+1 (208) 254-6996 essayswallet@gmail.com
  

Directions:  This article is relatively complex.  Focus on the introduction and the conclusions the write your journal essay.   I do not expect you to understand all the statistical lingo and tables.  Write a 250-word (you can go slightly over if needed) paper linking this article to family diversity discussed in Chapter 3.

Your journal essay will be graded on the following criteria: 

  1. Does the paper find that non-resident father involvement differs by race?  If so, how?
  2. In your personal experience, what external factors potentially explain the level of non-resident father involvement? How much of this is attributable to race?
  3. Define what the authors mean as social capital.  How important do you feel father involvement is for transmission of social capital across generations?  Are your observations the same as these authors? 
  4. How do the historic factors driving family diversity described in Chapter 3 relate to this article?

Each of these criteria are worth up to two points.  You will earn up to two additional points for college level writing that is free from misspellings and technical grammar errors.

R

acial and Ethnic Diversity in Nonresident Father Involvement

Author(s): Valarie King, Kathleen Mullan Harris and Holly E. Heard

Source: Journal of Marriage and Family , Feb., 2004, Vol. 66, No. 1 (Feb., 2004), pp. 1-21

Published by: National Council on Family Relations

Stable URL: http://www.jstor.com/stable/3599862

REFERENCES Linked references are available on JSTOR for this article: http://www.jstor.com/stable/3599862?seq=1&cid=pdf- reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Wiley and National Council on Family Relations are collaborating with JSTOR to digitize, preserve and extend access to Journal of Marriage and Family

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/termshttp://www.jstor.com/stable/3599862http://www.jstor.com/stable/3599862?seq=1&cid=pdf-reference#references_tab_contentshttp://www.jstor.com/stable/3599862?seq=1&cid=pdf-reference#references_tab_contents

VALARIE KING The Pennsylvania State University

KATHLEEN MULLAN HARRIS The University of North Carolina at Chapel Hill*

HOLLY E. HEARD Rice University**

Racial and Ethnic Diversity in

Nonresident Father Involvement

This study is the first to examine racial and ethnic diversity in nonresident father involvement for multiple domains of father involvement. Data come from a sample of 5,377 adolescents with nonresident fathers in the National Longitudinal Study of Adolescent Health (Add Health). In addition to contact, we explore more intensive types of involvement and qualities of the father- child relationship that tap key dimensions of social capital found to be especially important in promoting child well-being. We find racial/ ethnic differences for many aspects of father involvement, some of which can be explained by structural differences, especially father’s educa- tion and nonmarital childbearing. Our findings suggest that White youth with less educated fathers experience the greatest loss of social capital by living apart from their fathers.

Nearly one half of U.S. children will experience living without a biological father for some period during their childhood (Bianchi, 1990; Bumpass, 1984). The influence of nonresident fathers on off-

spring can be substantial, but it is highly variable

(Amato & Gilbreth, 1999). Patterns of father’s influence vary by race and ethnic diversity, which, in turn, are linked to socioeconomic status

differences as well as family history characteristics (e.g., nonmarital birth, single-parent status). We have developed a conceptual framework that inte- grates these factors to evaluate the extent to which

nonresident fathers affect offspring’s social capital. Research on nonresident parenting has been

criticized for the lack of attention to how involve-

ment by fathers varies by characteristics that represent the increasing diversity of families in the United States (Arditti, 1995). In particular, few studies examine the parenting behavior of nonresident fathers in ethnic minority groups. The few existing national studies are limited to Black-White comparisons regarding father visi- tation, although Hispanics occasionally are included as a third group. Findings are mixed, with inconsistent effects of race on visitation

reported in the literature (Cooksey & Craig, 1998). Some studies report that Black fathers visit more frequently than non-Blacks (e.g., King, 1994; Mott, 1990; Seltzer, 1991), but others find no difference (e.g., Seltzer & Bianchi, 1988). Two studies found that Hispanic fathers were most likely never to visit their children (King; Seltzer & Bianchi, 1988).

The first aim of this study is to contribute to our understanding of nonresident father involve- ment by examining diversity among several racial/ethnic groups for a variety of domains of father

involvement, documenting whether and where differences exist. We explore multiple ways that

Journal of Marriage and Family 66 (February 2004): 1-21 1

Department of Sociology, 211 Oswald Tower, The Pennsylvania State University, University Park, PA 16802 (vking@pop.psu.edu).

*Department of Sociology, CB# 3210, Hamilton Hall, University of North Carolina, Chapel Hill, NC 27599-3210.

**Department of Sociology, MS-28, P. O. Box 1892, Rice University, Houston, TX 77251-1892.

Key Words: children, divorce, ethnicity, fathers, race.

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

2 Journal of Marriage and Family

nonresident fathers are involved in the lives of their

adolescent children-tapping critical dimensions of social capital resources that fathers can provide to children despite living apart from them-which in turn have implications for understanding racial/ ethnic differences in child well-being. The second aim of this study is to ascertain

whether racial and ethnic differences in both the

amount and types of involvement are due mainly to socioeconomic and demographic differences that exist among the groups–differences in edu- cation and income and family history factors such as family structure, nonmarital births, and immi- gration. Aponte (1999) argues that much of what passes for cultural differences among families really stems from these structural factors, parti- cularly socioeconomic status, but that prior studies have not had the appropriate data to demonstrate this empirically. Does the weight of the evidence support this view with regard to nonresident father involvement?

The third aim of this study is to determine whether racial/ethnic differences in nonresident

father involvement vary by the father’s level of education. This possibility is ignored in the lit- erature, but research on men’s family-related attitudes and behaviors suggests important interactions between race/ethnicity and the father’s education (Cazenave & Leon, 1987; Lawson & Thompson, 1999). In particular, the literature suggests that educational differences in father involvement may be most pronounced among Whites and less so among minority fathers.

FATHERS AND SOCIAL CAPITAL

The increased number of fathers living apart from their children has led to concerns about the con-

sequences of this arrangement for child well- being. A number of studies have shown that children who grow up apart from their fathers are disadvantaged in many ways. Compared with children who grow up with both biological parents, children with nonresident fathers are more likely to engage in health-compromising behaviors such as drug and alcohol use, unpro- tected sex, and cigarette smoking; are less likely to graduate from high school and college; are more likely to experience teenage and/or nonmar- ital fertility; have lower levels of psychological well-being; have lower earnings; are more likely to be idle (out of school and out of work); and are

more likely to experience marital instability in adulthood (see review by Amato, 2000).

Adverse consequences can result because non- resident fathers invest less time and money in their children’s welfare than do resident fathers (McLanahan & Sandefur, 1994). Fathers who do not coreside with their children are less able to

transmit crucial economic, parental, and commun- ity resources that are instrumental to children’s healthy development. Income loss is a significant part of this disadvantage. Much of the remaining disadvantage results from a loss of social capital: parental guidance, time, attention, and social con- nections. This loss is linked to children’s cogni- tive and socioemotional development, greater high school dropout, lower college enrollment, and unstable employment and income in early adulthood (e.g., Furstenberg & Hughes, 1995). Further, this loss of social capital is not likely to be compensated for by other adult men in the child’s life. Stepfathers or other male partners of the mother are less likely to be committed to the child’s welfare than are biological fathers, and may compete with the child for the mother’s time and attention (McLanahan & Sandefur).

Social capital comes in two forms, and both are vital to child well-being. One form is inherent in father-child relations as fathers teach, nurture, monitor, and care for their children. In addition to

the time that fathers spend with their children, the

quality of the father-child relationship is a funda- mental source of social capital that is especially important for children’s school attainment and avoidance of risk behaviors (Astone & McLanahan, 1991; Harris & Ryan, 1999). Fathers who develop close affective bonds with children are more effective in monitoring, communicating, and teaching children. Living in different households makes it difficult for fathers to maintain affective bonds and to monitor their

children’s everyday activities (Amato, 1998). A second form of social capital is inherent in the relationship between parents and other indivi- duals and institutions in the community. These relationships provide access to information, assistance, opportunities, and other resources in the community that foster the healthy develop- ment of youth (Coleman, 1988).

Thus, when children live apart from their fathers, they have less access to parental resources in the form of social capital; they lose time and attention from the father; and they have reduced access to the father’s resources in the

community. Although many children experience

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 3

a decline in the quantity and quality of contact with their fathers after divorce, and although children born outside of marriage have even less contact with their nonresident fathers, a signifi- cant number of nonresident fathers still maintain

contact with their children (Amato & Rezac, 1994; Seltzer, 1991). The negative effects of divorce and nonmarital childbearing on children may be partly mitigated to the extent that non- resident fathers provide social capital. We examine racial/ethnic differences in several

key domains of father involvement that represent crucial measures of social capital. We focus pri- marily on the social capital inherent in the father- child relationship, including the time spent with fathers in shared activities, communication, and the quality of the father-child bond. We argue that each of these domains captures different aspects of father involvement, aspects that may represent different sources of social capital that have consequences for child well-being. Research indicates that the type of involvement

matters. Father contact by itself generally is not associated with positive outcomes for children, but more intensive types of involvement and qualities of the relationship (e.g., closeness) promote child well-being (Amato & Gilbreth, 1999; Marsiglio, Amato, Day, & Lamb, 2000). Another relevant dimension is whether fathers practice authoritative parenting. Many fathers take their children to res- taurants and movies, yet fail to engage in author- itative practices, such as helping with homework, talking about problems, or setting limits (Amato, 1987; Furstenberg, Nord, Peterson, & Zill, 1983). Although these other domains generally require a certain level of contact, they also indicate that a higher quality relationship exists between the father and his child. Therefore, research needs to

move beyond the single consideration of contact found in most studies to better understand the

multiple ways that nonresident fathers are involved in their children’s lives. Although this study does not explore all relevant dimensions of father involvement, it advances our understanding by including more diverse measures of fathering than have many prior studies of nonresident fathers (Hawkins & Palkovitz, 1999).

THE INFLUENCE OF DEMOGRAPHIC DIFFERENCES: SOCIOECONOMIC AND

FAMILY HISTORY FACTORS

A primary explanation for racial/ethnic differ- ences in nonresident father involvement is the

large differences in levels of income and educa- tion among U.S. racial/ethnic groups, rather than race or ethnicity per se (Aponte, 1999; Seltzer & Bianchi, 1988). On average, Whites and Asians enjoy the highest levels of income and education, whereas Blacks and Hispanics are most econom- ically disadvantaged (U.S. Bureau of Census, 1997).

Studies of nonresident father involvement con-

sistently demonstrate that income and education are associated with higher levels of father involve- ment (Furstenberg et al., 1983; Seltzer, 1991; Seltzer, Schaeffer, & Charng, 1989). Better edu- cated parents are more likely to conform to social expectations of close ties between parents and children despite separation (Seltzer & Bianchi, 1988). Nonresident parents with greater eco- nomic resources are better able to incur the

costs associated with visitation and participation in activities. In contrast, poverty and unemploy- ment may demoralize fathers who cannot provide financially for their children, leading them to withdraw from participating in their children’s lives (Doherty, Kouneski, & Erickson, 1998; Harris & Marmer, 1996). Mothers also may feel less compelled to foster or even allow the father’ s involvement if he does not contribute financially. Racial/ethnic differences in nonresident father

involvement may largely reflect socioeconomic differences.

Other demographic factors likely to be impor- tant in explaining racial/ethnic differences in non- resident father involvement include household

structure, nonmarital births, and migration. In the United States, Blacks have the largest percen- tage of family households with children that are headed by an unmarried woman, followed by Hispanics, Whites, and Asians (O’Hare, 1992). Significant differences exist in the percentage of births outside of marriage, from a high of 67% in 1990 for Blacks to a low of 5% for Chinese (U.S. National Center for Health Statistics, 1993). Thus, Black children are especially likely to be living in circumstances associated with lower levels of father involvement (e.g., born outside of marriage, living in nonparental households; Seltzer & Bianchi, 1988). In contrast, White chil- dren live in circumstances more conducive to

nonresident father involvement (e.g., born within marriage, living in custodial mother households).

Family experiences of migration, which are particularly salient for Hispanic and Asian families in the United States, negatively affect father contact (Del Carmen & Virgo, 1993;

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

4 Journal of Marriage and Family

Landale & Oropesa, 2001). The process of migra- tion and the differences between parents born and reared in another country and their U.S.-born children can be a source of intergenerational con- flict (Chilman, 1993; Ishii-Kuntz, 2000; McAdoo, 1978). Therefore, Hispanic and Asian adolescents may be more likely to argue with their fathers than are Whites or Blacks. Father-child conflict may be lower for Asians than for Hispanics, however, given the stricter control of emotions and cultural values emphasizing respect and discipline, which characterize Asian families (Demo & Cox, 2000; McAdoo; Wong, 1998). Frequent arguments may indicate that although nonresident fathers have contact with their children, the quality of the rela- tionship is reduced. Although an adolescent’s age is unlikely to

play a major role in explaining racial/ethnic dif- ferences in nonresident father involvement, we consider it because Hispanics and Blacks have a higher birth rate and younger age structure than Whites (U.S. Bureau of Census, 1997). In addi- tion, some studies report that the child’s age is positively related to nonresident father contact, although these comparisons are based on a larger age range than that of the current study (Seltzer, 1991). Certain activities with fathers may decline as the adolescent approaches adulthood and spends more time with peers than parents. How- ever, some topics of conversation, such as dating and romantic partners, may become more fre- quent as such issues are more relevant for older adolescents. Arguments also may become more frequent.

Thus, several important family history and socioeconomic factors may play an important role in explaining racial/ethnic differences in pat- terns of nonresident father involvement. If minor-

ity fathers are less involved, their children may have less access to the father’s social capital resources. However, this assumption of less involvement may not necessarily hold for all types of activities; nonstructural factors, such as cultural values, may influence the types of activ- ities that nonresident fathers engage in with their children. For example, the stronger role of reli- gion in the lives of Black Americans (Julian, McKenry, & McKelvey, 1994) may encourage Black fathers to attend religious services or to become involved in church-related activities

with their children more so than other groups. The strong role of Catholicism in Hispanic families may have a similar influence (Del Carmen & Virgo, 1993).

The role of ethnicity in the content of father- child communication is unclear. Discussions con-

cerning a child’s schooling may be similarly common among all ethnic groups given that chil- dren’s education is reported to be important to many, if not all, minority groups (McAdoo, 1978). Discussions concerning more personal topics, however, such as problems adolescents are having or issues related to dating or their social lives, may be least common in Asian families given their more formal relationships (McAdoo; Wong, 1998).

It should be noted that there can be significant variation within major racial/ethnic groups, such as Hispanics (e.g., Mexicans, Cubans, Puerto Ricans) and Asians (e.g., Chinese, Filipino). National surveys can rarely address this diversity given the small sample sizes of subgroups. We are able to make limited comparisons across several Hispanic and Asian subpopulations to examine such variation in nonresident father involvement.

THE INFLUENCE OF THE ADOLESCENT’S GENDER

Throughout our analyses, we pay particular atten- tion to differences by the adolescent’s gender. Nonresident fathers likely relate to sons and daughters differently, although this may depend on the type of involvement considered. Consis- tent with research on two-parent families finding that fathers are more involved with sons than

with daughters (e.g., Harris & Morgan, 1991) and that boys report closer relationships to their fathers than do girls (e.g., Youniss & Smollar, 1985), some studies of nonresident fathers report that sons enjoy longer and more frequent visits than daughters (e.g., Manning & Smock, 1999), and sons report being closer to their divorced fathers (e.g., King, 2002). However, other studies find little or no association between contact and

the child’s gender (e.g., Cooksey & Craig, 1998). In regard to activities, it is likely that fathers engage in certain activities such as sports more often with their sons than with their daughters, but may not necessarily differ in other activities such as how often they attend church with them. Although father involvement may vary by gen- der, we do not expect the adolescent’s gender to play a major role in explaining racial/ethnic dif- ferences in nonresident father involvement, as the probability of having a son or daughter does not vary by racial/ethnic group. The adolescent’s gender, however, may interact with race/ethnicity

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 5

in differentiating patterns of nonresident father involvement. For example, does the lower level of father contact among Hispanics found in prior research hold for both sons and daughters? Prior research has not examined interactive links

between the child’s gender and race/ethnicity on patterns of nonresident father involvement.

THE INTERACTION OF RACE/ETHNICITY AND

FATHER’S EDUCATION

Although not addressed in the research on non- resident fathers, research on men’s family-related attitudes and behaviors suggests that interactions may exist between race/ethnicity and the father’s level of education (Cazenave & Leon, 1987; Lawson & Thompson, 1999). In particular, the literature suggests that educational differences in father involvement may be most pronounced among Whites and less so among minority fathers.

Prompted by feminism, women’s rising work- force participation, and the profeminist men’s movement, the new fatherhood ideal that has been growing since the 1970s has led to increas- ing demands on fathers to spend more time with their children (Griswold, 1993). This cultural ideal is most fully accepted by the educated middle class and is promoted by social move- ments largely composed of educated Whites (Stag- genborg, 1998). Thus, the model of the involved father may be most salient to highly educated White fathers. The increasing participation of educated White women in the labor force, shoul-

dering the second shift, may lead to greater demands on husbands to become involved in

child care (Griswold). Historically, less educated White working-

class fathers spent less time with children than their middle-class counterparts as they spent longer hours at work and were involved in leisure patterns that more often took them away from the home (Griswold, 1993). The ideals of the new father and feminist demands for egalitarian gen- der relations are less salient in the working class, where gender relations are more traditional and expectations for fathers revolve more around breadwinning than caretaking.

There are reasons to expect that such educa- tional differences are less pronounced for minor- ities, and for Blacks in particular. Because Black women have been involved in the labor force

historically, Black men have been less likely to be the sole breadwinners, with implications for

marital roles and power relations (Landry, 2000; Zinn, 1990). Black marriages are more egalitar- ian than White marriages, with Black husbands more likely than White husbands to share house- work and child care. Greater equality between spouses in Black families has led to greater par- ticipation by Black fathers in child rearing that obscures educational differences (Demo & Cox, 2000; Lawson & Thompson, 1999; McLoyd, Cauce, Takeuchi, & Wilson, 2000; Taylor, 2000). We might therefore expect that less edu- cated Black men would be more involved with

their children than less educated White men, although it is less clear what the pattern will be for more educated men or for other ethnic groups.

Moreover, highly educated Black women may be less likely than their White counterparts to make additional demands on their husbands.

Efforts to change gender relations may be most salient for White women because gender oppres- sion is their main obstacle to status attainment. In

addition to gender oppression, however, Black women must confront race oppression along with their husbands; thus, issues of gender equal- ity are not their only concern (Cazenave & Leon, 1987). Therefore, we hypothesize that a father’s level of education will differentiate levels of non- resident father involvement more so for Whites

than for other minority fathers.

METHOD

Data

Data for this study come from the first wave of the National Longitudinal Study of Adolescent Health (Add Health). The full sample includes 20,745 high school and middle school students in 1995. Particularly important for the purposes of this study, many adolescent subpopulations were oversampled, including Blacks from well- educated families, and Chinese, Cuban, and Puerto Rican adolescents. When appropriate sam- ple weights are used, these data are a nationally representative sample of adolescents in Grades 7-12 in the United States. A parent or parent- figure (usually the resident mother) of each adolescent also was asked to complete a ques- tionnaire (n -17,670; see Bearman, Jones, & Udry, 1997 for a more detailed description of the data collection process).

From the main sample of 20,745 adolescents, we restricted our study to adolescents with valid sample weights who had a nonresident biological

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

6 Journal of Marriage and Family

father whom they knew was still living (n = 5,656). We further restricted the sample to adolescents who were 18 years old or younger (n = 5,480). An additional 103 cases were deleted because of missing information on race/ethnicity (n = 18) or because an adolescent was identified as other (n = 85) than White, Black, Hispanic, or Asian/Pacific Islander. This process yielded a final sample of 5,377 adolescents who had a nonresident father living elsewhere.

Measures

Nonresident father involvement. Measures of non- resident father involvement are based on adoles-

cent reports. Three measures assess contact. Single items indicate how often in the past 12 months (0=not at all, 5=more than once a week) the adolescent has stayed overnight with the father (M= 1.38, SD = 1.70), and how often the adoles- cent talked with the father in person, or on the telephone, or received a letter from him (M = 2.72, SD = 1.79). These two items were com- bined to assess whether the youth had any contact at all (either overnights or in person, phone, letter, or both) with the father (-1 yes, 86%; 0 =no). This allowed us to test whether differences exist

in the type of contact (overnight vs. in person, phone, letter) or the way that contact is measured (different levels of contact vs. any at all).

For adolescents who had any contact with their fathers in the past year (86%), additional single- item questions measured the types of activities they did together and the types of things they talked about in the past 4 weeks (1 =yes, 0 = no). Activities include shopping (27%); play- ing a sport (21%); attending religious services or church-related events (13%); going to a movie, play, museum, concert, or sports event (23%); and working on a school project (11%). These items were summed to form an activities scale

(a = .69; M -.95, SD – 1.30). Communication includes talking about some-

one the adolescent is dating or a party the ado- lescent went to (36%), a personal problem the adolescent is having (25%), the adolescent’s school work or grades (59%), and other things that the adolescent is doing in school (48%). These items were summed to form a talking

scale (a = .73; M- 1.68, SD= 1.41). Prior research (e.g., Amato & Rezac, 1994; Danziger & Radin, 1990) often combines father involve- ment measures (if more than one are even avail- able) into scales, but we explicitly tested for

differences between each of the individual activ-

ity and communication items (in addition to the scales) because racial/ethnic differences could vary for certain activities or discussion issues and not for others. For example, adolescents may be similarly likely to talk about school with their fathers, but racial/ethnic variation may be more likely in the frequency of talking about someone they are dating.

Two measures assess relationship quality. Argue is a single item regarding whether an ado- lescent had a serious argument about his or her behavior with his or her father in the past 4 weeks

(1 = yes, 14%; 0 = no) if they had any contact in the past year. Closeness is measured by a single item asked of all adolescents regarding how close they feel (1 =not close at all, 5 = extremely close) to their nonresident fathers (M=3.07, SD = 1.43).

Race/ethnicity. Four major racial/ethnic groups are examined and measured as a set of dummy variables: non-Hispanic White (n = 2,569), non- Hispanic Black (n = 1,729), non-Hispanic Asian (n= 193), and Hispanic (n=886). Adolescents who indicated that they were Hispanic or Asian were then asked to identify their specific ethnic backgrounds, which permitted a somewhat more refined analysis of father involvement across sev- eral Hispanic and Asian subpopulations. Hispanic ethnic groups include Mexicans (n = 349), Cubans (n = 150), Puerto Ricans (n = 203), Cen- tral/South Americans (n=167), and all other Hispanics (n 17). Asian subgroups include Chinese (n=43), Filipinos (n=94), and all other Asians (n = 56).

Socioeconomic status and family history factors. Mothers’ education and nonresident fathers’ edu-

cation are each represented as a set of dummy variables based on adolescent reports: less than high school (17% of mothers and 17% of fathers), high school graduate (36% and 36%), some col- lege (20% and 13%), college graduate (20% and 19%), and missing (7% and 15%). We sub- stituted the mother’s own report of her educa- tion if she filled out the parent survey and the adolescent report was missing. Household income is a set of dummy variables based on parent or parent-figure reports of total income from all sources in 1994 and pertains to the household the adolescent resides in: <$16,000 (24%); $16,000-32,999 (23%); $33,000-$60,000 (21%); >$60,000 (9%); and missing (23%).

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 7

Family structure is a set of dummy variables based on the adolescent’s household roster: bio-

logical mother and stepfather (34%), single mother (52%), and all others (14%). Marital birth is a set of dummy variables based on the marital and fertility histories in the parent survey that indicate whether the adolescent was born

within marriage (45%), outside of marriage (30%), or unknown (26%). This could not be accurately determined in a fair number of cases, particularly for adolescents living in nonparental households. Father is U.S. born is a set of dummy variables based on adolescent reports: yes (88%), no (11%), or missing (1%).

A category indicating missing responses (including adolescent responses of don’t know) was created for several of the above variables to

prevent further curtailment of sample size by allowing for the inclusion of the missing cases and correcting for any bias introduced by missing information. For example, an adolescent who reported little or no contact with his or her non- resident father was, not surprisingly, more likely to be unaware of his or her father’s education or

birthplace. Excluding such cases could lead to the overrepresentation of more involved fathers.

Adolescent’s age is a continuous variable ran- ging from 11 to 18 years (M=- 15.49, SD=- 1.74). Adolescent’s gender is a dichotomous variable (1 = boy, 47%; 0 = girl).

Analytic Strategy

We begin by examining patterns and variation in nonresident father involvement by the four major racial/ethnic groups under study. Then we reana- lyze our models of nonresident father involve- ment by racial/ethnic group in a multivariate framework, controlling for socioeconomic and family history factors. The importance of the adolescent’s gender also is considered. We then explore whether racial/ethnic group differences in nonresident father involvement are moderated

by the father’s level of education. Finally, we examine variation in nonresident father involve-

ment within Hispanic and Asian subpopulations. Ordinary least squares regression is employed

for the continuous involvement measures, and logistic regression is employed for the dichoto- mous measures. Analyses are conducted using the overall sample weight to correct for the dif- ferential probabilities of sample selection result- ing from factors such as the oversampling of minority groups. The SVY procedures in Stata

(Stata Corp., 1999) are used to adjust the standard errors of the model estimates for the clustered

and stratified design of Add Health (Chantala & Tabor, 1999; Johnson & Elliott, 1998). The racial/ethnic groups are represented as a series of dummy variables; differences between them are tested by changing the omitted category and reestimating the models for all comparisons.

RESULTS

Nonresident Father Involvement

by Race/Ethnicity

For illustrative purposes, we transformed the regression coefficients into predicted means (OLS models) and predicted probabilities (logis- tic models), and for each dependent variable we report where significant differences (at p < .05 and p <.01) are observed (Tables 1 and 2). Pat- terns of nonresident father involvement signifi- cantly differ by race/ethnicity for many of the measures under consideration.

White adolescents have significantly higher levels of both overnight visits and in-person, phone, or letter contact with their nonresident fathers compared to Black and Hispanic adoles- cents (Table 1). If we focus on whether children have any contact at all, however, a slightly dif- ferent picture emerges, indicating that how we measure contact has implications for exploring racial differences in involvement. Measured in

this way, Black-White differences disappear. Thus, Blacks are not more likely to have contact with their fathers as some prior research has reported (King, 1994; Mott, 1990; Seltzer, 1991). Consistent with other research, however (King; Seltzer & Bianchi, 1988), Hispanic adolescents stand out as being most likely not to have any contact (77%, compared with 87% of Whites, 86% of Blacks, and 92% of Asians).

Given that adolescents had contact with their

fathers in the past year, are racial/ethnic differ- ences apparent in the types of activities they do together? The general pattern is not one of some racial/ethnic groups standing out as being signifi- cantly higher or lower on most activities than other groups. Instead, particular groups are sig- nificantly higher or lower on a particular activity, and these patterns vary with the activity in ques- tion. White adolescents stand out on their com-

paratively high incidence of playing sports with their fathers (24%, vs. 17% of Blacks and 15% of Hispanics), but Black adolescents stand out on

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

8 Journal of Marriage and Family

TABLE 1. CONTACT AND ACTIVITIES BETWEEN ADOLESCENTS AND THEIR NONRESIDENT FATHERS PREDICTED BY RACE/

ETHNICITY (PREDICTED MEAN LEVELSa AND PREDICTED PROBABILITIESb, WEIGHTED)

Contact Activities

Activities

Overnighta Talka Any Contactb Shoppingb Sportsb Religiousb Movieb Projectb Scalea

White 1.55 2.85 .87 .29 .24 .11 .25 .10 .98

Black 1.09 2.55 .86 .24 .17 .17 .20 .12 .90

Hispanic 1.06 2.36 .77 .26 .15 .12 .23 .16 .92 Asian 1.44 2.78 .92 .30 .20 .17 .18 .11 .95

Fe 11.70** 10.33** 9.18** 1.61 4.67** 6.50** 1.57 3.00* 0.37

Differencesd W > B,H W>B,H H<W,B,A W>B, H B>W none H>W none p <.01

Differencesd W>B B>H p<.05

ne 5359 5364 5365 4553 4556 4555 4555 4553 4550

aMean levels derived from ordinary least squares regression. bPredicted probabilities derived from logistic regression. CF=F statistic for the overall model. dSignificant differences (at p <.01 and p <.05) between racial/ethnic groups summarized. W = White; B = Black; H = Hispanic; A = Asian. en = unweighted n. *p < .05. **p < .01.

their high rate of attending religious services with their fathers (17% vs. 11% of Whites), and His- panic adolescents stand out with the highest per- centage who have worked on a school project (16% vs. 10% of Whites). It is perhaps not surprising, then, that when all

of these activities are measured in an additive

scale, no significant differences emerge. High ratings on one activity are canceled out by lower ratings on another. Had we examined only the activities scale and not the patterns for the individual items, we would have reached a very different conclusion-namely, that signifi- cant racial/ethnic differences do not exist.

Instead, we see that although the overall level of

involvement in activities is similar, the types of activities that adolescents engage in with their nonresident fathers do vary by racial/ethnic group.

The items measuring what adolescents talk about with their fathers showed few differences

(Table 2), although talking about dating or a party is a significantly greater topic of conversa- tion for Whites (38%) and Hispanics (40%), com- pared to Blacks (31%) and especially Asians (17%). Asian and Hispanic adolescents were sig- nificantly more likely to report that they had a serious argument about their behavior with their nonresident fathers than were White or Black

adolescents (Asians = 26%, Hispanics = 19%,

TABLE 2. COMMUNICATION AND RELATIONSHIP QUALITY BETWEEN ADOLESCENTS AND THEIR NONRESIDENT FATHERS PREDICTED BY RACE/ETHNICITY (PREDICTED MEAN LEVELSa AND PREDICTED PROBABILITIESb, WEIGHTED)

Communication Relationship Quality

Datingb Problemsb Gradesb Schoolb Talking Scale Argueb Closenessa

White .38 .24 .58 .49 1.70 .14 3.04 Black .31 .24 .60 .47 1.63 .14 3.16

Hispanic .40 .29 .60 .45 1.74 .19 3.02 Asian .17 .20 .55 .43 1.37 .26 3.07

FP 5.10** 1.52 0.32 0.73 1.31 3.85* 1.23

Differencesd p < .01 W, H > B, A none none none none none Differencesd p < .05 H, A > W, B nC 4555 4555 4552 4548 4547 4555 5361

aMean levels derived from ordinary least squares regression. bPredicted probabilities derived from logistic regression. cF=F statistic for the overall model. dSignificant differences (at p < .01 and p < .05) between racial/ethnic groups summarized. W = White; B = Black; H = Hispanic; A = Asian. en = unweighted n.

*p <.05. **p <.01.

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 9

Blacks = 14%, Whites = 14%). Despite differ- ences in contact and arguments, levels of reported closeness to fathers are fairly similar across many racial/ethnic groups, a finding that is consistent with prior literature that reports a loose link between levels of father contact and

ratings of closeness by children (e.g., Furstenberg & Nord, 1985).

The demographic-composition hypothesis that predicts least involvement by minority fathers is not fully supported. Rather, minority fathers stand out as being more involved in certain activ- ities, whereas White fathers are more involved in other activities. White fathers have the highest and Hispanic fathers the lowest levels of contact with noncustodial children, but minority fathers are involved in other ways that can be beneficial for their children. In particular, minority fathers are engaged in the types of activities that provide crucial sources of social capital that are more strongly associated with child well-being, such as working on school projects and attending reli- gious services.

Multivariate Models

In preliminary analyses, we examined how the racial/ethnic groups differed on socioeconomic status and family history factors using a cross- tabular analysis with tests of independence based on the x2 value (mean levels were compared for the adolescent’s age). The findings for our sample of adolescents with nonresident fathers are consistent with general racial/ethnic trends in the larger population (results not shown; all results referred to and not shown are available

from the first author upon request). For example, Whites and Asians have higher levels of educa- tion and income compared with Blacks and His- panics. White adolescents are most likely to be living with a mother and stepfather, whereas Blacks are most likely to be living with a single mother. Other family types are more common among Asians and Blacks than Whites or His- panics. White adolescents are most likely to have been born within marriage, whereas Black ado- lescents are least likely. Hispanic and Asian fathers are significantly less likely to have been born in the United States. Hispanic adolescents are somewhat younger but not significantly so, reflecting the younger age distribution of this population. All of these differences, except for the adolescent’s age and gender, are significant (p <.001), suggesting that these factors could

account for the differences reported in Table 1. In addition, these groups differ in their propensity to have missing information on some of these characteristics, indicating the importance of modeling these differences instead of simply deleting such cases from the sample. For exam- ple, Hispanic adolescents are least likely to know their nonresident fathers’ level of education, which is probably related to their being more likely to not have any contact with them.

However, controlling for socioeconomic and family history factors has only a modest effect on the associations between race/ethnicity and our measures of father involvement. Tables 3

and 4 report the race/ethnicity regression coeffi- cients from the final multivariate model estimat-

ing our measures of father involvement, but intermediate models also were tested to ascertain

the factors that had the most influence (results not shown). Most of the differences between Tables I and 3 and between Tables 2 and 4 result from

controlling for socioeconomic status factors, the father’s education in particular. Controlling for nonmarital births sometimes further reduces or increases racial/ethnic differences.

White adolescents still have the highest levels of overnight visits and levels of in-person, phone, or letter contact, although differences are reduced (Table 3). The higher incidence of playing sports among Whites is still evident in comparison to Hispanics, but is no longer significantly different from Blacks. Blacks now also exhibit a higher frequency of working on school projects with their fathers compared to Whites.

Some racial/ethnic differences actually become stronger in the communication domain (Table 4). Hispanics are now significantly more likely to talk about dating and about problems compared to Whites and score highest on the overall talking scale. The addition of controls reduces differences in adolescent reports of hav- ing a serious argument with his or her nonresi- dent father, which further analysis shows is related to foreign-birth status of the father. Father- adolescent arguments are more likely when the father is foreign born, which is more common for Asian and Hispanic youth. The addition of controls increases differences in ratings of closeness to fathers with Black adolescents, reporting that they are closer than White adolescents.

Thus, in some cases, the lower socioeconomic status of Blacks and Hispanics (especially father’s education) and the greater likelihood of nonmarital childbearing (especially among

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

10 Journal of Marriage and Family

TABLE 3. CONTACT AND ACTIVITIES BETWEEN ADOLESCENTS AND THEIR NONRESIDENT FATHERS PREDICTED BY

RACE/ETHNICITY, NET OF CONTROLSa (UNSTANDARDIZED REGRESSION COEFFICIENTS, WEIGHTED)

Contact Activities

Overnightb Talkb Any Contactc Shoppingc Sportsc Religiousc Moviec Projectc Activities Scaleb

Blackd -.27** -.09 .12 -.11 -.18 .58** -.08 .36* .04 Hispanicd -.20 -.25* -.49** -.08 -.44* .06 .06 .64** .01 Asiand .15 -.08 .71 .16 -.07 .30 -.42 .05 -.02

P 15.96** 19.45** 7.69** 6.03** 8.61** 3.74** 6.85** 4.04** 8.80**

F groupf 3.74* 1.60 4.90** 0.41 1.81 6.98** 0.79 4.30** 0.09 Differencesg p < .01 W > B H < W, B, A none B>W none H>W none Differencesg p <.05 W>H W>H B>H B>W h 5359 5364 5365 4553 4556 4555 4555 4553 4550

aAll models include controls for father’s education, mother’s education, household income, family structure, marital birth, father is U.S. born, and adolescent’s age and gender. bOrdinary least squares regression. cLogistic regression. dReference category: White. eF= F statistic for the overall model. fF = F statistic for the racial/ethnic group effect; Ho: all groups are equal. gSignificant differences (at p < .01 and p < .05) between racial/ethnic groups summarized. W= White; B = Black; H = Hispanic; A = Asian. hn = unweighted n. *p < .05. **p < .01.

Blacks) depresses levels of involvement, and once these differences are controlled, overall levels of involvement for minority fathers increase. In other cases, the controls have little effect, indicating that demographic differences explain only part of the observed racial/ethnic differences.

We do not report on the significance of each of the control variables because they are not a major focus of this study. These results, however, are consistent with prior research and are easily summarized (see also Table 7). Higher levels of father’s education are positively related to father involvement in all domains except argu-

ments. Adolescents who do not know their fathers’

level of education report lower levels of involve- ment. Income is less important than education, but it is positively related to father involvement when it is significant (e.g, talking scale). After other controls, the level of mother’s education has few significant effects. Older adolescents report lower levels of father involvement compared to younger adolescents except for a higher frequency for some of the talking items (e.g., dating, problems). Few consistent differences exist between adolescents in

single-mother families and stepfather families. Youth with U.S.-born fathers are less likely to argue with their nonresident fathers than are

TABLE 4. COMMUNICATION AND RELATIONSHIP QUALITY BETWEEN ADOLESCENTS AND THEIR NONRESIDENT FATHERS PREDICTED BY RACE/ETHNICITY, NET OF Controlsa (UNSTANDARDIZED REGRESSION COEFFICIENTS, WEIGHTED)

Communication Relationship Quality

Dating’ Problemsc Gradesc Schoolc Talking Scaleb Arguec Closenessb

Blackd -.06 .21 .25* .08 .10 .07 .28** Hispanicd .53** .59** .23 -.03 .27* .14 .12 Asiand -.94* -.03 -.10 -.27 -.24 .53 .09 Fe 6.04** 3.75** 4.24** 5.13** 6.78** 2.64** 12.21** F groupf 6.44** 4.07** 1.84 0.43 3.06* 0.85 6.02** Differencesg p < .01 H > W, B, A H>W none none B>W

Differencesg p <.05 W > A B>W H > W, A hn 4555 4555 4552 4548 4547 4555 5361

aAll models include controls for father’s education, mother’s education, household income, family structure, marital birth, father is U.S. born, and adolescent’s age and gender. bOrdinary least squares regression. cLogistic regression.dReference category: White. eF = F statistic for the overall model. fF = F statistic for the racial/ethnic group effect; Ho: all groups are equal. gSignificant differences (at p < .01 and p < .05) between racial/ethnic groups summarized. W = White; B = Black; H = Hispanic; A = Asian. hn = unweighted n. *p < .05. **p < .01.

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 11

youth with foreign-born fathers. Although similar in many respects, boys are closer to their nonresident fathers than are girls, and they report a greater frequency of overnight visits, play- ing sports, and going to movies or sports events.

The Moderating Influence of Gender

We examined whether the racial/ethnic differ-

ences reported in Tables 3 and 4 were moderated by the adolescent’s gender by adding interaction terms between racial/ethnic group and adolescent gender, and controlling for father’s education, mother’s education, household income, family structure, marital birth, whether father is foreign born, and child’s age (results not shown). We also tested whether we could reject the null hypothesis that the set of interaction terms was equal to zero (see F int statistic reported in the Appendix). In most cases, the null hypothesis was not rejected, indicating the models were similar for boys and girls. However, significant interac- tions existed in 4 of the 16 models: in-person, phone, or letter contact; any contact; talking about dating or a party; and closeness. For ease of presentation and illustration, we created a new set of dummy variables representing eight groups comprised by the cross-classification of race/ ethnicity by adolescent’s gender. These eight groups were compared net of controls in a series of regression models. Group differences were tested by changing the omitted category and reestimating the models for all comparisons. This procedure allowed us to test for an overall group effect (whether to reject the null hypothesis that all the groups were equal). Finally, we report the significance of the overall F value for the full model. The results for the four significant inter- actions are summarized in the Appendix.

The higher levels of in-person, phone, or letter contact among White youth and their nonresident fathers is particularly pronounced among White boys, who enjoy significantly greater contact than White girls. The finding that Hispanic youth are least likely to have had any contact with their fathers in the past year is a consequence of low contact among Hispanic girls; Hispanic boys report levels of contact that are not significantly lower than most other youth. In addition, Asian boys are most likely to report having had any contact.

The finding that Hispanics are most likely to talk about dating holds for both Hispanic boys and girls. Significant gender differences in the frequency of talking about dating exist, however,

for Whites (girls talk about this more than boys) and Asians (boys talk about this more than girls; Asian girls are least likely to talk to their fathers about dating). Black youth report being closest to their fathers. Although boys are significantly clo- ser to their nonresident fathers than girls are (overall and within racial/ethnic groups), Black girls are as close to their fathers as White, His- panic, and Asian boys are.

The Moderating Influence of Father’s Education

To test whether race/ethnic effects vary by the father’s education, we dichotomized father’s edu-

cation into high school or less and some college or more. We first examined models similar to

those in Tables 3 and 4, adding interaction terms between racial/ethnic group and father’s education, controlling for mother’s education, household income, family structure, marital birth, whether father is foreign born, and child’s age and gender (results not shown). As we did for the analyses of adolescent gender, we tested whether we could reject the null hypothesis that the set of interaction terms was equal to zero (see F int statistic reported in Tables 5 and 6). In many cases, the null hypothesis was rejected, indicating an improved model fit by including the interactions.

A new set of dummy variables was created representing eight groups comprised by the cross-classification of race/ethnicity by father’s education. These groups were compared in a regression framework, net of controls (Tables 5 and 6). Group differences were tested by chang- ing the omitted category and reestimating the models for all comparisons. In addition to the overall F value for the full model, we tested for an overall group effect (whether to reject the null hypothesis that all of the groups were equal).

White fathers fall at the two extremes, with the

lowest levels of father involvement reported for White fathers with a high school education or less and the highest levels of involvement often reported for White fathers with higher levels of education. In contrast, minority fathers exhibit fewer significant differences by education, with levels of involvement that are often between the

two extremes exhibited by White fathers. For example, White adolescents with highly educated fathers enjoy significantly higher levels of over- night visits and in-person, phone, or letter contact compared to Black and Hispanic adolescents (Table 5).

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

TABLE 5. CONTACT AND ACTIVITIES BETWEEN ADOLESCENTS AND THEIR NONRESIDENT FATHERS PREDICTED BY RACE/ETHNICITY AND FATHER’S EDUCATION, NET OF CONTROLSa (UNSTANDARDIZED REGRESSION COEFFICIENTS, WEIGHTED)

Contact Activities

Overnightb Talkb Any Contactc Shoppingc Sportsc Religiousc Moviec Projectc Activities Scaleb

White colleged .66** .71** 1.26** .57** .62** .72** .67** .7** .49** Black hsd -.22* .04 .10 -.09 -.06 .85** .07 .78** .15 Black colleged .07 .18 .49* .19 .15 .84** .34 .49* .23* Hispanic hsd .01 -.14 -.44 .12 .01 .54 .23 .99** .21 Hispanic colleged .12 .13 .55 .47* -.01 .18 .46* .94** .29* Asian hsd .09 -.20 .10 .86* .75 .06 .26 .70 .40 Asian colleged .58 .34 1.33 .25 -.01 .81* -.33 .30 .10 F 9.24** 8.59** 3.09** 5.73** 7.34** 2.73** 4.62** 3.52** 7.19″* F int’ 2.90* 4.67** 2.70* 1.80 2.33 3.92* 2.43 3.68* 5.27** F groupg 8.53** 9.19** 5.51** 4.33** 2.61* 3.64** 3.99** 2.90** 4.90** Differencesh Wc > Whs, Bhs, Wc > Whs, Bhs, Wc > Whs, Wc > Whs, Bhs; Wc > Whs, Whs < Wc, Wc > Whs, Bhs Whs < Wc, We > Whs, p < .01 Bc, Hhs Bc, Hhs Bhs, Bc, Hhs; Ahs > Bhs Bhs Bhs, Bc Bhs, Hhs, Hc Bhs

Hhs < Bc

Differencesh Wc > Hc; Wc > Hc, Ahs We > Ahs; Wc > Bc; Wc > Bc, Hhs Whs < Ac Wc > Ac; Whs < Bc Wc > Bc; p < .05 Bhs < Whs, Bc > Whs; Ahs > Whs, Hc > Whs Whs < Bc, Hc

Bc, Ac Hhs < Bhs, Hc, Hhs; Hc > Whs, Ac Bhs

ni 4557 4563 4563 4030 4025 4031 4031 4029 4027

aAll models include controls for mother’s education, household income, family structure, marital birth, father is U.S. born, and adolescent’s age and gender. bOrdinary least squares regression. cLogistic regression. dReference category: White hs (hs – high school or less; college = some college or more). eF = F statistic for the overall model. fF– F statistic for the interaction effects of racial/ethnic group and father’s education; Ho: all interactions are zero. “F= F statistic for the group effect; Ho: all groups are equal. hSignificant differences (at p < .01 and p < .05) between groups summarized. Whs = White high school or less; Wc = White some college or more; Bhs = Black high school or less; Bc = Black some college or more; Hhs = Hispanic high school or less; Hc = Hispanic some college or more; Ahs -= Asian high school or less; Ac = Asian some college or more.’n = unweighted n. *p < .05. **p < .01.

h,

G O X

O C,

b op

CL

k2j b

h,

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 13

Regarding activities, highly educated White fathers are most likely to participate in shopping, sports, and movies. In most cases, they are not significantly different from minority fathers with similarly high levels of education, but they do exhibit significantly higher levels of involvement compared to less educated White fathers. In con- trast, minority fathers do not show significant differences by education. The interaction effect is modest but consistent. Moreover, the higher rates of attending religious services for Blacks and of having worked on a school project for Blacks and Hispanics result from the comparison to White adolescents whose fathers have a high school education or less.

The higher frequency of talking about prob- lems previously found for Hispanic adolescents is most pronounced when they are compared to White adolescents who have less educated fathers

(Table 6). Significant differences are now revealed for talking about school work or grades, with the lowest incidence reported for less edu- cated White fathers. Black, Hispanic, and Asian fathers do not show significant differences by education. Highly educated Black and White fathers are more likely to talk about school with their adolescents than their less educated counter-

parts, but Hispanic and Asian fathers show no difference by education. The overall talking scale shows more and less educated White fathers

at the two extremes. The higher ratings of close- ness given by Black adolescents in comparison to Whites is found only for the comparison to White adolescents with less educated fathers.

Consistent with our hypothesis, the father’s level of education differentiates levels of involve-

ment more so for Whites than for minority fathers. Among highly educated fathers, race and ethnic differences are relatively minor, but among less educated fathers, White fathers are the least involved with their adolescent

children-they do less, talk less, and develop less close bonds with their children. Overall, these results indicate that the social capital that is lost when fathers do not reside with their

children remains unavailable to White youth with less educated fathers and potentially further disadvantages the well-being of such youth. Moreover, it is mainly White youth with less educated fathers who are at risk, as education does not seem to differentiate the extent to

which minority fathers provide social capital resources through their involvement with non- resident children.

The Influence of Gender

We examined three-way interactions between race/ethnicity, father’s education, and the adoles- cent’s gender. The three-way interaction was sig-

nificant (F= 3.26, p < .05) only for the model predicting overnight visits (results not shown), suggesting that the results presented in Tables 5 and 6 are similar for boys and girls with one exception.

Previous results revealed that White adoles-

cents, particularly those with highly educated fathers, had the most overnight visits with their nonresident fathers. In addition, boys generally had more overnight visits than girls. Results from the three-way interaction model reveals that White adolescents with highly educated fathers have the most overnight visits, with White boys staying over significantly more frequently than White girls. Although they have fewer overnight visits than White boys with highly educated fathers, White boys who have less educated fathers generally have more overnight visits than boys and girls in other racial/ethnic groups who also have less educated fathers. Although White girls with less educated fathers have fewer overnight visits than White boys with less educated fathers, White girls are not significantly different from other girls who also have less educated fathers. In addition, among Hispanics with less educated fathers, boys have signifi- cantly more overnight visits with their fathers than girls. The difference in overnight visits between Hispanic boys and girls with highly educated fathers is not significant, however.

Differences Between Hispanics and Between Asians

Thus far, we have seen variation in patterns of nonresident father involvement among the major racial/ethnic groups, yet we know that substantial variation exists within these groups as well. We also examined differences in nonresident father

involvement within the Hispanic and Asian popu- lations in our study, controlling for socio- economic and family history factors. Given the smaller subsample sizes, however, it was not possible to test adequately for interactions with the father’s education. In general, Cuban youth tend to be significantly more involved with their nonresident fathers compared to Mexicans, Puerto Ricans, and Central/South Americans (results not shown but see Table 7). They enjoy the most contact, participate in more activities,

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

TABLE 6. COMMUNICATION AND RELATIONSHIP QUALITY BETWEEN ADOLESCENTS AND THEIR NONRESIDENT FATHERS PREDICTED BY RACE/ETHNICITY AND FATHER’S EDUCATION, NET OF CONTROLSa (UNSTANDARDIZED REGRESSION COEFFICIENTS, WEIGHTED)

Communication Relationship Quality

Dating’ Problemsc Gradesc Schoolc Talking Scaleb Arguec Closenessb

White colleged .49** .36** .64** .66** .49** .02 .53** Black hsd .07 .27 .41** .19 .21.02 .30** Black colleged .26 .40* .70** .55** .43** -.10 .56** Hispanic hsd .57** .74** .50* .26 .45** .09 .31* Hispanic colleged .79** .78** .45 .17 .48** -.09 .23

Asian hsd -.38 .17 .35 .13 .12 .05 .45* Asian colleged -.94″ -.01 .30 .21 -.03 .74 .17 F” 4.76** 2.64** 3.07** 3.53** 4.05** 2.72** 8.05** F int’ 1.97 0.88 2.51 2.60 3.86* 0.59 7.13** F groupg 6.08** 3.69** 4.89** 3.49** 7.30** 0.46 7.93**

Differencesh p < .01 Wc, Hc > Whs, Bhs, Ac; Whs < Wc, Whs < Wc, Wc > Whs, Bhs; Whs < We, Be, none Whs < We, Bhs, Bc Hhs > Whs, Ac; Ac < Bc Hhs, Hc Bhs, Bc Bc > Whs Hhs, Hc

Differencesh p < .05 Hhs > Bhs; Hc > Bc; Whs < Bc Whs < Hhs Be > Bhs Whs < Bhs; Whs < Hhs, Ahs; Ac < Whs, Bhs Wc > Bhs, Ac We > Bhs, Ac;

Bc > Bhs

n 4031 4031 4028 4026 4025 4031 4563

aAll models include controls for mother’s education, household income, family structure, marital birth, father is U.S. born, and adolescent’s age and gender. bOrdinary least squares regression. cLogistic regression. dReference category: White hs (hs – high school or less; college – some college or more). eF – F statistic for the overall model. ‘F F statistic for the

interaction effects of racial/ethnic group and father’s education; Ho: all interactions are zero. gF= F statistic for the group effect; Ho: all groups are equal. hSignificant differences (at p < .01 and p < .05) between groups summarized. Whs – White high school or less; We — White some college or more; Bhs – Black high school or less; Bc= Black some college or more; Hhs – Hispanic high school or less; He = Hispanic some college or more; Ahs – Asian high school or less; Ac = Asian some college or more. ‘n – unweighted n. *p < .05. **p < .01.

4

,

0

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 15

report higher levels of closeness, and have the fewest arguments. Our comparison of Chinese, Filipinos, and all other Asians revealed that Filipinos generally tend to have higher levels of father involvement (e.g., for any contact, talking about school work and grades), whereas Chinese fathers have the lowest (e.g., for shopping, talk- ing about school more generally). Overall, how- ever, there are more similarities than differences

between Asian subgroups.

Summary

Table 7 provides a summary of the major find- ings from this study by indicating the group that stands out as having either the most or least involved nonresident fathers (both extremes are occasionally noted) on each measure of father involvement. The first column summarizes the

findings regarding the influence of race/ethnicity and father’s education. For example, the pattern of White fathers falling at the two extremes (with the lowest levels of father involvement reported for White fathers with a high school education or less and the highest levels of involvement reported for White fathers with high levels of education) is easily identified. The next column summarizes the few significant gender interac- tions. Then we summarize the differences found

within Hispanic and Asian subgroups. Finally, we note the significant control predictors for each measure of father involvement.

CONCLUSIONS

Our first aim was to ascertain whether and where racial/ethnic differences in nonresident father involvement exist. We conclude that differences

exist for many, but not all, aspects of nonresident father involvement. It is noteworthy that many of these differences might be less apparent if only broad or global measures of involvement are examined. Instead, the diversity across race/eth- nicity is revealed by examining different dimen- sions of involvement and the different types of activities and topics of discussion that nonresi- dent fathers engage in with their children. No one racial or ethnic group stands out as being signifi- cantly higher or lower on father involvement. Instead, particular groups are significantly higher or lower on certain activities, and these patterns vary with the activity in question.

Our second aim was to ascertain whether racial/ethnic differences in nonresident father

involvement are due mainly to the large socio- economic and demographic differences that exist among the groups. In some cases, these charac- teristics explained racial/ethnic differences in nonresident father involvement, but in other cases they did not. Of the factors considered, socioeconomic circumstances, and the father’s education in particular, are the most influential in explaining racial/ethnic differences. Whether the youth was born within marriage was also influential in explaining some of these differ- ences. The lower education of Black and His-

panic fathers, and the greater likelihood of nonmarital childbearing, especially among Blacks, is linked to lower levels of involvement. Once these differences are controlled, overall levels of involvement for minority fathers increase. Family structure, whether the father was U.S. born, and the adolescent’s age and gender were least important in explaining racial/ ethnic differences. Aponte’s (1999) assertion that structural factors, and socioeconomic status in particular, largely account for racial/ethnic differ- ences between families is partially supported. However, the finding that race/ethnicity remains significant despite controls for structural factors suggests that nonstructural influences, such as cultural values or practices, are also important.

Overall, differences by adolescent gender were limited. Gender did, however, influence certain aspects of nonresident father involvement, and sometimes moderated the effects of race/ethnicity on father involvement. Where gender differences existed, boys generally reported more involve- ment with their nonresident fathers than did

girls (e.g., more overnight visits) and reported being closer to them (although Black girls reported being as close to their fathers as White, Hispanic, and Asian boys did).

Our third aim was to determine whether racial/

ethnic differences vary significantly by father’s education. We discovered that the father’s level

of education interacts with race/ethnicity to further differentiate patterns of nonresident father involvement. White fathers fall at the two

extremes, with the lowest levels of father involve- ment reported for White fathers with a high school education or less and the highest levels of involvement reported for White fathers with high levels of education. In contrast, minority fathers exhibit fewer significant differences by education, with levels of involvement that are between the two extremes exhibited by White fathers. Education, therefore, mainly differentiates

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

TABLE 7. SUMMARY OF MAJOR FINDINGS

Influence of Influence of Race/Ethnicity Within Hispanic Within Other Significant Race/Ethnicity and Father’s Educationa and Adolescent’s Genderb Groupsc Asian Groupsd Predictorse

Most Least Most Least Most Least Most Least

Contact

Overnight White college White college Cuban No differences Marital birth, boys’ younger, male

Talk White college White boys Hispanic girls Cuban No differences Single mother, marital birth, younger

Any White college Hispanic high Asian boys Hispanic girls Cuban Filipino Marital birth, younger school

Activities

Shopping White college No interaction Cuban Chinese Marital birth, younger Sports White college No interaction No differences No differences Younger, male Religious White high school No interaction Cuban No differences Younger Movie White college No interaction Cuban No differences Marital birth,

younger, male Project White high school No interaction Cuban No differences Younger Scale White college No interaction Cuban No differences Marital birth,

younger, male Communi-

cation

Dating White college Asian college Hispanic boys Asian girls No differences No differences Income, marital and Hispanic and girls birth, older

Problems White high school No interaction No differences No differences Stepfather, older Grades White high school No interaction No differences Filipino Younger School White college and No interaction No differences Chinese Income, younger

Black college Scale White high school No interaction No differences No differences Income,

marital birth

Relationship quality Argue No differences No interaction Cuban No differences Single mother,

non-U.S born

Closeness White high school Boys and Cuban No differences Less educated mother, Black girls marital birth,

younger, male

aBased on Tables 5 and 6. bBased on Appendix. CComparison of Cubans, Mexicans, Puerto Ricans, and Central/South Americans. dComparison of Filipinos, Chinese, and other Asians. eSignificant control variables (p < .05) in multivariate models in Tables 3 and 4. fBased on three-way interaction between race/ethnicity, father’s education, and adolescent gender.

Cr\

“v O X

h,

O

rY

frQ (b

;=S

~Lt?l C1

h,

This content downloaded from �������������69.137.123.55 on Sun, 16ff on Thu, 01 Jan 1976 12:34:56 UTC

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 17

involvement by White fathers. A more fruitful approach than focusing solely on whether demographic factors explain racial/ethnic differ- ences is to shift attention toward understanding how race/ethnicity and demographic factors interact with one another. The failure to examine

this interaction may account for the inconsistent findings on racial differences in nonresident father contact reported in previous research. A key finding of this study is that the social

capital resources that adolescents gain from non- resident fathers vary in important ways by race and ethnicity. White adolescents with highly educated fathers, boys in particular, enjoy the highest levels of contact with nonresident fathers. White adoles-

cents with highly educated fathers also occasionally report engaging in more activities with their non- resident fathers than minority youth do, although this is not uniformly the case. Minority youth with highly educated fathers are similar in terms of involvement beyond contact, providing crucial social capital to youth with nonresident fathers. Hispanic youth, particularly girls, lose access

to the father’s social capital more so than other youth because their fathers have the lowest levels of contact. When their nonresident fathers main-

tain contact, however, Hispanic youth benefit from the social capital resources inherent in com- munication with fathers. Black youth similarly benefit from close emotional bonds with their

fathers. These are precisely the types of involve- ment that have a greater association with child well-being than mere contact (Amato & Gilbreth, 1999; Marsiglio et al., 2000). Even the activity differences favor minority well-being. The higher rates of working on school projects for Hispanics and Blacks and the higher rates of religious activ- ities for Blacks are the more intensive types of involvements found to promote adolescent well- being. These forms of involvement contrast with father involvement in sports activities and watch- ing movies-forms of involvement that charac- terize highly educated White families. Despite their greater likelihood of becoming a nonresi- dent father, in many cases, minority men are just as involved (or more involved) nonresident fathers as Whites (see also Mott, 1990).

Our findings suggest that White youth with less educated fathers may be at greatest risk of sustained loss of social capital, as their nonresi- dent fathers are less involved than minority fathers and highly educated White fathers. Although their levels of contact are not signifi- cantly less than those of minority youth, they

stand out as having the lowest levels of talking with their fathers, and they report being less close to them. Our findings suggest that strong differ- ences in child well-being by father’s education might exist within White families, with children

born to highly educated White fathers being bet- ter off than those born to less educated White

fathers. These differences may not exist in minor- ity families where youth with both highly edu- cated and less educated fathers might fare more similarly, at least in ways that are linked to fathers’ provision of social capital. Our findings also are consistent with research that finds more

negative consequences of family disruption for Whites than for Blacks or Hispanics (McLanahan & Sandefur, 1994). Perhaps part of this differ- ence stems from the lower involvement of White nonresident fathers. It remains for future research

to explore directly the implications of these pat- terns of father involvement for child well-being.

Hispanic youth, especially girls, also may be at greater risk given their lower levels of contact with their fathers. One reason Hispanic fathers have less contact may be that they migrate for work purposes and are more likely to live away from their families. This is likely to be especially true for Mexican and Central/South American

fathers, and indeed, we find their levels of contact

to be among the lowest. Cuban families are more likely to live in ethnic enclaves (Suarez, 1998), which may mean that when families split up, fathers are more likely to remain in the same community, fostering their involvement with nonresident children. Nonresident Cuban fathers

can provide the forms of social capital inherent in father-child relations, and their connections to other individuals and broader social networks in

the community can continue and facilitate con- nections and opportunities for their children in the school, work, and social environments. Such findings underscore the importance of under- standing the contexts in which nonresident fathers and their children live-especially by eth- nic group-and how they influence patterns of involvement.

Future research would benefit from further

attention to understanding racial/ethnic differ- ences in fathering more broadly. Our study is unable to determine the extent to which the racial/ethnic differences in father involvement

reported here are specific to nonresident fathers or whether they reflect broader differences in family practices between fathers of different backgrounds.

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

18 Journal of Marriage and Family

Future research also would benefit from a life

course perspective on fathering that would focus on fathering over time (e.g., pre- and postdivorce) and across the lives of children (i.e., not limited to adolescents). Given the increasing orientation toward peers, adolescence may be a particularly challenging time for many nonresident fathers to maintain an active presence in their children’s lives, although such involvement may be particu- larly crucial as offspring navigate adolescence and the transition to adulthood. Yet little is

known about how nonresident parenting changes as children develop (Hawkins & Palkovitz, 1999). Further, little is known about how non- resident father involvement is influenced by other changes in family structure, such as the addition of a stepfather. This study is limited by examining the relationships between nonresident fathers and their children at a single point in the child’s life.

Children living in diverse types of situations also must be considered in future research. Most studies of nonresident father involvement are

based on samples of children living in resident mother households, yet, as this study demon- strates, many children (14%) with nonresident fathers live in other types of families as well. For example, some are living with grandparents or other relatives, some have foster parents, and others live in group homes. Such children have especially low levels of interaction with their fathers, yet we know little about these relation- ships. Further, there are significant racial/ethnic differences in the propensity to live in these types of households, with higher rates among Blacks and Asians. Including such children in our studies will provide a more accurate picture of nonresi- dent parenting.

Past research has demonstrated the many dis- advantages faced by children who grow up apart from their biological fathers (McLanahan & Sandefur, 1994). With half of all U.S. children facing this situation, increased attention is being turned toward understanding the role of nonresi- dent fathers in their children’s lives and the ways that involvement by nonresident fathers can pro- mote child well-being. This study is the first to examine racial and ethnic diversity in nonresident father involvement among several racial/ethnic groups. In contrast to most prior studies that rely on mother reports, which likely underesti- mate levels of nonresident father involvement

(Seltzer & Brandreth, 1994), our study uses recent data from a nationally representative sam- ple of adolescents, providing an important con-

tribution to the literature. We explored multiple ways that nonresident fathers are involved in their children’s lives, going beyond the more common single consideration of contact, and tapping critical dimensions of social capital resources that fathers can provide in time, atten- tion, guidance, emotional support, and connec- tions to information and opportunities in the social environments of youth. Our findings have important implications for understanding racial/ ethnic differences in nonresident father involve-

ment and point to certain groups of children, most notably White youth with less educated fathers, who experience sustained loss of father’s social capital, and in turn are at risk for further disadvantage in their lives.

NOTE

This research was supported by funding (R01 HD043384) from the National Institute of Child Health and Human

Development (NICHD) to Valarie King, principal investi- gator. We also gratefully acknowledge support to King as a Brookdale National Fellow, to Harris through grant P01 HD31921 from NICHD, and from core funding (R24 HD41025) from NICHD to the Population Research Insti- tute, The Pennsylvania State University. This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by grant POI1-HD31921 from NICHD, with cooperative funding from 17 other agencies. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Persons inter- ested in obtaining data files from Add Health should con- tact Add Health, Carolina Population Center, 123 W. Franklin Street, Chapel Hill, NC 27516-2524 (http:// www.cpc.unc.edu/addhealth/contract.html). Earlier versions of this article were presented at the 1999 annual meeting of the American Sociological Association (Chicago) and the 1999 Add Health Users Workshop, National Institutes of Health (Bethesda, MD). We thank Paul Amato, Alan Booth, and Linda Burton for their helpful comments.

REFERENCES

Amato, P. R. (1987). Family process in one-parent, stepparent, and intact families: The child’s point of view. Journal of Marriage and the Family, 49, 327-337.

Amato, P. R. (1998). More than money? Men’s con- tributions to their children’s lives. In A. Booth & A. Crouter (Eds.), Men in families (pp. 241-278). Hillsdale, NJ: Erlbaum.

Amato, P. R. (2000). The consequences of divorce for adults and children. Journal of Marriage and Family, 62, 1269-1287.

Amato, P. R., & Gilbreth, J. G. (1999). Nonresident fathers and children’s well-being: A meta-analysis. Journal of Marriage and the Family, 61, 557-573.

Amato, P. R., & Rezac, S. J. (1994). Contact with nonresident parents, interparental conflict, and

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

Nonresident Father Involvement 19

children’s behavior. Journal of Family Issues, 15, 191-207.

Aponte, R. (with Beal, B. A. & Jiles, M. E.) (1999). Ethnic variation in the family: The elusive trend toward convergence. In M. B. Sussman, S. K. Steinmetz, & G. W. Peterson (Eds.), Hand- book of marriage and the family (2nd ed., pp. 111-141). New York: Plenum.

Arditti, J. A. (1995). Noncustodial parents: Emergent issues of diversity and process. Marriage and Family Review, 20, 283-304.

Astone, N., & McLanahan, S. (1991). Family structure, parental practices, and high school completion. American Sociological Review, 56, 309-320.

Bearman, P. S., Jones, J., & Udry, J. R. (1997). The National Longitudinal Study of Adolescent Health: Research Design. Retrieved December 4, 2003, from http://www.cpc.unc.edu/projects/addhealth/ design.html.

Bianchi, S. (1990). America’s children: Mixed pro- spects. Population Bulletin, 45(1). New York: Popu- lation Reference Bureau.

Bumpass, L. L. (1984). Children and marital disruption: A replication and update. Demography, 21, 71-82.

Cazenave, N. A., & Leon, G. H. (1987). Men’s work and family roles and characteristics: Race, gender, and class. In M. S. Kimmel (Ed.), Changing men: New directions in research on men and masculinity (pp. 244-262). Newbury Park, CA: Sage.

Chantala, K., & Tabor, J. (1999). Strategies to perform a design-based analysis using the Add Health Data. Retrieved May 5, 2003, from http://www.cpc.unc. edu/projects/addhealth/files/weightl .pdf.

Chilman, C. S. (1993). Hispanic families in the United States: Research perspectives. In H. P. McAdoo (Ed.), Family ethnicity: Strength in diversity (pp. 141-163). Newbury Park, CA: Sage.

Coleman, J. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94(Suppl.), S95-S 120.

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, 187-200.

Danziger, S. K., & Radin, N. (1990). Absent does not equal uninvolved: Predictors of fathering in teen mother families. Journal of Marriage and the Family, 52, 636-642.

Del Carmen, R., & Virgo, G. N. (1993). Marital dis- ruption and nonresidential parenting: A multicul- tural perspective. In C. E. Depner & J. H. Bray (Eds.), Nonresidential parenting: New vistas in family living (pp. 13-36). Newbury Park, CA: Sage.

Demo, D. H., & Cox, M. J. (2000). Families with young children: A review of the research in the 1990s. Journal of Marriage and the Family, 62, 876-895.

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

Furstenberg, F. F., Jr., & Hughes, M. E. (1995). Social capital and successful development among at-risk youth. Journal of Marriage and the Family, 57, 580-592.

Furstenberg, F. F., Jr., & Nord, C. W. (1985). Parenting apart: Patterns of childrearing after marital disruption. Journal of Marriage and the Family, 47, 893-904.

Furstenberg, F. F., Jr., Nord, C. W., Peterson, J. L., & Zill, N. (1983). The life course of children of divorce: Marital disruption and parental contact. American Sociological Review, 48, 656-668.

Griswold, R. L. (1993). Fatherhood in America: A history. New York: Basic Books.

Harris, K. M., & Marmer, J. K. (1996). Poverty, pater- nal involvement, and adolescent well-being. Journal of Family Issues, 17, 614-640.

Harris, K. M., & Morgan, S. P. (1991). Fathers, sons, and daughters: Differential paternal involvement in parenting. Journal of Marriage and the Family, 53, 531-544.

Harris, K. M., & Ryan, S. (1999, August). Parental and neighborhood influences on adolescent risk behavior. Paper presented at the conference, “Par- enting and the Child’s World: Multiple Influences on Intellectual and Social-Emotional Develop- ment,” NIH campus, Bethesda, MD.

Hawkins, A. J., & Palkovitz, R. (1999). Beyond ticks and clicks: The need for more diverse and broader

conceptualizations and measures of father involve- ment. Journal of Men’s Studies, 8, 11-32.

Ishii-Kuntz, M. (2000). Diversity within Asian Amer- ican families. In D. H. Demo, K. R. Allen, & M. A. Fine (Eds.), Handbook of family diversity (pp. 274-292). New York: Oxford University Press.

Johnson, D. R., & Elliott, L. A. (1998). Sampling design effects: Do they affect the analyses of data from the National Survey of Families and House- holds? Journal of Marriage and the Family, 60, 993-1001.

Julian, T. W., McKenry, P. C., & McKelvey, M. W. (1994). Cultural variations in parenting: Perceptions of Caucasian, African-American, Hispanic, and Asian- American parents. Family Relations, 43, 30-37.

King, V. (1994). Variation in the consequences of nonresident father involvement for children’s well-

being. Journal of Marriage and the Family, 56, 963-972.

King, V. (2002). Parental divorce and interpersonal trust in adult offspring. Journal of Marriage and Family, 64, 642-656.

Landale, N. S., & Oropesa, R. S. (2001). Father involve- ment in the lives of mainland Puerto Rican chil-

dren: Contributions of nonresident, cohabiting and married fathers. Social Forces, 79, 945-968.

Landry, B. (2000). Black working wives: Pioneers of the American family revolution. Berkeley: Univer- sity of California Press.

Lawson, E. J., & Thompson, A. (1999). Black men and divorce. Thousand Oaks, CA: Sage.

Manning, W. D., & Smock, P. J. (1999). New families and nonresident father-child visitation. Social

Forces, 78, 87-116. Marsiglio, W., Amato, P., Day, R. D., & Lamb, M. E.

(2000). Scholarship on fatherhood in the 1990s and beyond. Journal of Marriage and the Family, 62, 1173-1191.

McAdoo, H. P. (1978). Minority families. In J. H. Stevens, Jr. & M. Mathews (Eds.), Mother/ child father/child relationships (pp. 177-195).

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

20 Journal of Marriage and Family

Washington, DC: National Association for the Education of Young Children.

McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent: What hurts, what helps. Cam- bridge, MA: Harvard University Press.

McLoyd, V. C., Cauce, A. M., Takeuchi, D., & Wilson, L. (2000). Marital processes and parental socialization in ?tlsb> families of color: A decade review of

research. Journal of Marriage and the Family, 62, 1070-1093.

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

Demography, 27, 499-517. O’Hare, W. P. (1992). America’s minorities-The demo-

graphics of diversity. Population Bulletin, 47(4). Washington, DC: Population Reference Bureau.

Seltzer, J. A. (1991). Relationships between fathers and children who live apart: The father’s role after separa- tion. Journal of Marriage and the Family, 53, 79-101.

Seltzer, J. A., & Bianchi, S. M. (1988). Children’s contact with absent parents. Journal of Marriage and the Family, 50, 663-677.

Seltzer, J. A., & Brandreth, Y. (1994). What fathers say about involvement with children after separation. Journal of Family Issues, 15, 49-77.

Seltzer, J. A., Schaeffer, N. C., & Chamg, H. (1989). Family ties after divorce: The relation- ship between visiting and paying child support. Journal of Marriage and the Family, 51, 1013-1032.

Staggenborg, S. (1998). Gender, family, and social movements. Thousand Oaks, CA: Pine Forge.

Stata Corporation. (1999). STATA reference manual, release 6. College Station, TX: Stata Press.

Suarez, Z. E. (1998). Cuban-American families. In C. H. Mindel, R. W. Habenstein, & R. Wright, Jr. (Eds.), Ethnic families in America: Patterns and variations (pp. 172-198). Upper Saddle River, NJ: Prentice Hall.

Taylor, R. L. (2000). Diversity within African American families. In D. H. Demo, K. R. Allen, & M. A. Fine (Eds.), Handbook offamily diversity (pp. 232-251). New York: Oxford University Press.

U.S. Bureau of Census. (1997). Statistical abstract of the United States: 1997. Washington, DC: U.S. Government Printing Office.

U.S. National Center for Health Statistics. (1993). Mor- bidity and mortality weekly report (Vol. 42, No. 20). Washington, DC: U.S. Government Printing Office.

Wong, M. G. (1998). The Chinese-American family. In C. H. Mindel, R. W. Habenstein, & R. Wright, Jr. (Eds.), Ethnic families in America: Patterns and variations (pp. 284-310). Upper Saddle River, NJ: Prentice Hall.

Youniss, J., & Smollar, J. (1985). Adolescent relations with mothers, fathers, and friends. Chicago: Univer- sity of Chicago Press.

Zinn, M. B. (1990). Family, feminism, and race in America. Gender and Society, 4, 68-82.

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

APPENDIX. FATHER INVOLVEMENT PREDICTED BY RACE/ETHNICITY AND ADOLESCENT’S GENDER, NET OF CONTROLSa (UNSTANDARDIZED REGRESSION COEFFICIENTS, WEIGHTED)

Talkb Any Contactc Datingc Closenessb

White boyd .17* -.03 -.28* .38** Black girld .08 .08 -.24 .36* Black boyd -.10 .14 -.14 .58** Hispanic girld -.35* -.82** .24 -.02 Hispanic boyd .01 -.15 .53* .63** Asian girld -.30 .18 -1.97** .08 Asian boyd .36 1.71* -.46 .46 Fe 17.00** 7.48** 5.58** 11.41** F intf 4.53** 3.11* 4.16** 2.92* F groupg 2.84** 4.03** 5.16** 13.00** Differencesh Wboy > Hgirl, Bboy; Hgirl < Wgirl, Wboy, Bgirl, Bboy, Hboy, Aboy Hboy > Wboy, Bgirl, Bboy > Wgirl, Bgirl, Hgirl; p < .01 Hgirl < Aboy, Bgirl Bboy, Agirl; Agirl < Wgirl, Bgirl > Wgirl, Hgirl;

Wboy, Bgirl, Bboy, Hgirl Wboy, Hboy > Wgirl, Hgirl

Differencesh Wboy > Wgirl; Hgirl < Agirl; Aboy > Wgirl, Hgirl > Bgirl, Wboy; Bboy > Wboy; Hboy > Agirl p < .05 Hgirl < Wgirl Wboy, Bgirl, Bboy, Hboy Hboy > Wgirl; Agirl < Aboy;

Wgirl > Wboy

ni 5364 5365 4555 5361

aAll models include controls for father’s education, mother’s education, household income, family structure, marital birth, father is U.S. born, and adolescent’s age. bOrdinary least squares regression. cLogistic regression. dReference category: White girl. eF= F statistic for the overall model. fF= F statistic for the interaction effects of racial/ethnic group and adolescent’s gender; Ho: all interactions are zero. gF=F statistic for the group effect; Ho: all groups are equal. hSignificant differences (at p < .01 and p <.05) between groups summarized. Wgirl = White girl; Wboy = White boy; Bgirl = Black girl; Bboy = Black boy; Hgirl = Hispanic girl; Hboy = Hispanic boy; Agirl= Asian girl; Aboy= Asian boy. n = unweighted n. *p < .05. **p <.01.

O ;s

~1

(h ;S cU

C~j

Y

1 O ~1 Ch

Ch ;S

Clr

This content downloaded from �������������69.137.123.55 on Sun, 16 Aug 2020 13:40:57 UTC�������������

All use subject to https://about.jstor.org/terms

  • Contents
    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18
    • 19
    • 20
    • 21
  • Issue Table of Contents
    • Journal of Marriage and the Family, Vol. 66, No. 1 (Feb., 2004), pp. 1-266
      • Front Matter
      • Fatherhood
        • Racial and Ethnic Diversity in Nonresident Father Involvement [pp. 1-21]
        • When Stepfathers Claim Stepchildren: A Conceptual Analysis [pp. 22-39]
        • Paternal Involvement with Children: The Influence of Gender Ideologies [pp. 40-45]
        • Role of the Father-Adolescent Relationship in Shaping Adolescents’ Attitudes toward Divorce [pp. 46-58]
      • Divorce
        • Dollars, Dependency, and Divorce: Four Perspectives on the Role of Wives’ Income [pp. 59-74]
        • Interactions between Cultural and Economic Determinants of Divorce in the Netherlands [pp. 75-89]
        • Did Divorces Decline after the Oklahoma City Bombing? [pp. 90-100]
      • Gender, Housework, and Power
        • Reconsidering the Division of Household Labor: Incorporating Volunteer Work and Informal Support [pp. 101-117]
        • Cohabitation and Housework: The Effects of Marital Intentions [pp. 118-125]
        • The Power of Older Women and Men in Egyptian and Tunisian Families [pp. 126-146]
      • Adolescents in Families
        • Family Time and the Psychosocial Adjustment of Adolescent Siblings and Their Parents [pp. 147-162]
        • Parenting Practices as Moderators of the Relationship between Peers and Adolescent Marijuana Use [pp. 163-178]
        • Welfare Reform and Teenage Pregnancy, Childbirth, and School Dropout [pp. 179-194]
      • Of General Interest
        • Disclosure and Relationship Satisfaction in Families [pp. 195-209]
        • Cohabitation and Children’s Family Instability [pp. 210-219]
        • Gender, Preloss Marital Dependence, and Older Adults’ Adjustment to Widowhood [pp. 220-235]
        • Remarriage, Unmarried Cohabitation, Living Apart Together: Partner Relationships Following Bereavement or Divorce [pp. 236-243]
        • Substance Use and Early Marriage [pp. 244-257]
      • Book Reviews
        • Review: untitled [pp. 258-259]
        • Review: untitled [pp. 259-261]
        • Review: untitled [pp. 261-262]
        • Review: untitled [pp. 262-263]
        • Review: untitled [pp. 263-264]
        • Review: untitled [pp. 264-266]
      • Back Matter