1 Strengthening Unmarried Families: Could Enhancing Couple Relationships Also Improve Parenting? Marcia J. Carlson Columbia University Sara S. McLanahan Princeton University Policy makers propose to promote healthy marriage among low-income unmarried cou- ples by providing services to improve relationship skills. This article uses data from the Fragile Families and Child Wellbeing Study to explore whether enhancing parents’ relationship skills may have spillover effects on parent-child relationships. Drawing on findings that relationship quality is positively associated with parenting among married couples, the study examines whether a similar link holds for unmarried couples. A positive association is observed between parents’ relationship quality at the time of a baby’s birth and parenting about 1 year later for both mothers and fathers. Findings indicate that the association between relationship quality and parenting is not affected by marital status. Further, among unmarried couples, neither coresidence nor birth order affects the association. The confluence of trends in marriage and fertility in recent decades has led to a rise in nonmarital childbearing. Today, 36 percent of all children are born to unmarried couples, while the percentages are even higher among African American and Hispanic couples: 46 per- cent and 69 percent, respectively (Hamilton et al. 2005). In response to the high rates of nonmarital childbearing and concern about chil- dren’s well-being, the recent reauthorization of the Temporary Assis- tance for Needy Families program included annual federal expendi- tures of $150 million for programs to promote healthy marriage among ( June 2006). Social Service Review 2006 by The University of Chicago. All rights reserved. 0037-7961/2006/8002-0004$10.00
2 298 Social Service Review unwed parents through the Healthy Marriage Initiative (U.S. Depart- ment of Health and Human Services 2006). The primary focus of these programs is to enhance couples’ relationship skills and increase union stability (Dion et al. 2003). Research now shows that well-designed premarital and postmarital intervention programs strengthen couple relationships, enhance marital stability, and reduce divorce (Stanley et al. 1995; Stanley, Blumberg, and Markman 1999; Stanley 2001). Policy makers hope such programs will do the same for unmarried parents. Research using multiple data sources shows that, contrary to conventional wisdom, unmarried parents highly value marriage and desire marriage for themselves (Lichter, Batson, and Brown 2004; Gibson-Davis, Edin, and McLanahan 2005; Waller and McLanahan 2005). Relationship education may therefore help unmarried couples realize their own intentions. Relationship education programs may also indirectly increase the quality of parenting. An extensive literature shows that marital quality and relationship quality are linked to positive parenting and outcomes for children (Howes and Markman 1989; Cummings and Davies 1994; Erel and Burman 1995; Aldous, Mulligan, and Bjarnason 1998; Emery 1999; Lindahl and Malik 1999; White 1999; Kitzmann 2000; Krishna- kumar and Buehler 2000; Orbuch, Thornton, and Cancio 2000). Most of these studies, however, use small samples of married couples that are disproportionately white and middle income (Erel and Burman 1995; Karney and Bradbury 1995). Thus, the link between relationship quality and parenting remains largely unexamined for unmarried, minority, and low-income parents. This article examines the association between relationship quality and parenting by using data from a new longitudinal survey of both married and unmarried couples that gave birth in large U.S. cities between 1998 and 2000. Both biological parents were interviewed shortly after the birth of their child and again when the child was approximately 1 year old, enabling the research to consider the as- sociation between relationship quality and parenting during early childhood. The study focuses on couples that were romantically in- volved at the time of their baby’s birth and in the subsequent year; this includes couples that were married, cohabiting, and visiting (ro- mantically involved but living apart). Two measures of relationship quality and two measures of parenting are used. The study investigates whether there are differences by marital status in how relationship quality is linked to parenting. Among unmarried parents, it considers whether relationship status (cohabiting or visiting) and birth order (first or subsequent birth) affect the association between relationship quality and early parenting.
3 Couple Relationships 299 Theoretical Perspectives and Previous Research Relationship Quality and Parenting Family theory rests on the notion that the family is a social system in which dyadic relationships (mothers and fathers, parents and children, siblings) between family members affect both individuals and other family relationships (Belsky 1984; Bronfenbrenner 1986; Erel and Bur- man 1995; Cox and Paley 1997). The marital relationship is of partic- ular importance and has been viewed historically as central to nuclear family dynamics (Cummings and O’Reilly 1997). Married couples with positive relationships (i.e., relationships with high mutual support and empathy and with low conflict) are more satisfied with their relation- ships and more likely to stay together than couples with poor rela- tionships (Cowan et al. 1994; Gottman 1994). According to the “spill- over hypothesis” (Erel and Burman 1995, 109), there is emotional spillover of positive affect or stress and overload from one dyadic family relationship to another. The empirical research indeed provides strong evidence for a positive association between the quality of marital and parent-child relationships (Cox et al. 1989; Erel and Burman 1995; Aldous et al. 1998; Lindahl and Malik 1999; White 1999; Kitzmann 2000; Krishnakumar and Buehler 2000; Orbuch et al. 2000). Most studies of marital quality and parenting focus on negative marital func- tioning (conflict, hostility, or aggression) rather than on positive mar- ital quality (but see White 1999 for an exception). Parenting (or parents’ role in the parent-child relationship) can be broadly categorized into positive behaviors (i.e., warmth, engagement, and responsiveness) and negative behaviors (i.e., punitiveness, harsh- ness, and hostility). It is widely agreed that positive parenting is bene- ficial, and negative parenting is detrimental, for children’s healthy de- velopment (Collins et al. 2000; Tamis-LeMonda, Bornstein, and Baumwell 2001; Boyle et al. 2004). Although there are some known gender differences in parenting a new baby (with women being more psychologically and physically involved in the child’s daily care [Cowan et al. 1985; Nomaguchi and Milkie 2003]), parental behavior by both mothers and fathers is shown to affect young children’s development (Kochanska, Aksan, and Carlson 2005). What about Unmarried Parents? Although the vast majority of research on the correlates and conse- quences of partner relationship quality focuses on marital relationships, some studies in nonmarital contexts show that the quality of the rela- tionship between the parents has important consequences. Most such studies focus on noncustodial fathers after divorce or on nonresident
4 300 Social Service Review fathers generally (both never-married and divorced fathers). The studies show that the quality of the relationship between parents after divorce is a key predictor of whether noncustodial fathers visit (Thompson and Laible 1999) and financially support (Teachman 1991) their children. Yet only a handful of known studies explore the link between relation- ship quality and fathering among unmarried parents as a group. This lack of research is likely due to the fact that data are limited on rela- tionship quality among unmarried parents. Using a sample of unmar- ried, urban, African American fathers, Rebekah Coley and P. Lindsay Chase-Lansdale (1999) find that a close mother-father relationship en- courages paternal involvement with young children. Marcia Carlson and Sara McLanahan (2004) find a positive association between emotional supportiveness in the mother-father relationship and fathers’ involve- ment around the time of a nonmarital birth. Since spillover may occur across varying family relationships, there is likely a positive association between the quality of a couple’s relationship and the quality of parenting among unmarried parents, as there is among married parents. At the same time, since marriage represents a significant commitment between partners, a commitment that has both legal and social ramifications, there may be differences by marital status in how relationship quality and parenting are linked. However, the di- rection of the differences is not clear. On the one hand, the association between relationship quality and parenting might be stronger among unmarried couples, as parental roles in nonmarital relationships are less scripted by social norms than those roles in marital settings. Thus, the nature of the parent-child relationship may be more contingent on relationship quality. On the other hand, the association between rela- tionship quality and parenting may be stronger among married couples because they invest more in the relationship (as evidenced by their legal commitment) than their unmarried counterparts do; their roles as part- ners and parents may be perceived as a larger part of their identity and may be more tightly linked. While a priori expectations are uncertain, this article tests whether and how marriage may moderate the association between relationship quality and parenting. Does coresidence matter?— Among unmarried couples that are roman- tically involved at the time of their baby’s birth, the link between re- lationship quality and parenting may vary between couples living to- gether and those living apart (in so-called visiting relationships). Recent research shows that unmarried visiting and cohabiting couples differ in various demographic characteristics and that visiting couples typically have lower-quality relationships than cohabiting couples (Osborne 2005). Coresidence allows both parents to interact regularly with the child, while a nonresident parent faces greater barriers to frequent in- teraction. Therefore, the quality of the relationship between parents
5 Couple Relationships 301 when one is nonresident may be more influential for parenting than the quality of the relationship between coresiding parents. Research suggests that the time shortly after Are first births different?— a baby’s birth is a critical period in which the quality of the marital relationship has particularly strong influence on the development of parent-child relationships (Cox et al. 1989). The birth of a child is a major life event, bringing with it a new parental identity and new re- sponsibilities. Relationship quality typically declines after a baby’s birth `re 2000), though (Cowan and Cowan 1992; Shapiro, Gottman, and Carre the extent of change varies (Belsky and Rovine 1990). Higher-order births create new demands on parents’ time and economic resources. Such demands, in turn, affect the parental relationship. However, the first birth represents the most important transition, as each parent in- dividually and the couple together assume new parental roles (Cowan and Cowan 1992). Using a large multiethnic sample of married and unmarried couples, this article examines four research questions. The first is whether par- ents’ relationship quality (both positive and negative measures) is pos- itively related to parenting behavior (both positive and negative mea- sures). The second is whether there are differences between married and unmarried parents in how relationship quality is linked to parent- ing. The third and fourth are if, among unmarried parents, coresidence and first-birth status, respectively, affect how relationship quality is as- sociated with parenting. Other Variables Previous research points to a number of variables that are related to parents’ relationship quality and parenting. These are discussed only briefly here, as they are not the major focus. They are mostly used as background variables that help to properly specify the models, eliminating potentially spurious relationships. Parents’ older age is associated with a greater postbirth decline in marital satisfaction (Cowan and Cowan 1992), lower-quality marital interactions (Frosch, Mangelsdorf, and McHale 1998), and more effective parenting (Day, Peterson, and McCracken 1998). Race and ethnicity may be associated with parenting practices (Berlin et al. 1995; Brooks-Gunn, Klebanov, and Duncan 1996). Growing up with both parents until adulthood likely affects both relationship qual- ity and parenting, since parenting practices affect children both directly (modeling) and indirectly (emotional well-being and beliefs; Gable, Bel- sky, and Crnic 1995). As noted above, the first birth is more significant than later births for shifting individual parental identities and responsi- bilities, as well as for affecting dyadic couple interaction (Cowan and Cowan 1992). Because highly educated parents are more likely than less educated parents to prepare for parenthood by reading books or taking
6 302 Social Service Review courses, they may develop stronger parenting skills than less educated parents (Simons et al. 1990). Mothers’ employment may increase stress and decrease the time and energy they have to invest in both their partners and parental roles (Belsky 1984). Poverty is also shown to adversely affect parenting behavior (McLoyd 1990). Parents in better physical health, and those who are free from sub- stance abuse problems, may have greater energy and capacity for positive relationships as partners and parents. Having more than one child in the household may diminish that energy. Religiosity is positively related to paternal involvement (King 2003) and to expressive mothering as well (Wilcox 1998). Whether the mother receives prenatal care in the first trimester of pregnancy is a crude indicator of early parenting be- havior, since prenatal care is nearly universally available (through the Medicaid program) to those who seek it; controlling for receipt of pre- natal care provides a more conservative estimate of how the quality of the relationship between parents at birth affects subsequent parenting. It is well known that child temperament affects family process (Lerner 1982), and destructive parenting behaviors may increase with children’s difficult temperament (Simons et al. 1990). Finally, the literature on coparenting (i.e., parents working together to raise their child) under- scores that one parent’s practices and behaviors are not independent of those of the other parent (Gable, Crnic, and Belsky 1994). Data and Methods This study uses data from the Fragile Families and Child Wellbeing Study, a new birth-cohort study of approximately 4,900 children born between 1998 and 2000 (including 3,712 children born to unmarried parents and 1,186 children born to married parents) in 20 cities with populations over 200,000. The Fragile Families Study used a stratified random sampling design; stratification was based on the public policy environments and local labor market conditions in the cities, and sam- pling occurred at three stages: cities, hospitals within cities, and births within hospitals (see Reichman et al.  for more information on study design). Baseline interviews with mothers and fathers were con- ducted shortly after their child’s birth. Mothers were interviewed in person in the hospital within 48 hours of the birth; fathers were inter- viewed either in the hospital or wherever they could be located. Follow- up interviews with both mother and father occurred when the child was about 1, 3, and 5 years old. Response rates for the baseline survey among eligible parents are 87 percent for unmarried mothers, 82 percent for married mothers, 75 percent for unmarried fathers, and 89 percent for married fathers. The 1-year follow-up interviews were completed with 90 percent of unmarried eligible mothers, 91 percent of married eligible
7 Couple Relationships 303 mothers, 70 percent of eligible unmarried fathers, and with 82 percent of eligible married fathers. The sample includes 2,803 couples that were in a romantic relation- ship at both surveys (baseline and 1 year) and did not have missing data on the relationship quality and parenting variables. The analyses use parallel samples of mothers and fathers who are the biological par- 1 The Fragile Families data are well suited for the ents of the focal child. study because they include multiple measures of relationship quality and parenting (positive and negative), because they include interviews with both mothers and fathers, and because they include a large sample of minority and unmarried parents. Variables The dependent variables are mothers’ and fathers’ parenting at the 1- year follow-up survey (when the child was 15 months old, on average). One positive and one negative measure of parental behavior were ex- amined for each parent: reports of positive engagement and frequency of spanking, respectively. A long-standing tenet of developmental psy- chology is that parental interaction and engagement has a substantial effect on children’s developmental trajectories (Maccoby and Martin 1983). Positive engagement here is measured by mothers’ and fathers’ reports of the number of days on which they engaged in the following five activities with their child in the week prior to the interview: play games like “peek-a-boo” or “gotcha,” play inside with toys, sing songs or nursery rhymes, read stories, and tell stories. Responses range from 0 to 7 days; these are averaged to yield a composite measure of engage- 2 p ment (Cronbach’s alpha .76 for fathers and .69 for mothers). The second measure assesses parenting through reports by mothers and fathers of the frequency with which they spanked the child in the month prior to the interview. A recent meta-analysis finds that corporal pun- ishment (of which spanking is the most common form) is negatively associated with child outcomes (Gershoff 2002). While there is some disagreement about when and to what degree spanking is detrimental (Baumrind, Larzelere, and Cowan 2002), spanking is rarely used with infants (Gershoff 2002); all children of parents in the current study are infants, and thus spanking by parents in this sample could be considered harsh parenting behavior. Parents were asked whether they spanked the child in the month before the interview because she or he was misbe- having or acting up; parents who replied affirmatively were asked whether spanking occurred “every day or nearly every day,” “a few times a week,” “a few times in this past month,” or “only once or twice.” Responses are coded into a five-category variable, ranging from 1 3 (never) to 5 (every day or nearly every day). Mothers’ and fathers’ reports about their parenting are positively correlated and statistically
8 304 Social Service Review significant, but the correlations are quite low, possibly indicating that parents behave somewhat independently. Among parents who are mar- ried at the time of the focal child’s birth, the correlations between mother and father are .24 for engagement and .21 for spanking; among parents who are unmarried, the corresponding figures are .13 for en- gagement and .25 for spanking. The key independent variable is parents’ relationship quality. Two aspects of relationship quality are measured. First, supportiveness re- flects the degree of positive affect in the mother-father relationship at the time of the focal child’s birth; supportiveness is represented by the mean score across four items. The items assess how frequently the other parent displays certain behaviors: (1) “is fair and willing to compromise when you have a disagreement,” (2) “expresses affection or love for you,” (3) “insults or criticizes you or your ideas” (coding reversed), and (4) “encourages or helps you to do things that are important to you.” Response choices are (1) never, (2) sometimes, and (3) often; responses are averaged into a composite score, with higher scores indicating greater supportiveness (alpha .55 for both fathers and mothers). A p measure of the change in supportiveness is also included, represented as the difference between an identical supportiveness measure at the baseline and 1-year surveys. A positive change score indicates that the relationship reportedly improved over time, and a negative score sug- gests that it attenuated. The second aspect of relationship quality is the frequency of conflict in the couple’s relationship. Parents are asked during the baseline in- terview about the frequency of conflict over six topics: money, spending time together, sex, the pregnancy, drugs, and being faithful. Again, response choices are (1) never, (2) sometimes, and (3) often. Responses are averaged into a composite score, with higher scores indicating more 4 p .61 for fathers, .62 for mothers). frequent conflict (alpha Unfor- tunately, a comparable measure of conflict is not included in the 1-year survey, so there is no change score on conflict. The correlations for mothers’ and fathers’ reports of relationship quality are again positive and statistically significant but mostly low; for married parents, the cor- relations are .24 for supportiveness and .45 for conflict. For unmarried parents, they are .21 for supportiveness and .31 for conflict. The fact that the supportiveness items concern the other parent’s behavior could contribute to the low correlation there, whereas the conflict measure reflects each parent’s report of the conflict between them; for both measures, the low reliability between mothers’ and fathers’ reports re- flects their different perspectives on the relationship. To reflect the intrinsically dyadic nature of couple relationships, a composite score is used for each relationship measure, taking the average of mothers’ and
9 Couple Relationships 305 fathers’ reports for supportiveness, change in supportiveness, and fre- 5 quency of conflict, respectively. All statistical models include the set of demographic and socioeconomic variables that may be related to both relationship quality and parenting. Unless otherwise indicated, identical measures from the baseline survey are used for both mothers and fathers. Mothers’ and fathers’ ages are specified as continuous variables. Each parent’s race and ethnicity are specified with a series of dummy variables: non-Hispanic black (reference category), non-Hispanic white, Hispanic, and other non-Hispanic race. A separate dummy variable indicates when the parents are of different races or ethnicities. Family background is represented by a dichotomy for whether each parent lived with both of his or her parents at age 15. A dummy variable measures whether the focal child is each parent’s first. Parents’ economic capabilities and current circumstances are mea- sured by the level of education, employment status, and household poverty status. Level of education is specified as four dummy variables: less than high school degree (reference category), high school degree, some college, and bachelor’s degree or higher. Employment status re- flects whether parents worked in the week (for fathers) or year (for mothers) prior to the survey. Household poverty status, or income-to- poverty ratio, is measured by dividing household income by the relevant federal poverty threshold in the baseline survey year (1998 to 2000) for the household in question. Three dummy variables represent being poor (0–.99), near poor (1–2.49), and not poor (2.5 and higher). Parents’ self-reported physical health status is included as a contin- uous variable; scores range from 1 (poor) to 5 (excellent). Parents’ reported problems with substance abuse are represented by a dummy variable; responses are coded as one if parents respond affirmatively that “drinking or drug use interfered with [their] work or personal relationships,” and otherwise as zero. Number of children in the house- hold is a continuous variable. The frequency of each parent’s religious attendance is a continuous variable, ranging from 1 (not at all) to 5 (once a week or more). A variable indicates that the mother received prenatal care in the first trimester of her pregnancy. A dummy variable represents whether the child is a boy. Whether the child is perceived to have a difficult temperament is represented by the average of three items from the Emotionality, Activity, and Sociability Temperament Sur- vey (Buss and Plomin 1984). At the 1-year survey, mothers report how often the focal child fusses and cries, gets upset easily, and reacts very strongly when upset (Buss and Plomin 1984; Mathiesen and Tambs 1999). Response choices range from 1 (not at all like my child) to 5 (very much like my child). Finally, the other parent’s engagement and spanking behaviors, respectively, are included.
10 306 Social Service Review Table 1 Sample Description by Marital Status at Baby’s Birth ( Proportions and Means ) Married Unmarried Mother Mother Father Father 23.78 31.92 26.23 29.60 Age at baby’s birth (mean) Race and ethnicity: 18.1 15.5 47.4 White non-Hispanic 47.3 26.0 47.8 Black non-Hispanic 24.0 49.9 20.8 31.7 32.2 Hispanic 20.2 6.6 2.4 7.8 Other non-Hispanic 2.4 Parents are of different race and ethnicity 11.4 11.4 12.5 12.5 65.2 Parent lived with both parents at age 15 39.1 41.6 67.6 Focal child is first birth 34.3 38.8 45.5 35.2 Level of education: Less than high school degree 36.8 38.6 11.3 12.0 22.7 35.1 36.3 High school degree 19.3 29.8 30.2 24.9 21.8 Some college Bachelor’s degree or higher 39.6 3.3 3.3 35.1 a Worked in the last year or week prior to survey 92.5 81.8 79.6 78.9 Household income-to-poverty ratio: 8.6 8.3 37.4 30.9 Poor (0–.99) Near poor (1–2.49) 24.0 22.8 38.7 38.5 Not poor (2.5 67.4 68.9 23.9 30.6 ) b Health status (mean) 3.87 3.95 4.09 4.07 2.3 2.6 6.1 Substance abuse problem 1.0 1.07 1.03 1.26 1.01 Number of children in household (mean) c Religious attendance (mean) 3.36 2.90 2.63 3.54 91.9 NA 81.3 Mother received prenatal care in first trimester NA 54.0 Child is a boy 52.2 52.2 54.0 d Child has difficult temperament (mean) 2.59 2.59 2.83 2.83 Number of cases ( N 732 732 1,351 1,351 ) All variables are from the baseline survey ( just after baby’s birth) and self- Note.— p not applicable. reported unless otherwise noted. NA a Mothers ( just after the time of birth) were asked whether they worked in the previous year; fathers were asked about whether they worked in the previous week. b Reports of health status range from 1 (poor) to 5 (excellent). c Reports of religious attendance range from 1 (never) to 5 (once a week or more). d From the 1-year survey, reported by mothers. Scores range from 1 to 5; a higher score indicates greater difficulty. Sample Table 1 provides descriptive information on the demographic, psycho- social, and economic characteristics of the mothers and fathers in the sample by marital status at the focal child’s birth. There are a number of differences between married and unmarried parents (all differences ! .05 p noted below are statistically significantly different at ). Married parents are, on average, about 5 to 6 years older than unmarried parents, less likely to be non-Hispanic black, and more likely to have lived with both parents at age 15. They also have much higher educational at- tainment. Only 11–12 percent of married mothers and fathers have less than a high school degree, compared to nearly two-fifths of unmarried
11 Couple Relationships 307 parents. Married parents are much more likely than unmarried parents to have graduated from college (more than one-third, compared to 3 percent of unmarried parents). Around 80 percent of both married and unmarried mothers reportedly worked in the year prior to the survey; 93 percent of married fathers and 80 percent of unmarried fathers worked in the previous week. Only 8–9 percent of married parents live in poverty, compared to 31–37 percent of unmarried parents. Parents generally report that they are in good health, with mean scores around 4.0 on a five-point scale. Admitted problems with substance abuse are rare; between 1 and 6 percent reported such problems. The mean num- ber of children in the household is just over one, although the mean is a little higher for unmarried mothers. Married parents reportedly attend religious services more frequently than unmarried parents. Ninety-two percent of married mothers say they received prenatal care in the first trimester of their pregnancy, compared to 81 percent of unmarried mothers. Just over half of all children are boys. Children of unmarried parents score somewhat higher on the measure that assesses whether a child has a difficult temperament. Results Table 2 presents descriptive statistics (means and standard deviations) on the relationship quality and parenting measures reported by married, cohabiting, and visiting parents at the time of the baby’s birth. Differ- ences across these three groups are tested. There is no statistically sig- nificant difference in supportiveness at the time of birth between mar- ried and cohabiting parents, but both groups report statistically significantly higher supportiveness than do visiting couples. With respect to conflict, married couples report statistically significantly less frequent conflict than do cohabiting couples, who in turn report statistically sig- nificantly less frequent conflict than visitors. At the same time, the sub- stantive differences by relationship status are in most cases quite small. All types of couples report that the level of supportiveness in the relationship declines in the year following their baby’s birth. Cohabitors experience a greater decline than visitors (with no statistically significant difference in the decline when comparing married couples to the other two groups). The pattern of relationship change appears to be similar across couples: Just over half report that supportiveness goes down be- tween the baseline and 1-year surveys (53 percent of married couples, 55 percent of cohabitors, and 52 percent of visitors); roughly one-fourth report that supportiveness stays the same (27 percent of married cou- ples, 23 percent of cohabitors, and 17 percent of visitors). Another quarter indicate that supportiveness goes up (20 percent of married couples, 22 percent of cohabitors, and 31 percent of visitors) over the time period (figures not shown in table).
12 a V V V 1 ! V V 1 C, V C, V ! C C 1 1 ! 1 ! C M M M M, C M M Significant Differences .30 .94 .91 .29 .27 SD 1.77 1.40 390 Visiting (V) .08 1.52 1.44 4.28 5.15 1.44 2.67 Mean .29 .87 .79 .29 .24 SD 1.66 1.35 961 Unmarried .12 Cohabiting (C) 1.43 4.56 1.34 5.11 1.40 2.72 Mean ; Shaffer 1995). .05 ! .29 .89 .83 .29 .25 SD 1.69 1.36 p Total 1,351 Table 2 .11 4.48 5.12 1.46 1.36 1.41 2.70 Mean .24 .77 .74 .26 .24 SD 1.60 1.35 732 .10 Married (M) 4.78 5.42 1.32 1.29 1.32 2.74 Mean Means on Relationship Quality and Parenting Measures by Marital Status ) N 1 year father average): Frequency of spanking Engagement Engagement Frequency of spanking Significant differences are tested with Bonferroni multiple-comparison tests ( Change in supportiveness, birth to Father’s parenting: Mother’s parenting: Frequency of conflict at time of birth a Supportiveness at time of birth Parenting at 1 year: Number of parents ( Parents’ relationship quality (mother-
13 Couple Relationships 309 Linear regression models are estimated to predict the two parenting measures (positive engagement and frequency of spanking) for mothers and fathers. We examine mothers and fathers separately in order to understand whether there are different dynamics in how relationship quality affects their parenting. Models for mothers and fathers are jointly estimated for each outcome using seemingly unrelated regressions (Zell- ner 1962), which allow the error terms to be correlated across mothers’ and fathers’ equations. To deal with missing data on six control variables with more than 10 missing observations, we assign the missing cases to the overall mean and include a flag variable to indicate the case has missing data on a particular variable. This approach allows us to utilize the full sample, but only cases with valid information affect the regres- sion estimates on particular variables. Table 3 reports results related to the first hypothesis, that parental relationship quality shows a positive and statistically significant associ- ation with parenting for both positive and negative measures. For both mothers and fathers, the level of supportiveness in the parents’ rela- tionship at the child’s birth is positively and statistically significantly associated with parent-child engagement. The change in supportiveness is also positively related to higher engagement. Together, these findings provide evidence that a supportive mother-father relationship is im- portant for positive parenting of very young children. Also, the change in supportiveness is negatively linked to spanking by mothers (but not fathers), indicating that when the relationship improves between par- 6 ents, the frequency with which mothers spank their child diminishes. With respect to conflict, parental conflict around the time of the baby’s birth is associated with lower paternal engagement. There is a margin- ally significant positive effect of conflict on maternal engagement, and it is evident from the estimates using mothers’ and fathers’ individual reports that this is driven entirely by fathers’ reports of conflict. Some literature suggests that the gender differentiation in roles subsequent to a new baby’s birth may yield conflict and disequilibrium in the couple relationship, with particularly negative consequences for men, given the mother-centered nature of child rearing (Chodorow 1978). This conflict could push mothers to compensate for the negative dy- namics in the couple relationship by investing more heavily in the parent-child relationship. There thus is evidence that both the positive and negative dimensions of parental relationship quality are important for parenting. Suppor- tiveness in the couple relationship at the time of birth and the change in supportiveness over the subsequent year are both significantly statis- tically associated with positive engagement by both mothers and fathers with their 1-year-old children. Yet, consistent with prior research (e.g., Erel and Burman 1995), there is also some evidence that the frequency of conflict affects parenting.
14 SE .07 .05 .05 .07 .05 .00 .09 .07 .02 .04 .05 .04 .05 .05 .08 .04 .04 Spanking .09 .01 .04 .00 .00 .02 .03 .01** .03 .01 .04 .21** .10* .08 .08 .02 .02 Coefficient SE .11 .01 .19 .15 .11 .04 .08 .10 .10 .09 .14 .11 .10 .18 .09 .14 .08 Father’s Parenting a .23* .06 .30* .23* .04 .06 .01 .03 .19* .10 .32** .71** .52** .20* .16 .22** .23 Engagement Coefficient SE .06 .09 .00 .07 .02 .05 .06 .05 .04 .09 .05 .05 .07 .05 .05 .04 .08 Year after Birth 1 Spanking .01 .04 .00 .01 .11 .01 .02 .18** .04 .05 .00 .18 .19** .21** .02 .09* .00 Coefficient SE .01 .12 .03 .08 .09 .06 .08 .08 .08 .15 .07 .09 .08 .11 .14 .12 .09 Table 3 Mother’s Parenting .15** .06 .01 .06 .04 .05 .07 .24** .06 .25** .20 .51** .42** .14 .06 .26* .20* Engagement Coefficient b Seemingly Unrelated Regressions of Parenting about less than high school): black non-Hispanic): p p ) c Bachelor’s degree or higher High school degree Some college White non-Hispanic Not poor (2.5 Near poor (1–2.49) Other non-Hispanic Hispanic Parents are of different race and ethnicity Age at baby’s birth Worked in the last year or week prior to survey Health status Focal child is first birth Level of education (ref. Parent lived with both parents at age 15 Race and ethnicity (ref. Supportiveness Household income-to-poverty ratio: Change in supportiveness Conflict Other characteristics: Parents’ relationship quality (mother-father average): 310
15 NA .02 .32 .02 .08 .01 .02 .04 .062 .03 2,083 NA .37** .02 .01 .03 .02 .14 .04 1.26** NA .03 .03 .04 .67 .03 .09 .073 .17 .07 2,083 NA .29** .08** .63 .06 .01 .01 .04 .04 .02 .01 .34 .02 .02 .05 .087 .05 .12 .04 2,083 .41** .88** .01 .09** .04* .03 .27* .01 .09** .02 .54 .03 .02 .109 .03 .08 .20 .06 .08 2,083 .19** .03 .07** .10** .03 .22 .08 .18* 2.27** e d ) N not applicable. p . NA . .01 .05 .10 ! ! ! p p p 2 Reports of religious attendance range from 1 (never) to 5 (once a week or more). Mothers (just after the time of birth) were asked whether they worked in the previous year; fathers were asked about whether they worked in the From 1-year survey, reported by mothers. Scores range from 1 to 5; a higher score indicates greater difficulty. Models for mothers and fathers are jointly estimated for engagement and spanking, respectively. Reports of health status range from 1 (poor) to 5 (excellent). Number of children in household Religious attendance Parents married at baby’s birth ** b e a R c Note.— d Substance abuse problem Other parent’s parenting Constant *. Mother received prenatal care in first trimester Child is a boy Child has difficult temperament Number of cases ( previous week. 311
16 312 Social Service Review With respect to the background and control variables, results suggest that white non-Hispanic mothers are reportedly more engaged with their children than black non-Hispanic mothers, and black non-Hispanic par- ents spank their children more frequently than parents in all other racial and ethnic groups. Results also suggest that, in couples of mixed race and ethnicity, both mothers and fathers demonstrate higher engage- ment, compared to parents with the same race and ethnicity. First-time mothers reportedly spank their children more frequently than mothers who have had prior children, and first-time fathers are more engaged than fathers who have other children. Education is positively related to engagement (but not spanking) among both mothers and fathers. There is no statistically significant effect of employment status on parenting, but fathers who are poor are statistically significantly more engaged than 7 Good maternal health is positively asso- their nonpoor counterparts. ciated with positive engagement, and maternal (and to a lesser extent paternal) substance problems are associated with more frequent spank- ing. The frequency of spanking reportedly diminishes, however, as the number of children in the mother’s household increases. Religious at- tendance is positively correlated with parental engagement for both mothers and fathers. Notably, there is no net difference in mothers’ or fathers’ parenting by marital status, once relationship quality and other variables that likely differentiate married from unmarried couples are controlled. Receipt of prenatal care during the first trimester of preg- nancy is positively associated with maternal engagement. Boys are more likely to be spanked by their mothers than girls. The difficulty of the child’s temperament is associated with maternal engagement and spank- ing; higher scores on the difficult temperament scale are associated with less interaction and more frequent spanking by mothers. Child tem- perament is not related to fathering. As expected, the other parent’s parenting is positively correlated with engagement and spanking, re- spectively, among both mothers and fathers. Subsequent analyses provide evidence about whether there are dif- ferences in how relationship quality is related to parenting between married and unmarried couples. The first section (marital status results) of table 4 shows these results. For neither mothers nor fathers is there any evidence of statistically significant interactions of supportiveness or conflict with marital status. Thus, the association between partner re- lationship quality and parenting appears to be similar for unmarried couples and married couples across all measures examined here. The final analyses explore whether associations between relationship quality and parenting differ among unmarried couples, depending on whether the couples cohabited (71 percent) or lived apart (29 percent) at the time of the baby’s birth, and depending on whether couples were having their first child or were having a second or later child. With respect to residential status, there is only one marginally statistically
17 Table 4 Seemingly Unrelated Regressions of Parenting with Interaction Effects for a Marital Status, Coresidence, and First Birth Engagement Spanking Coefficient Coefficient SE SE By marital status (full sample): Mother’s parenting: .43* Supportiveness .18 .11 .02 .23** .08 .14 .34* Change in supportiveness Conflict .09 .27* .14 .09 supportiveness .21 .29 .17 .18 # Married change in supportiveness Married .25 .14 .16 # .38 # .29 .25 .11 .16 Married conflict Father’s parenting: .79** .11 Supportiveness .10 .22 .17 .08 .53** Change in supportiveness .02 .17 .07 .08 Conflict .44* # supportiveness Married .36 .10 .17 .19 Married change in supportiveness .06 .31 .03 .15 # # conflict .31 .06 .15 Married .51 By cohabiting vs. visiting (unmarried sample only): Mother’s parenting: Supportiveness .12 .21 .32 .71* .27 .49** .17 Change in supportiveness .57* .58* .25 .07 .16 Conflict # Cohabiting .30 .39 .18 .25 supportiveness .20 # .26 .31 .36 Cohabiting change in supportiveness # conflict Cohabiting .30 .20 .19 .45 Father’s parenting: Supportiveness 1.20** .40 .02 .19 Change in supportiveness .75* .33 .13 .16 Conflict .31 .05 .15 .33 # supportiveness .57 .49 .11 .23 Cohabiting change in supportiveness .15 .26 .39 # .19 Cohabiting # .12 .37 .17 .18 Cohabiting conflict By first birth vs. later birth (unmarried sample only): Mother’s parenting: .55* .23 Supportiveness .14 .02 Change in supportiveness .17 .22* .11 .45** .22 .06 .11 Conflict .17 # supportiveness .14 .37 .13 First birth .23 First birth # change in supportiveness .18 .28 .01 .18 First birth # conflict .14 .29 .04 .19 Father’s parenting: Supportiveness .30 .22 .14 .76* .61** .02 .11 Change in supportiveness .23 .36 .24 Conflict .11 .10 First birth # supportiveness .15 .45 .26 .21 First birth # change in supportiveness .16 .35 .08 .16 First birth # .14 .35 .06 .16 conflict Each model includes all other independent variables shown in table 3 (age, race Note.— and ethnicity, family background, first birth, education, employment, poverty status, health status, substance problem, children in household, religious attendance, marital status at birth, prenatal care [mother models only], child sex, child temperament, and other parent’s en- gagement or spanking). a Models for mothers and fathers are jointly estimated for engagement and spanking, re- spectively. . p ! .10 ! .05 *. p ! .01 ** . p
18 314 Social Service Review significant interaction: change (improvement) in supportiveness is as- sociated with a greater decline in the frequency of spanking by mothers who live apart from the baby’s father (as compared to cohabiting moth- ers). With respect to parity, there are no statistically significant inter- actions. Relationship quality appears to be just as important for parents who have already had a child as it is for parents experiencing their first child’s birth. Discussion The analyses provide new evidence about how relationship quality is associated with parenting for couples with young children, and the re- sults have implications for current public policy initiatives. Reported relationship quality is found to be positively correlated with parenting for both mothers and fathers, corroborating a significant volume of previous research about positive spillover in marital relationships (Erel and Burman 1995; Aldous et al. 1998; Lindahl and Malik 1999; White 1999; Kitzmann 2000; Krishnakumar and Buehler 2000; Orbuch et al. 2000). However, this article extends these findings to a large multiethnic sample that includes unmarried parents who have recently had a child together. The findings about the importance of supportiveness for par- enting extend the literature on dyadic family relationships. Until now, that research focused primarily on married couples and the negative effects of marital discord. Results suggest that the correlation between relationship quality and parenting exists for both positive and negative dimensions of relationships. Conflict is found to affect parenting, but the stronger evidence concerns supportiveness within the couple rela- tionship. Supportive behaviors by one’s partner at the time of a baby’s birth, as well as an increase in the level of supportiveness over time, appear to promote mothers’ and fathers’ positive engagement in activ- ities with their young children. Both also attenuate the frequency with which mothers spank their children. The results also suggest that the effects of relationship quality on parenting are similar for mothers and fathers, whether the parents are married or not. Although consistency across family types makes sense theoretically, this is the first known study to provide evidence on the matter. At the same time, the descriptive results indicate that the level of relationship quality is slightly higher among married couples than unmarried couples. Thus, marital status does matter with respect to couple relationship quality, but within marital status (i.e., among mar- ried or among unmarried couples), the association between relationship quality and parenting appears to be similar. Further, among unmarried couples, cohabiting couples report higher-quality relationships than vis- iting couples, but there is no significant difference in the association of relationship quality and parenting for these two groups. In addition,
19 Couple Relationships 315 birth order does not affect the association between relationship quality and mothers’ or fathers’ parenting. The results indicate that the current initiatives to promote relation- ship quality and healthy marriage by increasing unmarried parents’ re- lationship skills might yield benefits for parenting behaviors and, hence, for children. If we assume that the associations are causal (a nontrivial assumption), the results suggest that two dimensions of couple rela- tionships are associated with specific parenting practices. Positive sup- port in relationships is associated with good parenting behaviors (e.g., reading to children) by both mothers and fathers, an increase in sup- portiveness is associated with less frequent harsh behavior by mothers (i.e., spanking), and conflict in a relationship is associated with lower engagement by fathers. These findings suggest that interventions might usefully focus on both increasing positive and reducing negative interactions within couples. That observation is underscored by longitudinal research of existing premarital education curricula (Stanley 2001). In order to strengthen couple relationships and facilitate marriage, potential interventions could include a range of marriage education programs designed to enhance communication, improve conflict management, and develop problem-solving skills (see Dion  for a summary of current mar- riage education programs relevant to the Healthy Marriage Initiative). Further, the results provide strong support for the idea that parenting programs would be most effective if they targeted both parents; until recently, most programs designed to improve parenting and child well- being have focused primarily on the mother-child dyad. Results also suggest that interventions potentially could be beneficial for all un- married couples that are in a romantic relationship at the time of the baby’s birth (and stay together over the first year), regardless of whether the parents are living together or not. Finally, while it is well known that the first birth represents an especially significant transition for cou- ples, the findings suggest that relationship quality is equally important for the parenting of mothers and fathers who have already had a child. Hence, results provide little reason to limit eligibility for relationship skills intervention programs to couples having a first birth. The analyses have several limitations. A major one is that it is not clear that the associations between relationship quality and parenting are causal. One could argue that both variables are due to a third factor, such as commitment to family, which is not measured. Similarly, it could be that the measures of mother-father and parent-child relationship quality simply reflect two dimensions of the same intrinsic attribute: the respondent’s interest in and ability to maintain healthy and positive family relationships. As such, findings would not represent causal effects but merely two correlated measures of the same latent construct. Al- ternatively, one could argue that causality is running in the opposite
20 316 Social Service Review direction, from parenting behavior to quality mother-father relation- ships. Without experimental data, these alternative explanations cannot be ruled out. However, controlling for child’s temperament diminishes the likelihood that the results are attributable only to the fact that rearing challenging children negatively affects both parent-child and mother-father relationships. Further, since each parent reports on the perceptions of relationship quality as well as on his or her own parenting, it is possible that the same respondent could overreport or underreport positive feelings of all kinds. This is sometimes referred to as “correlated response bias” (Glenn 1990, 827). Using a composite measure of mothers’ and fathers’ reports of relationship quality helps address this potential concern. A second limitation concerns the measures of relationship quality. Since these are reported right after the time of a baby’s birth, they may be affected positively or negatively by both the joys and stresses that can accompany the event of childbirth. Thus, the reliability and validity of the baseline measures could be questioned. Including a measure of the change in supportiveness between the time of birth and 1 year later provides a more accurate portrait of the developmental trajectory of couples’ relationship quality after their baby’s birth. To the extent that the baseline report represents an overestimate of actual quality, the change measure reflects both real change (as couples adjust to the challenges of parenting a young child) and a more realistic assessment after the warm feelings surrounding the time of birth have diminished. A third limitation concerns the parenting measures. While self-reports of parenting behavior are inferior to more objective direct observations, use of a large-scale survey provides information about a much broader and more representative group of (both married and unmarried) par- ents than would be possible using observational methods. Thus, to some extent, the study trades breadth for depth. Also, the Fragile Families Study includes unique information about both mothers and fathers, whereas previous studies typically rely on maternal reports about fa- thering. The 3- and 5-year waves of the Fragile Families Study included observational data on a subset of children. These data were collected by interviewers who directly observe mothers’ parenting in the child’s home and can provide a better idea of the validity of the self-reported measures. A final issue is that the sample is limited to couples that are roman- tically involved both at the time of a baby’s birth and 1 year later. The findings, thus, cannot be generalized to couples that do not fit this criterion because they broke up or because one or both parents were not interviewed at the 1-year survey. The authors’ analysis suggests that the couples in the Fragile Families Study that break up by the 1-year interview, as well as the couples lost to attrition, typically have lower-
21 Couple Relationships 317 quality relationships (less supportiveness and higher conflict) than cou- ples continuing in the relationship between baseline and the 1-year survey. Excluding these couples, as we do here, may downwardly bias the association between relationship quality and parenting (while up- wardly biasing the mean levels of relationship quality) as compared to what would be observed across a sample of all parents after a birth. This potential for bias does not pose a serious problem as long as one is clear about the population to which the findings can be generalized: couples that continued in a romantic relationship at least 1 year after a nonmarital birth; such couples are likely the most viable target for an intervention program. Conclusion This research provides new evidence about how couple relationship qual- ity is associated with parenting among a diverse sample of married and unmarried couples. Corroborating a vast literature on marital quality, results suggest that the association between partner relationship quality and parenting is similar for unmarried parents and married parents. To the extent that the association is causal, this research suggests that inter- vention programs designed to strengthen unmarried couple relationships and to encourage marriage could potentially further strengthen families by enhancing the parenting that children receive. References Aldous, Joan, Gail M. Mulligan, and Thoroddur Bjarnason. 1998. “Fathering over Time: Journal of Marriage and the Family What Makes the Difference?” 60 (4): 809–20. Baumrind, Diana, Robert E. Larzelere, and Philip A. Cowan. 2002. “Ordinary Physical Punishment: Is It Harmful? Comment on Gershoff (2002).” Psychological Bulletin 128 (4): 580–89. Belsky, Jay. 1984. “The Determinants of Parenting: A Process Model.” Child Development 55 (1): 83–96. Belsky, Jay, and Michael Rovine. 1990. “Patterns of Marital Change across the Transition Journal of Marriage and the to Parenthood: Pregnancy to Three Years Postpartum.” 52 (1): 5–19. Family Berlin, Lisa J., Jeanne Brooks-Gunn, Donna Spiker, and Martha J. Zaslow. 1995. “Examining Observational Measures of Emotional Support and Cognitive Stimulation in Black Journal of Family Issues 16 (5): 664–86. and White Mothers of Preschoolers.” Boyle, Michael H., Jennifer M. Jenkins, Katholiki Georgiades, John Cairney, Eric Duku, and Yvonne Racine. 2004. “Differential-Maternal Parenting Behavior: Estimating within- and between-Family Effects on Children.” Child Development 75 (5): 1457–76. Bronfenbrenner, Urie. 1986. “Ecology of the Family as a Context for Human Development: Developmental Psychology 22 (6): 723–42. Research Perspectives.” Brooks-Gunn, Jeanne, Pamela K. Klebanov, and Greg J. Duncan. 1996. “Ethnic Differences in Children’s Intelligence Test Scores: Role of Economic Deprivation, Home Envi- ronment, and Maternal Characteristics.” Child Development 67 (2): 396–408. Buss, Arnold H., and Robert Plomin. 1984. Temperament: Early Developing Personality Traits . Hillsdale, NJ: Erlbaum. Caldwell, Bettye M., and Robert H. Bradley. 1984. Home Observation for Measurement of the
22 318 Social Service Review . Little Rock: University of Arkansas at Little Rock, Center for Child De- Environment velopment and Education. Carlson, Marcia J., and Sara S. McLanahan. 2004. “Early Father Involvement in Fragile Conceptualizing and Measuring Father Involvement , edited by Randal Families.” 241–71 in D. Day and Michael E. Lamb. Mahwah, NJ: Erlbaum. Chodorow, Nancy. 1978. The Reproduction of Mothering: Psychoanalysis and the Sociology of Gender . Berkeley: University of California Press. Coley, Rebekah Levine, and P. Lindsay Chase-Lansdale. 1999. “Stability and Change in Paternal Involvement among Urban African American Fathers.” Journal of Family Psy- 13 (3): 416–35. chology Collins, W. Andrew, Eleanor E. Maccoby, Laurence Steinberg, E. Mavis Hetherington, and Marc H. Bornstein. 2000. “Contemporary Research on Parenting: The Case for Nature and Nurture.” American Psychologist 55 (2): 218–32. Cowan, Carolyn Pape, and Philip A. Cowan. 1992. When Partners Become Parents: The Big Life Change for Couples . New York: Basic. Cowan, Carolyn Pape, Philip A. Cowan, Gertrude Heming, Ellen Garrett, William S. Coysh, Harriet Curtis-Boles, and Abner J. Boles III. 1985. “Transitions to Parenthood: His, Hers, and Theirs.” Journal of Family Issues 6 (4): 451–81. Cowan, Philip A., Carolyn Pape Cowan, Marc S. Schulz, and Gertrude Heming. 1994. “Prebirth to Preschool Family Factors in Children’s Adaptation to Kindergarten.” 75–114 in edited by Ross D. Exploring Family Relationships with Other Social Contexts, Parke and Sheppard G. Kellam. Hillsdale, NJ: Erlbaum . Cox, Martha J., Margaret T. Owen, Jerry M. Lewis, and V. Kay Henderson. 1989. “Marriage, Child Development 60 (5): 1015–24. Adult Adjustment, and Early Parenting.” Cox, Martha J., and Blair Paley. 1997. “Families as Systems.” Annual Review of Psychology 48:243–67. Cummings, E. Mark, and Patrick Davies. 1994. Children and Marital Conflict: The Impact of Family Dispute and Resolution . New York: Guilford. Cummings, E. Mark, and Anne Watson O’Reilly. 1997. “Fathers in Family Context: Effects of Marital Quality on Child Adjustment.” 49–65 in The Role of the Father in Child . Edited by Michael E. Lamb. 3rd ed. New York: Wiley. Development Day, Randal D., Gary W. Peterson, and Coleen McCracken. 1998. “Predicting Spanking of Younger and Older Children by Mothers and Fathers.” Journal of Marriage and the Family 60 (1): 79–94. Dion, M. Robin. 2005. “Healthy Marriage Programs: Learning What Works.” Future of 15 (2): 139–56. Children Dion, M. Robin, Barbara Devaney, Sheena McConnell, Melissa Ford, Heather Hill, and Pamela Winston. 2003. “Helping Unwed Parents Build Strong and Healthy Marriages: A Conceptual Framework for Interventions.” Report no. 8785-401 to the U.S. De- partment of Health and Human Services. Mathematica Policy Research, Washington, DC. Emery, Robert E. 1999. Marriage, Divorce, and Children’s Adjustment . 2nd ed. Thousand Oaks, CA: Sage. Erel, Osnat, and Bonnie Burman. 1995. “Interrelatedness of Marital Relations and Parent- Child Relations: A Meta-analytic Review.” Psychological Bulletin 118 (1): 108–32. Frosch, Cynthia A., Sarah C. Mangelsdorf, and Jean L. McHale. 1998. “Correlates of Marital Behavior at 6 Months Postpartum.” 34 (6): 1438–49. Developmental Psychology Gable, Sara, Jay Belsky, and Keith Crnic. 1995. “Coparenting during the Child’s 2nd Year: A Descriptive Account.” Journal of Marriage and the Family 57 (3): 609–16. Gable, Sara, Keith Crnic, and Jay Belsky. 1994. “Coparenting within the Family System: Influences on Children’s Development.” Family Relations 43 (4): 380–86. Gershoff, Elizabeth Thompson. 2002. “Corporal Punishment by Parents and Associated Child Behaviors and Experiences: A Meta-analytic and Theoretical Review.” Psycho- logical Bulletin 128 (4): 539–79. Gibson-Davis, Christina, Kathryn Edin, and Sara S. McLanahan. 2005. “High Hopes but Even Higher Expectations: The Retreat from Marriage among Low-Income Couples.” Journal of Marriage and Family 67 (5): 1301–12. Glenn, Norval. 1990. “Quantitative Research on Marital Quality in the 1980s: A Critical Review.” Journal of Marriage and the Family 52 (4): 818–31.
23 Couple Relationships 319 What Predicts Divorce? The Relationship between Marital Processes and Gottman, John M. 1994. Marital Outcomes . Hillsdale, NJ: Erlbaum. Hamilton, Brady E., Stephanie J. Ventura, Joyce A. Martin, and Paul D. Sutton. 2005. “Preliminary Births for 2004.” Health E-Stats. Released October 28. U.S. Department of Health and Human Services, Center for Disease Control and Prevention, National Center for Health Statistics, Hyattsville, MD. Howes, Paul, and Howard J. Markman. 1989. “Marital Quality and Child Functioning: A Longitudinal Investigation.” Child Development 60 (5): 1044–51. Karney, Benjamin R., and Thomas N. Bradbury. 1995. “The Longitudinal Course of Marital Quality and Stability: A Review of Theory, Method, and Research.” Psychological Bulletin 118 (1): 3–34. Factor Analysis: Statistical Methods and Practical Kim, Jae-on, and Charles W. Mueller. 1978. Issues . Quantitative Applications in the Social Sciences, no. 14. Thousand Oaks, CA: Sage. King, Valarie. 2003. “The Influence of Religion on Fathers’ Relationships with Their Chil- Journal of Marriage and Family 65 (2): 382–95. dren.” Kitzmann, Katherine M. 2000. “Effects of Marital Conflict on Subsequent Triadic Family Interactions and Parenting.” 36 (1): 3–13. Developmental Psychology Kochanska, Grazyna, Nazan Aksan, and Jennifer J. Carlson. 2005. “Temperament, Rela- tionships, and Young Children’s Receptive Cooperation with Their Parents.” Devel- opmental Psychology 41 (4): 648–60. Krishnakumar, Ambika, and Cheryl Buehler. 2000. “Interparental Conflict and Parenting Family Relations 49 (1): 25–44. Practices: A Meta-analysis.” Lerner, Richard M. 1982. “Children and Adolescents as Producers of Their Own Devel- Developmental Review 2 (4): 342–70. opment.” Lichter, Daniel T., Christie D. Batson, and J. Brian Brown. 2004. “Welfare Reform and Marriage Promotion: The Marital Expectations and Desires of Single and Cohabiting Mothers.” Social Service Review 78 (1): 2–25. Lindahl, Kristin M., and Neena M. Malik. 1999. “Observations of Marital Conflict and Power: Relations with Parenting in the Triad.” Journal of Marriage and the Family 61 (2): 320–30. Maccoby, Eleanor E., and John A. Martin. 1983. “Socialization in the Context of the Family: Socialization, Personality, and Social Development , Parent-Child Interaction.” 1–101 in . New York: edited by E. Mavis Hetherington. Vol. 4 in Handbook of Child Psychology Wiley. Mathiesen, Kristin S., and Kristian Tambs. 1999. “The EAS Temperament Questionnaire— Journal Factor Structure, Age Trends, Reliability, and Stability in a Norwegian Sample.” of Child Psychology and Psychiatry 40 (3): 431–39. McLoyd, Vonnie C. 1990. “The Impact of Economic Hardship on Black Families and Child Children: Psychological Distress, Parenting, and Socioemotional Development.” 61 (2): 311–46. Development Nomaguchi, Kei M., and Melissa A. Milkie. 2003. “Costs and Rewards of Children: The Effects of Becoming a Parent on Adults’ Lives.” 65 (2): Journal of Marriage and Family 356–74. Orbuch, Terri L., Arland Thornton, and Jennifer Cancio. 2000. “The Impact of Marital Quality, Divorce, and Remarriage on the Relationships between Parents and Their Children.” Marriage and Family Review 29 (4): 221–46. Osborne, Cynthia. 2005. “Marriage Following the Birth of a Child among Cohabiting and Visiting Parents.” 67 (1): 14–26. Journal of Marriage and Family Reichman, Nancy E., Julien O. Teitler, Irwin Garfinkel, and Sara S. McLanahan. 2001. “Fragile Families: Sample and Design.” 23 (4): 303–26. Children and Youth Services Review Shaffer, Juliet Popper. 1995. “Multiple Hypothesis Testing.” Annual Review of Psychology 46: 561–84. Shapiro, Alyson Fearnley, John M. Gottman, and Sybil Carre `re. 2000. “The Baby and the Marriage: Identifying Factors That Buffer against Decline in Marital Satisfaction after the First Baby Arrives.” Journal of Family Psychology 14 (1): 59–70. Simons, Ronald L., Les B. Whitbeck, Rand D. Conger, and Janet N. Melby. 1990. “Husband and Wife Differences in Determinants of Parenting: A Social Learning and Exchange Model of Parental Behavior.” Journal of Marriage and the Family 52 (2): 375–92.
24 320 Social Service Review Family Relations 50 (3): Stanley, Scott M. 2001. “Making a Case for Premarital Education.” 272–80. Stanley, Scott M., Susan L. Blumberg, and Howard J. Markman. 1999. “Helping Couples Fight for Their Marriages: The PREP Approach.” 279–303 in Preventive Approaches in Couples Therapy , edited by Rony Berger and Mo Therese Hannah. New York: Brunner/ Mazel. Stanley, Scott M., Howard J. Markman, Michelle St. Peters, and B. Douglas Leber. 1995. “Strengthening Marriages and Preventing Divorce: New Directions in Prevention Re- 44 (4): 392–401. Family Relations search.” Tamis-LeMonda, Catherine S., Marc H. Bornstein, and Lisa Baumwell. 2001. “Maternal Responsiveness and Children’s Achievement of Language Milestones.” Child Devel- 72 (3): 748–67. opment Social Problems Teachman, Jay D. 1991. “Contributions to Children by Divorced Fathers.” 38 (3): 358–70. Thompson, Ross A., and Deborah J. Laible. 1999. “Noncustodial Parents.” 103–23 in Par- enting and Child Development in “Nontraditional” Families , edited by Michael E. Lamb. Mahwah, NJ: Erlbaum. U.S. Department of Health and Human Services. 2006. “Welfare Reform Reauthorized: Healthy Marriage, Fatherhood Initiative Approved; Work Requirement Strengthened.” News Release, February 8. U.S. Department of Health and Human Services, Adminis- tration for Children and Families, Washington, DC. http://www.dhhs.gov/news/press/ 2006pres/20060208.html. Waller, Maureen R., and Sara S. McLanahan. 2005. “‘His’ and ‘Her’ Marriage Expectations: Determinants and Consequences.” Journal of Marriage and Family 67 (2): 53–67. White, Lynn. 1999. “Contagion in Family Affection: Mothers, Fathers, and Young Adult Children.” Journal of Marriage and the Family 61 (2): 284–94. Wilcox, W. Bradford. 1998. “Conservative Protestant Childrearing: Authoritarian or Au- thoritative?” 63 (6): 796–809. American Sociological Review Zellner, Arnold. 1962. “An Efficient Method of Estimating Seemingly Unrelated Regres- sions and Tests of Aggregation Bias.” 57 Journal of the American Statistical Association (298): 348–68. Notes An earlier version of this article was presented at the annual meetings of the American Psychological Association, Chicago, August 2002, and the American Sociological Associ- ation, Chicago, August 2002. This research is supported by a grant to Marcia Carlson from the Demographic and Behavioral Sciences Branch of the National Institute of Child Health and Human Development (K01HD042776). The Fragile Families and Child Wellbeing Study is funded by a grant from the National Institute of Child Health and Human Development (R01HD36916) and a consortium of private foundations. The authors thank several anonymous reviewers, Jeanne Brooks-Gunn, Scott Coltrane, Ariel Kalil, Steven Nock, Robert Plotnick, Scott Stanley, and seminar participants at Columbia University, Johns Hopkins University, Princeton University, the University of Chicago, and the Uni- versity of Michigan for very helpful comments. 1. Attrition analyses by marital status indicate the characteristics of cases ( ) N p 613 dropped from the sample because the father or mother was not interviewed at the 1-year survey. Compared to the included sample, parents excluded from the sample are more likely to be nonwhite, poor, unemployed (except married fathers), have lower education, have a substance problem, and have children with more difficult temperaments (measure described subsequently). Also, except for married fathers, the reported baseline parental relationship quality is lower for the excluded cases than for those that remain in the sample. Some of these cases dropped due to attrition would have been excluded from the sample anyway, because the couple broke up between the birth and the 1-year follow- up; hence, the comparison of excluded cases with those in the sample represents an upper bound of the differences between the two groups. 2. At the beginning of data collection, an alternate set of responses for these questions
25 Couple Relationships 321 was used (for about 20 percent of the total sample): five categorical response choices, ranging from 1 (never) to 5 (every day). When repeating these response choices multiple times for each item proved too cumbersome, the survey was changed to ask parents to report the actual number of days in the week prior to the survey when they engaged in each activity; responses range from 0 to 7 days. For this article, the categorical responses to numbers of days are assigned as follows: never p 0; once or twice per month p 1; several times per month 2; several times per week p 4; every day p 7. p 3. The current measure is similar to that in the Home Observation Measurement of the Environment; Bettye Caldwell and Robert Bradley (1984) developed the measure, and it is included in the National Longitudinal Survey of Youth, Child Supplement. 4. The low reliabilities on the relationship quality measures suggest that the items are not measuring uniform concepts of supportiveness and conflict, respectively. At the same time, one factor clearly emerges out of each set of relevant items, and the factor loadings of each item are above 0.3, the cut point to be considered substantial (Kim and Mueller 1978). 5. We also estimated models using mothers’ and fathers’ individual reports of relation- ship quality. Not surprisingly, we find that each parent’s report of relationship quality is strongly related to his or her own reported parenting: all but one of the 12 coefficients (three relationship quality measures by two parenting outcomes for both mothers and fathers) is statistically significant (one is marginally statistically significant). There is far less effect of parents’ reports of relationship quality on the other parent’s parenting behavior. The only coefficients that are statistically significant (at ) are as follows: p ! .05 fathers’ report of conflict is associated with greater maternal engagement, and fathers’ report of supportiveness is linked to more frequent maternal spanking. 6. To test the robustness of these findings, logistic regression models were estimated using dichotomous measures of spanking. The pattern of results is very similar to those of the linear regression models. 7. This counterintuitive finding may be attributable to the host of independent variables that are correlated with poverty status; in bivariate models, nonpoor fathers are much more engaged than their poor counterparts (with no difference observed between near- poor and poor fathers).
UCare Connect + Medicare Provider and Pharmacy Directory Introduction This Provider and Pharmacy Directory includes information about the provider and pharmacy types in UCare Connect + Medicare and li...More info »
APA GUIDELINES for Psychological Practice with Boys and Men AUGUST 2018 APA | Guidelines for Psychological Practice with Boys and Men IMore info »
2018 ELECTION DIRECTORY for the 2019-2020 MINNESOTA LEGISLATURE Minnesota House of Representatives and Minnesota Senate Jan. 16, 2019More info »
United States Department of Agriculture Are Healthy Foods Really Economic More Expensive? Research Service It Depends on How You Economic Information Measure the Price Bulletin Number 96 May 2012 Andr...More info »
2018 Physical Activity Guidelines Advisory Committee Scientific Report To the Secretary of Health and Human ServicesMore info »