243424342434243524352435243524352435243524352435243524352435

2437

2438

2440244024402440244024402440244024402440244024402441244124412441

2441

24422442244224422442244224422442244224422442244224442444244424442444244424442444244424442444244424452445244524452445

2445

2446244624462446244624462446244624462446244624462448244824482448244824482448244824482448244824482448244924492449244924492449244924492449244924492449

2450

24512451245124512451245124512451245124512451245124512452245224522452245224522452245224522452245224522454245424542454245424542454245424542454245424542455245524552455

2455

24562456245624562456245624562456245624562456245624572457245724572457245724572457245724572457245724582458245824582458245824582458245824582458245824602460246024602460246024602460246024602460246124612461

2461

24622462246224622462246224622462246224622462246424642464246424642464246424642464246424642464246624662466246624662466

2466

24672467

2467

2468246824682468246824682468246824682468

2468

247024702470247024702470247024702470247024702470247124712471247124712471

2471

24722472247224722472247224722472247224722472247324732473247324732473247324732473247324732474247424742474247424742474247424742474247424742474247524752475247524752475247524752475247524752475247524762476

2476

24772477247724772477247724772477247724772477

2477

247824782478

2478

2479247924792479247924792479247924792479247924792480248024802480248024802480248024802480248024812481248124812481248124812481

2481

248424842484

2484

2486248624862486248624862486248624862486248624872487248724872487248724872487248724872487248724872489

2489

249024902490249024902490249024902490249024902490249124912491249124912491249124912491249124912491249224922492249224922492249224922492

2492

2493249324932493249324932493249324932493249324932494

2494

24952495249524952495249524952495249524952495249524972497249724972497

2497

2501250125012501250125012501250125012501250125032503250325032503250325032503250325032503250325052505250525052505

2505

25072507

2507

2508

2508

250925092509

2509

2511251125112511251125112511251125112511251125112513251325132513251325132513251325132513251325132516

2516

251925192519251925192519251925192519251925192519

2519

2520252025202520252025202520252025202520252025202520252125212521252125212521252125212521252125212521252225222522

2522

2523252325232523

2523

2525

2525

2526

2526

252825282528252825282528252825282528252825282528

25292530

25312531253125312531253125312531253125312531253225322532253225322532253225322532253225322532

2533253325332533253325332533253325332533253325332534253425342534253425342534253425342534253425342535253525352535

2535

2536253625362536253625362536253625362536253625362536253725372537253725372537253725372537253725372537253825382538253825382538253825382538253825382540254025402540254025402540254025402540254025402541254125412541254125412541254125412541

2541

25422542

2542

25442544254425442544254425442544254425442544

2545

25472547254725472547254725472547254725472547254725472548254825482548254825482548254825482548254825482550255025502550255025502550255025502550255025502550255125512551255125512551255125512551255125512551255225522552255225522552255225522552255225522552255225532553255325532553255325532553255325532553255325532554255425542554255425542554255425542554255425542554

2555

255625562556255625562556255625562556255625562556

2557

25582558255825582558255825582558255825582558255925592559255925592559255925592559255925592559

2560

256125612561256125612561256125612561256125612561256225622562256225622562256225622562256225622562

2563

25642564256425642564256425642564

2564

256525652565

2565

256525652565

2565

2567256725672567256725672567256725672567256725672570257025702570257025702570

2570

257025702570257025712571257125712571257125712571257125712571257125722572257225722572257225722572257225722572257225742574257425742574257425742574257425742574257425752575257525752575257525752575257525752575257525762576257625762576

2576

257725772577

2577

257925792579257925792579257925792579257925792580258025802580258025802580258025802580258025812581

2581

2582

2583258325832583258325832583258325832583258325832584258425842584258425842584258425842584258425842587258725872587258725872587258725872587258725872589258925892589258925892589258925892589258925892590259025902590259025902590259025902590259025912591259125912591259125912591

2591

2592

2594259425942594259425942594259425942594259425942595259525952595259525952595259525952595259525962596259625962596259625962596259625962596259625962598259825982598259825982598259825982598259825982599259925992599259925992599259925992599259925992599260026002600260026002600260026002600260026002600260126012601260126012601260126012601260126012601260226022602260226022602

2602

2604

26062606260626062606260626062606260626062606260626062607260726072607260726072607260726072607260726072608260826082608260826082608

2608

2609260926092609260926092609260926092609260926092609261026102610261026102610261026102610261026102610

2611

26122612261226122612261226122612261226122612261326132613261326132613261326132613261326132613261426142614261426142614

2614

26162616261626162616261626162616261626162616261626192619261926192619261926192619261926192619262026202620262026202620262026202620262026202620262026212621262126212621262126212621

2621

262126212624262426242624262426242624262426242624262426272627262726272627262726272627262726272627262726292629262926292629262926292629262926292629262926292630263026302630

2630

2631263126312631263126312631263126312631

2631

2632263226322632

2632

263326332633

2633

2634263426342634

2634

2635263526352635263526352635263526352635263526352635263626362636263626362636263626362636

2636

26372637263726372637263726372637

2637

26382638263826382638263826382638263826382638263826382639263926392639

2639

26412641264126412641264126412641264126412641264126422642264226422642264226422642264226422642264226422644264426442644264426442644264426442644264426442645264526452645264526452645264526452645

2645

264626462646264626462646264626462646264626462646264626472647264726472647264726472647264726472647265026502650265026502650265026502650265026502650

265126522653265626572657265726572657265726572657

2657

2659265926592659265926592659265926592659265926592660266026602660266026602660266026602660

2660

2662266226622662266226622662

2662

2663266326632663266326632663266326632663

2663

26642664266426642664266426642664266426642664

2664

266726672667266726672667266726672667266726672668266826682668266826682668266826682668266826722672267226722672267226722672267226722672267226742674267426742674267426742674267426742674

2676

26782678

2678

267926792679267926792679267926792679267926792679267926802680268026802680268026802680

2680

2682268226822682268226822682

2682

26832683

2683

268426842684268426842684268426842684268426842684268426862686268626862686268626862686268626862686

2687

2688268826882688268826882688268826882688268826882689

2689

2690269026902690269026902690269026902690269026922692269226922692269226922692269226922692269326932693

2693

269526952695269526952695269526952695269526952695

2697

2698269826982698269826982698269826982698269826982699269926992699269926992699269926992699269927002700270027002700270027002700

2700

27012701270127012701270127012701270127012701

2701

2703

270527052705270527052705270527052705270527052705270527082708

2708

2709270927092709270927092709270927092709

2709

271127112711271127112711271127112711271127112711271227122712271227122712271227122712271227122712271327132713

2713

2715

2715

2716

27172717

2717

271827182718271827182718271827182718271827182719271927192719271927192719271927192719271927192719

2720

272127212721

2721

272227222722272227222722272227222722

2722

2723272327232723272327232723272327232723272327262726

2726

2727272727272727272727272727

2727

2728

27292729272927292729272927292729272927292729272927292730273027302730273027302730273027302730273027302731273127312731273127312731273127312731273127312731273227322732273227322732273227322732273227322734273427342734273427342734273427342734273427352735273527352735273527352735273527352735273527352736273627362736273627362736273627362736273627372737273727372737273727372737273727372737273727382738273827382738273827382738273827382738

2738

27392739273927392739

2739

27392739273927392739273927402740274027402740274027402740

2740

27422742

2742

2745

2745

27462746274627462746

2746

2747

274827482748274827482748274827482748274827482749274927492749274927492749274927492749274927502750275027502750275027502750275027502750

2751

2752275227522752275227522752275227522752275227552755

2755

2756275627562756275627562756275627562756275627562756275727572757275727572757275727572757

2757

276127612761276127612761276127612761276127612761276127642764

27642765

27682768276827682768276827682768276827682768276827682769276927692769276927692769276927692769276927692770277027702770277027702770

2770

27712771277127712771277127712771277127712771

2772

2773277327732773277327732773

2773

277427742774277427742774277427742774

2774

27812781

2781

27832783278327832783278327832783278327832783278327842784278427842784278427842784278427842784278427842785278527852785278527852785278527852785278527852786278627862786278627862786278627862786278627862790279027902790279027902790279027902790279027902790279227922792279227922792279227922792279227922792279227942794279427942794279427942794279427942794279427962796279627962796279627962796279627962796279627962799279927992799279927992799279927992799279927992799280128012801280128012801280128012801280128012801280328032803280328032803280328032803280328032803280428042804

2804

2805280528052805280528052805280528052805280528052807280728072807280728072807280728072807280728072810

2810

2811

2812281228122812281228122812281228122812281228122813281328132813281328132813281328132813281328132814281428142814281428142814281428142814281428162816281628162816

2816

2818

28212821282128212821282128212821282128212821282128222822282228222822282228222822282228222822282228232823282328232823282328232823

2823

2824282428242824282428242824

2824

28262826

2826

28272827282728272827282728272827282728272827282728282828

2828

28292829282928292829282928292829282928292829282928322832283228322832283228322832

2832

283228322832283328332833283328332833283328332833283328332833

2834

283528352835283528352835283528352835

2835

28362836283628362836283628362836283628362836283628382838283828382838283828382838283828382838283928392839283928392839283928392839283928392839284028402840284028402840284028402840284028402840284228422842284228422842284228422842284228422845284528452845284528452845284528452845284528452848284828482848284828482848284828482848284828492849284928492849

2849

28502850

2850

285128512851285128512851285128512851285128512851285228522852285228522852285228522852285228522852

2853

28552855285528552855285528552855285528552855285528552857

2857

28592859285928592859285928592859285928592859

2859

28602860286028602860286028602860286028602860286028612861286128612861286128612861

2861

2862286228622862286228622862286228622862286228622863286328632863286328632863286328632863286328632863286428642864

2864

2865286528652865286528652865286528652865286528652866286628662866286628662866286628662866286628662870287028702870287028702870287028702870287028702871287128712871287128712871287128712871287128712871287228722872287228722872287228722872287228722872287328732873287328732873287328732873287328732873287328752875287528752875287528752875287528752875287528762876287628762876287628762876287628762876287628772877287728772877

2877

287828782878287828782878287828782878287828782878

2881

288228822882288228822882288228822882

2882

288428842884288428842884

2884

2885288528852885288528852885288528852885288528852885288728872887288728872887288728872887288728872887

2892

289328932893

2893

28952895289528952895289528952895

2895

28972897289728972897289728972897289728972897289728972898289828982898289828982898289828982898289828982899289928992899289928992899289928992899289929032903

2903

2904290429042904290429042904290429042904290429042906290629062906290629062906290629062906290629072907290729072907

2907

2908290829082908290829082908290829082908290829092909290929092909290929092909290929092909290929092910291029102910291029102910291029102910291029102911291129112911291129112911291129112911291129112911291229122912291229122912291229122912291229122912291329132913291329132913291329132913

291329152916

29182918291829182918291829182918291829182918292029202920292029202920292029202920292029202920292229222922292229222922292229222922292229222922292429242924

2924

29252925292529252925292529252925292529252925292529252928292829282928

2928

292929292929292929292929

29292930

293229322932293229322932293229322932293229322932293629362936293629362936293629362936293629362936

2937

29382938293829382938293829382938293829382938

2939

2941294129412941294129412941294129412941

2941

2942294229422942

2942

29452945294529452945294529452945294529452945294529452946294629462946294629462946294629462946294629462946294729472947294729472947294729472947294729472947294829482948294829482948294829482948294829482948295529552955295529552955295529552955295529552955295629562956295629562956295629562956295629562956

2958

29592959295929592959295929592959295929592959295929602960296029602960296029602960296029602960296029652965296529652965296529652965296529652965296529672967296729672967296729672967

2967

296829682968

2968

2968296829682968296829682968

2969

297229722972297229722972

2972

29732973297329732973

2973

297529752975

2975

2976297629762976297629762976297629762976

2976

297729772977297729772977297729772977297729772977297829782978297829782978297829782978297829782978297829802980298029802980298029802980298029802980298029812981298129812981298129812981298129812981298129832983298329832983298329832983298329832983298329832986298629862986298629862986298629862986298629862987298729872987

2987

29892989

2989

299029902990299029902990299029902990299029902990299029922992299229922992299229922992299229922992299329932993299329932993299329932993299329932993

29952995299529952995299529952995

2995

299629962996

2996

2998299829982998299829982998

2998

2999

30013001300130013001

3001

3002300230023002300230023002300230023002300230023003300330033003300330033003300330033003300330033008300830083008300830083008300830083008300830093009300930093009300930093009300930093009300930103010301030103010301030103010301030103010301030113011301130113011301130113011301130113011301130143014301430143014301430143014301430143014301430143015301530153015301530153015301530153015301530153016301630163016301630163016301630163016

3016

3017301730173017301730173017

3017

3019301930193019301930193019

3019

302030203020302030203020302030203020302030203021302130213021302130213021302130213021302130213021

3022

3023

3026302630263026302630263026302630263026302630263027

3027

3029

303030303030303030303030303030303030303030303031303130313031

3031

303230323032303230323032303230323032303230323036303630363036303630363036303630363036303630373037303730373037303730373037303730373037303730373038

3038

304030403040304030403040304030403040304030403040304130413041

3041

30433043304330433043304330433043

3043

30433043

3044

304630463046304630463046304630463046304630463046

3047

3049304930493049304930493049304930493049304930493050305030503050305030503050305030503050305030503051305130513051305130513051305130513051

3051

30523052305230523052305230523052305230523052305230543054305430543054305430543054305430543054305430543056305630563056305630563056305630563056305630563057305730573057305730573057305730573057305730583058305830583058305830583058305830583058306030603060306030603060306030603060306030603061306130613061306130613061

3061

3062306230623062306230623062306230623062306230623063306330633063306330633063306330633063306330633064306430643064306430643064306430643064306430643064

3065

306830683068306830683068306830683068306830683068

3070

3071307130713071307130713071307130713071307130713075

3075

307630763076307630763076307630763076307630763077

3077

30783078

3078

30793079307930793079307930793079307930793079307930803080308030803080

3080

308130813081308130813081308130813081308130813081

3082

308330833083308330833083308330833083308330833085

3085

308630863086308630863086308630863086308630863086308830883088308830883088308830883088308830883088

3088

30893089308930893089308930893089308930893089

3089

3090

30913091309130913091

3091

3093

3093

3094

30963096309630963096309630963096309630963096309830983098309830983098309830983098309830983098309930993099309930993099309930993099309930993099310031003100

3100

31013101310131013101310131013101310131013101310131023102310231023102310231023102310231023102310231023103310331033103310331033103310331033103310331063106310631063106

3106

310731073107310731073107310731073107310731073108

3108

3109310931093109310931093109310931093109310931093112311231123112311231123112311231123112311231123115311531153115311531153115311531153115

3115

311531163116311631163116311631163116

31163118

31193119311931193119311931193119311931193119312031203120312031203120312031203120312031203120312131213121312131213121312131213121312131213121312131223122

3122

31233123

3123

31243124312431243124312431243124312431243124312431243126

3126

31283128312831283128312831283128312831283128312931293129312931293129312931293129312931293129

3131

31323132313231323132313231323132313231323132313331333133

3133

313431343134313431343134313431343134

3134

31353135

3135

31413141314131413141314131413141314131413141314131423142314231423142314231423142314231423142314331433143314331433143314331433143314331433143314331453145314531453145

3145

31483148314831483148314831483148314831483148314831493149314931493149314931493149314931493149314931503150315031503150315031503150

3150

31513151315131513151315131513151315131513151

3154

315831583158315831583158315831583158315831583158315931593159315931593159315931593159315931593159

3160

316131613161316131613161316131613161316131613161316131623162316231623162316231623162316231623162316231633163316331633163316331633163316331633163

3165

316731673167316731673167316731673167316731673167316831683168316831683168

3168

317131713171317131713171317131713171317131713171

31723172

3172

31733173317331733173317331733173317331733173

3173

31743174317431743174317431743174317431743174317431773177317731773177317731773177317731773177

3179

31813181

3181

3181

3181

3183

3183

31843184318431843184318431843184

3184

3186318631863186

3186

318831883188318831883188318831883188318831883189318931893189318931893189318931893189318931903190319031903190319031903190319031903190319031913191319131913191319131913191319131913191319131913192319231923192319231923192319231923192319231923195319531953195

3195

31963196319631963196319631963196319631963196319731973197319731973197319731973197319731973198319831983198319831983198319831983198319832013201320132013201320132013201

3201

320132013201320232023202320232023202320232023202320232023204320432043204320432043204320432043204

3204

32053205

3205

32063206320632063206320632063206320632063206320632073207320732073207320732073207320732073207320832083208320832083208320832083208320832083211321132113211

3211

3212321232123212321232123212321232123212321232123215321532153215321532153215321532153215321532153215

3216

321732173217321732173217

3217

32183218321832183218321832183218321832183218321832193219321932193219321932193219321932193219322032203220322032203220322032203220322032203220322332233223322332233223322332233223322332233223322332253225

3225

3228322832283228

3228

32293229322932293229322932293229

3229

32293229322932293230323032303230323032303230323032303230323032303231323132313231323132313231323132313231323132313237323732373237323732373237323732373237323732373238

3238

3239323932393239323932393239323932393239323932403240324032403240324032403240324032403240324032423242

3242

32463246324632463246324632463246324632463246324732473247324732473247324732473247324732473247

3248

3249324932493249324932493249324932493249324932503250325032503250325032503250

3250

32503250325032513251325132513251325132513251325132513251325132513252325232523252325232523252

3252

3253

3253

32543254325432543254325432543254325432543254325432553255325532553255325532553255325532553255325532563256325632563256325632563256325632563256

3256

32583258325832583258325832583258325832583258

3258

3259325932593259325932593259325932593259325932603260326032603260326032603260326032603260326032603261326132613261326132613261326132613261326132613261326532653265326532653265

3265

326632663266326632663266326632663266326632663266

3267

3268326832683268326832683268326832683268326832683269326932693269326932693269326932693269326932693270327032703270327032703270327032703270327032703273327332733273327332733273

3273

327532753275327532753275327532753275327532753275327732773277327732773277327732773277327732773277327932793279327932793279

3279

3280328032803280328032803280328032803280

3280

32813281

3281

3282328232823282328232823282328232823282328232823282

3284

32873287328732873287328732873287328732873287

3287

328832883288328832883288328832883288328832883288

3288

328932893289

3289

3290329032903290329032903290329032903290329032903291329132913291329132913291329132913291329132913292

32923293

329332933293

3293

32953295329532953295329532953295329532953295329532963296329632963296329632963296329632963296

3298

329932993299

3299

33003300330033003300330033003300330033003300

3300

3302330233023302

3302

3306330633063306330633063306330633063306330633063306

33073309

331033103310331033103310331033103310331033103311331133113311

3311

331233123312331233123312331233123312

3312

33133313

3313

33143314331433143314331433143314331433143314331433143315331533153315

3315

3315331533153315331533153315331833183318

3318

3319

3320

3325332533253325332533253325332533253325332533263326

3326

3327332733273327332733273327332733273327

3327

33273328332833283328332833283328332833283328332833283328332933293329332933293329332933293329332933293330333033303330

3330

33313331333133313331333133313331333133313331333233323332333233323332333233323332333233323332

3333

33343334333433343334333433343334333433343334

3336

33373337333733373337333733373337333733373337333733383338333833383338333833383338333833383338333833393339333933393339333933393339333933393339334033403340334033403340334033403340334033403340334033413341334133413341334133413341334133413341334133423342334233423342334233423342

3342

334233423342334333433343334333433343334333433343334333433343334333443344334433443344334433443344334433443344334433443346334633463346334633463346334633463346334633463347

3347

33503350335033503350335033503350335033503350335233523352335233523352335233523352335233523352335533553355335533553355335533553355335533553357335733573357335733573357335733573357

3357

3358335833583358335833583358335833583358

3358

336033603360336033603360336033603360336033603360

3362

3363336333633363336333633363336333633363336333633365

3365

336633663366336633663366336633663366336633663366336733673367336733673367336733673367

3367

336833683368

3368

3369336933693369336933693369336933693369336933703370337033703370337033703370

3370

337133713371337133713371337133713371337133713371

3372

33723372337233723372337233723372337233723373337333733373337333733373337333733373

3373

337533753375337533753375337533753375337533753375337533763376

3376

33783378337833783378

3378

33793379337933793379337933793379337933793379337933793380338033803380338033803380338033803380338033813381338133813381338133813381338133813381338233823382338233823382338233823382338233823382338333833383338333833383338333833383338333833383338333853385338533853385338533853385338533853385338733873387338733873387338733873387338733873387338833883388338833883388338833883388338833883389

3389

3391339133913391339133913391339133913391339133913391339233923392339233923392339233923392339233923392339233933393339333933393339333933393

3393

33933393339433943394339433943394339433943394

3394

339533953395339533953395339533953395339533953395339633963396339633963396339633963396

3396

33983398339833983398

3398

33993399339933993399339933993399339933993399339933993400340034003400

3400

3403340334033403340334033403340334033403340334033403

34053406

340734073407340734073407340734073407340734073407341034103410341034103410

3410

3413341334133413341334133413341334133413341334133415

3415

341634163416341634163416341634163416341634163416341634183418341834183418341834183418

3418

341934193419341934193419341934193419

3419

34203420

3420

3421342134213421342134213421342134213421342134213422342234223422342234223422342234223422342234223424

3424

3425

3425

34273427342734273427342734273427342734273427342734283428342834283428342834283428342834283428342834293429

3429

343134313431343134313431343134313431343134313431343334333433

3433

34343434343434343434343434343434343434343434343434343435343534353435343534353435343534353435343534353437343734373437343734373437343734373437343734373437343834383438343834383438343834383438343834383438343834413441344134413441344134413441344134413441344134413444344434443444344434443444

3444

344534453445344534453445344534453445344534453445344734473447344734473447344734473447344734473447344734483448344834483448344834483448344834483448344834493449344934493449344934493449

3449

3450345034503450345034503450345034503450345034503451345134513451345134513451345134513451345134513451

3452

3455345534553455345534553455345534553455345534563456345634563456345634563456345634563456345634563458345834583458345834583458345834583458

3458

34583459345934593459345934593459345934593459345934593460346034603460346034603460346034603460346034603460

3461

3462346234623462346234623462346234623462346234623463346334633463346334633463346334633463346334633466346634663466346634663466346634663466346634663466346934693469346934693469346934693469346934693469347234723472347234723472347234723472347234723472347334733473347334733473347334733473347334733473347434743474

3474

34753475347534753475

3475

3476347634763476347634763476347634763476

3476

34763477347734773477347734773477347734773477

3477

347934793479347934793479347934793479347934793480348034803480348034803480348034803480348034803481348134813481348134813481348134813481348134813481348234823482348234823482348234823482348234823482348534853485348534853485348534853485348534853487348734873487348734873487348734873487348734883488348834883488348834883488348834883488348834903490349034903490

3490

349134913491349134913491349134913491349134913492349234923492349234923492349234923492349234923492349334933493349334933493349334933493349334933494

3494

3495349534953495349534953495349534953495349534953501350135013501350135013501350135013501350135013502350235023502

3502

3503350335033503350335033503350335033503

3503

3504350435043504

3504

3505350535053505

3505

3506350635063506350635063506350635063506350635063507350735073507

3507

351235123512351235123512351235123512351235123512

3512

351535153515351535153515351535153515351535153515351535163516351635163516351635163516351635163516351635183518351835183518351835183518351835183518351935193519351935193519

3519

352135213521352135213521352135213521352135213521352235223522352235223522352235223522352235223522

3522

35253525352535253525352535253525352535253525

3525

352635263526352635263526352635263526

3526

352735273527352735273527352735273527352735273527352935293529

3529

35303530353035303530353035303530353035303530353035303532353235323532353235323532353235323532353235323532353335333533353335333533353335333533353335333534353435343534353435343534353435343534353435373537

35373538

353935393539353935393539353935393539

3539

3540354035403540354035403540354035403540354035413541

3541

3542

3543354335433543354335433543354335433543354335433545354535453545354535453545354535453545354535453545354635463546354635463546354635463546354735473547354735473547354735473547

3547

35483548354835483548354835483548354835483548355035503550355035503550

3550

3551355135513551355135513551355135513551355135513552355235523552

3552

3553355335533553355335533553

3553

35543554355435543554

3554

355535553555355535553555355535553555

3555

355635563556

3556

35573557355735573557355735573557355735573557355735573560356035603560356035603560356035603560356035603560356235623562356235623562356235623562356235623563356335633563

3563

3566356635663566356635663566356635663566356635663566356835683568356835683568356835683568356835683568356935693569356935693569356935693569356935693569357035703570357035703570357035703570357035703571357135713571357135713571357135713571357135743574357435743574357435743574357435743574357435753575357535753575357535753575357535753575357535753577357735773577357735773577

3577

3578357835783578357835783578357835783578357835783578358135813581358135813581

3581

3582358235823582358235823582358235823582358235823583358335833583358335833583

3583

358435843584358435843584358435843584358435843584358535853585358535853585358535853585358535853585358735873587

3587

358835883588358835883588358835883588358835883588

3591

35923592

3592

35933593359335933593359335933593

3593

359435943594359435943594359435943594359435943595359535953595359535953595359535953595359535953595359735973597359735973597359735973597359735973597359835983598359835983598359835983598359835983598

35993600360036003600360036003600360036003600360036003601

3601

360336033603360336033603360336033603360336033604360436043604360436043604360436043604360436043605360536053605360536053605360536053605360536053605360636063606

3606

360736073607360736073607360736073607360736073607360836083608360836083608360836083608360836083608

3609

3610361036103610361036103610361036103610

3610

361036113611361136113611361136113611361136113611361136153615361536153615361536153615361536153615361536163616

3616

36193619

3619

3620362036203620362036203620362036203620362036203620362236223622362236223622362236223622362236223622

3624

3625362536253625362536253625362536253625362536253625

3626

3627

3627

362836283628362836283628362836283628362836283628362836293629362936293629362936293629362936293629362936303630363036303630363036303630363036303630

3630

3631363136313631363136313631363136313631363136323632363236323632363236323632363236323632363236323633363336333633363336333633363336333633

3633

3634363436343634363436343634363436343634

3634

363536353635

3635

363636363636

3636

3637

3640

3641

3641

364236423642364236423642364236423642364236423642

3644

3644364436443644364436443644364436443644364436443647364736473647364736473647364736473647364736473647364836483648

3648

3649

36503650365036503650365036503650365036503650365136513651365136513651365136513651365136513651365336533653365336533653365336533653

3653

3656365636563656365636563656

3656

3657365736573657365736573657365736573657365736573658365836583658365836583658365836583658365836583658

3659

3660366036603660366036603660366036603660366036603661366136613661366136613661

3661

3666366636663666366636663666366636663666366636663667366736673667366736673667366736673667366736673668366836683668366836683668366836683668

3668

36693669366936693669366936693669366936693669366936703670367036703670

3670

3671

3671

3672367236723672367236723672367236723672367236723672

3675

3677367736773677367736773677367736773677367736773678367836783678367836783678367836783678367836783679367936793679367936793679367936793679367936793679368036803680368036803680368036803680368036803680368236823682368236823682368236823682368236823682

3683

3684368436843684368436843684368436843684368436843685368536853685368536853685368536853685

3685

3689368936893689368936893689368936893689368936903690369036903690369036903690369036903690369036913691369136913691369136913691369136913691369136913693369336933693369336933693369336933693369336933694369436943694369436943694

3694

3695369536953695369536953695369536953695369536953697369736973697369736973697369736973697369736973697

3698

3700

370437043704370437043704370437043704370437043704370637063706370637063706370637063706

3706

37083708370837083708370837083708370837083708370837083709370937093709370937093709

3709

371137113711

3711

3712

37133713371337133713371337133713371337133713371337163716

3716

37183718

3718

37203720372037203720

3720

372137213721372137213721

3721

3722372237223722372237223722372237223722372237223723372337233723372337233723372337233723372337233723372437243724

3724

3726

3726

37293729372937293729372937293729372937293729372937303730373037303730373037303730373037303730

3733

37343734373437343734373437343734373437343734373437353735373537353735

3735

37363736

3736

373737373737373737373737373737373737373737373738373837383738373837383738373837383738373837383739

3739

37403740374037403740374037403740374037403740374037413741374137413741374137413741374137413741374137413742374237423742374237423742374237423742374237423743

3743

37443744374437443744374437443744374437443744374437443745374537453745374537453745374537453745374537453746374637463746

3746

37483748374837483748374837483748374837483748374837493749374937493749374937493749374937493749

3750

375137513751375137513751375137513751375137513751

3752

3753375337533753375337533753375337533753375337533753375437543754375437543754375437543754375437543754375537553755375537553755375537553755375537553755375737573757375737573757375737573757375737573757

3758

3759375937593759375937593759375937593759375937593759376137613761376137613761376137613761376137613761376237623762376237623762376237623762376237623762376337633763376337633763376337633763376337633763376437643764376437643764376437643764376437643764376437653765376537653765376537653765376537653765376537663766376637663766376637663766376637663766376637663767376737673767376737673767376737673767376737673767376837683768376837683768376837683768376837683768377137713771377137713771377137713771377137713771377137733773377337733773377337733773377337733773377337743774377437743774377437743774377437743774377437763776377637763776377637763776

3776

37783778377837783778377837783778377837783778377837793779377937793779377937793779377937793779377937803780378037803780378037803780378037803780

3780

378137813781378137813781378137813781378137813781

3784

378537853785378537853785378537853785378537853785378737873787378737873787378737873787378737873787378837883788378837883788378837883788378837883788379037903790

37903792

379537953795379537953795379537953795379537953795

3797

37983798379837983798379837983798379837983798379937993799379937993799379937993799379937993800380038003800380038003800380038003800380038003801380138013801380138013801380138013801380138013802380238023802380238023802380238023802

3802

38023803

3803

38043804380438043804380438043804380438043804

3804

38053805380538053805380538053805380538053805380638063806380638063806380638063806380638063806

3807

380838083808380838083808380838083808380838083808381038103810381038103810381038103810381038103810381138113811381138113811381138113811381138113811

3812

38133813381338133813

3813

3815

3816381638163816381638163816381638163816381638163818

3818

3819

382038203820382038203820

3820

382138213821382138213821382138213821382138213821

3823

3825382538253825382538253825382538253825382538253826382638263826382638263826382638263826382638273827382738273827382738273827382738273827382738293829382938293829382938293829382938293829382938303830383038303830

3830

38323832383238323832383238323832383238323832383438343834383438343834383438343834383438343835383538353835383538353835383538353835383538353836383638363836383638363836383638363836383638373837383738373837

3837

38383838383838383838383838383838

3838

3839383938393839383938393839383938393839383938393840384038403840384038403840384038403840384038423842

3842

384838483848384838483848

3848

38493849384938493849384938493849384938493849384938503850385038503850385038503850385038503850385038513851

3851

385338533853

38533856

38573857385738573857385738573857385738573857385738573858

3858

38603860386038603860

3860

386138613861

3861

386238623862

3862

3863

3865386538653865

3865

386838683868386838683868386838683868386838683868387038703870387038703870387038703870387038703870387038723872387238723872387238723872

3872

387338733873387338733873387338733873387338733874387438743874387438743874387438753875387538753875

3875

387738773877387738773877

3877

38793879387938793879387938793879387938793879387938813881388138813881

3881

3882388238823882388238823882388238823882388238833883388338833883388338833883388338833883388338843884388438843884388438843884388438843884388438843886388638863886388638863886388638863886

3886

388738873887388738873887

3887

3888

38893889

3889

389038903890389038903890389038903890389038903890389138913891389138913891389138913891

3891

3892389238923892389238923892389238923892389238923894389438943894389438943894389438943894389438943894389538953895389538953895389538953895389538953895389738973897389738973897389738973897389738973898389838983898389838983898389838983898

3898

389938993899389938993899389938993899389938993899

3901

390239023902390239023902390239023902390239023902390339033903390339033903390339033903390339033903390539053905390539053905390539053905390539053905

3906

39093909390939093909390939093909390939093909391039103910

3910

3912391239123912391239123912391239123912391239123913391339133913391339133913391339133913391339133913391539153915391539153915391539153915391539153915

39163917

39183918391839183918391839183918391839183918391839193919391939193919391939193919391939193919391939223922392239223922392239223922392239223922392439243924392439243924392439243924392439243924392539253925392539253925392539253925392539253925392639263926392639263926392639263926392639263926392739273927

3927

392939293929

3929

393039303930393039303930393039303930393039303930393039333933393339333933393339333933393339333933393339343934393439343934

3934

393539353935393539353935393539353935393539353936393639363936393639363936393639363936393639393939393939393939393939393939393939393939393939413941394139413941394139413941394139413941394139433943394339433943394339433943394339433943

3943

394539453945394539453945394539453945394539453946394639463946394639463946394639463946394639463947394739473947394739473947394739473947394739483948394839483948394839483948394839483948394939493949394939493949394939493949

3949

3950395039503950395039503950

3950

3951395139513951395139513951395139513951395139523952

3952

3953

395439543954395439543954395439543954395439543954395439553955395539553955395539553955395539553955395539573957

3957

39583958

3958

395939593959395939593959395939593959395939593961396139613961396139613961396139613961396139623962396239623962396239623962

3962

3963396339633963396339633963396339633963396339663966396639663966396639663966396639663966

3970

39713971397139713971397139713971397139713971

3971

39723972397239723972397239723972397239723972397239743974397439743974397439743974397439743974397439743976397639763976397639763976397639763976397639763978397839783978397839783978397839783978397839783979397939793979397939793979397939793979397939793980398039803980398039803980398039803980398039803981398139813981398139813981

3981

39833983398339833983398339833983398339833983398339833984398439843984398439843984

3984

3985398539853985398539853985398539853985398539853985398639863986398639863986398639863986398639863986398739873987398739873987398739873987398739873987398739883988398839883988398839883988398839883988399039903990399039903990399039903990399039903990399039923992399239923992

3992

399439943994

3994

3995399539953995399539953995399539953995399539953995

3997

399939993999399939993999

3999

4002400240024002400240024002

4002

400340034003

4003

400540054005400540054005400540054005400540054005400640064006400640064006400640064006400640064007400740074007400740074007

4007

40084008400840084008400840084008400840084008400840084009400940094009400940094009400940094009

4009

401140114011401140114011401140114011

4011

401340134013

4013

4014401440144014

4014

401540154015401540154015401540154015401540154015401640164016401640164016401640164016401640164017401740174017401740174017401740174017

4017

40184018401840184018401840184018401840184018

4020

4021402140214021402140214021402140214021402140214023402340234023402340234023402340234023

4023

402440244024402440244024

4024

40274027402740274027402740274027402740274027402740304030

4030

403140314031403140314031

4031

4032

403340334033403340334033403340334033403340334033

4036

4039

4039

4041

4041

4044404440444044404440444044404440444044404440444045404540454045404540454045404540454045

4045

404540464046404640464046404640464046404640464046404640464047404740474047

4047

405040504050405040504050405040504050405040504050405040514051405140514051405140514051405140514051405140524052405240524052405240524052405240524052405240534053405340534053405340534053405340534053405440544054

4054

405540554055

4055

4056405640564056405640564056405640564056405640564057405740574057405740574057405740574057405740574059405940594059405940594059405940594059405940594063406340634063406340634063406340634063406340634063

4064

406540654065406540654065406540654065406540654067

4067

4068

4069406940694069406940694069406940694069

4069

407040704070407040704070407040704070407040704070407140714071407140714071407140714071407140714071407340734073407340734073407340734073407340734073407340744074

4074

4075407540754075407540754075407540754075407540754075407640764076407640764076407640764076407640764076407640774077407740774077407740774077407740774077407840784078407840784078407840784078407840784080408040804080408040804080

4080

4081408140814081

4081

408340834083408340834083408340834083408340834084408440844084408440844084408440844084408440844084408540854085408540854085408540854085408540854085408640864086408640864086408640864086408640864086408640874087408740874087408740874087408740874087408740884088408840884088408840884088

4088

408940894089408940894089408940894089408940894089408940904090409040904090409040904090409040904090409040954095

4095

409740974097409740974097409740974097

4097

4098409840984098

4098

4099409940994099409940994099409940994099409940994099410341034103410341034103410341034103410341034103410441044104410441044104410441044104410441044104410441074107410741074107410741074107

4107

41074107410841084108410841084108410841084108410841084110411041104110411041104110411041104110411041104113411341134113411341134113411341134113411341144114

4114

41154115411541154115411541154115411541154115

4116

411741174117411741174117411741174117411741174117411841184118411841184118411841184118411841184119411941194119411941194119

4119

4120412041204120412041204120

4120

412141214121412141214121412141214121412141214123412341234123

4123

4124412441244124412441244124412441244124412441254125412541254125

4125

41274127412741274127412741274127

4127

41274127

4128

4129412941294129412941294129412941294129

4129

413041304130413041304130413041304130413041304130413041314131413141314131413141314131413141314131413141324132

4132

4135413541354135413541354135413541354135

4135

41364136413641364136413641364136413641364136413741374137

4137

414041404140414041404140414041404140

4140

41434143414341434143414341434143414341434143414341444144

4144

414941494149414941494149414941494152

4152

415541554155415541554155415541554155415541554155415641564156

4156

41574157415741574157415741574157415741574157415741584158415841584158415841584158415841584158

4158

415941594159415941594159415941594159415941594160416041604160416041604160416041604160416041604160416341634163416341634163416341634163416341634163

4165

416641664166416641664166416641664166416641664166416741674167416741674167416741674167416741674167416841684168416841684168

4168

41694169416941694169

4169

4170417041704170417041704170417041704170417041704174417441744174417441744174417441744174417441744175

4175

417641764176417641764176417641764176417641764176

4177

418141814181418141814181418141814181418141814181418241824182418241824182418241824182418241824182418241854185

4185

4188418841884188418841884188418841884188418841884191419141914191419141914191419141914191419141914191419241924192419241924192419241924192419241924192419341934193419341934193419341934193419341934193419441944194419441944194419441944194

4194

41954195419541954195419541954195419541954195419541954196419641964196419641964196419641964196419641974197419741974197419741974197419741974197419741974198419841984198

4198

419941994199419941994199419941994199419941994199420042004200420042004200420042004200420042004200

4201

4202420242024202

4202

4203420342034203420342034203420342034203420342034204420442044204420442044204420442044204420442044205420542054205420542054205420542054205420542054206420642064206420642064206420642064206420642064207420742074207420742074207420742074207420742104210421042104210421042104210421042104210421042114211421142114211

4211

421542154215421542154215421542154215421542154215421742174217421742174217421742174217421742174217421842184218421842184218421842184218421842184218421942194219421942194219421942194219421942194219422442244224422442244224422442244224422442244224422442254225422542254225422542254225422542254225422642264226

4226

422642264226422642264226

4226

422842284228422842284228422842284228422842284228

4229

423042304230423042304230423042304230423042304231

4231

423342334233423342334233423342334233423342334233

4234

4235

4235

4237

4238

4239423942394239423942394239423942394239423942394239424142414241424142414241424142414241424142414241424242424242424242424242424242424242424242424242424242434243424342434243424342434243

4243

42444244424442444244424442444244424442444244424442454245424542454245424542454245424542454245424542464246424642464246424642464246424642464246424742474247

4247

4249424942494249424942494249424942494249424942494250425042504250425042504250425042504250425042504251425142514251425142514251

4251

42534253425342534253425342534253425342534253425742574257425742574257425742574257425742574259425942594259

4259

4260426042604260426042604260426042604260426042604260426142614261426142614261426142614261426142614261426242624262426242624262426242624262426242624262426242634263426342634263426342634263426342634263426342654265

4265

4266

4266

4266426642664266426642664266426642664266426642674267426742674267426742674267426742674267426742674268426842684268426842684268426842684268426842684268

4270

4271427142714271427142714271427142714271427142714271427242724272427242724272427242724272427242724272

4273

42754275427542754275427542754275

4275

4275427542754277427742774277427742774277427742774277427742794279427942794279427942794279427942794279427942794281428142814281428142814281428142814281428142814282428242824282428242824282428242824282428242834283

4283

4284

4285

4285

428642864286

4286

4286428642864286428642864286428642884288428842884288428842884288428842884288429042904290429042904290429042904290429042904291429142914291429142914291429142914291429142914291429342934293429342934293429342934293429342934293429442944294429442944294429442944294429442944294429442974297429742974297429742974297429742974297429742984298429842984298429842984298429842984298

4298

43004300

4300

43014301430143014301430143014301430143014301430243024302430243024302430243024302430243024303

4303

430443044304430443044304

4304

4305430543054305430543054305430543054305430543054306430643064306

4306

43074307430743074307430743074307430743074307430743084308430843084308430843084308430843084308430843114311431143114311431143114311

4311

4312

4313

4314

4314

4315431543154315431543154315431543154315431543154317

4317

4318

431943194319431943194319431943194319431943194321432143214321432143214321432143214321432143214322432243224322432243224322432243224322432243224323432343234323432343234323432343234323432343234324432443244324432443244324432443244324432443244324

4326

4327432743274327432743274327432743274327432743274327432843284328432843284328432843284328432843284328432943294329432943294329432943294329432943294329433243324332433243324332433243324332433243324333433343334333

4333

43344334433443344334

4334

43354335433543354335433543354335

4335

43374337433743374337433743374337433743374337433743384338433843384338433843384338433843384338

4338

43414341434143414341434143414341

4341

434343434343

4343

434643464346434643464346434643464346434643464347434743474347434743474347434743474347

4347

4347434843484348434843484348434843484348434843484348434943494349434943494349434943494349434943494349

4349

435143514351435143514351435143514351435143514351435143524352435243524352435243524352435243524352435243544354435443544354435443544354435443544354435443554355435543554355435543554355435543554355435543564356435643564356435643564356435643564356

4356

43574357435743574357435743574357435743574357435743584358435843584358435843584358435843584358435843584361436143614361436143614361436143614361436143614361

4362

4364

4365436543654365436543654365436543654365436543654365

4366

4367436743674367436743674367436743674367436743674367

4368

43694369

4369

43704370437043704370437043704370437043704370437043704371437143714371437143714371437143714371437143714371437343734373437343734373437343734373437343734373

43754376

43774377437743774377437743774377437743774377437743774378437843784378437843784378437843784378437843784378437943794379

4379

438143814381438143814381438143814381438143814381438243824382438243824382438243824382438243824382

4383

4386438643864386438643864386438643864386438643864388438843884388438843884388438843884388438843894389438943894389438943894389

4389

4390439043904390439043904390439043904390

4390

4391439143914391439143914391439143914391439143914391439343934393439343934393439343934393439343934393

4393

4394

43944395

43974397439743974397439743974397439743974397439743974398

4398

44004400440044004400440044004400440044004400

4401

4402

4402

4403440344034403440344034403440344034403440344034409440944094409440944094409440944094409440944094410441044104410441044104410441044104410441044104412441244124412441244124412441244124412441244144414441444144414441444144414441444144414441444164416441644164416

4416

4417441744174417441744174417441744174417441744174417441844184418441844184418441844184418441844184418441944194419

4419

4420

4420

442144214421442144214421

4421

4423

4423

442444244424442444244424

4424

4425442544254425442544254425442544254425442544264426

4426

442744274427442744274427442744274427442744274427442844284428442844284428442844284428442844284428443044304430443044304430443044304430

4430

4431

443344334433443344334433443344334433443344334434443444344434443444344434443444344434

4434

44364436443644364436443644364436443644364436443644364441444144414441444144414441444144414441444144424442444244424442444244424442444244424442

4442

4443

44464446444644464446444644464446444644464446444644474447444744474447444744474447444744474447

4448

445044504450445044504450445044504450445044504450445144514451445144514451

4451

4453

445444544454445444544454445444544454445444544454445644564456445644564456445644564456445644564456

4457

445844584458445844584458445844584458445844584458446044604460446044604460446044604460446044604460446244624462446244624462

4462

44634463

4463

446344634463446344634463446344634463446344644464

4464

446644664466446644664466446644664466446644664466446644684468446844684468446844684468446844684468446844694469446944694469446944694469446944694469447144714471447144714471447144714471447144710.0

5.1

.15

Pre

gibo

n's

dbet

a

0 1000 2000 3000 4000 5000id

85

VariableMultipartner Fertility HistoryNo MPF children (ref)MPF (mother) 1.687 1.647MPF (father) 3.108 *** 3.073 **MPF (both) 2.554 ** 2.525 **Relationship Status at Child's BirthMarried (ref)Cohabiting 2.286 * 2.316 *Dating 5.970 *** 6.233 ***DemographicsMother's race

White (ref)Black 1.178 1.167Hispanic/Other race 0.713 0.709

Father's race differs from mother 1.070 1.088Mother's religiousness 1.145 1.132Mother's age 0.935 * 0.936 *Father is 10 years older/younger than mother 0.695 0.709Mother's substance abuse problem 0.849 0.846Mother lived with both parents at age 15 1.536 1.534Father ever in jail (mother's report) 1.544 † 1.503 †

Socioeconomic StatusMother's education

<High school 0.881 0.901High school/GED 0.616 0.626Some college 0.629 0.632College or more (ref)

Mother worked in year before birth 0.705 0.712Father worked in week before birth 1.149 1.155Mother on welfare 0.766 0.769Mother or father owns home 0.952 0.946Relationship CharacteristicsYears known each other 1.031 1.031Mother feels supported (1-3) 0.327 ** 0.326 **Mother reports conflict (1-3) 1.347 1.350Mother reports violence 0.623 0.610Mother distrusts men 0.859 0.867Mother has pro-marriage attitude 1.118 1.150Time IntervalsInterval 1 (months 1-3) (ref)Interval 2 (months 4-6) 0.402 † 0.426 †

Interval 3 (months 7-9) 0.446 * 0.455 *Interval 4 (months 10-12) 0.664 0.679Interval 5 (months 13-15) 0.404 † 0.413 †

Interval 6 (months 16-18) 0.215 *** 0.220 ***Interval 7 (months 19-21) 0.394 0.403Interval 8 (months 22-24) 0.547 0.559Interval 9 (months 25-27) 0.510 0.521Interval 10 (months 28-30) 0.270 * 0.276 *Interval 11 (months 31-33) 0.476 0.486Interval 12 (months 34-36) 0.128 *** 0.131 ***Interval 13 (months 37-50) 0.231 * 0.235 *Model Chi-Square (df) 1158.64(39) *** 1177.58(42) ***Pseudo-R 2

0.118 0.119Note: Estimates based on weighted data. Fit statistics based onunweighted data. †p < .10, *p < .05, **p < .01, ***p < .001

Original Model 4 (from Table 3)

Model 4.2 (from sensitivity

analysis 1)

Table 5. Odds Ratios from Discrete-Time Event Hist ory Model Including Cases with Missing Interviews at Wave 3 (Sensitivit y Analysis 1)

APPENDIX B

86

VariableMultipartner Fertility HistoryNo MPF children (ref)MPF (mother) 1.687 1.803MPF (father) 3.108 *** 2.754 **MPF (both) 2.554 ** 2.200 **Relationship Status at Child's BirthMarried (ref)Cohabiting 2.286 * 2.212 *Dating 5.970 *** 4.773 ***DemographicsMother's race

White (ref)Black 1.178 1.363Hispanic/Other race 0.713 0.832

Father's race differs from mother 1.070 0.732Mother's religiousness 1.145 1.153Mother's age 0.935 * 0.940 *Father is 10 years older/younger than mother 0.695 0.734Mother's substance abuse problem 0.849 1.204Mother lived with both parents at age 15 1.536 1.361Father ever in jail (mother's report) 1.544 † 1.663 †

Socioeconomic StatusMother's education

<High school 0.881 0.753High school/GED 0.616 0.632Some college 0.629 0.647College or more (ref)

Mother worked in year before birth 0.705 0.693Mother on welfare 0.766 0.715Mother (or father) owns home^ 0.952 1.040Relationship CharacteristicsYears known each other 1.031 1.022Mother feels supported (1-3) 0.327 ** 0.364 **Mother reports conflict (1-3) 1.347 1.119Mother reports violence 0.623 0.623Mother distrusts men 0.859 0.992Mother has pro-marriage attitude 1.118 1.031Time IntervalsInterval 1 (months 1-3) (ref)Interval 2 (months 4-6) 0.402 † 0.388 †

Interval 3 (months 7-9) 0.446 * 0.421 *Interval 4 (months 10-12) 0.664 0.640Interval 5 (months 13-15) 0.404 † 0.384 *

Interval 6 (months 16-18) 0.215 *** 0.219 ***Interval 7 (months 19-21) 0.394 0.376Interval 8 (months 22-24) 0.547 0.513Interval 9 (months 25-27) 0.510 0.476Interval 10 (months 28-30) 0.270 * 0.251 **Interval 11 (months 31-33) 0.476 0.439Interval 12 (months 34-36) 0.128 *** 0.116 ***Interval 13 (months 37-50) 0.231 * 0.190 *Model Chi-Square (df) 1158.64(39) *** 1133.78(41) ***Pseudo-R 2 0.118 0.116

data. ̂ Only mothers are included in Model 4.3. †p < .10, *p < .05, **p < .01, ***p < .001

Original Model 4 (from Table 3)

Table 6. Odds Ratios from Discrete-Time Event Hist ory Model Excluding all Father Information (Sensitivity Analysis 2)

Model 4.3 (from sensitivity

analysis 2)

Note: Estimates based on weighted data. Fit statistics based on unweighted

APPENDIX C

87

REFERENCES

Acs, G. (2005). Is the ring the thing? Child well-being and the transition from cohabitation to marriage. Paper presented at the 2005 Annual Meeting of the Association for Public Policy Analysis and Management, Washington, DC.

Administration for Children and Families. (2006). Regional map of ACF healthy marriage grantees. Retrieved May 24, 2009, from http://www.acf.hhs.gov/healthymarriage/pdf/june18_list_marriageedprograms.PDF

Allison, P. D. (1982). Discrete-time methods for the analysis of event histories. Sociological Methodology, 13, 61-98.

Allison, P. D. (1984). Event history analysis: Regression for longitudinal event data. Newbury Park, CA: SAGE Publications.

Allison, P. D. (1995). Survival analysis using SAS: A practical guide. Cary, NC: SAS Institute.

Allison, P. D. (2002). Missing data. Thousand Oaks, CA.

Amato, P. R., & Booth, A. (1997). A generation at risk: Growing up in an era of family upheaval. Cambridge, MA: Harvard University Press.

Amato, P. R., & Cheadle, J. (2005). The long reach of divorce: Divorce and child well-being across three generations. Journal of Marriage and the Family, 67(1), 191-206.

Amato, P. R., & Hohmann-Marriott, B. (2007). A comparison of high- and low-distress marriages that end in divorce. Journal of Marriage and Family, 69(3), 621-638.

Amato, P. R., Johnson, D. R., Booth, A., & Stacey, J. R. (2003). Continuity and change in marital quality between 1980 and 2000. Journal of Marriage and Family, 65, 1-22.

Andersson, G. (2002). Children's experience of family disruption and family formation: Evidence from 16 FFS countries. Demographic Research, 7, 343-364.

88

Avellar, S., & Smock, P. J. (2005). The economic consequences of the dissolution of cohabiting unions. Journal of Marriage and Family, 67, 315-327.

Axinn, W. G., & Thornton, A. (1992). The relationship between cohabitation and divorce: Selectivity or causal influence? Demography, 29, 357-374.

Becker, G. (1991). A treatise on the family. Cambridge, MA: Harvard University Press.

Binstock, G., & Thornton, A. (2003). Separations, reconciliations, and living apart in cohabiting and marital unions. Journal of Marriage and Family, 65, 432-443.

Bitler, M. P., Gelbach, J. B., Hoynes, H. W., & Zavodny, M. (2004). The impact of welfare reform on marriage and divorce. Demography, 41(2), 213-236.

Bouchard, G. (2005). Adult couples facing a planned or an unplanned pregnancy: Two realities. Journal of Family Issues, 26(5), 619-637.

Bracher, M., Santow, G., Morgan, S. P., & Trussell, J. (1993). Marriage dissolution in Australia - models and explanations. Population Studies, 47, 403-425.

Bronfenbrenner, U. (1979). The ecology of human development. Cambridge, MA: Harvard University Press.

Brown, S. L. (2000). Union transitions among cohabitors: The significance of relationship assessments and expectations. Journal of Marriage and the Family, 62, 833-846.

Bulanda, R. E., & Manning, W. D. (2008). Parental cohabitation experiences and adolescent behavioral outcomes. Population Research and Policy Review, 27, 593-618.

Bumpass, L., & Lu, H. (2000). Trends in cohabitation and implications for children's family contexts in the United States. Population Studies, 54, 29-41.

Carlson, B. L. (2008). Fragile families and child wellbeing study: methodology for constructing mother, father, and couple weights for core telephone public survey data waves 1-4. Washington, DC: Mathematica Policy Research, Inc.

89

Carlson, M., & Furstenberg, F. F. (2006). The prevalence and correlates of multipartnered fertility among urban U.S. parents. Journal of Marriage and Family, 68, 718-732.

Carlson, M., & Furstenberg, F. F. (2007). The consequences of multi-partnered fertility for parental involvement and relationships (Working Paper #2006-28-FF): Center for Research on Child Wellbeing.

Carlson, M., McLanahan, S., & England, P. (2004). Union formation in fragile families. Demography, 41, 237-261.

Catlett, B. S., & Artis, J. E. (2004). Critiquing the case for marriage promotion: How the promarriage movement misrepresents domestic violence research. Violence Against Women, 10(11), 1226-1244.

Center for Research on Child Wellbeing. (2003). Union formation and dissolution in fragile families (Fragile Families Research Brief, No. 14). Princeton, NJ: Author.

Center for Research on Child Wellbeing. (2005a). Introduction to the Fragile Families core public use data: Baseline, one-year, and three-year files. Princeton, NJ: Author.

Center for Research on Child Wellbeing. (2005b). Young children's behavioral problems in married and cohabiting families (Fragile Families Research Brief, No. 33). Princeton, NJ: Author.

Center for Research on Child Wellbeing. (2008). Introduction to the Fragile Families public use data: Baseline, one-year, and three-year core telephone data. Princeton, NJ: Author.

Centers for Disease Control and Prevention. (2000). Building data systems for monitoring and responding to violence against women: Recommendations from a workshop (MMWR 2000; 49 No. RR-11). Atlanta, GA: Centers for Disease Control and Prevention, US Department of Health and Human Services.

Charles, P., Orthner, D. K., Jones, A., & Mancini, D. (2006). Poverty and couple relationships: Implications for welfare policy. Marriage and Family Review, 39, 27-52.

90

Charles, P., & Perreira, K. M. (2007). Intimate partner violence during pregnancy and one-year post-partum. Journal of Family Violence, 22, 609-619.

Cherlin, A. J. (2005). American marriage in the early twenty-first century. The Future of Children, 15, 33-55.

Cherlin, A. J. (2009). The marriage-go-round. New York: Alfred A. Knopf.

Cherlin, A. J., Cross-Barnet, C., Burton, L. M., & Garrett-Peters, R. (2008). Promises they can keep: Low-income women's attitudes toward motherhood, marriage, and divorce. Journal of Marriage and Family, 919-933.

Cleves, M. A., Gould, W. W., & Gutierrez, R. G. (2004a). Describing the distribution of failure times. In An introduction to survival analysis using Stata (pp. 7-17). College Station, TX: Stata Corporation.

Cleves, M. A., Gould, W. W., & Gutierrez, R. G. (2004b). Model building using stcox. In An introduction to survival analysis using Stata (pp. 157-174). College Station, TX: Stata Corporation.

Corcoran, M., & Adams, T. (1997). Race, sex, and the intergenerational transmission of poverty. In G. J. Duncan & J. Brooks-Gunn (Eds.), Consequences of growing up poor (pp. 461-517). New York: Russell Sage Foundation.

Cowan, C. P., & Cowan, P. A. (2000). When partners become parents: The big life change for couples. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Cowan, C. P., Cowan, P. A., Pruett, M. K., & Pruett, K. (2007). An approach to preventing coparenting conflict and divorce in low-income families: Strengthening couple relationships and fostering fathers' involvement. Family Process, 46(1), 109-121.

Cowan, P. A., & Cowan, C. P. (1988). Changes in marriage during the transition to parenthood: must we blame the baby? In G. Y. Michaels & W. A. Goldberg (Eds.), The transition to parenthood: Current theory and research (pp. 114-154). New York: Cambridge University Press.

Cutrona, C. E., Russell, D. W., Abraham, W. T., Gardner, K. A., Melby, J. N., Bryant, C., et al. (2003). Neighborhood context and financial strain as predictors of marital

91

interaction and marital quality in African American couples. Personal Relationships, 10, 389-409.

DeBell, M. (2008). Children living without their fathers: Population estimates and indicators of educational well-being. Social Indicators Research, 87(3), 427-443.

DeMaris, A. (2000). Till discord do us part: The role of physical and verbal conflict in union disruption. Journal of Marriage and the Family, 62, 683-692.

Dew, J. (2008). The gendered meanings of assets for divorce. Journal of Family and Economic Issues, 30(1), 20-31.

Dion, M. R., Devaney, B., McConnell, S., Ford, M., Hill, H., & Winston, P. (2003). Helping unwed parents build strong and healthy marriages: A conceptual framework for interventions. Washington, DC: Mathematica Policy Research, Inc.

Duncan, G. J., & Brooks-Gunn, J. (1997). The effects of poverty on children. The Future of Children, 7, 55-71.

Dunifon, R., & Kowaleski-Jones, L. (2002). Who's in the house? Race differences in cohabitation, single parenthood, and child development. Child Development, 73, 1249-1264.

Edin, K., & Kefalas, M. (2005). Promises I can keep: Why poor women put motherhood before marriage. Berkeley: University of California Press.

Ellwood, D. T., & Jencks, C. (2001). The growing differences in family structure: What do we know? Where do we look for answers? (Unpublished manuscript). Harvard University.

England, P., & Edin, K. (2007). Unmarried couples with children: Hoping for love and the white picket fence. In P. England & K. Edin (Eds.), Unmarried Couples with Children. New York: Russell Sage Foundation.

Fein, D., Burstein, N. R., Fein, G. G., & Lindberg, L. D. (2003). The determinants of marriage and cohabitation among disadvantaged Americans: Research findings and needs. Bethesda, MD: Abt Associates Inc.

92

Foster, E. M., & Kalil, A. (2007). Living arrangements and children's development in low-income white, black, and Latino families. Child Development, 78(6), 1657-1674.

Gibson, C., Edin, K., & McLanahan, S. (2003). Higher hopes but even higher expectations: The retreat from marriage among low-income couples (Working Paper 2003-06-FF). Princeton, NJ: Center for Research on Child Wellbeing.

Grinstein-Weiss, M., Zhan, M., & Sherraden, M. (2006). Saving performance in individual development accounts: Does marital status matter? Journal of Marriage and Family, 68, 192-204.

Grogan-Kaylor, A. (2004). The effect of corporal punishment on antisocial behavior in children. Social Work Research, 28, 153-162.

Guo, G. (1993). Event-history analysis for left-truncated data. Sociological methodology, 23, 217-243.

Guo, S. (Forthcoming). Survival analysis: A practical guide to social work research. New York, NY: The Oxford University Press.

Guzzo, K. B. (2009). Marital intentions and the stability of first cohabitations. Journal of Family Issues, 30, 179-205.

Guzzo, K. B., & Furstenberg, F. F. (2007a). Multipartnered fertility among American men. Demography, 44(3), 583-601.

Guzzo, K. B., & Furstenberg, F. F. (2007b). Multipartnered fertility among young women with a nonmarital first birth: prevalence and risk factors. Perspectives on sexual and reproductive health, 39(1), 29-38.

Harknett, K., & Knab, J. (2007). More kin, less support: Multipartnered fertility and perceived support among mothers. Journal of Marriage and Family, 69, 237-253.

Harknett, K., & McLanahan, S. (2004). Racial and ethnic differences in marriage after the birth of a child. American Sociological Review, 69, 790-811.

93

Hawkins, A. J., Fawcett, E. B., Carroll, J. S., & Gilliland, T. T. (2006). The marriage moments program for couples transitioning to parenthood: Divergent conclusions from formative and outcome evaluation data. Journal of Family Psychology, 20, 561-570.

Hill, H. D. (2007). Steppin' out: Infidelity and sexual jealousy among unmarried parents. In P. England & K. Edin (Eds.), Unmarried Couples with Children. New York: Russell Sage Foundation.

Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic regression (2nd ed.). New York: John Wiley & Sons, Inc.

Hsueh, A. C., Morrison, K. R., & Doss, B. D. (2009). Qualitative reports of problems in cohabiting relationships: Comparisons to married and dating relationships. Journal of Family Psychology, 23(2), 236-246.

Jackson, A. P., Tienda, M., & Huang, C. (2001). Capabilities and employability of unwed mothers. Children and Youth Services Review, 23, 327-351.

Jones, A. (2008). Strengthening couple relationships among low-resourced, newly forming families. Paper presented at the 1st National Research Conference on Child and Family Programs and Policy. Bridgewater State University, Bridgewater, MA.

Koo, H. P., Suchindran, C. M., & Griffith, J. D. (1984). The effects of children on divorce and re-marriage: A multivariate analysis of life table probabilities. Population Studies, 38, 451-471.

Kutner, M. H., Nachtsheim, C. J., & Neter, J. (Eds.). (2004). Applied linear regression models (4th ed.): McGraw-Hill Irwin.

Lehrer, E. L. (2004). Religion as a determinant of economic and demographic behavior in the United States. Population and Development Review, 30, 707-726.

Lehrer, E. L., & Chiswick, C. U. (1993). Religion as a determinant of marital stability. Demography, 30, 385-404.

Lerman, R. I. (2002). Married and unmarried parenthood and economic well-being: A dynamic analysis of a recent cohort. Washington, DC: Urban Institute.

94

Lewin, A. C. (2005). The effect of economic stability on family stability among welfare recipients. Evaluation Review, 29, 223-240.

Lichter, D. T., & Graefe, D. R. (2001). Finding a mate? The marital and cohabitation histories of unwed mothers. In L. L. Wu & B. Wolfe (Eds.), Out of wedlock: Causes and consequences of nonmarital fertility (pp. 317-343). New York: Russell Sage Foundation.

Lichter, D. T., Kephart, G., McLaughlin, D. K., & Landry, D. J. (1992). Race and the retreat from marriage: A shortage of marriageable men? American Sociological Review, 57, 781-799.

Lichter, D. T., Qian, Z., & Mellott, L. M. (2006). Marriage or dissolution? Union transitions among poor cohabiting women. Demography, 43, 223-240.

London, A. S., & Parker, W. M. (2009). Incarceration and living arrangements: Findings from the National Health and Social Life Survey. Journal of Family Issues, 30(6), 787-812.

Lopoo, L. M., & Western, B. (2005). Incarceration and the formation and stability of marital unions. Journal of Marriage and Family, 67, 721-734.

Magnuson, K. A., & Gibson-Davis, C. M. (2007). Child support among low-income noncustodial fathers. In P. England & K. Edin (Eds.), Unmarried Couples with Children (pp. 228-251). New York: Russell Sage Foundation.

Manning, W. D., & Landale, N. S. (1996). Racial and ethnic differences in the role of cohabitation in premarital childbearing. Journal of Marriage and Family, 58, 63-77.

Manning, W. D., & Smock, P. J. (2002). First comes cohabitation and then comes marriage?: A research note. Journal of Family Issues, 23, 1065-1087.

Manning, W. D., Smock, P. J., & Majumdar, D. (2004). The relative stability of cohabiting and marital unions for children. Population Research and Policy Review, 23, 135-159.

95

Martin, J. A., Hamilton, B. E., Sutton, P. D., Ventura, S. J., Menacker, F., Kirmeyer, S., et al. (2007). Births: Final Data for 2005 (No. Vol. 56, No. 6). Hyattsville, MD: National Center for Health Statistics.

McLanahan, S. (1997). Parent absence or poverty: Which matters more? In G. J. Duncan & J. Brooks-Gunn (Eds.), Consequences of growing up poor (pp. 35-48). New York: Russell Sage Foundation.

McLanahan, S. (2009). Fragile families and the reproduction of poverty. The ANNALS of the American Academy of Political and Social Science, 621, 111-131.

McLanahan, S., Garfinkel, I., Reichman, N., Teitler, J., Carlson, M., & Audigier, C. N. (2003). The fragile families and child wellbeing study: Baseline national report. Princeton, NJ: Bendheim-Thoman Center for Research on Child Wellbeing.

McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent: What hurts, what helps. Cambridge, MA: Harvard University Press.

Meyer, D. R., Cancian, M., & Cook, S. T. (2005). Multiple-partner fertility: Incidence and implications for child support policy. Social Service Review, 79, 577-601.

Mincy, R. B. (2002). Who should marry whom: Multiple partner fertility among new parents (Working Paper #02-03 FF). New York: Center for Research on Child Wellbeing.

Moffitt, R. (2000). Policy effects on family structure and fertility: Welfare benefits and female headship in U.S. time series. American Economic Review, 90, 373-377.

Monte, L. M. (2007). Blended but not the Bradys: Navigating unmarried multiple partner fertility. In P. England & K. Edin (Eds.), Unmarried couples with children (pp. 183-203). New York: Russell Sage Foundation.

Oppenheimer, V. K. (1994). Women's rising employment and the future of the family in industrial societies. Population and Development Review, 20, 293-342.

Oppenheimer, V. K. (2003). Cohabiting and marriage during young men's career-development process. Demography, 40, 127-149.

96

Orbuch, T. L., Veroff, J., Hassan, H., & Horrocks, J. (2002). Who will divorce: A 14-year longitudinal study of black couples and white couples. Journal of Social and Personal Relationships, 19, 179-202.

Orthner, D. O., Jones-Sanpei, H., & Williamson, S. (2004). The resilience and strengths of low-income families. Family Relations, 53(2), 159-167.

Osborne, C. (2005). Marriage following the birth of a child among cohabiting and visiting parents. Journal of Marriage and Family, 67, 14-26.

Osborne, C., Manning, W. D., & Smock, P. J. (2004). Instability in fragile families: The role of race-ethnicity, economics, and relationship quality (Working Paper 2004-17-FF). Princeton, NJ: Center for Research on Child Well-Being.

Osborne, C., Martin, J. A., & Smock, P. J. (2007). Married and cohabiting parents' relationship stability: A focus on race and ethnicity. Journal of Marriage and Family, 69, 1345-1366.

Pagani, L., Boulerice, B., & Tremblay, R. E. (1997). The influence of poverty on children's classroom placement and behavior problems. In G. J. Duncan & J. Brooks-Gunn (Eds.), Consequences of growing up poor (pp. 311-339). New York: Russell Sage Foundation.

Parke, M. (2003). Are married parents really better for children? What research says about the effects of family structure on child well-being (Policy Brief No. 3). Washington, DC: Center for Law and Social Policy.

Qian, Z., & Preston, S. H. (1993). Changes in American marriage, 1972 to 1987: Availability and forces of attraction by age and education. American Sociological Review, 58, 482-495.

Reardon-Anderson, J., Stagner, M., Macomber, J. E., & Murray, J. (2005). Systematic review of the impact of marriage and relationship programs. Retrieved October 14, 2006, from www.urban.org/publications/41142.html

Rector, R. E., Johnson, K. A., Fagan, P. F., & Noyes, L. R. (2003). Increasing marriage will dramatically reduce child poverty (CDA 03-06). Washington, DC: The Heritage Foundation.

97

Reichman, N. E., Teitler, J. O., Garfinkel, I., & McLanahan, S. S. (2001). Fragile families: Sample and design. Children and Youth Services Review, 23, 303-326.

Riggio, H. R., & Weiser, D. A. (2008). Attitudes toward marriage: Embeddedness and outcomes in personal relationships. Personal Relationships, 15(1), 123-140.

Rose, R. A., & Fraser, M. W. (2008). A simplified framework for using multiple imputation in social work research. Social Work Research, 32(3), 171-178.

Rosenfeld, M. J. (2002). Measures of assimilation in the marriage market: Mexican Americans 1970-1990. Journal of Marriage and Family, 64, 152-162.

Saltzman, L. E., Johnson, C. H., Gilbert, B., & Goodwin, M. M. (2003). Physical abuse around the time of pregnancy: An examination of prevalence and risk factors in 16 states. Maternal and Child Health Journal, 7, 31-43.

Sassler, S., Cunningham, A., & Lichter, D. T. (2009). Intergenerational patterns of union formation and relationship quality. Journal of Family Issues 30(6), 757-786.

Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7(2), 147-177.

Schwartz, C. R., & Mare, R. D. (2005). Trends in educational assortative marriage from 1940 to 2003. Demography, 42, 621-646.

Seltzer, J. A. (2004). Cohabitation in the United States and Britain: Demography, kinship, and the future. Journal of Marriage and Family, 66, 921-928.

Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and event occurrence. New York: Oxford University Press.

Smock, P. J. (2000). Cohabitation in the United States: An appraisal of research themes, findings, and implications. Annual Review of Sociology, 26, 1-20.

South, S. J. (2001). The variable effects of family background on the timing of first marriage: United States, 1969-1993. Social Science Research, 30, 606-626.

98

StataCorp. (2007). ST: Survival analysis and epidemiological tables, release 10. College Station, TX: StataCorp Lp.

Stewart, S. D., Manning, W. D., & Smock, P. J. (2003). Union formation among men in the U.S.: Does having prior children matter? Journal of Marriage and Family, 65, 90-104.

Straus, M. A. (1979). Measuring intrafamily conflict and violence: The conflict tactics (CT) scales. Journal of Marriage and Family, 41, 75-88.

Teachman, J. D. (2002). Childhood living arrangements and the intergenerational transmission of divorce. Journal of Marriage and the Family, 64(3), 717-729.

Teachman, J. D. (2003). Premarital sex, premarital cohabitation, and the risk of subsequent marital dissolution among women. Journal of Marriage and Family, 65, 444-455.

Teachman, J. D. (2008a). Complex life course patterns and the risk of divorce in second marriages. Journal of Marriage and Family, 70, 294-305.

Teachman, J. D. (2008b). The living arrangements of children and their educational well-being. Journal of Family Issues, 29(6), 734-761.

Teachman, J. D., Tedrow, L. M., & Hall, M. (2006). The demographic future of divorce and dissolution. In M. A. Fine & J. H. Harvey (Eds.), Handbook of divorce and relationship dissolution (pp. 59-82). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers.

Thomas, S., & Sawhill, I. (2005). For love and money? The impact of family structure on family income. The Future of Children, 15, 57-74.

Thornton, A., Axinn, W. G., & Hill, D. H. (1992). Reciprocal effects of religiosity, cohabitation and marriage. American Journal of Sociology, 98, 628-651.

Twenge, J. M., Campbell, W. K., & Foster, C. A. (2003). Parenthood and marital satisfaction: A meta-analytic review. Journal of Marriage and Family, 65, 561-573.

99

UCLA Academic Technology Services. (2008). FAQ: What are pseudo R-squareds? Retrieved May 19, 2008, from http://www.ats.ucla.edu/stat/mult_pkg/faq/general/Psuedo_RSquareds.htm

Upchurch, D. M., Lilliard, L. A., & Panis, C. W. A. (2001). The impact of nonmarital childbearing on subsequent marital formation and dissolution. In L. L. Wu & B. Wolfe (Eds.), Out of wedlock: Causes and consequences of nonmarital fertility (pp. 344-402). New York: Russell Sage Foundation.

Ventura, S. J. (2009). Changing patterns of nonmarital childbearing in the United States (NCHS data brief, no 18). Hyattsville, MD: National Center for Health Statistics.

Wen, M. (2008). Family structure and children's health and behavior data from the 1999 national survey of America's families. Journal of Family Issues, 29(11), 1492-1519.

Wilcox, W. B., & Wolfinger, N. H. (2006). Unpacking the faith factor: Norms, decency, and relationship quality among urban parents (Working Paper #2006-18-FF). Princeton, NJ: Center for Research on Child Wellbeing.

Willett, J. B., & Singer, J. D. (1993). Investigating onset, cessation, relapse, and recovery: Why you should, and how you can, use discrete-time survival analysis to examine event occurrence. Journal of Consulting and Clinical Psychology, 61, 952-965.

Wilmoth, J., & Koso, G. (2002). Does marital history matter? Marital status and wealth outcomes among preretirement adults. Journal of Marriage and the Family, 64, 254-268.

Wu, Z., & Balakrishnan, T. (1995). Dissolution of premarital cohabitation in Canada. Demography, 32, 521-532.

Zlotnick, C., Johnson, D. M., & Kohn, R. (2006). Intimate partner violence and long-term psychosocial functioning in a national sample of American women. Journal of Interpersonal Violence 21(2), 262-275.