Contents

• Introduction
• Highlights of the Research Brief
• Method
• Detailed Findings
• Conclusions

Introduction

Over the course of their lives, men father children in multiple circumstances. For some, all of their births are in marriage. For some, all are in cohabiting unions. And for some, all are born while the man is not sharing any type of living arrangement with the mother. There are also complex combinations of marital and unwed births for some men.

The purpose of this research brief is to estimate the patterns of men's fertility. By taking a cross-section of American men, it is possible to provide estimates of how many men have each type of fertility pattern. As will be seen, the pattern of men's fertility is strongly associated with other characteristics such as race, education, age at first birth, family income, and labor force attachments.

This is not intended to be a study of completed fertility (i.e., of men's entire fertility histories). Rather, it is a descriptive cross-section of American fathers ages 15 to 44 as of 2002. For most of the analysis, however, fathers who have had at least two births are the focus.

The primary concern in this research brief is how decisions about unions and union formation are reflected in male fertility (fatherhood) patterns. A husband's wife may have a child shortly after a divorce. In this case, the birth is recorded as non-marital. Non-marital births may be to men who are not living with a partner ("single") or to men who are currently living with someone in a domestic cohabiting relationship ("cohabiting"). Births recorded as "marital" are those that occurred while the man was legally married to the mother. Some other possible patterns to consider include the following: A man and his girlfriend have a child while living together. They subsequently marry and have another child. This man has, therefore, both a cohabiting and a marital birth. Another man fathers a child with a woman he does not live with. They subsequently form a cohabiting household and have another child. This man has both a "single" and a "cohabiting" birth. Finally, a married man and his wife have a child. They divorce when she is pregnant with their second child and she subsequently gives birth. He has, then, a marital and a single birth. As will be seen, these and several other patterns of fatherhood are not uncommon in the U.S.

This research brief presents national estimates of marital and unmarried fatherhood patterns for men 15 to 44 years of age in the United States in 2002. The results are taken from Cycle 6 of the National Survey of Family Growth (NSFG). The primary focus is on the variety of circumstances in which men are fathers over their lives. It addresses questions like "How many men have only marital births?" or "How many men have both marital and unmarried births?"

The NSFG is a nationally representative cross-sectional survey that was administered by female interviewers in the selected person's home. The sample used in this research brief consists of 4,928 males 15 to 44 years of age. Details of the sampling and administrative designs related to male respondents are available in Martinez, et. al., 2006.⁽¹⁾ Complete immediate access to documentation (questionnaires, variable definitions, frequencies, codebooks, etc.) is at: http://nsfg.icpsr.umich.edu/cocoon/WebDocs/NSFG/public/index.htm.

[ Go to Contents ]

Highlights of the Research Brief

• Most American fathers (59.6%) have had children only while they were legally married to the mother. Almost a quarter of fathers (24.3%) have had both marital and unwed births. This means that about eight in ten fathers (83.9%) have had at least one marital birth. The remaining 16.1% of men have had only unwed births. The three groups of fathers appear distinct, and subsequent tables in this research brief will show that they differ on a wide range of circumstances. (Figure 1)
• There are distinct racial and ethnic differences in fatherhood patterns. Most (71.6%) non-Hispanic white fathers have had all of their births within marriage, compared to about half (48.9%) of Hispanic fathers and a quarter (25.9%) of black fathers. Hispanic and black fathers are between 3 and 4 times more likely to have births only while cohabitating than are non-Hispanic white fathers (13.9%, 9.6% and 3.5% respectively.) Black fathers are much more likely to have only unwed births (44.9%) than are either Hispanic fathers (24.1%) or White non-Hispanic fathers (6%). (Table 2)
• A father's marital status at the time his first child is born appears highly related to his marital status for subsequent births. Almost all men (92.2%) who were married when their first child was born have had only marital births. And almost half (45.6%) of men whose first birth was unmarried have had only unmarried births. Significantly, this pattern is unrelated to race/ethnicity. Among each racial grouping, no more than 15% of men whose first birth is marital also have had unmarried births. Being married at the time one's first child is born is a strong predictor that all subsequent births are likely to be marital births. This pattern has also been found among mothers.⁽²⁾ (Table 3)
• Marital first births occur much later in men's lives. On average, married men were 25.6 years of age when their first child was born. Men who have had only single births begin fatherhood at 22.5. Other complex types of fatherhood are associated with even younger ages at first birth. (Table 4)
• Later ages at first birth correspond with later ages at first sexual intercourse. On average, men who have had only marital births began having sexual intercourse a year later than men who have had only cohabiting births or both cohabiting and single births (at 16.5 years of age compared to 15.3 and 15.2 years). (Table 4)
• Fatherhood patterns are related to years of completed schooling, family income, health care coverage, and the number of months a man worked in the past year. On average, men with only married births have higher levels on all these measures. (Tables 6, 7, and 8) Current labor force status follows a similar pattern. Only 4.5% of men with all marital births are out of the labor force, that is, neither working nor seeking work. By contrast, 19.6% of men with only cohabiting births are out of the labor force. Lower rates of health insurance coverage among fathers of unwed births may also be related to these labor force patterns.

[ Go to Contents ]

Method

The NSFG is a sample survey of the adult population in America. It is weighted in this analysis to provide national population estimates. The 4,928 men in the Cycle 6 NSFG represent the 61.1 million men 15 to 44 years of age in the household population of the United States in 2002. On average, each man in the survey represents about 12,000 men in the population. Fertility, as used in this research brief, refers solely to children who are the biological offspring of the male reported to the interviewer during the survey. Adopted or foster children are not included. Overall, about 53% of men reported fathering no children, 17% reported fathering one child, 16% reported two, and 15% reported three or more children.

All biological births reported by the male were recorded as one of three types. If the man was legally married to the mother at the time she gave birth, then the birth is recorded as married. If the father was cohabiting with the mother at the time she gave birth, then the birth is recorded as cohabiting. And if the father was neither legally married to, nor living with the mother at the time she gave birth, the birth is recorded as single. Both co-habiting and single births are recorded as unwed births.

[ Go to Contents ]

Detailed Findings

Table 1 provides estimates of male fertility patterns. A little over half (53.2%) of American males (15-44) have had no biological children, while over one in four (28.3%) has had one or more children only while married. Table 1 illustrates the basic point that male fertility is complex and includes combinations of marital statuses and living arrangements. For example, while 3.5% of American men are fathers only to children born while they were not in any type of co-residential relationship ('single'), a similar fraction (3.4%) have had children both while married and while single.

We now shift attention to fathers only. Once fathers are the focus, there are obvious self-selection differences that reflect the choices men make about childbearing. For example, men who have all their children in marriage begin childbearing at later ages. Anything related to the timing of events in a man's life (e.g., completion of schooling), therefore, will likely be reflected in fertility patterns. To deal with this, we focus on men who probably have had most of the children they will father.

Figure 1. Fathers' Fertility Patterns

Figure 1 and subsequent tables and figures are restricted to men with at least two biological children.⁽³⁾Figure 1 combines several of the categories in Table 1 ('married and cohabiting', and 'married and single' are combined as 'married and unwed'). The majority of American fathers (59.6%) have had children only while they were legally married. And a sizable number (24.3%) have had both marital and unwed births. This means that about eight in ten fathers (83.9%) have had at least one marital birth. The remaining 16.1% of fathers have had only unwed births. Less than one in ten fathers (7.2%) have had births only while cohabiting, another 2.6% only while single, and 6.4% both cohabiting and single births.

Table 2 and Figure 2 include fathers of two or more children and show differences in fertility patterns by race and Hispanic identity. The first row of results indicates that black fathers are the least likely to have only marital births. Whereas 25.9% of black fathers have only marital births, 48.9% of Hispanic men and 71.6% of white men have only marital births. Individuals shown as "Other" include Asian, Pacific Islander, Alaska Native and American Indian. It was necessary to combine these groups due to limited sample sizes. About half of fathers (48.3%) in this combined group had only marital births.

Figure2: Father's Fertility Pattern by Race & Hispanic Identity

The last column provides national totals for comparison purposes. This makes it possible to draw simple comparisons to national estimates for any group. For example, as the last column indicates, 7.2% of all fathers have had only cohabiting births. Hispanic and "Other" men, however, have relatively higher rates (13.9% and 14.5%, respectively). White men have relatively lower rates (3.5%).

Comparably large fractions of all groups of men have both married and unwed births. Overall 24.3% of all fathers have had this pattern of fertility. This fraction does not differ dramatically among race/ethnic groups.

Hispanic and "Other" men have higher rates of cohabiting births relative to black and non-Hispanic white fathers. One in five Hispanic fathers (21.6%) and 28.1% of "Other" fathers have had at least one birth while cohabiting (cohabiting only + cohabiting and single). Black fathers have the highest rates of single births. A little more than one-third (35.3%) of black fathers have had at least one birth while single (single + cohabiting and single). White non-Hispanic fathers are most likely to have married births. Most of these fathers (93.9%) have had at least one marital birth (married only + married and unwed).

Figure 3 (also restricted to fathers of 2 or more children) begins to address the question of sequencing of births. By focusing on men's first births, this figure indicates that the large majority (64.6%) occur to men who are legally married to the mother. Recall that, overall (Figure 1) 59.6% of fathers have only marital births. This similarity in marital status at first birth, and overall births to date is not coincidental. Rather, it reflects a fundamental difference in men's fertility as shown in Table 3.

Figure 3. Father Married to Mother at Time of His First Child's Birth

Returning to the "married only" row, the "yes" column shows that almost all fathers (92.2%) whose first birth was a marital birth reported only married births. Another 7.8% have at least one birth while not married. In contrast, the "no" column indicates that of all fathers who were unmarried when his first child was born, about half (54.4%) subsequently have at least one marital birth. These results suggest that a man's first birth is related to the pattern of births he eventually has. If his first child is born in marriage, so are almost all of his children. If his first child is born outside of marriage, only half of such men eventually go on to have at least one marital birth.

Analyses (not shown) bolster this conclusion. Regardless of race, subsequent births following a marital first birth are largely marital births. The percentage of fathers who go on to have unwed births following a marital first birth are as follows: Hispanic (7%), Non-Hispanic White (7.2%), Non-Hispanic black (14.1%) and Non-Hispanic other (12.0%). In sum, despite large racial differences in overall fertility patterns, men whose first birth is marital appear to have a different fertility trajectory than other men regardless of race or ethnicity.

In Table 4 and Figure 4 averages and medians are presented for fathers' current age, his age at the time of his first birth, and his age at first sexual intercourse. On average (Table 4, bottom row), fathers have their first birth at 24.1 (median 23) years. They first have sexual intercourse at 16. These averages, however, differ according to men's fertility patterns. Men who have had only marital births have their first child at an older age (25.6). The youngest age at first birth is found for men who have had cohabiting and single births (21.3) or married and unwed births (21.5).

Figure 4: Fathers Ages at Interview, First Birth, and First Sexual Intercourse (Median Age)

Men with only marital births begin having sex about one year later (average 16.5 years) than men in any other fertility pattern. The youngest age at first intercourse is found for men who have had both cohabiting and single births (15.2) (note: these are averages, and medians show more limited differences).

The men in this analysis were selected because they are fathers with at least two children. The average age of men in this analysis (36.9 years) is about a year older than the average age of all fathers in the NSFG survey. The multiple birth criteria do introduce some selectivity in the sample. Overall, fathers with multiple births are about one year younger at their first birth than all fathers in the NSFG, and this pattern is most pronounced for only-marital and only-cohabiting fathers. While fathers with multiple children are older at the time of interview and younger at the time of first birth, there is little difference between fathers with multiple children and all fathers with respect to the timing of first sexual intercourse. (Analysis not shown.)

Table 5 reports current marital status of fathers by their fatherhood patterns. A man's current marital status is clearly related to his marital status at the time his children were born. Almost two thirds (68.5%) of currently married men have had only marital births. More generally, among those who are, or who have been married (widowed, divorced, or separated) most fathers have had only marital births (of course only 49.5% of currently separated men fit this description). Many men in these groups may have cohabited at an earlier time. In contrast, half (48.5%) of currently cohabiting fathers have had only births while cohabiting. Among fathers who have never married, the largest group (38.2%) have had all their children while not living with the mother (single). A smaller fraction of never-married fathers (30.8%) had all children while cohabiting. The balance (31%) had children in both living arrangements.

Table 6 reports average education (years of schooling), median total income of the father's family (individual income was not gathered in the NSFG) and labor force attachment (percentage who worked all of the past 12 months). The first entry shows that fathers of only marital children have 13.4 years of education on average (compared with 12.79 for the national average of these fathers). The lowest level of educational attainment is recorded for men who reported only cohabiting, or cohabiting and single births (11.4 and 11.6 years, respectively).

Total family income is a problematic measure because NSFG does not record individual incomes.⁽⁴⁾ Accordingly, these median incomes should be understood to pertain to the family in which a father is living, be it with his partner or alone (men in single-person households, therefore, will generally have lower reported incomes). The highest median family income is among fathers who have had all their children while married ($45,000) and the lowest is for men who have had their children while cohabiting only or single only. Family income roughly tracks labor force attachment, with fathers who have only had married births having the highest labor force attachment (86.2% worked all 12 months in the past year) and men who have had only cohabiting births the lowest (52.3%).

Table 7 reports labor force status at the time of the interview. Nationally, 77.2% of fathers aged 15 to 44 were working full time, and 7.7% were working part time. Only 7.2% were out of the labor force (not working, or not looking for work). Fathers with at least one marital birth (married only, or married and unwed) have the highest rates of full-time employment (80.9% and 76.9%, respectively). The lowest rates of full time employment were reported for fathers who had both cohabiting and single births (64.5%). The fraction of men not in the labor force was highest among those who reported only cohabiting births (19.6% not in the labor force) and cohabiting and single births (17.4%).

Table 8 focuses on responses to a question about whether the father had lacked medical coverage for any period in the past year. Nationally, almost a third (29.3%) of all fathers answered Yes to this question. However, over half of fathers with only cohabiting births said they lacked health insurance coverage. Fathers of only married births, in contrast, were the least likely to lack insurance coverage (21.8%). Lower rates of health insurance coverage among fathers of unwed births may also be related to the labor force patterns identified above.

[ Go to Contents ]

Conclusions

Fatherhood occurs in varied contexts for American men. While most fathers limit their childbearing to marriage, many make other choices. The variety of fertility patterns shown in this research brief is related to transitions over the life course (e.g., age at first sex, age at first birth) and involvement in economic and social systems. Generally, fathers who have children born in marriage appear to fare better on most measures of achievement (education, labor force attachment) and access (e.g., health care coverage). Few men appear to transition from married to unmarried births. Once a man has a marital birth, almost all subsequent births are also marital. Among men whose first birth is out of marriage, slightly more than half have a subsequent marital birth. These findings parallel those for women, suggesting that for American parents, having one's first birth within marriage is a strong predictor that all subsequent births are likely to be marital births.

The findings highlight the large demographic, economic, and social differences between the two types of fathers. But they do not imply cause and effect relationships. We cannot know whether successful interventions to encourage unmarried men to delay having a child until after marriage would result in more men reflecting the educational, economic, or social patterns now seen among men who self-select themselves to be married fathers. To the extent that marital births are associated with better outcomes for children and adults, however, this research brief suggests that any efforts to reduce unwed births to men should concentrate on delaying first births until after marriage since these are so strongly predictive of subsequent births.

[ Go to Contents ]

Endnotes

1. Gladys M. Martinez, Anjani Chandra, Joyce C. Abma, Jo Jones, and William D. Mosher. 2006. "Fertility, Contraception, and Fatherhood: Data on Men and Women from Cycle 6 (2002) of the 2002 National Survey of Family Growth. National Center for Health Statistics, Centers for Disease Control and Prevention, Vital and Health Statistics, Series 23, Number 26.

2. These results were obtained by calculating Table 3 separately for each racial/ethnic group. For results about women, see "Historical and Life Course Trajectories of Non-marital Childbearing" (chapter 1) in Out of Wedlock: Causes and Consequences of Non-marital Fertility. Lawrence L. Wu and Barbara Wolfe (Eds.), Russell Sage Foundation, 2001.

3. Among men aged 40 and older in the NSFG (who have presumably had almost all the children they will ever have), 67% have had 2 or fewer children (i.e., 33% have more than 2). By restricting the analysis to men with at least two children, therefore, we account for the large majority of birth histories for men.

4. Median incomes are reported because average (mean) and median values differed. For years of schooling, median and mean values were very similar.