Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Dec 1.
Published in final edited form as: Fam Relat. 2011 Dec;60(5):586–601. doi: 10.1111/j.1741-3729.2011.00672.x

Relationship-Specific Investments, Family Chaos, and Cohabitation Dissolution Following a Non-marital Birth

Claire M Kamp Dush 1
PMCID: PMC3212101  NIHMSID: NIHMS310938  PMID: 22081737

Abstract

Predictors of two types of cohabitation dissolution, dissolution with a continued romantic relationship and without (i.e. breakup), were examined using data from mothers cohabiting at the time of a non-marital birth in the Fragile Families and Child Wellbeing Study (n = 1624). Life tables indicated 64% of unions dissolved within 5 years; of these, 76% broke-up. Black mothers had the highest rates of dissolution. Maximum likelihood discrete-time event history results revealed that younger mothers were more likely to experience cohabitation dissolution into a breakup. Fewer relationship-specific investments and more family chaos were also associated with greater risk of cohabitation dissolution into a breakup. Mothers’ multipartnered fertility and fewer relationship-specific investments were associated with greater risk of cohabitation dissolution with a continued romantic relationship. Post-dissolution, mothers who maintained a romantic relationship were more likely to reenter a union with their former partner while mothers whose union broke-up most often remained so.

Keywords: cohabitation, cohabitation dissolution, break-up, relationship-specific investments, chaos

As the US divorce rate rose in the 1960s and 1970s, family scholars began to examine the predictors of divorce and its effects on adults and children. The increase in cohabitation in the past thirty years (Fields & Casper, 2001) has led to the rise in a different kind of union dissolution – cohabitation dissolution. Like marriage, many of these cohabiting unions involve children. Indeed, 41% of all births in the US in 2008 were to unmarried women (Hamilton, Martin, & Ventura, 2010), and recent estimates are that about two-fifths of non-marital births in the US are to cohabiting couples (Carlson & McLanahan, 2002). Because these unions are often unstable (Lichter, Qian, & Mellot, 2006) it is of particular concern that family scholars know very little about the process of cohabitation dissolution, particularly among the most vulnerable of cohabiting couples – couples with children.

The first goal of this study was to provide estimates of cohabitation dissolution rates within the first five years of a nonmarital birth by type of dissolution - cohabitation dissolution as a breakup and cohabitation dissolution with a continuing romantic relationship. The second goal was to identify demographic, relationship, and family predictors of each type of cohabitation dissolution among parents. The final goal was to compare, after the initial cohabitation dissolution, the relationship status of formerly cohabiting mothers by type of cohabitation dissolution two to four years post-dissolution. Longitudinal data from the Fragile Families and Child-Well Being Study (n = 1624) were used to examine these associations in a large sample of low-income cohabiting families with a recent birth. Additional knowledge regarding the predictors of cohabitation dissolution among low-income parents could aid in the design and implementation of interventions to either prevent cohabitation dissolution or support children and parents experiencing it.

A Family Systems Perspective on Relationship Transitions

Emery and Dillon (1994) posited that that the central task of relationship change is the redefining of relationship boundaries. Borrowing from family systems theory (Minuchin, 1974), relationship boundaries were defined as the explicit or implicit rules that govern family relationships. Boundaries in marriage develop over time through role expectations, discussions of the roles and rules, and the behaviors of family members (Emery & Tuer, 1993). Thus, at the conclusion of a marriage, particularly when children are involved, boundaries must be redrawn and new rules to govern roles and behaviors of family members must be established.

Relationship boundaries after cohabitation dissolution must also be redefined. However, initial relationship boundaries in cohabiting unions may be unclear. Family sociologists (Cherlin, 2004; Manning & Lamb, 2003; Nock, 1995) have argued that despite its increase, cohabitation is still an “incomplete institution” whereby family members do not have established norms for how they should treat one another. Indeed, there is a lack of norms for transitioning into cohabitation; at least some cohabiting couples experience the phenomenon of “sliding versus deciding” whereby couples move through transitions such as moving in together, engagement, and even parenthood without fully considering or discussing the implications (Manning & Smock, 2005; Stanley, Rhoades, & Markman, 2006). Couples move through these phases rapidly (Sassler, 2004), and once the transition to cohabitation occurs, barriers to exiting the union increase and relationship inertia (Stanley, et al.) can propel a couple into other relationship transitions such as a transition to parenthood without reflection on the implications of these transitions. Overall, this absence of norms, expectations, and discussions creates boundaries that are ambiguous in terms of the obligations and rights of each cohabiting partner and any children in the family. Because norms for cohabitation are unclear it follows that the norms for its couple-negotiated dissolution would be unclear. Qualitative data from the Time, Love, and Cash study that included a small subsample (n = 75) of families in the Fragile Families and Child-Wellbeing Study found that 27% of unmarried couples who broke up during the study got back together by the study’s end (Reed, 2007). Because on-again/off-again relationships have been found to be associated with fewer positives such as love and understanding from a partner, and more negatives such as communication problems (Dailey, Pfiester, Jin, Beck, & Clark, 2009), it is important that union dissolutions that end in a relationship status other than as a breakup be explored.

Cohabiting unions could dissolve in at least three different ways. First, the cohabiting couple could stop living together full-time, but maintain their romantic relationship while living together some of the time, a relationship status referred to as “visiting” (Carlson & McLanahan, 2004). Second, couples could stop living together but continue their romantic relationship as a dating relationship. For example, Reed (2007) found that after dissolution, fathers usually moved back in with their own parents, relatives, or friends, and fathers stayed in these households while attempting to reconcile with the mother of their child. During this period, the couple may still be romantically involved as they work through problems, even though they are no longer living together. Third, couples could end their romantic relationship all together. Even if formerly cohabiting couples did maintain a romantic relationship, they still may ultimately breakup. Using data from the Detroit metropolitan area, Binstock and Thornton (2003) found that 10% young adults whose cohabiting union dissolved eventually reconciled (defined as moving back in together), but these reconciliations were often short lived and followed by a second separation.

Predictors of Cohabitation Dissolution after a Nonmarital Birth

Rusbult (1980) argued that intimate relationship commitment and stability was a function of the comparison of the perceived rewards and costs of the relationship to the quality of alternatives and investments already made in the relationship. In particular, commitment increased with the passage of time because the investment of resources in the relationship increased the costs of leaving the relationship. Extending the investment model (Rusbult) to cohabitation dissolution, factors that increase investments in the cohabiting union should be associated with maintaining a romantic relationship while factors that decrease investments should be associated with ending the romantic relationship. Investments include initially extraneous resources such as shared possessions that become connected to the relationship as well as resources that each partner puts directly into the relationship (Rusbult, 1983).

Relationship factors

Relationship-related factors are associated with the level of investment in the relationship. Specifically, relationship quality is positively associated with relationship investments (Rusbult, 1980) and low quality cohabiting unions are more likely to dissolve (Brown, 2000; Osbourne, Manning, & Smock, 2007; Waller & McLanahan, 2005), perhaps because of reduced investments. Relationship-specific investments are those “investments whose returns depend on the relationship’s continuation” (Crawford, 1990, p. 561). Having another child with the father of the focal child could be a sign of greater investment in the relationship as children are generally considered relationship-specific investments (England & Kilbourne, 1990). Lillard and Waite (1993) found that couples who were more likely to divorce were less likely to have a child, and less likely to have an additional child if already parents. Hence, having an additional child may be indicative of a willingness to invest further in the relationship; indeed Osbourne et al. found that having a new child within three years of the birth decreased the odds of marital or cohabitation dissolution. Having a joint account or credit card or sharing money may also signify a greater relationship-specific investment in the relationship. Treas (1993) suggested that individuals who expect their relationship to dissolve may have more incentive to keep their money separate from their partner, and found that individuals in marriages in which at least one partner was remarried, and who may expect that their marriage may not be permanent, were less likely to combine money in a joint account. Couples that have joint bank accounts report higher levels of relationship quality (Addo & Sassler, 2010). Cohabiting parents, whose relationships are less stable (Lichter et al., 2006), are less likely than married parents to combine all their money (Kenney, 2004). Finally, getting married also represents an increased relationship-specific investment in the relationship as marriage is a legal, formal commitment.

Family chaos

Family chaos could also decrease investments in the relationship. Fiese and Winter (2010) suggested that two dimensions of family chaos include “the disruptions of daily activities and the associated felt experience (hectic, out of control)” (p. 50). Recently, several dimensions were suggested as indicators of the construct including residential mobility (Hertzman, 2010; Lichter & Wethington, 2010), nonstandard and shifting work hours, inflexible, stressful, and family-hostile work environments (Repetti & Wang, 2010), and unreliable and changing childcare settings (Corapci, 2010). These factors are often experienced by low-income couples (Wachs & Evans, 2010), and in the Fragile Families and Child Wellbeing Study used for this study, 33% of cohabiting mothers were impoverished, and their average annual household income was just $26,548 (Kalil & Ryan, 2010). Low-income couples with higher levels of chaos may have less time and energy to invest in their relationship (Fiese & Winter).

In particular, residential mobility is stressful for families and has been tied to negative outcomes for low-income children (Simpson & Fowler, 1994) and a greater risk of divorce (Boyle, Kulu, Cooke, Gayle, & Mulder, 2008). Residential mobility may increase family chaos as families have to adjust to a new residence and often new neighborhoods (Lichter & Wethington, 2010). Inflexible work environments increase family chaos because parents in these environments have difficulty dealing with childcare problems or child illness (Perry-Jenkins, 2005) while at risk of repercussions at work or losing their job. Parents have been found to be less stressed when they have autonomy, schedule control, fewer job demands, greater job security, and a supportive workplace (Galinsky, Bond, & Friedman, 2010). Shifting work schedules and working multiple jobs can cause extra stress as family time and routines become difficult to maintain when shifts change from week to week or occur during family mealtimes or bedtime (Staines & Pleck, 1983; Barnett, 2006).

Inconsistent childcare can also contribute to family chaos because when childcare falls through, parents must make adjustments to their schedules, and for low-income families, their job may be at risk (Henly & Lyons, 2000). Multiple childcare arrangements can also lead to family chaos as parents juggle work and securing a childcare provider for each time childcare is needed. In a low-income sample, those with multiple arrangements almost universally had irregular schedules with their childcare providers, leading to a hectic, recurring process of securing childcare from the available providers for needed times (Henly & Lyons). Overall, families that have more chaos have less time to spend together and less time to establish and maintain family rituals and routines that are linked to positive family outcomes (Fiese & Winter, 2010), and thus have less time to invest in their relationship with their partner.

General investments

Other factors such as education and age (Becker, 1985) that are associated with overall human capital, or general investments (England & Kilbourne, 1990), increase investments in the relationship because more resources are available to be invested. Indeed, Lichter et al. (2006) found that women with older partners were less likely to dissolve their cohabiting union and more likely to continuously cohabit. Employed individuals invest more in their relationships because employment is associated with more income and resources as well as higher psychological well-being (Brown, et al., 2003), which in turn is associated with higher relationship quality (Fincham, Beach, Harold, & Osbourne, 1997). Indeed, cohabiting unions were more likely to dissolve when the male partner was unemployed (Smock & Manning, 1997) and when he had lower earnings (Sassler & McNally, 2003). Further, because with the passage of time investments increase, those partners who have spent more time in their relationship have invested more than those who have spent less (Rusbult, 1980). Brown (2000) found that cohabiting unions that were longer in duration were less likely to dissolve.

As was previously highlighted, the birth of an additional child may be associated with greater investments (England & Kilbourne, 1990), thus those mothers that had a higher order birth may have lower risk of dissolution. Because the nonmarital fertility rate is at a historic high and nonmarital unions are less stable (Lichter et al., 2006), parents that have a nonmarital birth may have already had a previous child with a different person, a situation known as multipartnered fertility. Multipartnered fertility is associated with lower relationship quality (Hill, 2007) which may lead to fewer relationship investments; indeed, Monte (2011) found that fathers’ multipartnered fertility increased the risk of cohabitation dissolution using data from the Fragile Families and Child Wellbeing study.

Racial Differences

There are racial disparities in rates of cohabitation dissolution (Lichter et al., 2006); the cohabiting unions of African American women are more likely to dissolve. Further, there are racial disparities in the nonmarital birth rate in the US; an overwhelming 73% of African American children were born to unmarried mothers in 2009 (Hamilton et al., 2010), though African American women are less likely to be cohabiting at their nonmarital birth compared to other races (Manlove, Ryan, Wildsmith, & Franzetta, 2010) and are more likely to dissolve their cohabiting union compared to other races ( Brines & Joyner, 1999). Hispanics are either more or just as supportive of marriage as non-Hispanic Whites (Oropesa & Gorman, 2000), but also tend to be more accepting of cohabitation even in the absence of plans to marry (Oropesa, 1996). Hispanics are more likely than are non-Hispanic Whites to treat cohabitation as a replacement for marriage (Landale and Fennelly, 1992) and thus Hispanics will likely have more stable unions.

Hypotheses

Because investments serve as a barrier to dissolving a relationship (Rusbult, 1983), they should also serve as a barrier to breaking-up and promote the maintenance of a romantic relationship even after cohabitation dissolution. Thus, I hypothesized that couples with lower relationship quality, fewer relationship- specific investments, greater family chaos, younger ages, less education, unemployment, shorter lengths of time the mother knew the father at the time of her pregnancy, a higher-order birth, and multipartnered fertility, would be more likely to dissolve their cohabitating union, and would also be more likely to dissolve their union into a breakup rather than a continued romantic relationship. I also expected that unions of African American women would be more likely to dissolve than the unions of non-Black/non-Hispanic women while the unions of Hispanic women would be less likely to dissolve than the unions of non-Black/non-Hispanic women.

Method

Data came from the Fragile Families and Child Well-being Study, a study of unwed mothers and fathers and their children. The baseline data included mothers (n = 4898) and fathers who had children in the US between 1998 and 2000; the sample for this paper was limited to those mothers who reported they had been cohabiting at the birth of their child (n = 1733). Both parents were interviewed in the hospital shortly after their child’s birth (Wave 1) with follow-up interviews conducted when the child was one (Wave 2), three (Wave 3), and five (Wave 4) years old. A cohabiting union at birth was defined if the mother reported she was 1) not married, 2) living with the focal child’s father, and 3) romantically involved on a steady basis or in an on-again and off-again relationship with the focal child’s father. Of the 1733 mothers cohabiting at birth, 80 were lost due to attrition, and 29 because their relationship status was unable to be coded at later waves. The final sample size was 1624 mothers.

Variables

Dependent variables

The main dependent variable was an indicator of the status of the cohabiting union. At waves 2, 3, and 4, mothers were first asked “What is your relationship with [the father of the focal child] now? Married, romantically involved, separated/divorced, just friends, or not in any kind of relationship?”, then asked “Are you and [father of the focal child] currently living together all or most of the time, some of the time, rarely, or never?” At waves 2, 3, and 4, cohabitation was defined as 1) married or romantically involved and 2) living together all or most of the time. Note that at birth, all couples were unmarried and cohabiting, but cohabiting couples that married at follow-up were retained in the sample. Cohabitation dissolution was divided into two categories. Cohabitation dissolution with a continuing romantic relationship was coded if the mother reported that she was 1) romantically involved and 2) living together some of the time, rarely, or never. Cohabitation dissolution as a breakup was coded if the mother reported that her relationship with the father of the focal child was separated/divorced, just friends, or not in any kind of relationship. For the final set of analyses, the post-dissolution status for those mothers who were followed up after their initial union dissolution was coded similarly but with marriage was distinguished from cohabitation.

The duration of the union was calculated from the birth of the focal child to the initial observation of the cohabitation dissolution or the final observation point if the mother was censored, that is, if no dissolution occurred prior to the final observation. The date of cohabitation dissolution was measured by the year and month of dissolution if a breakup occurred. However, the date at which couples’ full-time cohabitation ended was not asked of those couples whose cohabitation ended in a continued romantic relationship. For these couples (n = 218), and for couples for whom the date of dissolution was missing, the interview date was used as the date of dissolution (n = 285).

Independent variables

Time-invariant independent variables were measured at baseline except where noted. Age was measured in years. Mothers’ race/ethnicity was coded as a series of dummy variables; categories included Black (non-Hispanic), Hispanic, and non-Black/non-Hispanic. The length of time the mother knew the father at the time of the pregnancy was measured in years. Mother’s higher order birth was measured as a dummy variable indicating whether the mother had an older child at the time of the birth. Multipartnered fertility was measured at Wave 2; mothers’ and fathers’ multipartnered fertility were dummy variables where 1 indicated that the parent had a child with a previous partner. Time-varying independent variables were measured at each wave and included father’s employment status; 1 = father was working and/or in-school in the week prior to the interview, 0 = the father was not working or in-school the week prior to the interview.

Relationship quality

Mother’s relationship quality was a time-varying variable and the mean of a four-item measure at baseline (α = .56) and a six-item measure at waves 2, 3, and 4 (α was .72, .77, and .75 respectively). Mothers rated the frequency that fathers engaged in certain behaviors on a scale from 0 (never) to 2 (often) with sometimes at the midpoint. Behaviors assessed at each wave were “he is fair and willing to compromise when you have a disagreement”, “he expresses affection or love for you”, “he insults or criticizes you or your ideas” (reverse coded), and “he encourages or helps you to do things that are important to you”. At waves 2, 3, and 4, two additional behaviors were assessed – “he listens to you when you need someone to talk to” and “he really understands your hurts and joys”.

Relationship-specific investments

Relationship-specific investments was a time-varying count variable that was the sum of four indicators of investments at waves 2, 3, and 4. A new child with the father of the focal child was recorded using the household roster at each phase if a mother reported 1) a biological child that was younger than the focal child and 2) that the father of the focal child was the biological father of this child. A marriage was recorded if the mother reported that she and the father married since the previous wave. Joint checking account or credit card was recorded if the mother reported either a joint checking account or credit card with the father of the focal child. Pooling money was recorded if the mother reported putting either some or all of her money together with the father in response to the question “Couples handle money differently. Which of the following do you do? Keep your money separate, put some of your money together but keep the rest separate, or put all of your money together”. The investments count variable ranged from 0 (no investments) to 4 (all four investments).

Family chaos

Family chaos was a time-varying count variable that was the sum of 8 indicators of chaos at waves 2, 3, and 4. Moving since the previous wave was recorded if the mother reported any moves since the previous wave. Work stress was recorded by answering always or often to “My shift and work schedule (cause/caused) extra stress for me and my child always, often, sometimes, or never”. Childcare problems was recorded if always or often responses were given for “Where I (work/worked) it (is/was) difficult to deal with child care problems during working hours always, often, sometimes, or never?”. Inflexible work schedule was recorded for a sometimes or never response to “In my work schedule I (have/had) enough flexibility to handle family needs always, often, sometimes, or never”. Nonstandard work schedule was recorded if the mother answered yes to “At your primary job, (do/did) you regularly work different times each week?”. Multiple jobs was coded as an affirmative answer by the mother to “Some people work more than one regular job. Was there ever a time in the last 12 months that you worked more than one regular job at the same time?”. Change in childcare was recorded as answering nonzero to “How many times have you changed your childcare arrangements since (the previous wave)? By changes I mean, for example, that your child got a new babysitter, or started going to a new family child care program or day care center.” Childcare fell through was recorded as answering nonzero to “Approximately how many times in the past month did you have to make special arrangements because your usual child care arrangement fell through? Please include times when your child care provider(s) (was/were) sick or unavailable due to a holiday or vacation.”

Analytic Plan

The first goal was to provide life table estimates of cohabitation dissolution rates. Multiple-decrement life tables estimated the likelihood of cohabitation dissolution into a romantic relationship or a breakup (Preston, Heuveline, & Guillot, 2001). Unions were tracked for a maximum of six years if the mother was cohabiting with the biological father of the focal child at birth (some children were six years old at Wave 4). Censoring occurred when the mother was cohabiting at the final observation or are at her final observation prior to attrition. The life tables were run separately by race and the log-rank test for the equality of survival curves (Mantel, 1966) was used. Dissolution rates are reported through five years.

The second goal was to examine correlates of the type of cohabitation dissolution. Maximum likelihood discrete-time event history models were used to compare types of cohabitation dissolution (Allison, 1984). The event, cohabitation dissolution, was measured within discrete points of time (i.e. between waves) because the exact date of cohabitation dissolution was not available for all respondents. Respondents contributed person-years to the file beginning with the birth of the focal child until 1) the exact year of dissolution, 2) the year of the interview where the dissolution was reported if the exact dissolution date was missing, or 3) the final interview year. Following Allison’s strategy for analyzing discrete-time data, multinomial logistic models of the competing risks of cohabitation dissolution into a romantic relationship or breakup were run. The model took the following form:

log(PitjPit0)=α+∑m=1Mβmxmij+∑n=1Nβnxnij(t−1),j=1,2 (1)

where Pitj was the conditional probability of cohabiting mother i at year t since the focal child’s birth experiencing event j (j = 1 for cohabitation dissolution into a romantic relationship or j= 2 for breakup) versus no event occurring Pit0 at year t. When year t was either the year at which the mother was censored or experienced a dissolution, the dependent variable was coded 0 = censored (no dissolution observed), 1 = cohabitation dissolution into a romantic relationship, and 2 = breakup. For person-years prior to the year the mother was censored or her union dissolved, the dependent variable was coded as 0. α is an indicator of time included in the model to control for time dependence (Allison, 2006). The model also included m time-invariant predictors measured at baseline and n time-varying predictors measured at the wave prior to t because the time-varying predictors predicted transitions into cohabitation dissolution between t – 1 and t. When the lag between waves was longer than a year (i.e. between waves 2 and 3 and waves 3 and 4), the time-varying predictor at both times t and t – 1 was the value reported at the previous wave that occurred at t – 2. The final set of analyses compared the relationship status of formerly cohabiting mothers after the initial observation of dissolution by type of cohabitation dissolution. z tests of the equality of proportions were run to test for group differences.

Overall, little data were missing. In models that included only the demographic variables, including age, race, education, and fathers’ employment status, 57 respondents were missing data. In models that included family and relationship characteristics, 163 respondents were missing data on one or more of the family and relationship variables (multipartnered fertility, how long the mother knew the dad at birth, relationship quality, relationship investments, and family chaos), or about 14% of the respondents. Allison (2008) recommended listwise deletion when attrition was less than 15%, thus that strategy was employed.

Results

Descriptive Statistics and Multiple-Decrement Life Table Results

Descriptive statistics

Descriptive statistics are presented in Table 1 by cohabitation dissolution status. About 40% of the sample cohabited until their final interview, while 57% dissolved their union – a majority breaking-up rather than maintaining romantic involvement. Mothers were about 24 years old and fathers about 27 years old. Mothers and fathers who continually cohabited were significantly older than mothers and fathers whose unions dissolved. Black mothers made up 40% of the full sample and were significantly over represented in the cohabitation dissolution group. Hispanic mothers, at 34% of the full sample, were significantly overrepresented in the continuously cohabited group as were non-Black/non-Hispanic mothers. About 40% of mothers had less than a high school education, only 26% had any college. At baseline, 83% of fathers were employed; a significantly greater proportion of fathers who continually cohabited were employed compared to those whose union eventually dissolved. Overall, mothers were satisfied with their relationship at baseline - on a scale from 0 to 2, the mean was 1.71. The length of time the mother knew the father prior to the birth was around 4 years. Not surprisingly, mothers whose union eventually dissolved were significantly less satisfied at baseline compared to mothers whose union remained intact. About 64% of the mothers had an older child at the time of the birth. About 40% of the mothers and 34% of the fathers had multipartnered fertility, or previous children with a different partner; mothers and fathers whose union eventually dissolved had higher rates of multipartnered fertility.

Table 1.

Descriptive Statistics at Baseline for the Full Sample and by Cohabitation Dissolution Status

Full sample
Continually cohabited
Dissolved
Test statistics for group differences
M SD % Range M SD % M SD % t χ2
Dissolved cohabiting union 56.96
 Type of dissolution
  Romantic relationship 13.42
  Breakup 43.53
Variables Measured at Wave 1
Mother’s age 24.20 5.48 15–43 24.69 5.69 23.83 5.30 3.14**
Father’s age 26.98 6.85 16–66 27.52 6.74 26.58 6.91 2.72**
Mother’s race
  Black 43.97 30.76 53.95 86.89**
 Hispanic 34.11 43.63 26.92 49.49**
 non-Black/non-Hispanic 21.92 25.61 19.14 9.75**
Education
 Less than high school 39.17 37.16 40.69 2.08
 High school graduate 33.93 34.58 33.44 0.23
 Some college or more 26.40 27.55 25.54 0.82
Father employed 82.73 86.85 79.56 14.57**
Mother higher order birth 64.24 62.18 65.80 2.27
Length of time mother knew father at pregnancy 4.09 4.19 0–36 4.17 3.96 4.03 4.35 0.64
Relationship quality 1.71 0.33 0–2 1.75 0.31 1.67 0.34 5.30**
N 1624 699 925
Variables Measured Beginning at Wave 2
Mother multipartnered fertility 40.83 38.87 44.03 2.74+
Father multipartnered fertility 34.10 30.38 40.25 10.69**
Relationship-specific investments 1.34 0.88 0–3 1.48 0.87 1.10 0.84 6.95**
 Married 20.09 24.92 12.19 25.21**
 Joint checking or credit card 25.85 28.72 21.14 7.75**
 Pooled money 80.00 86.32 69.65 43.34**
 New child with focal father 9.54 10.39 8.15 1.47
Family chaos 1.41 1.17 0–7 1.29 1.11 1.60 1.24 3.91**
 Moved 46.23 44.53 49.00 2.01
 Work stress 14.55 13.47 16.09 5.19*
 Childcare problems 14.17 12.78 16.19 3.28+
 Inflexible work schedule 27.63 25.77 30.28 6.88**
 Nonstandard work schedule 29.43 29.64 29.10 0.03
 Multiple jobs 7.64 5.32 11.44 13.26**
 Change in childcare 5.00 4.10 6.47 2.94+
 Childcare fell through 11.60 9.73 14.68 5.96*
N (excludes those who exited the sample before Wave 3) 1150 699 451
Open in a new tab
+

p ≤ 0.10

*

p < 0.50

**

p < 0.01

Relationship-specific investments and family chaos were first measured at Wave 2, thus the descriptive statistics reported excluded those mothers whose union dissolved before Wave 2. The overall mean number of relationship-specific investments was just over 1; the most common investment was pooled money (80%), followed by joint checking accounts or credit cards (26%), marriage (20%), and a new child with the focal father (10%). Mothers who continually cohabited reported significantly more relationship-specific investments at Wave 2 than did mothers whose union eventually dissolved. The mean level of family chaos was also just over one; the most common sources of family chaos were moving (46%), nonstandard (29%) and inflexible (28%) work schedules, work stress (15%), and childcare problems (14%). Overall, mothers whose cohabiting unions eventually dissolved reported significantly more family chaos at Wave 2 than did mothers whose union remained stable.

Life table results

A majority (64%) of all cohabiting unions dissolved within five years; most (46%) dissolved within the first three years. At the end of five years, just over three-quarters of those cohabiting unions that dissolved ended in a breakup, while the remainder of mothers maintained a romantic relationship with the father of the study’s focal child after cohabitation dissolution. Results from the log rank test for the equality of survival functions indicated significant race differences in the survival rates as well as in the type of dissolution that occurred; overall, the unions of Black mothers were significantly more unstable than the unions of Hispanic and non-Hispanic/non-Black mothers. Black mothers were significantly more likely to experience any type of cohabitation dissolution within 5 years (75%) compared to Hispanic (52%; χ2(1) = 44.70, p < .001) and non-Black/non-Hispanic mothers (57%; χ2(1) = 14.47, p < .001). In particular, Black mothers were significantly more likely to experience cohabitation dissolution into a continuing romantic relationship than Hispanic (χ2(1) = 30.03, p < .001) and non-Black/non-Hispanic (χ2(1) = 13.90, p < .001) mothers. Further, Black mothers were also more likely to experience the breakup of their cohabiting union compared to Hispanic (χ2(1) = 17.85, p < .001) and non-Black/non-Hispanic (χ2(1) = 3.96, p = .05) mothers.

Hispanic mothers were marginally significantly less likely to experience cohabitation dissolution (52%) than non-Black/non-Hispanic mothers (57%; χ2(1) = 2.70, p = .10). There were no differences by type of cohabitation dissolution between Hispanic and non-Black/non-Hispanic mothers. Finally, 28% of Black mothers’ union dissolved into a romantic relationship while 72% dissolved as a breakup. Less than one-fifth of cohabitation dissolutions experienced by Hispanic and non-Black/non-Hispanic mothers involved a continued romantic relationship; just over 80% of Hispanic and non-Black/non-Hispanic mothers whose union dissolved experienced cohabitation dissolution as a breakup.

Maximum Likelihood Discrete-Time Event History Analysis Results

Maximum likelihood discrete time event history models using multinomial logistic regression predicted the competing risks of types of cohabitation dissolution and are presented in Table 2. In models including only the demographic predictors, Black mothers were more likely to have either kind of cohabitation dissolution rather than continuously cohabit compared to non-Black/non-Hispanic mothers. Hispanic mothers were more likely than non-Black/non-Hispanic mothers to continuously cohabit rather than experience cohabitation dissolution as a breakup.

Table 2.

Discrete Time Event History Model Results using Multinomial Logistic Regression Predicting the Competing Risks of Cohabitation Dissolution into a Romantic Relationship versus Breaking-up

Continually cohabiting vs.
Continually cohabiting vs.
Romantic
Broken-up
Romantic
Broken-up
β SE RRR β SE RRR β SE RRR β SE RRR
Mother’s race
 Black 0.83*** 0.21 2.29 0.43*** 0.11 1.54 0.64 0.35 1.90 0.32 0.16 1.37
 Hispanic −0.31 0.24 0.74 −0.27* 0.12 0.76 −0.32 0.41 0.73 −0.17 0.17 0.85
 non-Black/non-Hispanic (excluded group)
Mother’s age 0.02 0.02 1.02 −0.02 0.01 0.98 −0.05 0.03 0.95 −0.03* 0.02 0.97
Father’s age −0.03 0.01 0.97 −0.01 0.01 0.99 0.00 0.03 1.00 −0.00 0.01 1.00
Education
 Less than high school 0.32 0.18 1.37 0.22* 0.10 1.24 0.27 0.31 1.31 0.07 0.14 1.07
 High school graduate (excluded group)
 Some college or more 0.28 0.19 1.32 −0.02 0.11 0.98 0.56 0.32 1.75 0.08 0.16 1.08
Father’s employed −0.62*** 0.17 0.54 −0.17 0.11 0.84 −0.24 0.31 0.79 −0.05 0.16 0.95
Family/relationship characteristics
 Mother higher order birth 0.35 0.18 1.41 −0.05 0.09 0.95 0.26 0.41 1.30 −0.09 0.17 0.92
 Mother multipartner fertility 0.72* 0.35 2.06 −0.03 0.17 0.97
 Father multipartner fertility 0.44 0.28 1.56 0.14 0.14 1.15
 Length of time mother knew father at pregnancy 0.01 0.03 1.02 −0.00 0.02 1.00
 Relationship quality −0.12 0.33 0.88 −0.30 0.16 0.74
 Relationship-specific investments −0.55*** 0.15 0.58 −0.26*** 0.07 0.77
 Family chaos −0.04 0.11 0.96 0.20*** 0.05 1.22
Observation time −0.27*** 0.06 0.76 −0.20*** 0.03 0.82 0.48*** 0.09 1.61 0.28*** 0.04 1.32
χ2 231.35*** 182.45***
df 18 30
Pseudo R2 0.04 0.06
Persons 1567 987
Person Years 5208 4239
Open in a new tab
*

p < 0.05

**

p < 0.01

***

p < 0.001

Mothers with less than a high school education were more likely to breakup than remain continuously cohabiting compared to mothers with a high school degree. In results not shown, mothers with less than a high school education were also more likely to breakup than remain continuously cohabiting compared to mothers with at least some college (β = 0.23, p = .04). Mothers with employed partners were less likely to dissolve their cohabiting union into a continued romantic relationship, and were more likely to continuously cohabit, compared to mothers with unemployed partners.

In results not shown, older mothers were more likely to see their union dissolve into a romantic relationship than to breakup (β = 0.04, p = .05) and mothers with employed partners were less likely to see their relationship dissolve into a romantic relationship than to breakup (β = −0.45, p = .02). Mothers who had a higher-order birth were less likely to experience cohabitation dissolution into a romantic relationship and more likely to experience cohabitation dissolution into a breakup (β = 0.39, p = .02).

Most demographic predictors were no longer significant after the addition of the family/relationship predictors, indicating that the family/relationship predictors mediated most of these associations. Older mothers were significantly more likely to continue to cohabit rather than dissolve their union into a breakup and mothers with multipartnered fertility were more likely to experience dissolution into a continued romantic relationship than continuously cohabit.

Mothers who reported more relationship-specific investments were less likely to dissolve their union into a romantic relationship, and were more likely to continually cohabit; the relative risk of dissolving a cohabiting union into a romantic relationship decreased by 42% with each additional relationship-specific investment. Further, mothers with fewer relationship-specific investments and more family chaos were significantly more likely to experience dissolutions as a breakup and less likely to continuously cohabit. For each additional unit of relationship-specific investments, the relative risk of cohabitation dissolution into a breakup decreased by 33%. For each additional unit of family chaos, the relative risk of cohabitation dissolution into a breakup increased by 22%. The coefficient for relationship quality did not reach significance. In results not shown, mothers who had multipartnered fertility were at greater risk of cohabitation dissolution into a romantic relationship rather than into a breakup (β = 0.75, p = .05). Further, mothers who reported more chaos were less likely to experience cohabitation dissolution into a romantic relationship and were more likely to breakup (β = −0.25, p = .04).

Across all models, time effects were varied. In the model with only the demographic predictors, mothers were more likely to remain cohabiting than to dissolve their union into a romantic relationship as time passed. In contrast, in the model that included the demographic as well as the family/relationship controls, mothers were more likely to dissolve their union into a romantic relationship or a breakup than to remain cohabiting as time passed. Further, in results not shown, unions were significantly more likely (β = −0.20, p = .04) to dissolve as breakups rather than as romantic relationships as time passed.

Sensitivity analyses

The advantages of the discrete time event history models were that all individuals who were censored prior to the final wave were included in the models and time-varying covariates were allowed. Allison (2006) argued that discrete time event history analysis is appropriate for situations in which “events [here, cohabitation dissolution] can occur at any time but measurement of time is not very precise” (pp. 63). However, because a non-trivial amount of data were missing on the exact date of dissolution, I also conducted multinomial logistic regression models on transitions out of cohabitation by wave 4 to check that results were not sensitive to model specification. In these models, each person contributed a single observation point, and time-varying independent variables were measured only at the wave immediately preceding the dissolution or final interview. All individuals who did not dissolve their union and were lost due to attrition prior to wave 4 were excluded from the model. Results (available from the author) were consistent with the results of this paper with a single exception; in the model with demographic controls only, employed fathers were more likely to continuously cohabit rather than breakup. All other results were robust to model specification.

Relationship Status Post Cohabitation Dissolution

The final analyses examined what followed cohabitation dissolution in terms of the parents’ relationship, and whether there were significant differences by type of cohabitation dissolution. For those who dissolved by waves 2 or 3, data from waves following the initial report of the dissolution (waves 3 or 4) were used to examine the relationship status of the mother and the biological father of the study’s focal child two to four years after the initial cohabitation dissolution. Figure 2 presents these findings along with results from a series of z tests that tested the equality of proportions. Mothers who remained romantically involved with the father were significantly more likely than mothers whose union ended in a breakup to be married to (7% versus 4%; z = 1.91, p = .06), cohabiting with (30% versus 9%; z = 6.66, p < .001), or romantically involved (13% versus 4%; z = 4.38, p < .001) with the father of the focal child two to four years later. Further, mothers whose unions ended but remained romantically involved were significantly less likely to be broken-up with the father two to four years later compared to mothers whose union ended in a breakup (50% versus 83%; z = −8.76, p < .001). Overall, breakups were more likely to endure than were cohabitation dissolutions that ended with the parents romantically involved.

Figure 2.

Figure 2

Open in a new tab

Relationship status two to four years after initial cohabitation dissolution by type of cohabitation dissolution (n = 676). Asterisks mark significant differences in relationship status following initial cohabitation dissolution by type of cohabitation dissolution on z tests of the equality of proportions. + p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001.

Discussion

The unions of low-income mothers who were cohabiting at the birth of their child often dissolved and quickly; 46% within three years and 64% dissolved within five. There were two “types” of cohabitation dissolution. Three-quarters of mothers’ romantic relationships ended after cohabitation dissolution; that is, the mothers were broken-up with the father of their child. There are clear disadvantages to the simultaneous end of a union and romantic relationship particularly when children are involved. The negative effects of divorce for children have been clearly documented (Amato, 2000) and cohabitation dissolution has similar impacts on children when it ends in a breakup (Wu, Hou, & Schimmele, 2008). Negative effects for mothers include a decrease in financial resources (Avellar & Smock, 2005) and mental health (Rhoades, Kamp Dush, Atkins, Stanley, & Markman, 2011). The breakup of these unions endured; less than 20% of couples rekindled their romantic relationship two to four years post-dissolution.

The remaining 25% of mothers whose cohabiting union dissolved maintained a romantic relationship with the father of their child. Thus, the end of coresidence did not necessarily mean the end of the parents’ intimate relationship. Of these, 37% moved back in with their former partner two to four years later, and another 13% remained romantically involved. Overall, about half of the mothers whose unions ended but continued romantic involvement were still romantically involved while the remainder was broken-up two to four years later. While a majority of cohabitating unions dissolved into a breakup, there was a minority that remained romantically involved and managed to maintain that involvement years later.

There were significant race differences in both the rate and type of cohabitation dissolution. Black mothers were more likely than Hispanic and non-Black/non-Hispanic mothers to dissolve their union consistent with previous research (Lichter et al., 2006), and when it dissolved, they were more likely to continue a romantic relationship with their former partner. Hispanics were less likely to dissolve their cohabiting union, perhaps due to Hispanic cultural ideals that suggest that cohabitation is a marriage-like relationship (Oropesa, 1996).

Following the investment model (Rusbult, 1980), I examined the role of relationship quality, relationship-specific investments, family chaos, and general investments in predicting cohabitation dissolution. Mothers who reported greater relationship-specific investments were more likely to continue their cohabiting union rather than experience either type of cohabitation dissolution. Pooling money, having a joint checking or credit card, marrying, and having another baby with their partner appeared to be indicative of a greater investment in the relationship, and these unions were more likely to endure. Unmarried parents likely invested more in those relationships that were higher quality and had a greater chance of enduring. Overall, the findings related to relationship-specific investments suggest that when couples invest in their relationship and are satisfied with it, they are more likely to maintain it.

Family chaos was also a significant predictor of cohabitation dissolution into a breakup. Families with more chaos have more stress and less time to spend together as a family (Fiese & Winter, 2010). The mothers who reported more family chaos were more likely to experience cohabitation dissolution as a breakup rather than either continuously cohabit or experience dissolution while remaining romantically involved. Mothers reported moving, non-standard, stressful, and inflexible work environments, and childcare problems, each of which is associated with more chaotic family life and a lower likelihood of creating and maintaining family routines, rituals, and time together (Corapci, 2010; Fiese & Winter; Repetti & Wang, 2010; Wachs & Evans, 2010). Without the time to invest in their relationship, these cohabiting unions were at greater risk of dissolving as a breakup with little chance of reconciliation. After dissolution, it is likely that family chaos increased further.

Turning to general investments, those with more human capital or resources to invest in their relationship were less likely to experience the breakup of their cohabiting union. Older and more educated mothers and those with employed partners were less likely to experience cohabitation dissolution, though few of the general investments remained significant after the addition of the family and relationship variables. Mothers’ multipartnered fertility was associated with a greater risk of cohabitation dissolution with a continued romantic relationship compared to either continuous cohabitation or cohabitation dissolution into a breakup. When unmarried mothers repartner, the new partner is often higher quality in terms of human capital indicators (Bzostek, McLanahan, & Carlson, 2010) and is involved with the mother’s child. Thus the fathers of the focal child in this study were likely involved with his partners’ older children, and these greater investments may have supported a continued romantic relationship because the father had invested in both the mother and child. Yet, the finding that mother’s multiparntered fertility is associated with greater risk of cohabitation dissolution with a continued romantic relationship over continuous cohabitation indicates that these families may still have relationship and family problems.

Time was an indicator of greater relationship investment because shared possessions, memories, and social contacts increased with time, and in the model with general investments, mothers were less likely to experience cohabitation dissolution with a continued romantic relationship as time passed. However, once the family and relationship characteristics were accounted for, and the couples who dissolved or left the sample by the child’s first birthday were dropped from the sample because pre-dissolution family and relationship indicators were unavailable, the passage of time was associated with a greater risk of dissolution. This result is consistent with Cherlin (1977) who found that children in their preschool years were a deterrent to divorce, but once children started school, they no longer were a deterrent. By wave 4, older children were moving out of preschool and into kindergarten. Future research should more closely examine the role of child age, or time, in union dissolution among unmarried parents.

Limitations

There were limitations to this study. First, the exact dates of dissolution were missing for some respondents. Though statistical methods (i.e. discrete time models) were used that account for this error and sensitivity analyses were performed, the models would have been better specified with exact date of union exit. Another limitation is that only the mothers’ reports of the relationship were used. Future research might compare mother and father reports to examine gender differences in perceptions of the couple relationship pre and post-dissolution. The measure of relationship quality used in this study had a low alpha reliability at baseline, and to better understand the dynamics associated with relationship quality in the process of cohabitation dissolution, more reliable survey measures as well as observational measures of couple interaction would be recommended. Another limitation was that all Hispanics were in a single category. Previous research has found differences by country; Mexican Americans tend to be more supportive of marriage, while Puerto Ricans support cohabitation even in the absence of plans to marry (Oropesa, 1996). Future research might distinguish Hispanics by country of origin or by foreign-born status.

A final limitation was that results were only generalizable to mothers having a nonmarital, cohabiting birth in urban settings in the US. Non-parents may be less likely to maintain their romantic relationship after cohabitation dissolution; there is likely less contact between ex-partners when a child is not involved, which may hinder the maintenance of a romantic relationship. Indeed, rates of reconciliation among the parents in this study were higher than those found by Binstock and Thornton (2003) in a population that included non-parents. Further, a higher income/education sample may have more resources invested in their union, and may be more likely to maintain their union or dissolve their union but maintain a romantic relationship. Lichter et al. (2006) found that poor women were more likely to dissolve their union and less likely to marry compared to non-poor women. Future research should examine the process of cohabitation dissolution among non-parents and across the socio-economic spectrum.

Conclusion

Cohabitation dissolution is a common process that families negotiate with little help; neither legal requirements nor social norms are in place to govern it. Few relationship-specific investments and high family chaos put couples at risk for cohabitation dissolution. Cohabiting partners, and in this study, parents, are on their own as they attempt to establish new relationship boundaries, family roles, and expectations, often in the face of increased family chaos. Prevention efforts that target work-life balance such as regular work schedules, flexible work policies, and high quality, affordable, and reliable childcare could help reduce stress and promote family time and regular family rituals and routines (Repetti & Wang, 2010; Fiese & Winters, 2010). Intervention efforts designed to aid families with the cohabitation dissolution process could help both adults and children weather these family structure transitions with fewer negative mental health and behavioral side effects.

Acknowledgments

I am grateful for support for this research from the National Institute of Child Health and Human Development (NICHD; 1K01HD056238). This paper and its contents are solely my responsibility and do not necessarily represent the official views of NICHD. I thank Wendy Manning, Daniel Lichter, Liz Peters, Elizabeth Cooksey, Sarah Schoppe-Sullivan, Melissa Curran, David Johnson, and anonymous reviewers for helpful comments on earlier drafts of this paper. The Fragile Families study was funded by a grant from NICHD (#R01HD36916) and a consortium of private foundations and other government agencies.

Footnotes

Persons interested in obtaining Fragile Families data should see http://www.fragilefamilies.princeton.edu/data.asp for further information.

References

  1. Addo F, Sassler S. Financial arrangements and relationship quality among low-income couples. Family Relations. 2010;59:408–423. doi: 10.1111/j.1741-3729.2010.00612.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allison PD. Event history analysis. Beverly Hills: Sage Publications; 1984. [Google Scholar]
  3. Allison PD. Unpublished workshop notes. Philadelphia, PA: Statistical Horizons; 2006. Notes from event history and survival analysis using Stata. [Google Scholar]
  4. Allison PD. Unpublished workshop notes. Philadelphia, PA: Statistical Horizons; 2008. Notes from missing data. [Google Scholar]
  5. Amato PR. Consequences of divorce for adults and children. Journal of Marriage and the Family. 2000;62:1269–1287. [Google Scholar]
  6. Avellar S, Smock PJ. The economic consequences of the dissolution of cohabiting unions. Journal of Marriage and Family. 2005;67:315–327. [Google Scholar]
  7. Barnett RC. Relationship of the number and distribution of work hours to health and quality-of-life (QOL) outcomes. In: Perrewé PL, Ganster DC, editors. Employee health, coping and methodologies. Oxford, UK: Elsevier; 2006. pp. 99–138. [Google Scholar]
  8. Becker GS. Human capital, effort, and the sexual division of labor. Journal of Labor Economics. 1985;3:S33–S58. [Google Scholar]
  9. Binstock G, Thornton A. Separations, reconciliations, and living aparat in cohabiting and marital unions. Journal of Marriage and Family. 2003;65:432–443. [Google Scholar]
  10. Boyle PJ, Kulu H, Cooke T, Gayle V, Mulder CH. Moving and union dissolution. Demography. 2008;45:209–222. doi: 10.1353/dem.2008.0000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Brines J, Joyner K. The ties that bind: Principles of cohesion in cohabitation and marriage. American Sociological Review. 1999;64:333–356. [Google Scholar]
  12. Brown SL. Union transitions among cohabitors: The significance of relationship assessments and expectations. Journal of Marriage and Family. 2000;62:833–846. [Google Scholar]
  13. Brown DW, Balluz LS, Ford ES, Giles WH, Strine TW, Moriarty DG, Croft JB, Mokdad AH. Associations between short- and long-term unemployment and frequent mental distress among a national sample of men and women. Journal of Occupational and Environmental Medicine. 2003;45:1159–1166. doi: 10.1097/01.jom.0000094994.09655.0f. [DOI] [PubMed] [Google Scholar]
  14. Bzostek SH, McLanahan, Carlson Mothers’ repartnering after a nonmarital birth. Fragile Families Work Paper 06-27-FF. 2010 doi: 10.1093/sf/sos005. Retrieved from: http://crcw.princeton.edu/workingpapers/WP06-27-FF.pdf. [DOI] [PMC free article] [PubMed]
  15. Carlson MJ, McLanahan SS. Fragile Families, Father Involvement and Public Policy. In: Tamis-LeMonda C, Cabrera N, editors. Handbook of father involvement: Multidisciplinary perspectives. Mahwah, NJ: Lawrence Erlbaum Associates; 2002. pp. 461–488. [Google Scholar]
  16. Carlson MJ, McLanahan SS. Early father involvement in fragile families. In: Day R, Lamb M, editors. Conceptualizing and measuring father involvement. Mahwah, NJ: Lawrence Erlbaum Associates; 2004. pp. 241–271. [Google Scholar]
  17. Cherlin AJ. The deinstitutionalization of American marriage. Journal of Marriage and Family. 2004;66:848–861. [Google Scholar]
  18. Cherlin AJ. The effects of children on marital dissolution. Demography. 1977;14:265–272. [PubMed] [Google Scholar]
  19. Corapci F. Child care chaos. In: Evans G, Wachs TD, editors. Chaos and its influence on children’s development: An ecological perspective. Washington, DC: American Psychological Association; 2010. [Google Scholar]
  20. Crawford VP. Relationship-specific investment. The Quarterly Journal of Economics. 1990;105:561–574. [Google Scholar]
  21. Dailey RM, Pfiester A, Jin B, Beck G, Clark G. On-again/off-again dating relationships: How are they different from other dating relationships? Personal Relationships. 2009;16:23–47. [Google Scholar]
  22. Emery RE, Dillon P. Conceptualizing the divorce process: Renegotiating boundaries of intimacy and power in the divorced family system. Family Relations. 1994;43:374–379. [Google Scholar]
  23. Emery RE, Tuer M. Parenting and the marital relationship. In: Luster T, Okagaki L, editors. Parenting: An ecological perspective. Hillsdale, NJ: Erlbaum; 1993. pp. 121–148. [Google Scholar]
  24. England P, Kilbourne B. Markets, marriages, and other mates: The problem with power. In: Friedland R, Robertson AF, editors. Beyond the marketplace: Rethinking economy and society. Hawthorne, NY: Aldine de Gruyter; 1990. pp. 163–189. [Google Scholar]
  25. Fields J, Capser LM. Current Population Report. Washington, DC: U.S. Census Bureau; 2001. America’s families and living arrangements; pp. P20–537. [Google Scholar]
  26. Fiese BH, Winter MA. Family stories and rituals. In: Bray J, Stanton M, editors. Handbook of family psychology. Oxford, UK: Blackwell Publishing; 2010. pp. 625–636. [Google Scholar]
  27. Fincham FD, Beach SRH, Harold GT, Osborne LN. Marital satisfaction and depression: Different causal relationships for men and women? Psychological Science. 1997;8:351–357. [Google Scholar]
  28. Galinsky E, Bond JT, Friedman DE. The role of employers in addressing the needs of employed parents. Journal of Social Issues. 2010;52:111–136. [Google Scholar]
  29. Hamilton BE, Martin JA, Ventura SJ. National Vital Statistics Reports. 16. Vol. 58. Hyattsville, MD: National Center for Health Statistics; 2010. Births: Preliminary data for 2008. [PubMed] [Google Scholar]
  30. Henly JA, Lyons The negotiation of child care and employment demands among low-income parents. Journal of Social Issues. 2000;56:683–706. [Google Scholar]
  31. Hertzman C. The role of temporal and spatial instability in child development. In: Evans GW, Wachs TD, editors. Chaos and its influence on children’s development: An ecological perspective. Washington, DC: American Psychological Association; 2010. pp. 113–131. [Google Scholar]
  32. Hill HD. Steppin’ out: Infidelity and sexual jealousy among unmarried Parents. In: England P, Edin K, editors. Unmarried couples with children. New York, NY: Russell Sage Foundation; 2007. pp. 104–132. [Google Scholar]
  33. Kalil A, Ryan RM. Mothers’ economic conditions and sources of support in Fragile Families. Future of Children. 2010;20:39–61. doi: 10.1353/foc.2010.0009. [DOI] [PubMed] [Google Scholar]
  34. Kenney C. Cohabiting couple, filing jointly? Resource pooling and U.S. poverty policies. Family Relations. 2004;53:237–247. [Google Scholar]
  35. Landale NS, Fennelly K. Informal unions among mainland Puerto Ricans: Cohabitation or an alternative to legal marriage? Journal of Marriage and Family. 1992;54:269–280. [Google Scholar]
  36. Lichter DT, Qian Z, Mellott LM. Marriage or dissolution? Union transitions among poor cohabiting women. Demography. 2006;43:223–240. doi: 10.1353/dem.2006.0016. [DOI] [PubMed] [Google Scholar]
  37. Lichter DT, Wethington E. Chaos and the diverging fortunes of American children: A historical perspective. In: Evans GW, Wachs TD, editors. Chaos and its influence on children’s development: An ecological perspective. Washington, DC: American Psychological Association; 2010. pp. 15–32. [Google Scholar]
  38. Lillard LA, Waite LJ. A joint model of marital childbearing and marital disruption. Demography. 1993;30:653–681. [PubMed] [Google Scholar]
  39. Manlove J, Ryan S, Wildsmith E, Franzetta K. The relationship context of nonmarital childbearing in the U.S. Demographic Research. 2010;23:615–654. doi: 10.4054/DemRes.2010.23.22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Manning WD, Lamb KA. Adolescent well-being in cohabiting, married, and single-parent families. Journal of Marriage and Family. 2003;65:876–893. [Google Scholar]
  41. Manning WD, Smock PJ. Measuring and modeling cohabitation: New perspectives from qualitative data. Journal of Marriage and Family. 2005;67:989–1002. [Google Scholar]
  42. Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemotherapy Reports. 1966;50(3):163–170. [PubMed] [Google Scholar]
  43. Minuchin S. Families and family therapy. Cambridge, MA: Harvard University Press; 1974. [Google Scholar]
  44. Monte LM. Multiple partner maternity versus multiple partner paternity: What matters for family trajectories. Marriage and Family Review. 2011;47:90–124. [Google Scholar]
  45. Nock SL. A comparison of marriages and cohabiting relationships. Journal of Family Issues. 1995;16:53–76. [Google Scholar]
  46. Oropesa RS. Normative beliefs about marriage and cohabitation: A comparison of non-Latino Whites, Mexican Americans, and Puerto Ricans. Journal of Marriage and Family. 1996;58:49–62. [Google Scholar]
  47. Oropesa RS, Gorman BK. Ethnicity, immigration, and beliefs about marriage as a ‘Tie that binds’. In: Waite L, Bachrach C, Hindin M, Thomson E, Thornton A, editors. Ties that bind: Perspectives on marriage and cohabitation. Hawthorne, CA: Aldine de Gruyter; 2000. pp. 188–211. [Google Scholar]
  48. Osborne C, Manning WD, Smock PJ. Married and cohabiting parents’ relationship stability: A focus on race and ethnicity. Journal of Marriage and Family. 2007;69:1345–1366. [Google Scholar]
  49. Perry-Jenkins M. Work in the working class: Challenges facing families. In: Bianchi SM, Casper LM, King BR, editors. Work, family, health, and well-being. Mahwah, NJ: Erlbaum; 2005. pp. 453–472. [Google Scholar]
  50. Preston SH, Heuveline P, Guillot M. Demography: Measuring and modeling population processes. Maiden, MA: Blackwell; 2001. [Google Scholar]
  51. Reed J. Anatomy of the breakup: How and why do unmarried couples with children break up. In: England P, Edin K, editors. Unmarried couples with children. New York, NY: Russell Sage Foundation; 2007. pp. 133–155. [Google Scholar]
  52. Repetti RL, Wang S. Work-family spillover. In: Reis HT, Sprecher SK, editors. Encyclopedia of human relationships. Thousand Oaks, CA: Sage Publications; 2010. pp. 1694–1697. [Google Scholar]
  53. Rhoades GK, Kamp Dush CM, Atkins DC, Stanley SM, Markman HJ. Breaking up is hard to do: The impact of unmarried relationship dissolution on mental health and life satisfaction. Journal of Family Psychology. 2011;25:366–374. doi: 10.1037/a0023627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Rusbult CE. Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology. 1980;16:172–186. [Google Scholar]
  55. Rusbult CE. A longitudinal test of the investment model: The development (and deterioration) of satisfaction and commitment. Journal of Personality and Social Psychology. 1983;45:101–117. [Google Scholar]
  56. Sassler S. The process of entering into cohabiting unions. Journal of Marriage and Family. 2004;66:491–505. [Google Scholar]
  57. Sassler S, McNally J. Cohabiting couple’s economic circumstances and union transitions: A re-examination using multiple imputation techniques. Social Science Research. 2003;32:553–578. [Google Scholar]
  58. Simpson GA, Fowler MG. Geographic mobility and children’s emotional/behavioral adjustment and school functioning. Pediatrics. 1994;93(2):303–309. [PubMed] [Google Scholar]
  59. Smock PJ, Manning WD. Cohabiting partners’ economic circumstances and marriage. Demography. 1997;34:331–342. [PubMed] [Google Scholar]
  60. Staines GL, Pleck JH. The impact of work schedules on the family. Ann Arbor, MI: Institute for Social Research; 1983. [Google Scholar]
  61. Stanley SM, Rhoades GK, Markman HJ. Sliding vs. deciding: Inertia and the premarital cohabitation effect. Family Relations. 2006;55:499–509. [Google Scholar]
  62. Treas J. Money in the bank: Transaction costs and the economic organization of marriage. American Sociological Review. 1993;58:723–734. [Google Scholar]
  63. Wachs TD, Evans GW. Chaos in context. In: Evans GW, Wachs TD, editors. Chaos and its influence on children’s development: An ecological perspective. Washington, DC: American Psychological Association; 2010. pp. 3–13. [Google Scholar]
  64. Waller MR, McLanahan SS. “His” and “her” marriage expectations: Determinants and consequences. Journal of Marriage and Family. 2005;67:53–67. [Google Scholar]
  65. Wu Z, Hou F, Schimmele CM. Family structure and children’s psychosocial outcomes. Journal of Family Issues. 2008;29(12):1600–1624. [Google Scholar]

RESOURCES