Young Disadvantaged Men as Fathers

Abstract

This article reviews the existing literature on young disadvantaged fathers’ involvement with children. It first outlines the predominant theoretical perspectives regarding father involvement among resident (married and cohabiting) biological fathers, resident social fathers (unrelated romantic partners of children’s mothers), and nonresident biological fathers. Second, it presents a brief discussion of the ways in which fathers contribute to childrearing. Third, it describes the socioeconomic characteristics of men who enter fatherhood at a young age, highlighting that they tend to be socioeconomically disadvantaged. Fourth, it reviews the empirical research on both antecedents of father involvement and patterns of involvement across father types. Finally, it describes the limitations of existing research and provides suggestions for future research and policy.

Both conventional wisdom and existing research suggest that father involvement is beneficial for children’s development and well-being. Yet, changes in men’s roles in family life over the past few decades, coupled with increased diversity in the structure of families, have both reshaped norms and expectations with regard to fatherhood and also resulted in many children being exposed to multiple types of father figures. Whereas fathers once primarily functioned as providers or breadwinners, the father’s role now encompasses a range of child-rearing activities, many of which require active engagement in children’s care (Cabrera et al. 2000). Furthermore, the father figures in children’s lives are now likely to consist of nonresident biological fathers and resident social fathers (used here to connote men who are married to or cohabiting with children’s mothers, but are not their biological fathers) as well as resident biological fathers. And, children born to (particularly young and unmarried) disadvantaged parents are more likely to experience multiple types of father figures than those born to older, more advantaged parents given high rates of union dissolution, relationship instability, and multi-partnered fertility among the former (Carlson and Furstenberg 2008; Meyer, Cancian, and Cook 2005).

In this paper, we first review the predominant theoretical perspectives regarding father involvement among resident (married and cohabiting) biological fathers, resident social fathers, and nonresident biological fathers. Second, we present a brief discussion of the ways in which fathers contribute to child rearing. Third, we describe the socioeconomic characteristics of men who enter fatherhood at a young age, highlighting that they tend to be socioeconomically disadvantaged. Fourth, we review the empirical literature on both antecedents of father involvement and patterns of involvement across father types.¹ Finally, we describe the limitations of existing research and provide suggestions for future research and policy.

How Might Biology, Marriage, Co-Residence, and Social Selection Influence Father Involvement?

Despite focusing on different aspects of family relationships, sociological, economic, and evolutionary perspectives on father involvement point to three consistent hypotheses, which suggest that, all else equal: (1) biological fathers will invest more in children than social fathers; (2) married fathers will invest more than unmarried fathers; and (3) resident fathers will invest more than nonresident fathers. In addition, young disadvantaged fathers are likely to invest less in children than older and more advantaged fathers both because the former are more likely to be either cohabiting or nonresident, as opposed to married or resident, and also because they generally have fewer resources through which to invest. Social selection is also likely to play a significant role with regard to variation in investment in children among men who are biological or social, married or unmarried, resident or nonresident, and younger or older fathers. Below, we discuss each of these hypotheses as well as the potential for social selection to influence father involvement.

Hypothesis 1: Biological fathers will invest more than social fathers

Sociological perspectives on father involvement explicitly consider the influence of biological ties, co-residence, and marriage on fathers’ investments in children, underscoring that different family types are subject to varying degrees of institutionalization, which is likely to play an important role regarding father investments. As such, biological ties are viewed as leading to higher levels of investment than social ties because biological fathers are both legally and normatively obligated to invest in their children; legal and normative expectations of social fathers are less clearly institutionalized or explicit (Cherlin 1978; Cherlin and Furstenberg 1994; Furstenberg and Cherlin 1991).

Whereas sociological perspectives attend to biology, marriage, and co-residence, evolutionary theories primarily emphasize the importance of biological ties for childbearing and child rearing, paying little explicit attention to marriage or co-residence. These perspectives suggest that, given an evolutionary interest in passing on their genes and assuring their children’s success (Daly and Wilson 2000; Emlen 1997), coupled with relatively high (financial and time) costs associated with parental investment, men will inherently strive to provide for their own biological offspring and, by comparison, invest less in social children (Daly and Wilson 1987; Case, Lin, and McLanahan 2001).²

Economic perspectives lead to a similar conclusion, but for still other reasons. Economic approaches to altruism (Becker 1974, 1991), for example, assume interdependent utility functions between family members, such that the utility of an individual family member is partially dependent on the utility of another family member. This implies that parents’ utilities increase with that of their children (and vice versa). While social fathers may engage in some level of altruism toward their partners’ children, the role of altruism among social fathers is likely to be more limited than that among biological fathers. Parents are also likely to make fewer investments in children from whom they expect lesser future returns (Becker 1993) or whom they perceive as having lesser endowments, and these expectations and perceptions may differ by biological status (Bergstrom 1997; Case, Lin, and McLanahan 2001). Biological fathers are likely to assume or expect that their relationship with their children will last throughout their lives and that their children will care for them as they age; the extent to which social fathers will share these assumptions or expectations is ambiguous. On average, then, social fathers are expected to invest less in children than biological fathers. In addition, young disadvantaged fathers, both biological and social, may have lesser expectations of future compensation from their children or less sanguine perceptions of their children’s endowments than their older and more advantaged counterparts; if so, then these men may invest less in children.

Hypothesis 2: Married fathers will invest more than unmarried fathers

Sociological theory suggests that (resident) married fathers will invest more in children than cohabiting or nonresident fathers given the institutional strength of marriage. That is, the legal and public aspects of marriage create an “enforceable trust” (Cherlin 2004) such that married parents are encouraged to make joint relationship-specific investments in children (England and Farkas 1986). As such, marriage and child rearing have been said to consist of a “package deal,” particularly for men (Furstenberg and Cherlin 1991; Townsend 2002). In contrast, cohabitation is marked by a lack of institutionalization, relatively few formal and legal obligations (Nock 1995), and considerable partnership instability, such that children of cohabiting parents are likely to experience multiple family transitions (Graefe and Lichter 1999) and to receive fewer paternal investments (Hofferth and Anderson 2003). Thus, sociological theory suggests that married biological or social fathers are likely to invest more than their cohabiting counterparts.³

Economic perspectives also have implications regarding marriage. To the extent that marriage represents a more formal commitment to an entire family (rather than simply to a spouse) than does cohabitation, married (biological or social) fathers may have a greater sense of economic altruism toward the children, may feel a greater sense of obligation toward them, or may have higher expectations regarding returns on their investments than cohabiting fathers.

Finally, although evolutionary perspectives do not explicitly address marital status, it is plausible that marriage among biological parents signifies a greater willingness on the part of a father to make long-term investments in a child (particularly if marriage is also associated with greater confidence in paternity) and/or a commitment to future childbearing with the child’s mother. If so, evolutionary perspectives may also, albeit implicitly, suggest that married biological fathers will invest more in children than cohabiting and nonresident biological fathers (Anderson, Kaplan, and Lancaster 2007).

Hypothesis 3: Resident fathers will invest more than nonresident fathers

Co-residence implies that children will, to some degree, share their fathers’ income and quality of life (Case, Lin, and McLanahan 2003). That is, fathers at least partially provide for their resident children simply by providing for themselves. Thus, resident fathers face lower costs of investing in children than nonresident fathers given economies of scale. In addition, co-residence lowers the transaction costs associated with investing in children by reducing the time and money needed to arrange and attend visits, as well as impeding the ability of the other parent to limit access to a child (Carlson and Furstenberg 2008). Resident fathers may also have greater incentives to invest in children than nonresident fathers given that they are better able to monitor the ways in which their investments are utilized (Weiss and Willis 1985). Finally, with the exception of child support, nonresident fathers are subject to relatively few legally enforceable obligations to invest in their children (although there are now normative expectations—though relatively ambiguously defined—that nonresident fathers play an ongoing role in their children’s lives). In short, co-residence lowers the price of investing in children and should thereby encourage investment. Again, both because young disadvantaged fathers are less likely to be co-resident and because they have fewer total resources than their older, more advantaged counterparts, they are likely to make lesser investments.

The role of social selection

Another possible explanation for variation in investments in children by father type is that the quantity and quality of such investments are driven by differences in the characteristics of the individuals who “select” into particular families (Amato 2005; Hofferth and Anderson 2003). For example, fathers may be more likely to cohabit than to marry or to be nonresident than resident because they are unattractive partners on a range of characteristics, including their willingness or ability to contribute economic resources to the family (see, e.g., Smock and Manning 1997); this may be particularly true of young disadvantaged fathers. Likewise, the characteristics of social fathers are likely to differ from those of biological fathers. For example, social fathers may have traits that are associated with lesser investments in children (e.g., problems with violence, drugs, or alcohol) to the extent that men with the most desirable fathering qualities will be in ongoing relationships with the mothers of their own biological children. Alternatively, however, mothers who enter into relationships with social fathers may choose a new partner who exhibits characteristics and behaviors that are likely to benefit their child(ren) (Bzostek, Carlson, and McLanahan 2007).

Prior studies of father involvement have primarily used two strategies to attempt to adjust for social selection factors: controlling for as many confounding covariates as possible and estimating fixed effects models that adjust for unobserved persistent characteristics while assessing either within-father change (see, e.g., Gibson-Davis 2008) in involvement over time or within-father differences (see, e.g., Hofferth and Anderson 2003) in investments in different children (e.g., biological and social children). Results suggest that social selection factors explain part, but not all, of the associations between father type and father involvement. Nonetheless, these strategies cannot be assumed to produce causal estimates of the effects of father type on investments in children. For the most part, what we know about social selection into family types suggests that on average social, unmarried, and nonresident fathers will invest less in children than biological, resident, and married fathers based on the preexisting characteristics of men who select into each group. Empirically, however, there are notable exceptions to this general pattern, which we discuss below.

What Do Fathers “Do” for Children?

As families have evolved in both form and nature, so has the social role of “father.” This role historically consisted primarily of providing financially for children and, until relatively recently, economic contributions were widely viewed as a father’s most important input into a household (Crockett, Eggebeen, and Hawkins 1993; McLanahan and Sandefur 1994). Fathers made these contributions in their role as co-resident breadwinner or, in the case of father absence, through formal or informal child support payments.⁴ In recent decades, however, direct and ongoing paternal involvement in childrearing has increasingly come to be seen as important for children of all ages (Hernandez 1993). Thus, today’s fathers are expected to engage in a wide range of direct and indirect childrearing activities (Hewlett 2000). These activities, often referred to as reflecting “investments” in children, have been described as falling into three critical dimensions of fathering: engagement, accessibility, and responsibility (Lamb 1987, 2000; Lamb et al. 1987). As outlined in Lamb (2000), engagement generally refers to shared father-child experiences in which fathers directly interact with their children (e.g., helping with homework, feeding, playing together); accessibility entails the father being physically present, available to, and able to monitor a child, but not actively engaged with the child (e.g., the father is in the same house or room as a child, but they are engaged in separate activities); responsibility includes taking an active role in child-rearing tasks and decisions regarding a child’s care and well-being (e.g., ensuring, arranging, or providing for children’s needs with regard to health, hygiene, and supervision). Fathers may further indirectly invest in their children by supporting (financially, emotionally, or otherwise) their children’s mothers, although this aspect of father involvement has received little scholarly attention (Hawkins and Palkovitz 1999; Marsiglio, Day, and Lamb 2000). To date, existing research suggests that children spend considerably more time engaged with their fathers in terms of play, companionship, and the receipt of personal care than in household work, social, or learning activities (Hofferth et al. 2002; Yeung et al. 2001).

Despite relatively sophisticated conceptualizations of father involvement, as discussed below, “most empirical work still employs fairly simple measures of selected aspects of engagement” (Nelson, 2004, p. 434). As such, our knowledge of the range of ways fathers contribute to child rearing—and how father involvement may differ by socioeconomic factors and a father figure’s biological, marital, and co-resident status vis-à-vis a child and his or her mother—has generally been limited to simple accountings of the amount of time fathers spend with children and the number and types of activities fathers directly engage in with children (Marsiglio, Amato, Day, and Lamb 2000). We know much less about fathers’ ongoing accessibility to their children, responsibility for their children’s care and well-being, or the warmth, sensitivity, and overall quality of their interactions with their children. On the whole, however, young disadvantaged fathers may have less capacity than older more advantaged fathers to invest in children across all of these domains, as they are likely to have fewer financial resources to (potentially) devote to child rearing (Danziger and Radin 1990; Furstenberg 1995) and higher levels of social, emotional, and behavior difficulties (Bunting and McAuley 2004).

Who Becomes a Young Father?

The age at which individuals become parents differs considerably by race and ethnicity and is positively correlated with socioeconomic status. Black and Hispanic men are more likely than white men to become young (e.g., under age 25) fathers (Hynes et al. 2008) and young fathers have lower average levels of education and employment, come from lower socioeconomic status families, and have greater psycho-emotional problems and levels of delinquent behaviors than men who did not experience early fertility (Bunting and McAuley 2004). In addition, data from the Fragile Families and Child Wellbeing Study (FFCW)—the first birth cohort study to collect extensive data on (and from) unmarried fathers—reveal that unmarried fathers are considerably younger and more disadvantaged than married fathers; they are less educated, less healthy, and less likely to be working; they are also more likely to have been incarcerated, to have a substance problem, to be black or Hispanic (and less likely to be white), and to have children by multiple partners (Carlson and McLanahan, forthcoming).

In short, the individuals who select into parenthood at young ages tend to be socioeconomically disadvantaged (Carlson and McLanahan forthcoming; Child Trends 2002; Hamilton, Martin, and Ventura 2009; Lerman 1993; Marsiglio 2000),⁵ to be unmarried, and to have characteristics that limit their ability to provide for their children economically. These same characteristics are associated with low levels of father involvement (Cooksey and Craig 1998; Danziger and Radin 1990; Furstenberg and Harris 1993; Roggman et al. 2002), suggesting that social selection may well play a role in explaining associations between father type and father involvement. Furthermore, as Nelson (2004) points out, declines in (particularly black) young disadvantaged men’s labor force participation and wages over last few decades (see also Sum, this volume) imply that those characteristics that are associated with early entry into fatherhood may be becoming more prominent in the U.S. population. This is likely to have important implications for both public policy and U.S. society given that differences in the socioeconomic characteristics of younger and older parents are likely to be associated with patterns of intergenerational transmission of inequality (McLanahan and Percheski 2008).

What Do We Know About Young Disadvantaged Fathers’ Involvement With Children?

In this section, we outline those factors that are associated with men’s investments in children. We then describe patterns of father involvement for resident biological fathers, resident social fathers, and nonresident biological fathers.⁶ We pay careful attention to what is known about differences in father involvement between married and cohabiting fathers. Whenever possible, we also highlight studies that are particularly relevant to young disadvantaged fathers. Finally, given high rates of incarceration among disadvantaged young men, coupled with the particular difficulties associated with continued involvement with children among incarcerated fathers, we discuss father involvement with regard to non-incarcerated and incarcerated nonresident fathers separately.

Antecedents of father involvement

Prior research points to multiple characteristics of fathers, mothers, children, and mother-father relationships that are associated with father involvement; consistent with what we know about social selection, many of these same factors are associated with age of entry into fatherhood and with father type. Father background characteristics that are positively correlated with involvement include accumulated human capital (e.g., educational attainment and employment) (Cooksey and Craig 1998; Danziger and Radin 1990; Hofferth et al. 2002; Landale and Oropesa 2001; Manning and Smock 2000; Manning, Steward, and Smock 2003), the quality of men’s experiences with their own fathers and other male role models (Cabrera et al. 2000; Hofferth 2003), and the extent to which men identify with the fathering role (Cabrera et al. 2000; Rane and McBride 2000). Father psycho-social characteristics such as a history of incarceration or physical violence are negatively associated with both ongoing involvement with children and payment of child support (Carlson and McLanahan, forthcoming; Carlson, McLanahan, and Brooks-Gunn 2008; Ryan, Kalil, and Ziol-Guest 2008), as is paternal re-partnering and subsequent fertility (Manning and Smock 2000). Finally, nonresident fathers who make (particularly informal) financial contributions to their children tend to be more involved with them, although the causal direction of this association has not been firmly established (Nepomnyaschy 2007).

Maternal characteristics such as whether a mother and child live with extended family and whether the mother has re-partnered or had subsequent children also appear to influence father involvement. Whereas re-partnering and new partner fertility are associated with decreased involvement (Tach, Mincy, and Edin, forthcoming), the association between whether the mother lives with kin and father involvement is likely to reflect the quality of the relationship between the father and the mothers’ co-resident relatives (Bunting and McAuley 2004; Danziger and Radin 1990; Krishnakumar and Black 2003).

Child characteristics such as age (Hofferth et al. 2002; Lamb 2000), gender (Lamb 2000; Lundberg, McLanahan, and Rose 2005), and health status (Reichman, Corman, and Noonan 2004) are also associated with father involvement. In general, both mothers and fathers are more involved with younger children than with older children; however, the ratio of mother to father involvement with children tends to decrease as children age, suggesting that relative to mothers, fathers spend more time with older than younger children (Lamb 2000). There is also some evidence that fathers are more involved with sons than daughters (Lamb 2000), although this may only hold for married fathers (Lundberg, McLanahan, and Rose 2005). Finally, fathers appear to be less involved with children who were born low birth weight, are disabled, or have developmental delays (Reichman, Corman, and Noonan 2004).

Involvement on the part of both resident and nonresident fathers has also been linked to the quality of the mother-father relationship and conflict therein (Bunting and McAuley 2004; Carlson and McLanahan 2004; Carlson, McLanahan, and Brooks-Gunn 2008; Ryan, Kalil, and Ziol-Guest 2008). Particularly salient in this regard is maternal gatekeeping, defined as the influence a mother exerts on father involvement with children either directly, by controlling access to children, or indirectly, by affecting paternal role identity (McBride et al. 2005), often through encouragement or criticism (Schoppe-Sullivan et al. 2008). Maternal gatekeeping behaviors are closely tied to the relationship quality of mothers and fathers as well as to mothers’ perceptions of paternal competence and the father’s role within the family (Fagan and Barnett 2003). Research suggests that maternal gatekeeping behaviors heavily influence father involvement among both resident and nonresident fathers (Allen and Hawkins 1999; Fagan and Barnett 2003; McBride et al. 2005; Schoppe-Sullivan et al. 2008).⁷

Father involvement among resident biological and social fathers

The vast majority of resident (biological and social) fathers spend time and engage in activities with children on a relatively consistent basis (see, e.g., Child Trends 2002). However, compared to mothers’ activities with children, fathers’ activities are much more likely to consist of play (including sports) and leisure activities than cognitively stimulating activities. Furthermore, whereas the majority of resident fathers do engage in “rule setting,” monitoring activities, and discipline, they do so much less frequently than do mothers (Child Trends 2002).

Most existing research suggests that (particularly married) resident biological fathers are more involved with children than all other father types (see, e.g., Hofferth et al. 2007). However, (married) resident biological fathers also tend to have greater human capital than other father types and are typically older (Hofferth and Anderson 2003; Manning and Brown 2006; Manning and Lichter 1996; Smock 2000). A large group of studies has also compared the involvement of married biological fathers to that of married stepfathers. In general, this literature finds that married stepfathers are less involved with stepchildren than are married biological fathers with their biological children (Amato and Sobolewski 2004; Coleman et al. 2000; Hetherington and Stanley-Hagan 1999; Nelson 2004). Married stepfathers are less likely to participate in activities with children (Cooksey and Fondell 1996; Thomson, McLanahan, and Curtin 1992), express positive feelings toward children (Thomson et al. 1992), be supportive of their children (Amato 1987), and exhibit monitoring and controlling behaviors toward (particularly adolescent) children (Amato 1987; Fisher et al. 2003; Hetherington and Jodl 1994).

Far less research has compared the involvement of married biological fathers to that of cohabiting biological fathers or that of married social fathers to that of cohabiting social fathers. Results from the handful of studies that directly compare married and cohabiting biological fathers suggest that the former are more likely to make financial investments in children (Landale and Oropesa 2001), spend more time engaged with children (Hofferth and Anderson 2003; Hofferth et al. 2007), engage more frequently in caregiving activities (Landale and Oropesa 2001), and exhibit slightly more cooperation in parenting with their children’s mothers (Berger et al. 2008) than are cohabiting biological fathers. However, they also engage in more frequent spanking (Gibson-Davis 2008). Existing evidence regarding differences in paternal warmth and responsibility for parenting between married and unmarried resident biological fathers has been quite mixed (Hofferth and Anderson 2003; Hofferth et al. 2007).

We are aware of only one study that directly compares involvement in child rearing between married and unmarried social fathers. Berger and colleagues (2008), using data from FFCW find that, even after controlling for a wide range of selection factors, married social fathers of young children are reported by mothers as being more engaged with children, sharing more responsibility for parenting, and cooperating more fully in parenting than their cohabiting counterparts. Mothers also trust married social fathers more so than cohabiting social fathers to care for children in their absence.⁸

Finally, a few recent studies have documented differences in father involvement between biological and social fathers, sometimes taking marital status into account. Findings from these studies tend to differ across data sources and by whether analyses are bivariate or multivariate, with multivariate analyses of FFCW data being most favorable toward social fathers.⁹ For example, bivariate analyses of father involvement across five large scale datasets suggest that married biological fathers have higher levels of engagement than both married and cohabiting social fathers; they also appear to express more warmth than married social fathers (Hofferth et al. 2007). Similarly, Hofferth and Anderson’s (2003) multivariate analyses using data from the Panel Study of Income Dynamics (PSID) suggest that, in general, married and unmarried social fathers are less involved with children than are married biological fathers. ¹⁰ In contrast, however, Berger and colleagues’ (2008) multivariate analyses using FFCW data suggest that (particularly married) resident social fathers engage in higher levels of shared responsibility for and cooperation in parenting than resident biological fathers, and Gibson-Davis’s (2008) multivariate analyses using FFCW data suggest that cohabiting social fathers have higher levels of engagement and instrumental support, and lower levels of spanking than married biological fathers.

On the whole, the existing literature leads us to conclude that biological fathers tend to be more involved than social fathers and also that married biological fathers are typically more involved with children than unmarried biological fathers. However, exceptions to this general pattern are common. In particular, analyses using FFCW data indicate that social fathers engage in relatively high quality behaviors relative to those of biological fathers. Given that the families in FFCW are more disadvantaged than those in most other national data sources, this may suggest that differences in the behaviors of biological and social fathers are smaller among disadvantaged families than among more advantaged families and/or that disadvantaged mothers who re-partner tend to do so with men who they perceive as investing in their children.

Father involvement among non-incarcerated nonresident biological fathers

For most families, the very nature of nonresident fatherhood means that a father will be less involved in his children’s day-to-day lives than will his resident counterparts, and that staying involved with his children, particularly in terms of face-to-face contact, will require considerably more effort. Clearly, children spend less time with nonresident than resident fathers. Although the majority of children with a nonresident father have some contact with him, recent estimates suggest that a substantial proportion—approximately 40 percent—do not. Furthermore, the remaining 60 percent see their nonresident father an average of only 69 days per year (Child Trends 2002). It is important to note, however, that estimates of whether and how much contact children have with their nonresident fathers range considerably across data sources and reporters (Argys et al. 2007). For example, Argys and colleagues (2007) using data from six large studies report that 45 percent to 62 percent of white fathers and 39 percent to 81 percent of non-white fathers had contact with children 5 years of age or younger during the prior year.¹¹

Current evidence also suggests that patterns of father-child contact tend to be relatively similar across divorced and unmarried nonresident fathers (although children born to unmarried fathers receive slightly less contact) (Argys et al. 2007), that there is considerable variation in involvement patterns within each of these nonresident father types, and that father contact tends to decrease over time for all types of nonresident fathers (Carlson and McLanahan, forthcoming; Lerman and Sorenson 2000). Given that children born to young disadvantaged fathers are disproportionately likely to be born outside of marriage, we focus on studies of father involvement following a nonmarital birth. Estimates from FFCW, for example, suggest that about 5 years after a nonmarital birth, 63 percent of nonresident fathers had seen their child during (approximately) the prior 2 years and 43 percent had seen their child more than once in the past month (Carlson, McLanahan, and Brooks-Gunn 2008). However, whereas most unmarried fathers are substantially involved with their children and their children’s mother around the time of the birth, with many engaged in romantic or cohabiting relationships with the mother, the majority of these relationships dissolve within 5 years. As a result, children born to unmarried parents are likely to be exposed to considerable relationship instability and to decreasing involvement from their nonresident fathers over time (Carlson and McLanahan, forthcoming). Although this pattern is common, it is important to note that there is also considerable variation in involvement trajectories and that many nonresident fathers exhibit steady or even increasing involvement with children over time (Lerman and Sorenson 2000; Ryan, Kalil, and Ziol-Guest 2008).¹²

In addition, nonresident fathers tend to package involvement with formal or informal cash or in-kind support. That is, multiple types of contributions to child rearing are likely to be bundled together such that children who have no contact with their fathers are also less likely to benefit from their fathers’ financial support (Nepomnyaschy 2007; Ryan, Kalil, and Ziol-Guest 2008). Such packaging may reflect both fathers’ motivations toward rearing their children and mothers’ expectations that fathers contribute financially to child rearing as a precondition of access to children (Danziger and Radin 1990; Edin 2000; Edin and Kefalas 2005).

Finally, nonresident father interactions with children tend to be concentrated in “recreational rather than instrumental” activities (Marsiglio, Amato, Day, and Lamb 2000, p. 1184). Much of the time nonresident fathers spend with children appears to be leisure-related; nonresident fathers are less likely than both mothers and resident fathers to provide cognitive support, monitoring, or supervision to children, or to enforce rules (Child Trends 2002; Furstenberg and Cherlin 1991; Hofferth et al. 2002; Yeung et al. 2001). Along these lines, Hofferth and colleagues (2002) report that the majority of nonresident fathers interviewed in the PSID-CDS report having relatively little influence over child-rearing decisions (although a fifth report having a great deal of such influence).

Despite considerable variation by data sources, on the whole, current evidence suggests that: (1) nonresident divorced and never married fathers have relatively similar levels of involvement; (2) there is considerable variation in both involvement levels and changes in involvement over time among nonresident fathers; (3) on average, children’s contact with nonresident fathers tends to decrease over time; and (4) nonresident fathers’ activities with children tend to be more leisure oriented than those of both mothers and resident fathers.

Father involvement among incarcerated nonresident biological father

Because most existing large-scale data sources are household based, incarcerated fathers tend to be excluded from their samples (although mothers are sometimes asked about these men’s whereabouts, characteristics, and behaviors). As such, we know relatively little about these men’s ongoing involvement with their children. Yet, they comprise an important subset of (particularly young and disadvantaged) nonresident fathers. It is well established that lower socioeconomic status and black men are incarcerated at high rates in the U.S. Indeed, whereas approximately 2.3 percent of all U.S. children have an incarcerated parent, black children are seven and a half times more likely than white children to experience parental incarceration (Glaze and Maruschak 2008). And, incarcerated men are likely to be absent from children’s lives for a considerable amount of time. The majority of incarcerated fathers are expected to spend more than 4 years in prison, with average sentence lengths of approximately 6 to 7 years among fathers in state custody and 8 to 9 years among those in federal custody (Mumola 2000).

By definition, the amount of involvement incarcerated men are able to have with their children is limited.¹³ Thus, both the criteria used to assess involvement and the normative expectations of involvement for incarcerated fathers are likely to differ from those for other nonresident fathers for multiple reasons. First, simply by being incarcerated, the father has relatively little choice as to how frequently (and when) he can see his children. Second, children may be prevented from visitation by the custodial mother without the father having much (if any) recourse. For example, a mother who has a poor relationship with her child’s incarcerated father may be unlikely to arrange visitation (Poehlmann 2005). Third, even if the mother is willing to arrange visitation, there are clear costs associated with doing so and some mothers may simply be unable to afford much if any visitation. Fathers are often incarcerated hundreds of miles from their homes (Mumola 2000). Thus visitation may require transportation expenses, lodging, and potentially days off work (which may result in lost wages) (Hairston 2001). As such, the distance between an incarcerated father’s home and the facility in which he is incarcerated is a strong predictor of any face-to-face contact with his children (Hairston 2008).

These factors suggest that contact between incarcerated men and their children will be limited. This conclusion is supported by existing empirical evidence: Mumola (2000), for example, finds that only 40 percent of fathers in state prison had weekly contact (phone, mail, or visits) with their children and that 57 percent had never had a visit from their child(ren) during their incarceration. Glaze and Maruschak (2008) report that approximately 30 percent of incarcerated fathers had weekly contact with their children and an additional 23 percent had contact at least monthly; however, 22 percent had no contact with their children during their current period of incarceration. Furthermore, of those fathers who have contact with their children during incarceration, only about 40 percent receive a face-to-face visit; instead, father-child contact is most commonly accomplished through the mail, with 69 percent of fathers reporting exchanging letters with their children during their sentence (Glaze and Maruschak 2008).

What Are the Major Limitations of Existing Research on Young Disadvantaged Men as Fathers?

As with any body of research, there are limitations to existing studies of father involvement. First and foremost are limitations of the available data. Although national surveys (FFCW, PSID-CDS, NSFG, Add Health) have increasingly included interviews with both resident and nonresident biological (and, less commonly, resident social) fathers in recent years, concerns continue to exist regarding the underrepresentation of men—particularly those who are most likely to be young disadvantaged fathers—due to sampling techniques that exclude active military personnel and institutionalized (including incarcerated) populations, as well as the use of households as sampling units (Carlson and McLanahan, forthcoming; Nelson 2004). Furthermore, concerns about systematic nonresponse, attrition, and missing data with regard to fathers are salient when considering estimates of father involvement, particularly if the least involved and most disadvantaged fathers are most likely to be lost from longitudinal studies. There is also evidence that (particularly low-income) men are unreliable reporters of their own fertility either because they do not know they are fathers or due to intentional misreporting (Nelson 2004).¹⁴ Despite these concerns, however, the field has made great strides toward collecting data directly from men and more datasets are striving to interview both resident and nonresident parents. These efforts should be applauded and expanded upon. In particular, more studies should interview social fathers in addition to biological fathers.

Second, better measurement of father involvement is necessary. Most studies focus on the amount of time children spend with (or with access to) their fathers and sometimes include the types and amounts of activities in which they mutually engage. Recent studies such as FFCW, Add Health, and the PSID-CDS include somewhat more detailed measures of paternal co-parenting, warmth, monitoring, and control. However, information on the “quality” of father-child relationships and interactions continues to lag behind that of mothers and children. Future research would benefit from examining the full range and types of parenting behaviors that have been considered with regard to mothers, including assessments of more “qualitative” aspects of parenting such as nurturance, sensitivity, permissiveness, authoritativeness, authoritarianism, responsiveness, emotional support, and cognitive stimulation. Future research should also investigate both the extent to which fathers indirectly support child rearing by emotionally supporting their children’s mothers (Marsiglio, Day, and Lamb 2000). These aspects of father involvement have been largely overlooked in existing work.

Third, categorizations of family “type” continue to be insufficient as family configurations are likely more fluid than can be accounted for by existing survey data. There is no common definition as to what truly constitutes a cohabiting relationship as opposed to an intensive visiting relationship. And, it is unclear whether it is more important to delineate families by co-residence, marriage, or both. For example, studies often group all nonmarital births together, whether parents are cohabiting, visiting, or not romantically involved. Yet, the quality and quantity of father involvement may differ considerably by these factors. Additional theoretical and qualitative work may help to better understand the full range of relationship trajectories families undergo and how they may be related to ongoing father involvement. Such information is difficult to collect via most surveys (which generally conduct interviews a year or more apart); however, there may be utility in asking sample parents to complete “relationship” calendars detailing their romantic relationships and the amount of time various partners spend in their households over time. Similar work history calendars have been used to study labor force participation and may provide a model for how to collect more detailed and reliable information on the fluidity of parental romantic relationships and living arrangements.

Fourth, studies of the determinants of father involvement continue to be plagued by difficulties—if not impossibilities—in being able to identify causal effects. Although existing work has shed a great deal of light on those factors that are correlated with father involvement, in the absence of strong evidence regarding causal relations caution must be exercised in creating interventions and policies intended to increase father involvement. As such, interventions should be carefully monitored and evaluated for their efficacy in achieving this goal (see Knox, Cowan, and Cowan, this volume).

Finally, there are several important research questions that have received scant attention to date. First, there has been little work on whether all fathers should be encouraged to be involved with their children and, if not, when father involvement should be discouraged (e.g., severe substance abuse or violence) and how. We know little about the circumstances in which father involvement may do children more harm than good. Second, existing research has paid little attention to the effects of fatherhood and various family configurations on men’s well-being. A considerable body of work focuses on mothers’ relationship trajectories and their influence on both mother and child well-being; a parallel literature related to fathers’ relationship trajectories has yet to be produced. Third, single father families have received little attention in the research literature despite being a growing demographic group. Thus, we know little about single fathers’ economic investments in and involvement with their children, nor about those of their children’s nonresident mothers. Fourth, additional research on incarcerated fathers’ relationships with their children is warranted, particularly because most interactions between children and their incarcerated fathers do not appear to be face-to-face. Finally, there has been a notable lack of attention to the “full package” of parenting behaviors to which children are exposed. Most studies focus on the behaviors of a particular parental figure rather than attempting to account for the full range of investments children may simultaneously receive from resident or nonresident mothers, resident or nonresident fathers, and various social parents. While it is important to understand the types of investments made by individual parental figures, what may ultimately matter for children is the quality and quantity of the full set of parenting behaviors to which they are exposed and by whom.

What Can We Conclude about Young Disadvantaged Men as Fathers?

Notwithstanding the limitations described above, the existing literature points to several conclusions regarding young fathers’ involvement with children. First, young fathers are generally more disadvantaged and less involved with children than are older fathers. Second, married biological fathers are more likely to be involved with their children than their unmarried counterparts (and young disadvantaged fathers are less likely to be married than older more advantaged fathers). Third, although much of the existing research indicates that resident biological fathers are more involved with children than resident social fathers, this pattern may be less pronounced among disadvantaged families than among more advantaged families. Thus, future research would benefit from more detailed information on the roles of social fathers in disadvantaged households, including information collected directly from these men. Fourth, considerably more research and policy attention are needed around issues of father involvement for incarcerated men. Finally, nonresident father involvement with children is much more focused on recreational activities than is the involvement of both resident fathers and mothers, suggesting that some families with nonresident fathers may benefit from policies and programs to help these men further develop as supportive parents. To the extent that father involvement benefits children (see Carlson and Magnuson, this volume), it may be important to encourage investments through a range of policy and programmatic options that are described in the subsequent chapters of this volume. These include child support policies (Cancian, Meyer, and Han), policies and programs related to education and transitions to work for young disadvantaged men (Heinrich and Holzer), policies and programs that encourage healthy family relationships and strengthen fatherhood (Knox, Cowan, and Cowan), and income support and labor market policies targeted at young men (Mincy).

Biographies

Lawrence Berger is Associate Professor of Social Work and Faculty Affiliate of the Institute for Research on Poverty at the University of Wisconsin–Madison. His research focuses on the ways in which economic resources, sociodemographic characteristics, and public policies affect parental behaviors and child and family well-being. His participation in this project was supported by NICHD grant K01HD054421.

Callie Langton is a doctoral student in Public Policy at the University of Wisconsin–Madison and the 2009–2010 Institute for Research on Poverty’s Dissertation Fellow. Her research focuses on how family characteristics and public policy affect childhood health and well-being.