Michele T Martin, Robert E Emery, Tara S Peris. Handbook of Contemporary Families. Editor: Marilyn Coleman & Lawrence H Ganong. Sage Publication. 2004.
The past 40 years have witnessed profound changes in the structure of American families. Single-parent families in particular have proliferated due to dramatic increases in divorce, nonmarital childbearing, and cohabitation, sparking vigorous debate about the consequences these family structures may have for children. Despite myriad efforts to understand this phenomenon, both from a policy standpoint and from a psychological perspective, findings have been largely inconsistent (Demo & Cox, 2000; Lipman, Boyle, Dooley, & Offord, 2002). In this chapter, we review key findings regarding outcomes for children raised in single-parent families, calling attention to their subtle but psychologically important responses. We then consider several factors that may contribute to these outcomes and discuss methodological and theoretical issues inherent to work in this area as well as how they might be addressed in future research.
Changes in American Families
Interest in single-parent families and their influence on children stems both from the increased occurrence of these households and from evidence that youth in these homes may be exposed to numerous environmental stressors that place them at risk for subsequent poor outcomes. Indeed, single-parent homes headed by women account for approximately 18% of all American families, with nearly half of these households living below the poverty level (Federal Interagency Forum on Child and Family Statistics (FIFCFS, 2000). Further, these trends are evident cross-culturally, with nearly half of all white children and two thirds of black children likely to live in single-parent households at some point in their childhoods (Teachman, Tedrow, & Crowder, 2000). In the context of these staggering statistics, however, an equally important issue pertains to the mechanisms by which single-parent households emerge. In this section, we briefly consider demographic changes in divorce, nonmarital childbearing, and cohabitation as important contributing factors.
The prevalence of divorce is well documented, with over 40% of children born in married households expected to experience their parents’ divorce (Schoen & Standish, 2000). Further, about one third of all births in the United States occur outside marriage, including 25.7% of all births to white mothers, 40.9% of births to Hispanic mothers, and 69.8% of births to African American mothers in 1996 (Ventura, Peters, Martin, & Maurer, 1997). In addition, although rates of teen pregnancy have declined somewhat in recent years, they nearly quadrupled between the 1960s and the early 1990s (Hogan, Sun, & Cornwell, 2000), thereby serving (in many cases) to create a generation of single-parent households. Similarly, rates of cohabitation have increased dramatically, with 25% to 50% of nonmarital childbirths involving children born to cohabiting parents. Although research on these families is just beginning to emerge, preliminary evidence suggests that they may be more fragile (Manning & Smock, 2002) and thus more likely to result in single-parent households. Collectively, these rapid shifts in family structure have converged to produce an increasing number of single-parent families.
The proliferation of one-parent households has in turn generated heated and often polemical debate about the consequences these family environments may hold fo children. Specifically, although some researchers raise strong concerns about the well-being of children in single-parent families (e.g., Popenoe, 1993; Wallerstein, Lewis, & Blakeslee, 2002), others argue vehemently in support of children’s adaptability to changing families (e.g., Demo, 1993). In this chapter, we take the position that the best evidence does not suggest that the truth lies in between these two extremes. Rather, the best research implies the need for a more nuanced approach to understanding the consequences of divorce and nonmarital childbearing for children. As we have argued elsewhere, we must at once recognize that (a) children in single-parent families face many challenging economic and family stressors; (b) despite the stress, the great majority of these children demonstrate considerable psychological resilience; (c) even many resilient children report painful feelings and memories about their growing-up experiences; and (d) many of the putative effects of single-parent families are, in fact, partially attributable to nonrandom selection, including selection influenced by correlated genetic factors (Emery, 1999; Emery, Waldron, Kitzmann, & Aaron, 1999; Laumann-Billings & Emery, 2000). As the literature on single-parent families is vast, our focus here is on the overarching pattern of findings that has emerged in the past 30 years. Further, because the great majority of childrenover 80%live with their single mothers, and because research in this area refers largely to female-headed households (Emery, 1998), we focus primarily on single-mother families. Finally, as the death of a parent may carry its own unique risks for child well-being, we use the term single-parent family to refer to situations in which both parents are living but only one remains in the home.
The Functioning and Well-Being of Children from Single-Parent Families
The risks that accrue to children in single-parent homes are well documented (Ackerman, D’Eramo, Umylny, Schultz, & Izard, 2001; Lipman et al., 2002). Indeed, much of the research on single-parent households has focused on the link between single-parent families and poor outcomes for children, indicating that children in these households may experience an array of problematic behaviors. In the following section, we summarize findings pertaining to child behavior problems in both the internalizing and externalizing spectra, outcomes related to academic competence, and teen pregnancy. Further, as there is evidence that the risks associated with life in a single-parent home may persist into adulthood, we extend this discussion to include outcomes related to mental health and relationship stability in adulthood. Critically, we note that most children are psychologically resilient, as documented by the absence of any of these difficulties. At the same time, however, we suggest that the absence of a psychological disorder is not the same as the absence of psychological distress, and we use this distinction as a springboard to elaborate upon methodological issues related to the assessment of child well-being.
Externalizing and Internalizing Behavior Problems
Of all of children’s psychological problems, parents’ marital status is most consistently and strongly associated with an increased risk for externalizing behavior (Amato & Keith, 1991b; Emery, 1982; Patterson, De Baryshe, & Ramsey, 1989). This pattern of findings, although often more pronounced for boys than for girls (Ackerman et al., 2001; Hetherington & Stanley-Hagan, 1999), has emerged cross-culturally and in studies employing a range of methodological designs. For example, in their intensive, multi-method study of a nonclinic sample of 4- to 6-year-olds, Hetherington, Cox, and Cox (1978) found that children from divorced families were disobedient, aggressive, demanding, and lacking in self-control compared to children in married-parent families. Findings such as these have also been replicated in large national studies conducted in both Britain and the United States and have been shown to persist into adolescence (Mott, Kowaleski-Jones, & Menaghan, 1997; Peterson & Zill, 1986). Moreover, they are not limited to studies of divorce but have been repeatedly borne out in research more broadly examining the impact of single-parent families on child well-being (Ackerman et al., 2001).
Research is much more equivocal in establishing single-parent status as a risk factor for children’s internalizing problems, such as depression, anxiety, and low self-esteem (Amato & Keith, 1991b). For example, a recent study found no difference in children’s depression between single-parent and two-parent families after family process variables were controlled (Lansford, Ceballo, Abbey, & Stewart, 2001). Hypothesized impairments in self-concept also have been studied, but differences typically are small in magnitude and not statistically significant (Amato & Keith, 1991b; Barber & Eccles, 1992). A similarly inconsistent pattern exists with regard to the link between family structure and child anxiety disorders (Dunn, O’Connor, & Levy, 2002; Tweed, Schoenbach, George, & Blazer, 1989).
Divorce and single parenthood are associated with worse performance on a variety of academic measures (Kurdek, Fine, & Sinclair, 1995; McLanahan & Sandefur, 1994; Zill, 1995). However, one methodological difficulty in this area has been that of distinguishing among different domains of academic outcome, with differing results emerging with regard to (a) standardized test scores, (b) grades and related indicators of performance in school, (c) various measures of misconduct in school, and (d) school completion and educational attainment.
Family structure differences on standardized test scores and school grades, although often statistically significant, are typically quite small in magnitude and may diminish when confounding environmental stressors (e.g., poverty) are carefully controlled (Zimiles & Lee, 1991). More substantial differences commonly are found on indices of misbehavior in school. In fact, one direct comparison of the three categories of school outcomes found effect sizes ranging from 0.10 to 0.20 standard deviation units for standardized test scores, 0.15 to 0.28 standard deviation units for GPA, and 0.20 to 0.36 standard deviation units for misbehavior (Zill, 1995). However, the largest and most meaningful differences associated with family status relate to school completion and educational attainment (Amato & Keith, 1991a). Numerous studies have documented that children from single-parent households are about twice as likely to drop out of school as children from married families across national samples, ethnic groups, and divorced versus never-married families and with socioeconomic status controlled for (McLanahan & Sandefur, 1994).
Although there is considerable support for the link between single-parent family status and teen pregnancy (Davis & Friel, 2001), the most compelling findings in this area come from work by McLanahan and Sandefur (1994). Specifically, their work analyzed data from five different national surveys and found that single parenthood was linked with a significant and substantial increased risk of teen pregnancy. Risk was slightly higher for children from never-married versus divorced families, but nevertheless children from divorced families were about twice as likely as children from married families to have a teen childbirth.
Although research has often focused on psychological problems, most children who experience living in a single-parent family do not get pregnant, drop out of school, or require treatment from a mental health professional. For example, Zill, Morrison, and Coiro (1993) found that twice as many 12- to 16-year-old children from divorced as married families received psychological counseling in their national sample of children: 21% versus 11%. However, this means that 79% of 12- to 16-year-old children from divorced families (versus 89% from married families) had coped with their parents’ divorce without receiving psychological help. Similar results emerge when one computes the inverse of risk in other prominent studies (e.g., McLanahan & Sandefur, 1994). Such evidence is an important reminder that most children are resilient in coping. Thus, it seems clear that the majority of children from single-parent families proceed along a relatively healthy child development trajectory as measured by key indicators of their academic, social, and psychological adjustment.
Psychological Distress or Pain
It is important to stress, however, that resilience is not the same as invulnerability; rather, resilience implies that children may bounce back from the stress of living in a single-parent family. In many cases, children may experience psychological difficulties that do not meet clinical criteria but are problematic nonetheless. For example, marital transitions can lead to difficult feelings and worries among children, even when these feelings are not severe enough to warrant a clinical diagnosis. Kurdek and Berg (1987) reported that, among a sample of children whose parents had been separated an average of 17 months, more than 40% agreed that “I can make my parents unhappy with each other by what I say or do.” Similarly, Laumann-Billings and Emery (2000) found that many of 99 well-functioning, resilient young adults from divorced families still reported painful feelings. For example, 49% said that they worried about big events like graduations or weddings when both of their parents would be there; 48% felt that they had a harder childhood than most other people; 44% said that their parents’ divorce still caused struggles for them; and 28% wondered if their father even loved them. Feelings of responsibility, guilt, and painful memories are psychologically important and can be difficult for children to cope with even if they do not lead to depression. In fact, we suspect that the high levels of pain may help to account for some of the different conclusions about the consequences of single parenting reached by clinicians and researchers. Researchers focus on children’s resilience, namely the absence of psychological problems found among most children from divorced and single-parent families (despite the increased risk). Clinicians note the emotional struggles often reported even by resilient children who have coped successfully with stress. Although little work to date has examined subclinical distress as it specifically relates to single-parent family status, we contend that this is an important area for future research and one that might be well suited to addressing previous patterns of inconsistent findings with regard to child outcome (e.g., internalizing behaviors).
Relatively little research has focused on the adaptive skills of children from single-parent families. Growing up in a single-parent family may disrupt the development of prosocial skills, or the converse may be true. For some children, divorce may promote positive development such as a greater sense of personal responsibility, self-esteem, and more gender-neutral attitudes and aspirations (Barber & Eccles, 1992). In this regard, Robert Weiss (1979) has suggested that divorce makes children grow up a little faster. Because of increased family demands, as well as changes in the family’s authority structure, children in single-parent families may have to assume responsibilities at an earlier age than their peers. This may cause them to become precociously competent in social and practical matters (Barber & Eccles, 1992; Weiss, 1979).
A study of a nonclinic sample of children living in two-parent families is suggestive in this regard (Block, Block, & Morrison, 1981). In this longitudinal study, a measure of parental agreement over child-rearing practices completed when children were 3 years old predicted lower levels of aggression in school among boys at ages 3, 4, and 7, but agreement tended to be associated with higher levels of externalizing among girls. In fact, higher levels of disagreement predicted some increases in social competence among the girls but not among boys (e.g., empathic relatedness, resourcefulness, protectiveness of others). When parents fight, girls may be more likely to respond to the stress with increased prosocial behavior, whereas boys may be more likely to respond with increased aggression and noncompliance.
Although intuitively appealing, such an interpretation is speculative. Despite some astute observation, the increased social competence assertion has yet to be adequately documented (Barber & Eccles, 1992). Moreover, if increased maturity is found among children whose parents have divorced, it is not clear whether this is a desirable outcome. The words increased maturity have favorable connotations, but demands for early competence may deprive children of the opportunity to engage in activities that have less immediate benefit but serve them well in the long run. Although the issue of increased or perhaps exaggerated social competence is still open to question, one issue seems increasingly clear. Asking children to perform more adult instrumental duties may not hinder their healthy development. However, children who take on adult emotional responsibilities, in particular attempting to meet their parents’ psychological needs, are more prone to depression during early adult life (Martin, 1995).
Adult Mental Health and Relationship Stability
A considerable body of research has linked growing up in a single-parent family with differences in adult psychological functioning. Studies using large, national samples have found a relationship between growing up in a single-parent family (due to death, divorce, or parents who never married) and depression in adulthood (Amato, 1991; Furstenberg & Teitler, 1994). Similarly, recent twin studies have found that parental loss (due to separation or death) is associated with increased risk for alcohol dependence and major depression in adulthood (Kendeler, Sheth, Gardner, & Prescott, 2002). On the whole, however, research tends to indicate that single-parent family status is more strongly related to social functioning in adulthood than to psychological disorders per se. In particular, research has linked living in a single-parent family with later differences in the quality of young adult romantic relationships. This link with premature or problematic intimate relationships is of growing interest as the children of the divorce boom cohort come of age. Interest is propelled by consistent evidence that children from divorced families are more likely to divorce themselves compared to adults who grew up in married families (Amato, 1996; Bumpass, Martin, & Sweet, 1991). Estimates vary across studies and ethnic groups (weaker transmission effects sometimes are reported for blacks: e.g., Haurin, 1992), but the increase is substantively important, typically ranging from a 25% to 50% increase in divorce risk. Although much of the work in this area to date has focused on children of divorce, we suggest that growing up with only one parent (regardless of the reason) may have substantial implications for adult functioning in romantic relationships, and we underscore the need for further examination of the long-term consequences of being reared in a single-parent household.
Functioning in Single-Parent Families
Although child adjustment has been more widely examined, research has increasingly focused on patterns of family interaction in single-parent households and on how these families may differ in their functioning from two-parent families. This line of research has faced the formidable challenge of untangling the effects of single-parent family status from the effects of the numerous other risk factors with which it is associated. From a methodological standpoint, careful examination of how these subtle, complex family dynamics mediate the link between family structure and child outcome will be essential in testing hypotheses about the specific influence of single-parent family status. Moreover, we note that many of these family stressors are important to children’s well-being in their own right (e.g., economic hardship), regardless of their relationship to family structure or mental health measures.
One of the most robust and striking differences between single-parent and dual-parent families is that single-parent families are more likely to be economically disadvantaged. Single-parent family status is strongly associated with poverty in the United States, nonmarital childbirth more so than divorce. In 2001, 28.6% of female-headed single-parent homes and 13% of male-headed single-parent households fell below the poverty level (U.S. Bureau of the Census, 2002). Further, single-parent family status explains a substantial proportion of racial differences in poverty. In 1995, for example, 43.2% of black children versus 16.6% of white children were living in poverty; much of the difference was attributable to the greater prevalence of single parenthood among blacks than whites. Among black children who lived with two parents, 14.8% fell below poverty cutoffs versus 61.6% who lived with a single mother. The comparable figures for white children were 9.9% for two parents and 44.2% for single mothers (U.S. Bureau of the Census, 1996).
Although both single mothers and fathers are more likely to experience financial strain than their counterparts in first-marriage households, on the whole, single mothers tend to fare worse in this regard. Hilton, Desrochers, and Devall (2001) reported that single fathers had better resources than single mothers, who tended to be less educated, maintain lower-status employment, and have fewer financial and social resources. In addition, economic strain may be particularly pronounced for teenage mothers, who have fewer social and economic supports and are at greater risk for psychological problems (Moffitt, 2002).
Economic hardship is a multifaceted stressor, and there is much to suggest that financial strain may negatively affect parents’ psychological well-being, in turn undermining effective parenting (Mistry, Vandewater, Huston, & McLoyd, 2002). Recent work in this area has focused on the interplay between economic hardship, parental emotional distress, and parenting skills, suggesting that these family dynamics may mediate the effect of economic strain on child development (Conger et al., 2002). As each of these family process variables has been independently linked to poor child outcomes, an important task for future research is to address further patterns of bidirectional influence among these parental variables as well as their collective influence on child adjustment.
However, it is important to recognize that financial difficulties may also exert direct effects on children by causing them to have to move into marginal housing, lose contact with friends, attend inadequate schools, cope with their parents’ worries about money, and face various threats to their physical safety, experiences that are taxing in their own right. Thus, economic strains have a substantial direct impact on children’s adjustment, regardless of the family context in which they occur (Conger, Reuter, & Conger, 2000). In particular, financial hardship has been linked to a host of poor child outcomes, including behavioral and academic difficulty as well as cognitive and physical impairment (McLoyd, 1998). Moreover, adding statistical controls for income reduces differences between the adjustment of children in married and single-parent families by about half for academic measures like school attainment and by a lesser amount for internalizing and externalizing problems (Duncan & Brooks-Gunn, 1997; McLanahan, 2000; McLanahan & Sandefur, 1994).
Task Strain and Parenting
Single mothers report higher levels of life stress compared to women in two-parent homes and may often feel that they have to fulfill both parental roles to their children. In addition, they often have sole responsibility for all household tasks, while simultaneously juggling work and family demands (Heath & Orthner, 1999). These multiple roles can lead to considerable task strain. Examination of this parenting stress and how it is managed is particularly important in light of recent research indicating that parenting variables mediate the link between economic hardship and child well-being (Mistry et al., 2002), as well as other research indicating that family process variables may mediate the effects of family structure on child well-being (Lansford et al., 2001).
Men and women who are residential parents commonly experience strain in fulfilling tasks traditionally assumed by the opposite-gender parent (Chase-Lansdale & Hetherington, 1990; Luepnitz, 1982). In addition, they may have less time alone, which is often beneficial in buffering the transmission of negative emotion from parent to child (Larson & Gillman, 1999). As single parents are likely to be under more stress, this may affect how much energy, time, and focus they can direct toward parenting, including tasks of both affection and discipline. Moreover, their experience of negative emotion may serve to undermine their ability to parent effectively.
Hetherington’s research (Hetherington, 1989; Hetherington, Cox, & Cox, 1982) has provided convincing evidence that single mothers demonstrate poorer parenting skills after marital separation than married mothers. They make fewer maturity demands, communicate less well, show less affection, and are less consistent and less effective in controlling their children. Boys in particular receive less positive feedback and more negative sanctions than daughters. These findings are critical, given research suggesting that the quality of parenting may account for much of the effect of family structure on child outcome (Amato & Fowler, 2002; Martinez & Forgatch, 2002), as well as findings indicating that ineffective parenting in single-parent homes is likely to persist and exacerbate over time (Loeber et al., 2000).
Affection and discipline, the two key domains of parenting, must be carefully negotiated in single-parent families (Emery, 1992, 1994; Emery & Tuer, 1993). Although increased closeness and affection often is a goal, some seemingly close relationships can be problematic. Some children in single-parent families assume inappropriate practical and perhaps emotional responsibilities (Hetherington & Kelly, 2000). One particular concern is that some seemingly close mother-child relationships may be a response to the parent’s needs, not the child’s. Teenagers are less well adjusted when they worry about and feel a need to take care of their parents (Buchanan, Maccoby, & Dornbusch, 1996). Emotional parentification also can be a special concern for younger children who appear to be well adjusted and responsible but who are judged as overburdened by clinicians (Byng-Hall, 2002; Johnston, 1990).
Family processes in father-headed single-parent households have not been studied often or in much detail, but many of the concerns of single mothers are surely shared by single fathers. Residential single fathers have fewer financial problems, but they struggle more with new parental and household responsibilities. Both male and female adolescents report feeling somewhat closer emotionally to their residential single mothers than their residential single fathers, although they report similar levels of parental control (Buchanan et al., 1996). Interestingly, authoritative parenting is associated with better child adjustment in father-residence families, but the relation is stronger in mother-residence families (Buchanan et al., 1996). These findings may be attributable either to selection factors or to difficulties in residential fathering. In either case, it appears that, on average, residential fathers have a few more problems with parenting than residential mothers.
Contact with Nonresidential Parent
Unfortunately, researchers consistently report a low amount of contact between nonresidential fathers and their children, on average (McLanahan & Carlson, 2002). In one of the more recent surveys, for example, approximately one third of divorced fathers saw their children only once or not at all in the past year. About 4 out of 10 fathers saw their children a few times a year up to three times a month, while about 25% saw them once a week or more (Seltzer, 1991). Contact is higher shortly after a marital separation, but it declines with time. In contrast to fathers, nonresidential mothers maintain somewhat greater contact with their children (Stewart, 1999). In one national survey, 30% of nonresidential mothers saw their children once a week or more (versus 25% of nonresidential fathers), and 42% of mothers (versus 55% of fathers) saw their children less than once a month to not at all (Zill, 1988). Furthermore, nonresidential mother-child contact may actually increase, not decrease, over time (Buchanan et al., 1996).
In general, research points to a reluctance of fathers to assume the single-parent role (Hamer & Marchioro, 2002), and thus the vast majority of studies in this area focus on nonresidential fathers. These fathers often have been described as Disneyland dads, and evidence does indicate that visiting fathers place less emphasis on discipline, chores, and schoolwork than fathers in two-parent families (Furstenberg & Nord, 1985). Visits normalize considerably over time, but divorced fathers continue to be less restrictive than married fathers (Hetherington et al., 1982). Compared to married fathers, adolescents rate divorced fathers as less involved, less democratic, and less consistent (Simons & Beaman, 1996). Further, recent evidence suggests that engaging in leisure activities with nonresident fathers as a form of parental involvement is not associated with increased child well-being (Stewart, 2003), underscoring the need for more substantive paternal involvement in children’s lives.
Research on nonresident mothers remains scant and has produced relatively inconsistent findings. For example, there is evidence to suggest that children living with their fathers following paternal remarriage experience greater disruption in the mother-child relationship than children who live with their mothers following maternal remarriage (Aquilino, 1994). However, other studies suggest that children may have an easier time preserving the mother-child relationship than the father-child relationship (Clarke-Stewart & Hayward, 1996). In addition, recent work indicates that nonresident mothers and fathers may be similar in terms of how they interact with their children, with both parents more inclined to participate in leisure activities versus school and disciplinary activities (Stewart, 1999).
Is the quantity of contact or the quality of the nonresidential parent-child relationship associated with difficulties or enhancements in children’s social, psychological, or academic functioning? A fair amount of research has been conducted on this topic in recent years, but the answer appears to be more complicated than the question. For example, one synopsis of the literature found 17 studies where better child adjustment was related to more father contact, 6 studies where better adjustment was associated with less contact, and 9 studies reporting no relation (Amato & Rezac, 1994).
The most likely explanation for the conflicting results is that a number of variables moderate the association between contact with the nonresidential parent and children’s psychological adjustment. One likely moderator is the definition of frequent contact. In one widely cited study, the most frequent of four levels of contact included fathers who saw their children 25 times a year or more, an average of about 2 days per month (Furstenberg, Morgan, & Allison, 1987). Much more frequent contact, perhaps up to the level of joint physical custody, might be necessary to observe benefits in children’s adjustment. Second, the consistency of visitation is another likely moderator, as inconsistent contact and missed visits may actually be harmful to children (Healy, Malley, & Stewart, 1990). Third, the quality of the nonresidential parent-child relationship is probably more important than the quantity of contact, as other research suggests (Barber, 1994; Clarke-Stewart & Hayward, 1996; Simons, Whitbeck, Beaman, & Conger, 1994). Fourth, and perhaps most important, frequent contact may be beneficial to children when interparental conflict is low but harmful when conflict is high (due to children’s increased exposure to the conflict). Some research also supports this possibility (Amato & Rezac, 1994; Healy, Malley, & Stewart, 1990). Other research suggests that parental conflict is a better predictor of child well-being than is marital status (Demo & Acock, 1996; Vandewater & Lansford, 1998).
Single-parent families have often found creative solutions in dealing with challenges of living. A frequent adaptation is the reliance on extended family networks, or kinship, to provide both practical and emotional support. George and Dickerson (1995) noted that sharing child care responsibilities among adult kin other than biological parents in African American families is a practice that can be traced to traditional African communities. Other extrafamilial resources that may provide support include day care personnel, teachers, and friends. Santrock and Warshak (1979) found that the amount of contact with adult caretakers outside the family (the noncustodial parent, babysitters, relatives, day care personnel) was positively related to a child’s functioning following parental divorce. The quality of the contact would appear to be particularly important (Jenkins & Smith, 1990).
Coping with single parenthood when it is due to divorce or an ended cohabiting relationship often takes several years to achieve (Morrison & Cherlin, 1995). Feelings of success in adapting to single parenthood are linked with a number of qualities, such as relinquishing anger at the noncustodial parent; developing informal support networks, including professional counseling before problems get out of hand; relating to children without relying on them for emotional support; and attaining a sense of confidence and pride in managing one’s family.
Single-parent families frequently undergo one or more marital or cohabiting transitions. These can take the form of legal marriages or, more often, cohabiting relationships. The remarriages or repartnerings are often temporary. In one national survey, 37% of children with a remarried parent later experienced a second divorce (Furstenberg, Peterson, Nord, & Zill, 1983).
Remarried families have fewer fixed assets, such as owning a home, than do first-marriage families (Thomson, 1994), but following remarriage single mothers regain disposable incomes comparable to their always-married counterparts (Duncan & Hoffman, 1985). Remarriage thus solves many financial problems created by divorce, and some have viewed it as a similar solution to other problems of divorced families. However, a growing body of research clearly indicates that remarriage is not a simple reconstitution of the two-parent family but is instead yet another difficult transition for biological parents, stepparents, and children (Booth & Dunn, 1994; Hetherington & Clingempeel, 1992).
On average, the combined benefits and costs of remarriage seem to about cancel each other out in terms of children’s mental health. Children in remarried families exhibit about the same levels of psychological problems as do children living with single, divorced parents (Amato, 1994; Zill, 1988). As with divorce, however, there are considerable individual differences in children’s adjustment to remarriage. Future research, especially prospective studies beginning before cohabitation, will be important in helping to untangle the impact of remarriage or repartnering on children.
Parents’ Mental Health
Parents experiencing mental health problems are more likely to be single, and studies of adult mental health invariably find a link between marital status and psychopathology. Married or remarried adults have fewer mental health problems than single or divorced adults. In fact, marital status has been found to be a better predictor of adult mental health than age, race, socioeconomic status, or childhood experience (e.g., Gove, Hughes, & Styles, 1983).
Once again, however, the question arises as to whether the adult mental health problems are consequences of divorce or whether the parental problems predate and perhaps precipitate the divorce. The mental health problems that have been implicated most are depression, antisocial behavior, major mental illness such as schizophrenia or bipolar disorder, and personality disorders. Of these difficulties, depression has received the most attention. However, questions of cause and effect are basic in considering the relation between depression and divorce. Although a marital separation is likely to trigger or exacerbate grieving (Emery, 1994), it is not clear whether syndromal depression is better viewed as a consequence or a cause of marital discord and divorce (Fincham, Beach, Harold, & Osborne, 1997; Gotlib & McCabe, 1990). If adult depression causes divorce, depression is unlikely to (directly) explain adverse psychological adjustment among children from divorced families.
Even if parental depression is a consequence of divorce, research suggests that parental conflict or marital status (Downey & Coyne, 1990; Emery, Weintraub, & Neale, 1982) or inadequate parenting (Simons & Johnson, 1996) is likely to mediate much of the link between parental depression and children’s adjustment. Larson and Gillman (1999) found that negative emotions such as anxiety and anger are transmitted from single mothers to their adolescent children. Finally, it is obvious, though commonly overlooked, that genetic factors may contribute both to parental depression and to children’s psychological problems, with divorce serving as an exacerbating or spurious variable.
Surely the most significant oversight in research on children and divorce is the relative absence of empirical research on the unique experiences of ethnic minorities. Research on national samples, which contain large number of minorities, generally has failed to find differences in the adjustment of white and black children from married versus divorced (or never-married or remarried) families (Fine, McKenry, Donnelly, & Voydanoff, 1992; McLanahan & Sandefur, 1994). One study, however, found that family structure was unrelated to psychosocial outcomes among African American adolescents (Salem, Zimmerman, & Notaro, 1998). Instead, their functioning was linked to living in a supportive, positive, and controlled family environment. Still, research examining different processes in ethnic minority families is sorely needed (McLoyd & Smith, 2002). For example, it has been suggested that extended family support buffers single African American mothers and their children from many of the strains of parenting alone (Wilson, 1989). However, some research with national samples suggests that kin support is neither widely available nor clearly beneficial to the adjustment of African American children (Jayakody, Chatters, & Taylor, 1993; McLanahan & Sandefur, 1994). Such negative evidence is far from conclusive, especially given the complexity and fluidity of the living arrangements of African American children (Dunifon & Kowaleski-Jones, 2002). Rather, it suggests that researchers need a better understanding of the experience of divorce among African Americans and other ethnic minorities before conceptual models of the consequences for children can be adequately tested.
Methodological and Theoretical Issues
Although single-parent households are beginning to receive increased empirical scrutiny, salient methodological and conceptual issues remain (see Demo & Cox, 2000, for review). First, much of the work in this area to date has been largely atheoretical. For the most part, such research has worked from the premise that children in two-parent households benefit from ongoing exposure to both same-sex and opposite-sex role models and that transitions from one family structure to another (i.e., divorce/separation) are sources of stress and anxiety for children (Demo & Cox, 2000). However, theoretical models of how family structure per se affects child adjustment have yet to be specified. As with the literature on divorce, which has frequently indicated that family factors associated with marital disruption (e.g., marital conflict, family cohesiveness) may account for much of the link between divorce and poor child outcome, it is critical to consider relevant correlates of single-parent family status (many of which have been discussed above) that may serve to mediate its effects.
Closely related to this issue is the need for more dynamic analyses of familial interaction. By and large, studies of family structure have been somewhat static methodologically, relying on cross-sectional reports of marital status at a single time point and thus neglecting the complex patterns of interaction that unfold over time (Schrag, Peris, & Emery, 2003). Though instructive, this approach undermines the importance of continuity that may be provided by other caregivers in the home environment (e.g., siblings, grandparents) and fails to consider how the developmental timing of changes in family structure may influence child adjustment. Further, such research often groups children on the basis of two-parent versus single-parent family status alone, a choice that masks critical differences in family resources, education, and child adjustment within each of these groups. Again, we underscore the need for longitudinal examination of family processes associated with single-parent households and for careful consideration of potential mediator and moderator variables in examining links between family structure and child outcomes. Having considered the literature in this area, we turn now to a discussion of how these points might be integrated into future research.
Careful Measurement of Outcome
Different sampling and measurement strategies may differentially influence the conclusions of clinical and empirical experts. Clinicians commonly emphasize the negative consequences of divorce reported by children in therapy (e.g., Kalter, 1990; Wallerstein & Blakeslee, 1989), but clinicians can overlook the similar struggles faced by children from married families, the resilience of those not in treatment, or, indeed, the strengths of their own clients. In contrast, demographers often highlight the small differences found between children from divorced and married families in large, representative samples (e.g., Allison & Furstenberg, 1989), but they run the risk of inappropriately accepting the null hypothesis of no differences, a particular problem for large surveys with limited measurement. Finally, some controversy about the consequences of divorce can be traced to the political agendas not only of politicians but also of family scholars (see Amato, 1993, and replies; Popenoe, 1993, and replies).
Many of psychologists’ empirical measures of children’s psychological health are not especially sensitive to the subtleties of emotional experience. These measures often assess a global, concrete outcome, such as a parent’s ratings of a child’s behavior problems or the number of years in school a child completes. Clearly, parent-rated behavior problems and level of educational attainment are important indices of children’s well-being. At the same time, the measures used in empirical research commonly assess children from the outside in: that is, from the perspective of adults. These objective measures may miss many of the concerns that are of emotional importance to children themselves, even among children who cope well on the outside despite their inner concerns.
Future studies should continue to work toward increased methodological sophistication. For example, conducting observational ratings of parenting behaviors may more accurately represent what the child experiences in terms of parenting than the parent’s self-report of his or her parenting behaviors. Researchers should examine the quality of relationships rather than simply assessing family structure variables or the quantity of contact (Demo, Fine, & Ganong, 2000). In addition, future studies should continue to examine more specific interactional variables that may be mediators for the effects of living in single-parent families on children’s adjustment.
As there have been such dramatic shifts in the numbers of children living in single-parent families, researchers should specifically examine cohort effects. Living in a single-parent family in this decade is likely to be a different experience than it was in the 1960s. Cohort effects may reflect decreasing social stigma and a decreasing sense of being different from other children. These changes may reduce negative impacts on children’s functioning.
Behavior genetics is a newly emerging field that is likely to contribute greatly to our understanding of human behavior. Genetically controlled personality characteristics may in part explain how growing up in a single-parent family is linked to certain outcomes. For example, people who graduate from high school or college have lower divorce rates than those who complete only part of either educational experience, suggesting that a personality characteristic may be linked both with completing school and remaining married (U.S. Bureau of the Census, 1992). Genetic influences contribute to divorce (McGue & Lykken, 1992), and intergenerational continuities can be explained as well by genes as by environments. Opponents of the genetic view can rightly argue that genes cannot account for cohort or cross-cultural differences in divorce rates. However, this argument does not preclude a genetic explanation. The normative threshold for single parenthood may be determined by (shifting) cultural standards, but individual differenceswho becomes a single parent, given a set thresholdmay be determined by one, probably several, genetically mediated personality characteristics and psychological disorders. Because of this important possibility, and the growing body of research indicating that gene-environment correlations do explain many putative environmental effects, it is essential that researchers use genetically informed designs in future studies of single parenting and of family influences more generally.
Research indicates that living in a single-parent family increases the risk for a number of negative outcomes for children. On average, children fare better in a two-parent family than in a single-parent family. At the same time, the great majority of children living in divorced and single-parent families function well despite the expected stress and sources of distress. Thus, it is important to recognize both the struggles and the strengths of single-parent families. We should be particularly cautious about labeling single parenthood as inevitably bad, lest we similarly label those segments of our society where single-parent families not only are common but represent a part of normative family life for the majority of children.