Consequences of Fatherhood for Young Men’s Relationships

Abstract

This paper examined how the onset and timing of the transition to fatherhood affects the type and quality of young men’s relationships with partners and parents. Data are drawn from the 1979 National Longitudinal Survey of Youth – Young Adult Survey and included young men (ages 18–31 years old in 2006) who varied on residential status with their children and timing of fatherhood (N = 1,931). Results indicated the effects of fatherhood varied across types of fathers, with residential fathers more likely to be in a committed but less satisfactory relationship regardless of timing of fatherhood. Nonresidential fathers were more likely to have close relationships with their mothers and fathers, but findings varied by timing of fatherhood and gender of parent. Implications of these findings are framed in terms of young men’s developmental readiness for multiple demands of first-time fatherhood.

The transition to parenthood is often experienced as a rewarding but difficult life event characterized by rapid changes in a couple’s relationship and new stresses (Belsky, 1984; Cowan & Cowan, 2000). In becoming a parent, young adults must negotiate extensive personal, familial, social, and professional changes. The transition to parenthood may be particularly difficult for young adults who make this transition earlier than what is normatively expected, as it can interrupt romantic, educational and career-related transitions to adulthood, further taxing limited social and economic resources (Sigle-Rushton, 2005).

Currently, more than one third of all births occur outside of marriage, with even higher proportions occurring among racial and ethnic minorities (Martin et al., 2006). Even so, more than 80% of non-marital couples are romantically involved at the birth of the child, and half of them are cohabiting (McLanahan et al., 2003). Although there is substantial research examining the consequences of early transitions to parenthood for young women, the consequences for young men have been less studied (Moore et al., 1995; Hofferth & Hayes, 1987). Consequences for men may differ in that they depend on the circumstances at the time of the transition.

Using a sample of young men followed from childhood into early adulthood and from which the timing of their first birth is known, this paper examines the consequences of fatherhood for young men’s romantic and familial relationships. We use the last wave of data available in 2006 to compare the types of relationships of young men who became fathers with those who had not. Because the consequences will vary for young men depending on whether the young man lived with the child and mother at birth and the age at which he became a parent, we divide first fatherhood by whether the father was residential or nonresidential and whether his transition was early (less than age 20) or on-time or later (age 20 and older).

Timing, Fatherhood, and Men’s Development

According to individual and family life cycle theories (McGoldrick & Carter, 1999), the transition to young adulthood is characterized by key developmental tasks, including completing schooling, becoming employed and gaining financial independence, developing intimate relationships, and maintaining adult relationships with one’s own parents. Societal roles such as being a parent are accompanied by a set of expectations and demands and necessitate resources for their successful enactment (Cooney, Pedersen, Indelicato, & Palkovitz, 1993). Role transitions across different social settings (e.g., intimate and family relationships, parenthood) require synchronization such that the demands associated with each role are managed effectively (Cooney et al., 1993; Elder, 1994). Developmental theorists posit that successful transition to the next life cycle stage is dependent in large part on achievements and skills acquired in previous stages (McGoldrick & Carter, 1999).

Societal norms guide the timing and order in which transitions into various social and family roles should occur (Hogan & Astone, 1986). As such, role transitions may be age-appropriate or “off-time” (Neugarten, Moore, & Lowe, 1965). Off-time transitions have been linked to negative social and developmental outcomes, and may adversely affect the role performance and timing of subsequent transitions due to inadequate preparation and social support (Hogan & Astone, 1986). Most of the early childbearing literature (e.g., Moore, et al. 1995; Hofferth & Hayes 1987) has defined “too early motherhood” as occurring during the teenage years compared with the twenties and older. Among fathers, early transition to parenthood has also generally been defined as prior to age 20 (Belsky, Jaffee, Caspi, Moffitt, & Silva, 2003; Jaffe, Caspi, & Moffitt, 2001). Even though the age of fatherhood has risen to about age 27, becoming a father during the twenties is considered normal and on-time (U.S. Census Bureau 2009). The majority of young fathers do not live with their children (Hofferth & Goldscheider forthcoming). In contrast, men who become fathers in their 30’s or later have demonstrated greater involvement with their children than even “on-time” fathers (Cooney et al., 1993; Heath, 1994).

Several demographic factors have been associated with the timing of parenthood. Mother’s age at first birth, family socioeconomic status (SES), and number of parental transitions (i.e., the number of times a father has left/re-entered the child’s household) are associated with their child’s early transition to fatherhood across early adulthood (Capaldi & Patterson, 1991; Pears et al., 2005). Early parenthood is more likely among children born to young mothers (Hardy, Astone, Brooks-Gunn, Shapiro, & Miller, 1998; Pears et al., 2005). Youth whose mothers are depressed are likely to transition to parenthood earlier (Moore et al., 1995). Higher SES in the family of origin, particularly parental educational attainment, is positively associated with adolescents’ expectations for and likelihood of later entry into parenthood (Crockett & Bingham, 2000; Pears et al., 2005). Young African Americans enter parenthood at earlier ages than Whites (U.S. Census Bureau 2009).

The premature adoption of roles and responsibilities associated with fatherhood could lead young men toward disadvantage later in adulthood (Sigle-Rushton, 2005). Early-timed transitions may have adverse effects on role performance because they are made prior to one’s developmental readiness and acquisition of skills and resources necessary for successful adaptation to first-time parenting (Hawkins et al., 1995; Pears, Pierce, Kim, Capaldi, & Owen, 2005). Additionally, early role transitions may leave the tasks associated with prior stages uncompleted (e.g., career and educational attainment), and may result in difficulty mastering the tasks associated with that role.

How young adults experience and negotiate the transition to parenthood often sets the stage for future coparental and parent–child relations. The earlier the young adult makes the transition to parenthood the greater the likelihood the couple will experience conflict and instability, as the psychological and interpersonal demands associated with maintaining a happy long-term romantic and coparenting relationship may be beyond the realm of what is developmentally normative (Collins, 2002). Not surprisingly, fathers who delay the transition to parenthood are more likely to be positively involved with their children, perhaps due in part to greater emotional maturity and the achievement of early career goals that allow them to successfully adapt to the multiple demands associated with the parent role (Cooney et al., 1993).

Transition to Fatherhood and Men’s Social Relationships

The challenges of parenthood can also facilitate interpersonal development and psychological growth, particularly among those young adults who feel supported and valued by significant others, such as parents and partners (Cowan & Cowan, 2000). Successful transition to fatherhood is associated with greater emotional maturity and connectedness with others, reflected in men’s shift from a more individual focus toward becoming more caring and concerned about relationships with children, partners and parents (Heath; 1994, Palkovitz, 2002; Settersten, 2007). Although an early transition to fatherhood has been associated with negative consequences for young men, being married (and, to a lesser extent, being in a stable cohabiting relationship) may reduce the negative effects of this off-time transition (Sigle-Rushton, 2005). In contrast, young men who become fathers but fail to establish committed romantic partnerships and who avoid parental responsibilities are more likely to suffer negative consequences for their well-being and development (Nock, 1998).

In general, affective closeness with parents increases when children move from adolescence to young adulthood (Aquilino, 1997; Rossi & Rossi, 1990). The transition to marriage is predictive of positive relationships between young adults and their parents (Aquilino, 1997; Clooney, 1997). However, the same is not the same for parenthood. The transition to parenthood has been associated with negative effects on intergenerational relationships, including higher levels of conflict (Kaufman & Uhlenberg, 1998) and decreased frequency of contact between adult children and their parents (Rossi & Rossi, 1990).

Heterogeneity in the quality of intergenerational relationships may be due, in part, to developmental experiences of the young adult, not only whether a given life course event or transition occurred (e.g., transition to marriage and/or parenthood) but also to whether that transition is normative in timing (Belsky et al., 2003). Examining the effects of onset and timing of life-course events and intergenerational relationships in young adulthood, Belsky et al. (2003) found intergenerational relationships characterized by greater warmth and less conflict when young adults remained childless at age 26. Additionally, young adults who were in serious romantic relationships had closer relationships with parents and more contact with their fathers, but only when the adult children were married (versus cohabitating). These life course transitions explained variation in young adult–parent relationships above and beyond that explained by family relationship quality during early adolescence. Overall, these findings indicate that the more normative the life course of the young adult with regards to intimate relationships and parenthood, the more positive and less conflicted their relationship is with their mother and father.

Goals of the Current Study

The goal of this paper is to examine how the onset and timing of the transition to fatherhood affects the type and quality of young men’s social and familial relationships. We expect that the impact of becoming a parent on young men’s relationships will vary on two dimensions: the young man’s residence with their children at birth and the timing of having their first child. Consistent with individual and family development and life course perspectives, we hypothesize that, for young men, becoming a father will predict greater commitment to one’s partner, whether marriage or cohabitation. Additionally, we hypothesize young men who (a) transition to parenthood “on-time” or later (in their 20’s and beyond); (b) were resident with their child at birth; and (c) are currently in a committed romantic relationship (i.e., marriage) will report the highest levels of happiness and least conflict in their romantic relationships, compared to all other father types. Similarly, we expect that young men who were resident with their child at birth and transitioned to fatherhood on-time or later will report greater levels of closeness with their mothers and fathers, compared to all other father types. We also expect that young men who resided with their child at birth and are currently in a committed romantic relationship with report greater levels of closeness with their mothers and fathers.

Methods

Data

The data come from the adult children born to women who participated in the 1979 National Longitudinal Survey of Youth. The NLSY79 is a nationally representative sample of 14–22 year-olds first surveyed in 1979 (i.e., the G1 generation). The NLSY79 obtained detailed information from respondents, including their demographic, employment, socioeconomic, marital, and family characteristics. Biennially since 1994, children of the original NLSY79 female cohort who were at least age 15 were interviewed as respondents through a separate, but linked study, the Young Adult Survey (i.e., the G2 generation). The Young Adult Survey provides longitudinal data on residential status, relationship with household members, father involvement, changes in family and household composition, closeness to parents, relationship with partners, and the socioeconomic characteristics of the children born to the original 1979 female youth cohort. We use data collected from the Young Adult Survey in 2006 when the children of the NLSY79 women had reached young adulthood.

For this study, we created a dataset with the young adult’s (G2) detailed reports of their relationships with partners, parents, and the community in 2006 together with information reported by the mother (G1) for earlier years. The sample was restricted to all young men born to NLSY79 mothers who were surveyed in 2006, were aged 18 and older, and were not enrolled in high school. The sample excluded 20 men missing residential and marital status at the time of their child’s birth, 24 missing the number of father transitions, and 64 missing data on the dependent variables assessing closeness to mother (62 men) and marital conflict/happiness (2 men). For 734 cases, young men were missing on the variable assessing closeness to father because they did not have a father to serve as a reference for the question. These men were dropped in the multivariate analyses examining father closeness, but were retained in regressions predicting all other outcome variables. The final sample size was 1,931 young men—1,247 of whom were in a partnership at the time they were surveyed.

Measures

Dependent Variables

This study examines the association between fatherhood status and six outcomes: 1) partnership status; 2) partnership type; 3) happiness in current relationship; 4) conflict in current relationship; 5) closeness to mother; and 6) closeness to father.

Relationship with Partner

We created a measure of partnership status in which respondents who were married, cohabiting, or with a regular girlfriend in 2006 were coded one; men who were not in a relationship were assigned zero. In order to examine partnership type and the likelihood of being in one type of relationship over another (or not in a relationship), we created four dichotomous variables indicating whether the respondent reported being married, cohabiting, with a girlfriend, or not in a partnership in 2006.

In addition to assessing relationship status, two measures were used to assess the quality of the partner relationship. Happiness in a relationship is a continuous measure based on responses to the questions about how happy respondents are in their current relationship. Married and cohabiting men were asked the following question, “Would you say your current relationship/marriage is…” Responses were rated on a 3-point scale where 1 indicated very happy and 3 indicated not very happy. Young men with a girlfriend were asked how happy they were with their girlfriend. Responses ranged from 1 (very happy) to 4 (very unhappy). The second and third response categories indicating respondents were “somewhat happy” and “somewhat unhappy” with their girlfriend were combined into one category because very few men reported being “somewhat unhappy” (less than 4 percent of the sample). Thus, a comparable measure of partnership happiness was created for men who were married/cohabiting or who had a girlfriend. Responses were reverse-coded so that higher scores indicated greater happiness. Conflict in a relationship is continuous measure based on responses to two items, “How frequently do you and your girlfriend/partner/spouse have arguments about showing affection toward each other?’ and “How frequently do you and your girlfriend/partner/spouse have arguments about leisure or free time?” Responses to items were based on a 4-point scale with lower scores indicative of less conflict. Respondents not currently in a partnership were coded to missing and dropped from analyses examining happiness and conflict within a relationship.¹

Relationship with Parents

We used men’s reports of closeness to one’s parents as an indicator of having a positive parental relationship. Closeness to one’s parents was measured by asking respondents how close they feel to their mother, father and/or stepfather, respectively. Items were rated on a 4-point Likert scale ranging from 1 to 4, with higher ratings indicative of greater closeness to one’s parent. If the respondent reported on closeness to both a biological father and stepfather, the biological father report was retained and used in the analysis.

Independent and Control Variables

Fatherhood Type

Dimensions of fatherhood status were based on both the timing of first childbearing and on residence at first birth. Annual waves of the Young Adult Survey were checked to determine when a young man resided with his child. Because births occurred between survey waves, residence was defined as of the first wave following childbirth. Residential fathers are young men living with their first child around the time of birth, whether or not the mother of those children was also in residence. The age of the young man and his child in 2006 were used to determine the father’s age when the child was born.

Five variables capture residential status and age of father at first birth simultaneously. They are: 1) young man was a resident father and less than age 20 at the birth of first child (3.8% of the current sample); 2) young man was a resident father and age 20 or older at the birth of first child (11.3%); 3) young man was a nonresident father and less than age 20 at the birth of first child (5.8%); 4) young man was a nonresident father and age 20 or older at the birth of first child (6.1%); and 5) young man was not a father in 2006, the omitted category in multivariate analyses (73.0%).

Background Characteristics and Controls

In all analyses we included a number of demographic and background characteristics of the young adult as control variables. Young adult’s age is a continuous measure based on the age of the respondent in 2006. All background characteristics were measured from birth to age 9. Race/ethnicity was based on the mother’s report from 1979 and is comprised of two dummy-coded indicators: Black and Hispanic. The reference category in multivariate analyses is non-Hispanic White. Maternal education is a continuous measure based on the mother’s highest grade of completed education reported during the time the young adult was age 0–9. The measure is an average value taken across all years for which there were complete data. Maternal depression is a continuous measure based on a 20-item scale from the Center for Epidemiologic Studies Depression Scale (CES-D) that asks about symptoms of depression. Responses to each item ranged from zero to three with three indicating more depressive symptoms; the average across items formed the scale. Mean replacement imputation was used in the case of 78 mothers who were missing information on this scale.

Using records from the household of the young man’s mother we were able to track the presence of the biological/step father while the young man was growing up. Using this information, we created the number of father transitions, which is a continuous measure based on the number of times the young adult’s biological/step father left or came back into the household when the child was aged 0–14. Because some individuals were missing several years of data, we created a transition indicator only if we had a minimum of three years of data. Twenty-four young men missing due to this criterion were dropped. The average income of the father is a continuous variable based on the mother’s total spouse/partner wage-and-salary income in the past calendar year. An average value was created across all years in which data were available when the young adult was aged 0–9. This variable, in 2006 dollars, was scaled by taking the natural log. An average across all spouses’ incomes was substituted for the 3% of the cases missing information on spouse income.

Analysis Plan

Logistic regression is used to examine the relationship between different fatherhood types and the likelihood of being in a partnership in 2006 and multinomial logistic regression is used to examine the likelihood of being in one type of partnership over another (e.g., being married versus having a girlfriend). Ordinary least squares regression is used to examine the degree to which father status and age at first birth are associated with subsequent parental closeness and relationship quality (among partnered men only). We applied sampling weights to all bivariate and multivariate analyses to adjust unweighted cases for the minority over-samples and year-to-year sample attrition (Center for Human Resource Research, 2004).

All regressions were estimated in Mplus using full information maximum likelihood for missing data. Because we weighted the multivariate regressions, we used the MLR estimator in Mplus, which produces maximum likelihood parameter estimates with standard errors and a chi-square test statistic that are robust to non-normality and non-independence of observations (Muthén and Muthén, 2007).

Results

Means, standard deviations, and percentages for demographic characteristics, partner status, and partner and parental closeness are listed in Table 1. Chi-square analyses and analyses of variance were used to compare types of fathers on demographic characteristics and dependent variables. Men who did not have children were the youngest of all men; they had a mother with more years of education and fewer depression symptoms, the income of their father was higher, and they had experienced the fewest number of father transitions (see Table 1). Results of chi-square analyses revealed a significant relationship between nonresidential father status and ethnic group membership, with more nonresidential fathers identifying as Black.

Table 1.

Means and Percentage Distributions of Young Adult and Maternal Characteristics by Young Adult’s Fatherhood Type

			Fatherhood Type
					Resident		Resident		Nonresident		Nonresident
	Total		Not Father		AFB<20		AFB 20+		AFB<20		AFB20+
	Mean	(SD)	Mean	(SD)	Mean	(SD)	Mean	(SD)	Mean	(SD)	Mean	(SD)	F/Chi-sq.	DF
Demographic Characteristics
Age of respondent	23.0	(3.2)	22.1	(3.0)	24.1	(2.9)	26.0	(2.4)	24.7	(2.5)	26.4	(2.2)	138.0 ***	4
Percentage black	18.0	—	14.0	—	17.9	—	15.2	—	49.9	—	41.7	—	124.1 ***	4
Percentage Hispanic	8.8	—	8.4	—	11.0	—	10.3	—	11.8	—	6.7	—	10.1 *	4
Family of Origin Characteristics
Mother’s education (in years)	12.1	(2.0)	12.4	(2.0)	10.6	(1.8)	11.5	(1.8)	10.8	(1.4)	11.2	(1.5)	27.9 ***	4
Mother’s depression score	10.6	(9.8)	10.1	(9.9)	12.5	(8.8)	11.1	(11.1)	13.7	(7.9)	12.2	(9.0)	6.9 ***	4
Income of father	3.1	(2.3)	3.4	(2.6)	2.2	(1.4)	2.3	(1.5)	2.2	(.9)	2.4	(1.4)	15.0 ***	4
Number of father transitions	1.3	(1.8)	1.2	(1.8)	2.2	(2.0)	1.3	(1.6)	1.7	(1.6)	1.4	(1.8)	5.8 ***	4
Partner Status
Percentage Married	16.1	—	7.7	—	34.8	—	61.0	—	17.6	—	20.7	—	330.1 ***	4
Percentage Cohabiting	14.6	—	10.6	—	26.9	—	23.6	—	30.1	—	23.7	—	115.1 ***	4
Percentage Regular girlfriend	34.8	—	41.6	—	10.8	—	4.8	—	26.1	—	32.6	—	126.0 ***	4
Percentage Not in a relationship	34.5	—	40.2	—	27.5	—	10.7	—	26.3	—	23.0	—	90.6 ***	4
Total percentage	100.0		100.0		100.0		100.0		100.0		100.0
Relationship with Partnera
Happiness in relationship	2.7	(.5)	2.7	(.5)	2.6	(.47)	2.7	(.49)	2.6	(.47)	2.6	(.48)	2.2	4
Conflict in relationship	2.0	(.9)	2.0	(.9)	1.9	(.67)	2.0	(.81)	2.2	(.70)	2.2	(.87)	4.5 **	4
Relationship with Parentsb
Closeness with father/stepfatherc	2.7	(1.1)	2.7	(1.1)	2.8	(.98)	2.4	(1.11)	2.4	(.95)	2.7	(1.05)	1.11	4
Closeness with mother	3.2	(.9)	3.2	(.9)	3.0	(1.01)	3.1	(.95)	3.4	(.69)	3.3	(.83)	1.58	4
N	1931		1329		77		202		173		150

^*p < .05

^**p < .01

^***p < .001

AFB = Age at first birth of chld.

^aExcludes young adults not in a partnership. Higher scores indicate greater happiness; lower scored indicate less conflict.

^bHigher scores indicate greater closeness.

^cExcludes 734 young men without a father.

Source: Authors’ calculations from the NLSY and NLSY-Young Adults Survey.

Men who did not have children were the most likely not to be in a relationship (see Table 1). Young fathers who lived with their children at birth were more likely to be married, whereas young fathers who did not live with their children at birth were more likely to be cohabitating with a partner in 2006. Men who were older and resident with the mother at first birth were highly likely to be married. Among those in partnerships, there does not appear to be a significant difference across residential father types in the level of happiness with the relationship. Regardless of their age at first birth, nonresidential fathers appear to report slightly more conflict in their relationships than those who were residential or had not become fathers.

Predictors of Partner Status and Relationship Quality

Results of logistic and OLS regression analyses for predicting partner status, partner type, and relationship happiness and conflict are listed in Tables 2–4. With the exception of residential fathers who were under age 20 at first birth, fathers were more likely than non-fathers to be in partnerships (see Table 2). Residential fathers who were older at the birth of their first child had the greatest odds of being in a relationship; they were 4.57 times as likely as men who were not fathers to be partnered.

Table 2.

Odds Ratios from Logistic Regression of Partner Status and Multinomial Logistic Regression of Partner Type on Young Adult’s Fatherhood Type

	All Men	Partnered Men
	Partner Status	Married vs. Girlfriend	Cohabiting vs. Girlfriend
	Odds Ratio	Odds Ratio	Odds Ratio
Resident father, less than age 20 at first birth	1.47	17.96 ***	7.23 ***
Resident father, aged 20+ at first birth	4.57 ***	34.70 ***	12.03 ***
Nonresident father, less than age 20 at first birth	1.73 *	3.41 **	2.94 ***
Nonresident father, aged 20+ at first birth	1.89 **	1.23	1.52
Not a father (omitted)
Age of young adult	1.05 **	1.48 ***	1.18 ***
Black	0.73 *	0.16 ***	0.58 *
Hispanic	0.71†	0.62	0.80
Maternal education	0.99	1.07	0.88 **
Maternal depression	1.00	1.01	1.01
Income of father	0.99	0.96	0.95
Number of transitions	1.08	1.04	1.07
N	1931	1247	1247

^†p<.10

^*p < .05

^**p < .01

^***p < .001

Note: Partner status is coded so that 1=married, cohabiting, or with girlfriend and 0=not partnered.

Table 4.

Unstandardized Coefficients from OLS Regression of Closeness to Mother and Father on Young Adult’s Fatherhood Type among Partnered Men

	Closeness to Mother				Closeness to Father
	Model 1		Model 2		Model 1		Model 2
	B	(SE)	B	(SE)	B	(SE)	B	(SE)
Resident father, less than age 20 at first birth	−0.10	(.17)	−0.11	(.17)	−0.11	(.27)	0.00	(.29)
Resident father, aged 20+ at first birth	0.01	(.10)	0.01	(.10)	−0.11	(.22)	−0.06	(.24)
Nonresident father, less than age 20 at first birth	0.27	(.11) *	0.26	(.11) *	−0.51	(.29)†	−0.43	(.30)
Nonresident father, aged 20+ at first birth	0.17	(.12)	0.16	(.12)	0.45	(.27)†	0.44	(.27)†
Not a father (omitted)
Age of young adult	−0.01	(.01)	−0.01	(.01)	−0.05	(.02) *	−0.05	(.02)†
Black	0.30	(.07) ***	0.29	(.07) ***	−0.07	(.16)	−0.07	(.16)
Hispanic	0.14	(.08)†	0.14	(.08)†	−0.37	(.14) **	−0.36	(.13) **
Maternal education	−0.03	(.02)†	−0.03	(.02)	−0.07	(.02) **	−0.08	(.02) **
Maternal depression	−0.01	(.00) *	−0.01	(.00) *	−0.01	(.01)	−0.01	(.01)
Income of father	0.03	(.02)†	0.03	(.02)†	0.00	(.02)	0.00	(.02)
Number of transitions	−0.01	(.02)	−0.01	(.02)	−0.15	(.04) ***	−0.14	(.04) ***
Married			−0.01	(.09)			−0.05	(.17)
Cohabiting			0.03	(.08)			−0.30	(.17)†
Intercept	3.71	(.36)	3.69	(.36)	5.21	(.62)	5.14	(.63)
R-squared	0.04		0.04		0.13		0.13
N	1247		1247		754		754

^†p<.10

^*p < .05

^**p < .01

^***p < .001

Note: Sample restricted to men currently in a partnership (i.e., married, cohabiting, or with a girlfriend). Higher scores indicate greater closeness.

Among partnered men, the odds of being married or cohabiting versus having a girlfriend were significantly higher for fathers, regardless of residential status and age at first birth (see Table 2). Being a resident father at the time of the birth, however, was associated with large odds of being in a marital or cohabiting relationship relative to having a girlfriend. Relative to nonfathers, resident fathers who were at least age 20 at first birth were 34.7 times as likely to report being married and 12 times as likely to report cohabiting versus having a girlfriend. Likewise, residential fathers who were young at the birth of their first child (i.e., less than age 20) relative to men who were not fathers were nearly 18 times as likely to be married and 7 times as likely to be cohabiting than to report having a girlfriend. The odds of being married or cohabiting versus having a girlfriend were slightly higher for nonresident fathers who were less than age 20 at first birth relative to non-fathers. There was no statistically significant difference in the odds of being married or cohabiting relative to having a girlfriend between nonresident fathers who were aged 20 or older at first birth and men who were not fathers.

Although becoming a father increased the likelihood of being in a relationship, the effects of this transition on the quality of men’s relationships were not uniformly positive. Among men in a partnership, being a resident father was significantly associated with less happiness with the relationship, but only for residential fathers who were young at first birth (see Model 1 in Table 3). However, after controlling for partnership type (Model 2), being a residential father, regardless of age at first birth, was associated with less happiness. Model 2 also shows that married men reported the highest levels of happiness.

Table 3.

Unstandardized Coefficients from OLS Regression of Happiness with Relationship and Overall Conflict in the Relationship on Young Adult’s Fatherhood Type among Partnered Men

	Happiness				Conflict
	Model 1		Model 2		Model 1		Model 2
	B	(SE)	B	(SE)	B	(SE)	B	(SE)
Resident father, less than age 20 at first birth	−0.17	(.09) *	−0.24	(.09) **	−0.11	(.13)	−0.15	(.13)
Resident father, aged 20+ at first birth	−0.06	(.05)	−0.14	(.05) **	0.12	(.09)	0.08	(.09)
Nonresident father, less than age 20 at first birth	−0.08	(.07)	−0.10	(.07)	0.21	(.11)†	0.18	(.11)†
Nonresident father, aged 20+ at first birth	−0.07	(.07)	−0.08	(.07)	0.26	(.13)†	0.25	(.14)†
Not a father (omitted)
Age of young adult	0.01	(.01)	0.00	(.01)	−0.01	(.01)	−0.01	(.01)
Black	−0.15	(.04) ***	−0.12	(.04) **	0.23	(.07) ***	0.25	(.07) ***
Hispanic	0.04	(.04)	0.05	(.04)	0.03	(.07)	0.04	(.07)
Maternal education	0.00	(.01)	0.00	(.01)	−0.02	(.02)	−0.02	(.02)
Maternal depression	0.00	(.00)	0.00	(.00)	0.00	(.00)	0.00	(.00)
Income of father	0.00	(.01)	0.00	(.01)	0.02	(.02)	0.02	(.02)
Number of transitions	−0.02	(.01)	−0.02	(.01)†	0.01	(.02)	0.01	(.02)
Married			0.19	(.04) ***			0.09	(.09)
Cohabiting			0.05	(.05)			0.10	(.08)
Intercept	2.67	(.18) ***	2.84	(.18) ***	2.29	(.34) ***	2.34	(.34) ***
R-squared	0.03		0.04		0.03		0.03
N	1247		1247		1247		1247

^†p<.10

^*p < .05

^**p < .01

^***p < .001

Note: Sample restricted to men currently in a partnership (i.e., married, cohabiting, or with a girlfriend). Higher scores indicate greater happiness and more conflict.

Being a nonresidential father, regardless of age at first birth, was associated with greater relationship conflict; the results are consistent across Models 1 and 2 but significant at only p<.10. Adding marital and cohabitation status (Model 2) did not change the relationship of residential type with conflict and neither of these variables was significantly related to conflict.

Predictors of Parent Closeness

Among partnered men, being a nonresidential early-timed father was significantly associated with greater closeness to one’s mother (see Table 4). This finding held even after controlling for partnership type. Interestingly, among nonresident fathers, being a young father at first birth was associated with less closeness to one’s father, whereas being an older father at first birth was associated with greater closeness. These results were significant at the p <.10 level. Once we control for the type of partnership, the effect of being an early-timed nonresident father declined. Men who cohabited with a partner were marginally less close to their fathers than those who had a girlfriend.

Discussion

This paper focused on young men’s transitions to fatherhood and how the timing of first fatherhood affected the quality of their relationships with other key adults — their partners and parents — during this stage. We also examined how the impact of the timing of this transition on young men’s relationships varied across those men who lived with their child and those men who did not live with their child at first birth. Although becoming a father was associated with being more likely to be in a committed romantic relationship, this transition had adverse consequences for the quality of these relationships. And while the transition to fatherhood was associated with greater closeness with young men’s mothers and fathers, these effects were stronger for nonresidential men, contrary to our predictions.

For most of the young men in this study, becoming a first-time father increased the likelihood of being in a later romantic relationship. However, residential fathers who transitioned early into fatherhood were no more likely to be in a current partnership than men who were not fathers. It may be that making an early transition to fatherhood failed to adequately prepare some young residential fathers to effectively balance the roles of being both a parent and a partner. As it is normative for even the happiest of couples to struggle to maintain time together as they adapt to new parenthood (Cowan & Cowan, 2000), it may be particularly difficult for young men who undertake the multiple responsibilities of being a resident father so early in their lives. Early fatherhood is likely to lead to less stability in romantic relationships; thus the lower likelihood of being in one at the 2006 survey date.

In addition to predicting partner status, becoming a father also increased the likelihood of being in a committed relationship among men who were residing with the child at birth and currently had a partner, consistent with our predictions. Most of these young men who were in a relationship in 2006 were more likely to be married or living with their partner: 51% of early-timed and 75% of on-time or later residential-at-birth fathers either married or cohabited with the biological mother of their child. These results are consistent with developmental and life course perspectives suggesting that successful transitioning to fatherhood fosters increased connectedness with others and greater positive attention and time to one’s relationships, including those with partners (Heath, 1994; Palkovitz, 2002; Settersten, 2007).

Nonresidential on-time or later fathers, however, were no more likely to be in a committed (versus casual) relationship than men who were not fathers. In contrast to residential fathers, significantly fewer nonresidential fathers were currently residing with their child’s biological mother (13% of early and 17% of on-time or later nonresidential fathers, respectively). In addition, being a young nonresidential father, regardless of timing of fatherhood, was marginally predictive of greater conflict in their romantic relationships. One of the primary tasks in becoming a nonresidential parent is effectively working with the custodial parent to rear their common child across households, so as to maintain involvement in the child’s life (McHale, Kuersten-Hogan, & Rao, 2004). It may be that nonresidential fathers’ efforts to maintain a positive coparenting relationship, while keeping them involved with their child, may compromise their efforts in maintaining satisfactory romantic relationships. This may be an especially difficult balancing act for young nonresidential fathers, who may not have adequately developed skills in managing relationship conflict while learning to be a father. Nonresidential on-time or later fathers may become comparatively less connected and more independent as they get older, and, as a result, more likely to be in casual and more conflicted relationships.

Although residential fathers are more likely to be in a committed relationship, these relationships may not be satisfying. Young residential fathers, regardless of the timing of their first child, were less happy in their relationships. Within the couple relationship, the transition to parenthood has been shown to be associated with longitudinal declines in relationship quality (Cowan & Cowan, 2000; 2002). Some decline is normal. However, young men’s increased attention to child care responsibilities may detract from time spent with their romantic partner and increase stress on the couple relationship over time, particularly for those men who transition earlier into fatherhood (Cowan & Cowan, 2000). However, residential fathers in a marital relationship appear to be happier than those in cohabiting or dating relationships. As marriage is a socially recognized institution, it confers social benefits that cohabiting relationships do not, such as greater connectedness with others and support in maintaining these connections (Sigle-Rushton, 2005; Stanley, Rhoades, & Markman, 2006). It is not surprising, then, that the transition to marriage is associated with greater satisfaction in young men’s relationships.

Contrary to our expectations, being an early-timed partnered nonresident father was associated with greater closeness with one’s mother. The introduction of children as new members of a family system generally results in increased intergenerational contact and support provision to new parents (Cowan & Cowan, 2000). As the men in this sample are the children of young mothers in the original NLSY cohort, it may be that these young men are the children of mothers who also had their child at an early age and, therefore, these mothers may be particularly sympathetic and understanding. Because these mothers were transitioning into grandparenthood as their young sons became early first-time fathers and partners, the higher levels of closeness may be reflective of these mothers actively embracing their role as grandmothers and welcoming new members — both infants and partners — into the family system.

The relationship between the young man’s residential fatherhood type and father closeness differs by age at first birth. Being a young father at first birth was associated with less closeness to one’s father, whereas being an older father at first birth was associated with greater closeness. However, once we accounted for the effects of young men’s relationship status, the effect of being an early-timed nonresident father was no longer significant. Although relationships with parents and extended family tend to intensify following the birth of the child, the quality of the relationships (whether previously close or more distant) does not often change (Cowan & Cowan, 2000; 2002). Perhaps the fathers of these young men, concerned that their sons may repeat their own patterns of transitioning in and out of the home if they are not in a committed partnership, may express their worry and disapproval in ways that feel distancing and unsupportive to their sons. It is also possible that these young men never had a close father-figure. Men who experienced more family structure change while growing up were both more likely to become nonresidential fathers (Hofferth & Goldscheider, forthcoming), and were less close to their fathers (Table 4).

The study had the advantage of using a nationally representative data set that included young men who were and were not fathers of varying residential status, ethnicities and socioeconomic status. It was also able to assess the timing of young fathers’ first birth, and link that event to later relationship outcomes. However, some limitations must be mentioned. The study lacked longitudinal information (e.g., change measures on relationship status and quality) on the course of these young men’s relationships; therefore, we were not able to make causal inferences regarding timing of fatherhood and later romantic relationship status and outcomes. Additionally, as the majority of our sample had not yet become fathers, we were not able to adequately assess covariates that may impact the quality of their romantic and familial relationships (i.e., age and gender of child).

Nonetheless, our findings indicate that the timing of the transition to fatherhood appears to affect romantic and intergenerational relationships in the life course of young men, such that those who achieve normative developmental milestones in the transition to parenthood (i.e., marriage) are more likely to have satisfying relationships. This may be particularly the case for young nonresidential fathers and their parents, as on-time and later nonresidential fathers who fail to pursue committed romantic relationships risk isolation from not only their current partners but their own fathers as well. As current father initiative programs focus on the father’s relationship with the mother, it is recommended that they also attend to the grandparents’ experience of this transition, as their adjustment may help facilitate a positive developmental trajectory for young fathers.