"Just like his dad": family background and residency with children among young adult fathers.

Drawing upon a conceptual framework proposed by Doherty and colleagues (1998), we examine the association betwe0n father residence during adolescence and early paternity and residency for a sample of young adult men. Out data are from the National Survey of Adolescent Males (Waves I, III)--a representative sample of young adult men in the U.S., aged 21 to 27 in 1995. Using multinomial logistic techniques, we find that living with a father during adolescence reduces the odds that young men experience fatherhood at an early age. In addition, living with a biological father as a teen, increases the odds that young adult fathers reside with their children. The association between living with a father as a teen and early paternity and child residency is attenuated by background characteristics and current education and work status.

Keywords: father absence, young fathers, early childbearing 

**********

Increasingly in the U.S., children are living all or part of their lives in households without fathers (Bianchi & Casper, 2000). Non-marital childbearing and divorce have changed the context within which men father and parent children. Although residing with children does not guarantee that men are actively involved fathers, non-residence certainly inhibits contact and father involvement with children (Cooksey & Craig, 1998; Furstenberg, 1988). Father residence, therefore, is an important ingredient facilitating father-child interaction (McLanahan & Sandefur, 1994)

In addition to inhibiting parental involvement, other consequences of father absence have been documented. Paternal absence negatively influences child well-being in part because children raised in female-headed homes have access to fewer economic and parental resources than children raised in two-parent homes (Brien & Willis, 1997; Furstenberg & Weiss, 2000; Nook, 1998). Children with absent fathers have lower educational attainment, leave school earlier (Amato, 1994; Hofferth, Reid, & Mott, 2001;Lang & Zagorsky, 2001; McLanahan & Sandefur, 1994) and are more prone to deviant behaviors such as substance use (Jaffee, Caspi, Moffit, Taylor, & Dickson, 2001) or violence (Mackey & Mackey, 2003).

Children raised in single-parent households are also more likely to have children outside of marriage and at young ages (Dearden, Hale, & Blankson, 1994; Furstenberg & Weiss, 2000; McLanahan & Sandefur, 1994). Early childbearing is related to low educational attainment and other high-risk behaviors such as drug use (Hofferth, Reid, & Mott, 2001; Thornberry, Smith, & Howard, 1997). A recent study by Pears, Pierce, Kim, Capaldi, and Owen (2005) concludes that early childbearing may be normative to the children of young parents. Thus, father absence is likely to perpetuate the generational pattern of early childbearing and the negative outcomes associated with it (McLanahan & Sandefur; Pogarsky, Thornberry, & Lizotte, 2006).

The formation of new families or stepparent relationships is generally easier when children are young relative to during adolescence (Hofferth & Anderson, 2003). In addition, adolescence is a crucial period of development when father absence is especially negative (McLanahan & Sandefur, 1994). Carlson (2006), in a study of youth aged 10 to 14, found greater behavioral problems among teens not living with two-biological parents relative to teens that were. Thus, to further explore the relationship between father absence and early family formation, we analyze data from a U.S. sample of men at a young age. In particular, we model whether or not respondents lived with their father during adolescence, and how father presence/absence is associated with respondents becoming fathers and residing with their own children.

Responsible Fathering

Our analyses draw upon a conceptual framework of responsible fathering proposed by Doherty, Kouneski, and Erikson (1998). Doherty and his colleagues define responsible fathering in terms of legal paternity, father residence or non-residence, economic support, and involvement in childrearing. They propose a conceptual model of influences on responsible fathering that include father factors, mother factors, the co-parental relationship, contextual factors, and child factors. This ecological framework perceives the welfare of fathers, mothers, and children as interrelated and interdependent. Drawing upon this conceptual model, we consider the influence of father factors and contextual factors on paternity and father residence with children.

Father Factors

Father factors identified by Doherty et al. (1998) include a man's relation with his own father. A father's presence or absence influences the models adult children later build upon in forming their own family relationships. Children are influenced by what they observe in their family of origin as they then create their own families as adults (Amato & Booth, 2001). Young men that grow up without a father present are likely to follow a pattern themselves of early childbearing and father absence.

Father absence influences the later family patterns of children by reducing parental monitoring and economic well-being. Children in father absent homes are more likely to live in poverty (Da Vanzo & Rahman, 1993; McLanahan & Sandefur, 1994), and receive less parental monitoring and more negative parenting compared to children in two-parent families (McLanahan & Sandefur). Indeed, McLanahan (2004) argues that over time children from father absent households have been losing more resources in terms of parental time and money than children from father present households.

One consequence of growing up without the involvement and resources of a father is the early initiation of sexual activity and early childbearing (McLanahan & Sandefur, 1994). Pears et al. (2005) found that poor parental monitoring increased the likelihood that young men engaged in antisocial behavior and this was associated with early entry into fatherhood. In addition, they found that low socio-economic status increased the likelihood of early childbearing (Pears, Pierce, Kim, Capaldi, & Owen, 2005).

In addition to early childbearing, growing up without a father is associated with non-residential parenting. For young men, growing-up without a father can reduce their exposure to positive role models of fathering (Pope & Englar-Carlson, 2001). McLanahan and Sandefur (1994) argue that nonresident fathers provide negative role models that encourage sons to be irresponsible fathers--they send the message that children are the responsibility of mothers, not fathers.

This leads us to our hypotheses:

Hypothesis (1): Young men that lived with a father during adolescence are less likely to become fathers at a young age than young men that did not live with a father.

Hypothesis (2): Conditional on having children, young fathers that lived with a father during adolescence are more likely to reside with their children compared to young fathers that did not live with a father during adolescence

In addition to the influence of living with a biological father as an adolescent, we also explore the influence of living with a stepfather. Past studies suggest that not only biological fathers, but also stepfathers in some circumstances influence children positively (Marsiglio, 2004). For example, Hofferth and Anderson (2003) found that children in step or blended families with married parents tared better than children with unmarried parents. White and Gilbreth (200), in a study of adolescents with both a stepfather and biological father, found a positive association between healthy adolescent outcomes and a quality relationship with the stepfather. In contrast, relationships with the nonresidential, biological father had positive effects on child outcomes, but were less consistent.

The presence of a stepfather generally provides increased economic support to children, however, parenting behaviors are less supportive in stepfather families compared to biological, two-parent families (Thomson, Hanson, & McLanahan, 1994). Thus, the lack of economic and parental support that accompanies father absence can be ameliorated to some degree by the presence of a stepfather. Therefore, we also consider the presence of a stepfather during adolescence in our study of early fatherhood.

Contextual Factors

Contextual factors identified by Doherty et al. (1998) in their conceptual model include factors such as employment opportunities, economic factors, and cultural expectations. In particular, low educational attainment is associated with both early fatherhood and father absence (Clarke, Cooksey, & Verropoulou, 1998; Kiernan, 1997; Thornberry, Smith, & Howard, 1997). Also the employment status and educational attainment of fathers is positively related to the amount of economic support they give their children, and economic support is influenced by the amount of contact men have with their children (Rangarajan & Gleason, 1998).

Another contextual factor influencing early parenting and father absence is religious affiliation and religiosity. Previous studies indicate that religious youth are less likely to initiate sexual activity and become parents at a young age relative to non-religious youth (Miller & Moore, 1990). Religiosity and religious affiliation have also been considered in studies of father contact with children (Cooksey & Craig, 1998).

In addition to the above contextual factors, we consider controls for race and ethnicity. Past studies demonstrate that the likelihood that men reside with their children varies by race. Black children are less likely to live with two biological parents than are non-blacks; however, nonresidential black fathers are more likely to remain in contact with their children than are white nonresidential fathers (Eggebeen, Crockett, & Hawkins, 1990; Mott, 1990). Also, Hispanic children are less likely to have absent fathers than are non-Hispanic white children (Clarke, Cooksey, & Verropoulou, 1998).

Informed by the conceptual model proposed by Doherty et al. (1998), we examine the relationship between father presence in adolescence, and the likelihood that young men become fathers and live with their children. We also include controls for contextual and demographic factors. Our focus is on early paternity and residency with children based on a sample of young men in their early twenties.

Data and Methods

Sample

Data from the National Survey of Adolescent Males (NSAM), 1988, 1990/91, and 1995 are used to model how residence with father of origin influences early paternity and residency with children among young adult men (Sonenstein, Pleck, Ku, & Turner, 2000). In 1988 a nationally representative sample of non-institutionalized adolescent males in the U.S. aged 15 to 19 was surveyed. A follow-up survey was given in 1990/91 and again in 1995. Background information on the family structure of the respondent at age 14 from the 1988 survey is used to predict the likelihood of fathering children by 1995. Recent studies underscore the importance of father presence during adolescence to positive child outcomes (Carlson, 2006).

In 1995 the young men were between the ages of 21 and 27. Of the 1,880 young men surveyed in 1988, 1,377 (73% of original sample) were surveyed in the 1995 follow-up. Our analyses are based on this sample of 1,377 young adult men, of which 421 (30%) reported fathering a child by 1995. Sample weights are used to adjust for cases lost between surveys.

Past studies have noted the incomplete reporting of fertility by men in surveys (Bachu, 1996; Rendall, Clarke, Peters, Ranjit, & Verropoulou, 1999). This raises questions about the quality of data available in the National Survey of Adolescent Males. However, as noted by Rendall et al., the more recent the retrospective histories the better the reporting, because the number of marital births from unions dissolved by the survey date is reduced. Since we focus on very young men interviewed every two to four years, we expect, given the recent nature of the retrospective histories in the NSAM, that the reporting of fertility is probably more complete than that of other national surveys of older men. But given the potential for incomplete reporting by nonresidential men, our results may underestimate the number of fathers, and particularly nonresidential fathers.

Model Specifications

Our primary model examines the effect of living with a father at age 14 on the likelihood of fathering a child and living with that child by 1995. We also model the association between other background and current characteristics and paternity and residency.

Our dependent variable measures paternity and residency with children as reported in 1995. It includes three categories: (1) never fathered children, (2) fathered children and resides with at least some of them, and (3) fathered children but does not reside with any of them. Of the residential fathers, only 8% reported not residing with all of their children.

The primary background variable of interest is family structure at age 14 and measures whether or not the respondent lived with a father (see Table 1). Two dichotomous variables measure living with a father at age 14. One indicates whether or not the respondent reported living with his biological or adoptive father, the other indicates whether or not he was living with a stepfather at age 14.

Other background characteristics include whether the mother was a teen parent, religious affiliation and attendance, and age. In addition, we include controls for race and ethnicity. These characteristics are based on information provided in the 1988 survey. A dichotomous variable measures whether or not the respondent's mother was under age twenty when she gave birth to her first child. Religious affiliation is based on the following categories: Protestant, Catholic, other religion, and no religious affiliation. Catholic is the reference category. In addition, a measure indicating whether or not the respondent reported being "born again" or fundamentalist Christian is also included Age is measured in years and indicates the respondent's age in 1988. Religious attendance in 1988 is measured on an interval scale as follows: (1) weekly, (2) once or twice a month, (3) less than monthly, or (4) never. Race and ethnicity is measured by four dummy variables: non-Hispanic Black, non-Hispanic White, Hispanic, and other non-Hispanic race. White is the reference category.

Current characteristics include education and work status as reported in 1995. Education is measured by three dummy variables: less than high school, a high school diploma only (or GED), some college or college graduate. High school only is the reference category. Employment status is a dichotomous variable indicating whether or not the respondent reported being currently employed.

The dependent variable is a trichotomy, therefore, we estimate the models using multinomial regression techniques. The equations express the log odds of one outcome versus another as a linear function of the independent variables. The coefficients represent the increase or decrease in the log odds of having children or not residing with children (versus residing with children) associated with a unit or category change in an independent variable. Odds ratios (exponent of the log odds) are reported in the tables.

Prior to presenting the multivariate results, descriptive statistics are presented in Table 1. The multivariate results are then presented in three steps: (1) the influence of father residence at age 14 is modeled, (2) father residence at 14 and the 1988 background characteristics and controls are modeled, and finally, (3) rather residence, background controls, and current characteristics measured in 1995 are modeled. This allows us to compare the effect of family structure on paternity and residency alone (Model 1), with background controls (Model 2), and finally also with current characteristics (Model 3).

Results

Descriptive statistics are presented in Table 1. The first column gives percentage distributions for the entire sample of young men. The second, third, and fourth columns provide percentages by father residence age 14 (our primary independent variable). About three-fourths of the young men reported living with their father at age 14; and an additional seven percent reported living with a stepfather. Thirty percent of the young men reported that their mother gave birth as a teenager. A much higher percent (42%) of young men not living with a biological of stepfather at age 14 reported that their mother was a teen parent.

Background characteristics measured in 1988 show the sample distribution by religious affiliation. Most of the respondent's reported they were Protestant with less than ten percent indicating they had no religious affiliation as a teen. Differences in religious affiliation by family structure are not large. About 20% of the respondents reported being born again or fundamentalist Christian. Religious attendance was lower among respondents reporting that did not live with a biological or adoptive father at age 14. Race and ethnicity measures indicate that overall 73% of respondents were White and 15% were African American. In contrast, a much higher percentage of respondents not living with a father at age 14 were Black (34%), and fewer were White (54%).

Current characteristics (measured in 1995) indicate that 62% of the respondents reported having some college or more education, and 11% reported not finishing high school. In contrast, among the young men not living with a biological or stepfather at age 14, only 41% reported some college and 22% had not finished high school. Current employment patterns were similar. Overall 85% of the respondents reported that they were currently working in 1995 compared to 78% of those not living with a biological or stepfather at age 14.

In terms of fathering children, 24% of the respondents reported that they had fathered a child by 1995. Seventeen percent reported being fathers and residing with their children. A higher percentage of young men not living with a biological or stepfather at age 14 reported being a father in 1995 (39%); twenty-two percent reported being a father and residing with their children. These bivariate percentages suggest that not living with a father as a teen is associated with early paternity. To further explore these associations, we turn to our multivariate results.

The results of the multivariate analyses are presented in Table 2. For each model, the first column shows the odds of not having children relative to residing with children, and the second column gives the odds of nonresidence with children compared to residence. Thus, odds in the first column indicate paternity, and odds in the second, residency Model 1 includes only the father residence at age 14 variables. The first column shows that living with a biological father at age 14 increases the likelihood that the respondent did not have children relative to living with children he has fathered. Living with a stepfather at age 14 had no influence on early fatherhood. Thus, living with a biological or adoptive father as an adolescent reduced the odds of becoming a father as a young adult.

In contrast, living with a father, even a stepfather, as an adolescent is associated with residing with children versus nonresidency (Model 1, column 2). Young fathers that lived with their biological father as a teen were 2.8 times more likely to reside with their children (1/.35=2.86 reciprocal odds) than not reside with them. Living with a stepfather is also associated with child residency as a young adult. Respondents that lived with a stepfather as a teen were four times more likely to reside with their children than not (1/.25 = 4.00 reciprocal odds).

Once controls for background factors in 1988 are added to the model, the influence of father residence at age 14 continues to influence the likelihood of early fatherhood, but no longer significantly influences residency with children (Model 2). Once background factors are included, respondents that lived with their biological or adoptive father at age 14 were still 56% more likely to be childless relative to be residing with children. However, residency as opposed to nonresidency with children is no longer influenced by living with a biological or stepfather at age 14 once controls for race and ethnicity are included in the analyses.

In terms of the background factors, Protestant religious affiliation is associated with early fatherhood relative to Catholic affiliation, but religious affiliation has no influence on residency versus nonresidency with children. The same pattern is found among fundamentalist Christians. Born again fundamentalism is associated with early fatherhood, but not residency versus nonresidency with children. Religious attendance as a teen has no influence on later paternity or residency.

Controls for age and race and ethnicity indicate that older age is associated with paternity, but not residency. Hispanic young men are 53% less likely to be childless (as opposed to residing with children), relative to White young men. The only background factor in Model 2 predictive of residing with children is race. Black fathers are 7.2 times more likely to not reside with their children, relative to White fathers.

As shown in Model 3, current characteristics are strongly associated with paternity. Higher education reduces the likelihood that young men become fathers early and lower education is associated with early fatherhood. Young men with some college were three times more likely to be childless than those with only a high school degree. Respondents currently employed were 2.5 times more likely to be a father than those not working (1/.39 = 2.56 reciprocal odds). Current education had no significant influence on residency, but young men reporting current work were 4.5 times more likely to be residing with their children than not (1/.22 = 4.54 reciprocal odds).

Discussion

Drawing upon a conceptual model proposed by Doherty, Kouneski, and Erickson (1998), we consider the influence of father factors and contextual factors on responsible fathering in terms of residence with children. We concur with Doherty et al. when they state that "fathering is uniquely sensitive to contextual influences, both interpersonal and environmental" (p. 289).

From our examination of father factors, we find that living with a father as a teen is associated with early fatherhood. Indeed, we find support for our hypothesis that living with a father during the teen years reduces the likelihood that young men become fathers at a young age (hypothesis 1). The influence of living with a father as a teen on early fatherhood is attenuated by the inclusion of current education and work status. In particular, higher education is associated with delayed fatherhood and this is supported by other research (Thornberry, Smith, & Howard, 1997). Pears et al. (2005) also conclude that youth that do not succeed academically experience earlier family transitions and have fewer opportunities for higher education.

In addition to the influence of fathers, we find that teen parenting is intergenerational. Young men were more likely to be fathers if their mothers had given birth as a teen. This concurs with the findings of Pears et al. (2005) regarding the timing of entry into fatherhood among young men. Thus, the family formation patterns of both fathers and mothers appear to independently influence the timing of fatherhood among young men.

If young men do father children, we find some support for our hypothesis that young fathers are more likely to live with their children if they lived with their own father or a stepfather as a teen (hypothesis 2). The association between living with a father or stepfather at age 14 and later child residency is attenuated when measures of race are included. As in other studies, we found that African American fathers were less likely to live with their children than were White fathers. This finding supports previous research regarding racial differences in father residence (Eggebeen, Crockett, & Hawkins, 1990; Mott, 1990).

In general, living with a father during adolescence influences the later paternity and residency of sons in two ways. First, living with a father as a teen inhibits early paternity, and second it encourages residency with children when young men do become fathers. In contrast, living with a mother that began childbearing as a teen is associated with early paternity among young men. As proposed by Doherty et al. (1998), responsible fathering (in terms of later childbearing and residency with children) is sensitive to the interpersonal context (such as own father relationships), as well as the environmental context (i.e., educational and employment opportunities). These findings are important because past studies conclude that early fatherhood is associated with negative outcomes for both children and young fathers (Brown & Eisenberg, 1995; Nock, 1998). If young men do father children, recent research indicates that father involvement matters more in the well-being of children if the father resides with the children (Carlson, 2006).

Family policies, therefore, need to provide greater support to young men raised with absent fathers during adolescence. Positive fathering role models can help them develop needed parenting and relationship skills (Carlson & McLanahan, 2004; Marsiglio, Day, & Lamb, 2000). In particular, young men raised in father absent households that later father children outside of marriage should be targeted. Family education and support is needed to encourage the formation of stable partner and child relationships within the context of residing together, especially when young men have not had positive family role models as teens.

Conclusions drawn from this study, however, need to take into consideration limitations of the data. Father residency during adolescence is measured only at age 14 and thus provides only a cross-sectional view of family structure during the teen years. In addition, our data do not consider the quality of the relationship between fathers and sons. Responses regarding paternity are based on retrospective histories. However, given that the time frame between the 1988 1990/91, and 1995 waves of the National Study of Adolescent Males is relatively short, we are optimistic that paternity reporting is more complete than in other national surveys of adult men (Bachu, 1996; Rendall et al., 1999).

Another study limitation is our inability to separate causation from selection. In particular, the influence of biological fathers versus stepfathers may be due to selectivity. Men who are stepfathers may differ from biological fathers in important ways that influence early fatherhood and residency among the children they live with (Hofferth & Anderson, 2003). Even among biological fathers, there may be selective characteristics that influence residency versus nonresidency with children. Unobserved characteristics such as alcohol use or violence may be biasing our results. We cannot conclude definitively that our results are causal and not due to selection. There is some research, for example McLanahan and Sandefur (1994), that allows for correlation between error terms and still finds significant effects of family structure on child outcomes. These findings bolster the argument that such findings are not due only to selection.

Further research, however, is needed to model not only the likelihood of residing with children, but also the quality of fathering and father involvement across generations. It is apparent that fathering matters, but further understanding is needed to break intergenerational patterns of nonresidential fathering. Fathering research and policies need to consider interventions, particularly to high risk young fathers that will help them effectively parent, otherwise the increasing pattern of single, female-headed households is likely to continue.

References

Amato, P. (1994). Father-child relations, mother-child relations, and offspring psychological well-being in early adulthood. Journal of Marriage and the Family, 56, 1031-1042.

Amato, P., & Booth, A. (2001). The legacy of parents' marital discord: Consequences for children's marital quality. Journal of Personality and Social Psychology, 81,627-638.

Bachu, A. (1996). Fertility of American men. (Population Division Working Paper, No. 14). Washington, DC: U.S. Census Bureau.

Bianchi, S. J., & Casper, L. M. (2000). American families. Population Bulletin, 55, 3-42.

Brien, M. J., & Willis, R. J. (1997). Costs and consequences for the fathers. In R. A. Maynard (Ed.), Kids having kids: Economic costs and social consequences of teen pregnancy. Washington, DC: The Urban Institute Press.

Brown, S. S., & Eisenberg, L. (Eds). (1995). The best intentions: Unintended pregnancy and the well-being of children and families. Washington, DC: National Academy Press.

Carlson, M. J., &. McLanahan, S. S. (2004). Early father involvement in fragile families. In R. D. Day & M. E. Lamb (Eds.), Conceptualizing and measuring father involvement (pp. 241271). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Carlson, M. J. (2006). Family structure, father involvement, and adolescent behavioral outcomes. Journal of Marriage and Family, 68, 137-154.

Clarke, L., Cooksey, E. C., & Verropoulou, G. (1998). Fathers and absent fathers: sociodemographic similarities in Britain and the United States. Demography, 35, 217-228.

Cooksey, E. C., & Craig, P. H. (1998). Parenting from a distance: The effects of paternal characteristics on contact between nonresidential fathers and their children. Demography, 35(2), 187-200.

Da Vanzo, J., & Rahman, M. O. (1993). American families: Trends and correlates. Population Index, 59, 350-386.

Dearden, K., Hale, C., & Blankson, M. (1994). Family structure, function, and the early transi-tion to fatherhood in Great Britain. Journal of Marriage and the Family, 56,844-852.

Doherty, W. J., Kouneski, E. F., & Erikson, M. F. (1998). Responsible fathering: An overview and conceptual framework. Journal of Marriage and the Family, 60,277-292.

Eggebeen, D. J., Crockett, L. J., & Hawkins, A. J. (1990). Patterns of adult male co-residence among young children of adolescent mothers. Family Planning Perspectives, 22(5), 219223.

Furstenberg, F. F., Jr. (1988). Good dads--bad dads: Two faces of fatherhood. In J. L. Palmer & I. V. Sawhill, (Eds.), The changing American family and public policy (pp. 193-218). Washington, DC: The Urban Institute Press.

Furstenberg, F. F., & Weiss, C. (2000.) Intergenerational transmission of fathering roles in at risk families. Marriage and Family Review, 29, 181-200.

Hofferth, S., Reid, L., & Mott, F. L. (2001). The effects of early childbearing on schooling over time. Family Planning Perspectives, 33, 259-267.

Hofferth, S., & Anderson, K. G. (2003). Are all dads equal? Biology versus marriage as a basis for paternal investment. Journal of Marriage and Family, 65,213-232.

Jaffee, S., Caspi, A., Moffit, T., Taylor, A., & Dickson, N. (2001). Predicting early fatherhood and whether young fathers live with their children: Prospective findings and policy reconsiderations. Journal of Child Psychology and Psychiatry, 42,803-815.

Kiernan, K. E. (1997). Becoming a young parent: A longitudinal study of associated factors. The British Journal of Sociology, 48, 406-428.

Lang, K., & Zagorsky, J. L. (2001). Does growing up with a parent absent really hurt? The Journal of Human Resources, 36,253-273.

Mackey, W. C., & Mackey, B. (2003). The presence of fathers in attenuating young male violence: Dad as a social palliative. Marriage and Family Review, 35, 63-75.

Marsiglio, W., Day, R. D., & Lamb, M. E. (2000). Exploring fatherhood diversity: Implications for conceptualizing father involvement. In H. E. Peters, G. W. Peterson, S. Steinmetz, & R. Day (Eds.), Fatherhood: Research, interventions and policies (pp. 269-293). New York: Haworth Press.

Marsiglio, W. (2004). When stepfathers claim stepchildren: A conceptual analysis. Journal of Marriage and Family, 66, 22-39.

McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent. Cambridge, MA: Harvard University Press.

McLanahan, S. (2004). Diverging destinies: How children are faring under the second demographic transition. Demography, 41,607-627.

Miller, B., & Moore, K. A. (1990). Adolescent sexual behavior, pregnancy, and parenting: Research through the 1980s. Journal of Marriage and Family, 52, 1025-1044.

Mott, F. L. (1990). When is a father really gone? Paternal-child contact in father-absent homes. Demography, 27, 499-517.

Nock, S. L. (1998). The consequences of premarital fatherhood. American Sociological Review, 63(2), 250-263.

Pears, K. C., Pierce, S., L., Kim, H. K., Capaldi, D. M., & Owen, L. D. (2005). The timing of entry into fatherhood in young, at-risk men. Journal of Marriage and Family, 67, 429-447.

Pogarsky, G., Thornberry, T. P., & Lizotte, A. J. (2006). Developmental outcomes for children of young mothers. Journal of Marriage and Family, 68, 332-344.

Pope, M., & Englar-Carlson, M. (2001). Fathers and sons: The relationship between violence and masculinity. The Family Journal: Counseling and Therapy for Couples and Families, 9, 367-374.

Rangarajan, A., & Gleason, P. (1998). Young unwed fathers of AFDC children: Do they provide support? Demography, 35, 175-186.

Rendall, M. S., Clarke, L., Peters, H. E., Ranjit, N., & Verropoulou, G. (1999). Incomplete reporting of men's fertility in the United States and Britain: A research note. Demography, 36, 135-144.

Schoen, R., & Standish, N. (2001). The retrenchment of marriage: Results from marital status life tables for the United States, 1995. Population and Development Review, 27, 553-563.

Sonenstein, F. L., Pleck, J. H., Ku, L., & Turner, C. F. (2000). National Survey of Adolescent Males, Old Cohort, Waves 1-3, 1988, 1990-91, 1995 (Data Set P1-P4, Carley, M. L., & Lang, E. L., Archivists). The Urban Institute. Los Altos, CA: Sociometrics Corporation, Data Archive on Adolescent Pregnancy and Pregnancy Prevention.

Thomson, E., Hanson, T. L., & McLanahan, S. S. (1994). Family structure and child well-being: Economic resources vs. parental behaviors. Social Forces, 73, 221-242.

Thornberry, T. P., Smith, C. A., & Howard, G. J. (1997). Risk factors for teenage fatherhood. Journal of Marriage and the Family, 59,505-522.

White, L., & Gilbreth, J. G. (2001). When children have two fathers: Effects of relationships with stepfathers and noncustodial fathers on adolescent outcomes. Journal of Marriage and Family, 63,155-167.

Renata Forste, Department of Sociology, Brigham Young University; Jonathan Jarvis. Department of Sociology, University of Hawaii at Manoa.

Dr. Jay Fagan served as Action Editor for this article.

Correspondence concerning this article should be sent to Renata Forste, Department of Sociology, Brigham Young University, 2032 JFSB, Provo, UT 84602. Electronic mail: [e-mail]

Table 1
Background and Current Characteristics of Young Men by Fatherhood
Status (Percentages)

Characteristics                      Total sample    Lived with
                                                       father
                                                       Age 14
Relationship with father
Lived with father age 14             74
Lived with stepfather age 14          7

Background characteristics (1988)
Mother teen parent                   30              28
Religion
  Protestant                         44              44
  Catholic                           34              35
  Other religion                     13              14
  No religion                         9               8
Born again or fundamentalist         21              22
Religions attendance
  Weekly                             52              56
  Once or twice a month              16              15
  Less than monthly                  16              14
  Never                              16              15
Age
  Mean                               16.9            16.9
  (Std)                              (1.36)          (1.34)
Race and ethnicity
  Black                              15              10
  White                              73              77
  Hispanic                            9              10
  Other race/ethnicity                3               3

Current characteristics (1995)
Education
  Less than high school              11               9
  High school diploma only           28              26
  Some college or more               62              65
Currently working                    85              86
Residence with children
  No children                        76              79
  Non-resident father                 7               4
  Resides with children              17              16
[N]                                  [1377]          [918]

Characteristics                      Lived with      Other family
                                      father           Age 14
                                      Age 14
Relationship with father
Lived with father age 14
Lived with stepfather age 14

Background characteristics (1988)
Mother teen parent                   24              42
Religion
  Protestant                         50              44
  Catholic                           30              29
  Other religion                     11              13
  No religion                        8               14
Born again or fundamentalist         11              19
Religions attendance
  Weekly                             30              45
  Once or twice a month              22              16
  Less than monthly                  31              18
  Never                              17              22
Age
  Mean                               16.6            17.2
  (Std)                              (1.23)          (1.42)
Race and ethnicity
  Black                              12              34
  White                              81              54
  Hispanic                            5               8
  Other race/ethnicity                3               3

Current characteristics (1995)
Education
  Less than high school               4              22
  High school diploma only           32              38
  Some college or more               74              41
Currently working                    84              78
Residence with children
  No children                        79              61
  Non-resident father                 4              17
  Resides with children              17              22
[N]                                 [91]           [368]

Source: National Survey of Adolescent Males (1988, 1995)
Note: Percentages are weighted to reflect population parameters;
N's are unweighted.

Table 2
The Likelihood Young Men Reside with their Children by Background and
Current Characteristics (Multinomial Odds)Characteristics

                                        Model (1)

                                   No kids       Nonresid.
                                     vs.            vs.
                                   Resident       Resident
Relationship with father
Lived with father age 14         1.74 **         0.35 ***
Lived with stepfather age 14     1.62            0.25 *
Background characteristics
Mother teen parent
Religion
Protestant
Catholic
Other religion
No religion
Bom again or fundamentalist
Religious attendance
Age
Race and ethnicity
Black
White
Hispanic
Other race/ethnicity
Current characteristics
Education
Less than high school
High school diploma only
Some college or more
Currently working
-2LL                             30.99          1207.64
Chi-square (df)                  51.70(4)        232.92
Cox & Snell Pseudo R-square        .037            0.156
[M]                              [1377]         [1377]

                                        Model (2)

                                   No kids       Nonresid.
                                     vs.            vs.
                                   Resident       Resident
Relationship with father
1346.12
Lived with father age 14         1.56 *         0.62
Lived with stepfather age 14     1.26           0.40
Background characteristics
Mother teen parent               0.44 ***       1.03
Religion
Protestant                       0.54 **        0.64
Catholic                         1.00           1.00
Other religion                   0.98           0.91
No religion                      0.97           0.81
Born again or fundamentalist     0.67 *         0.62
Religious attendance             0.90           1.01
Age                              0.74 ***       0.86
Race and ethnicity
Black                            1.08           7.18 ***
White                            1.00           1.00
Hispanic                         0.47 **        1.23
Other race/ethnicity             0.80           1.77
Current characteristics
Education
Less than high school
High school diploma only
Some college or more
Currently working
-2LL
Chi-square (df)                  355.71
Cox & Snell Pseudo R-square      0.228
[M]                              [1377]

                                        Model (3)

                                   No kids       Nonresid.
                                     vs.            vs.
                                   Resident       Resident
Relationship with father
Lived with father age 14         1.29           0.66
Lived with stepfather age 14     0.84           0.43
Background characteristics
Mother teen parent               0.60 **        0.94
Religion
Protestant                       0.51 ***       0.60
Catholic                         1.00           1.00
Other religion                   0.92           0.83
No religion                      1.05           0.71
Born again or fundamentalist     0.68 *         0.65
Religious attendance             0.95           1.02
Age                              0.72 ***       0.88
Race and ethnicity
Black                            1.11           5.86 ***
White                            1.00           1.00
Hispanic                         0.48 **        1.15
Other race/ethnicity             0.90           1.54
Current characteristics
Education
Less than high school            0.47 **        0.84
High school diploma only         1.00           1.00
Some college or more             3.11 ***       0.74
Currently working                0.39 ***       0.22 ***
-2LL
Chi-square (df)                  355.71
Cox & Snell Pseudo R-square      0.228
[M                               [13771

Source: National Survey of Adolescent Males (1988, 1995)
Note: Sample statistics are weighted; N's are unweighted. * p<.05
** p<.01 *** p<.001