The Relationship Between Maternal Gatekeeping, Paternal Competence, Mothers' Attitudes About the Father Role, and Father Involvement JAY FAGAN MARINA BARNETT Temple University This study explored the relationships between maternal gatekeeping, mothers'perceptions of father competence, mothers' attitudes about the father role, and amount of father involvement. The sample consisted of 30 nonresidential and 72 residential fathers. The results of path analysis revealed that residential status of the father had a direct link to mothers' gatekeeping behavior. Father competence was indirectly and directly linked to amount of father involvement with children. Gatekeeping mediated the relationship between father competence and involvement. Maternal gatekeeping was causally linked to amount of father involvement. Keywords: maternal gatekeeping; father involvement; parental competence; attitudes about fathers Researchers of parenting have become increasingly aware of the importance of studying parent-child relationships in the context of the coparent-ing relationship. Now more than ever, researchers recognize that in addition to parents' having direct influences on their children, they also have indirect influences by way of their relationship to other adults in the child-rearing network (Marsiglio, Amato, Day, & Lamb, 2000). Coparenting has been conceptualized in a variety of ways. Several researchers have focused on the support that parents give to each other when they are advancing a parenting goal (Gable, Belsky, & Crnic, 1995). Other researchers have concentrated on interparental agreement (Vaughn, Block, & Block, 1988), parental interaction focused on discussing child-rearing issues (Ahrons, 1981), child-related disputes between parents (Jouriles et al., 1991), or the parenting alliance (McBride & Rane, 1998), defined as the ability of a parent to acknowledge, respect, and value the parenting roles and tasks of the partner (Cohen & Weissman, 1984). Yet others have JOURNAL OF FAMILY ISSUES, Vol. 24 No. 8, November 2003 1020-1043 DOI: 10.1177/0192513X03256397 O 2003 Sage Publications 1020 Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1021 referred to the role that maternal gatekeeping plays in the relationship between mothers and fathers (Arendell, 1996; Doherty, Kouneski, & Erickson, 1998). In the divorce literature, researchers have given a good deal of attention to parenting conflict and support between former spouses (Ahrons & Miller, 1993; Ahrons & Wallisch, 1987; Madden-Derdich & Leonard, 2000). The focus of this study is on maternal gatekeeping. The major question addressed here is, What influences mothers to engage in gatekeeping behavior, and how are these influences and gatekeeping associated with father involvement? Based on the work of scholars such as Allen and Hawkins (1999) and Madden-Derdich and Leonard (2000), we defined gatekeeping as mothers' preferences and attempts to restrict and exclude fathers from child care and involvement with children. In this article, we draw from family systems theory to conceptualize the relationship between maternal gatekeeping and father involvement with children. According to family systems theory, all systems erect boundaries that affect the flow of information and energy between the environment and the system orbetween various parts of the system (Klein & White, 1996). Boundaries are defined as the invisible barriers that regulate the amount of contact between individuals (Nichols & Schwartz, 1998). A critical purpose of boundaries is to protect the autonomy of the family and its subsystems by managing proximity and hierarchy (Minuchin, 1974; Minuchin & Fishman, 1981). Many factors come into play when mothers and fathers engage in boundary making relative to children. A key factor is the flow of information that occurs between spouses concerning child-rearing practices and styles, child care involvement, and child-related decisions (Nichols & Schwartz, 1998). Family researchers have made several interesting observations that are relevant to understanding mother-father boundaries. Mothers seem to play a pivotal role in facilitating the father-child relationship (Arendell, 1996; Deutsch, Lussier, & Servis, 1993; Lewis, Feiring, & Weinraub, 1981; Marsiglio, 1995). Walker and McGraw (2000) have observed that there is ample evidence suggesting that mothers actively promote relationships between children and fathers. Wives were found to be more influential in involving their husbands in parenting than were husbands in involving their wives in parenting (Belsky, 1979). Even when mothers and fathers get divorced, the mother's support is a key factor in the degree to which fathers participate in coparenting interaction (Braver & O' Connell, 1998; Madden-Derdich & Leonard, 2000). Others have observed that some mothers exert considerable influence over fathers by limiting their involvement with children (Doherty et al., 1022 JOURNAL OF FAMILY ISSUES / November 2003 1998). Several researchers have found that mothers believe they have primary responsibility for the home and child care (Baber & Monaghan. 1988). Attitudes such as these may occur because women partially define themselves by their ability to influence the domestic domain (Kranichfeld, 1987; LaRossa, 1997). Rutter and Schwartz (2000) have suggested that because women have not been able to accumulate influence readily within the social structure, they have instead sought to obtain power within the family structure. In all likelihood, maternal gatekeeping is only partially explained by women's positions within the social structure. Mothers' and fathers' personality and behavioral characteristics also need to be considered. Hochschild (1989) reported that wives tended to restrict husbands' access to children when the men did not adequately participate in housework. Wives may also exclude husbands from becoming involved in a range of family tasks when they perceive men's domestic standards to be too low (Hawkins, Marshall, & Meiners, 1995). In actuality, very little is known about maternal gatekeeping. To date, only a few researchers have conducted quantitative studies in which maternal gatekeeping was operationally defined and measured. DeLuccie (1995) defined maternal gatekeeping as attitudes about the importance of the role of the fathers' and mothers' satisfaction with father involvement. DeLuccie (1995) hypothesized that maternal attitudes and satisfaction with father involvement have a "gatekeeping" function in relation to frequency of father involvement. She found that these gatekeeper variables are significantly associated with father involvement with children. Furthermore, they serve a mediating role between amount of father involvement and mothers' satisfaction with social support, with employment status, and with marriage, as well as with age of the child. Allen and Hawkins (1999) may have been the first researchers to develop an instrument designed specifically to measure gatekeeping. Their definition comprised four components: having a high standard for housework and child care, enjoying control overfamily tasks, having an identity that is contingent on making sure children are well groomed and keeping a clean house, and having a traditional attitude that women enjoy and find it easier to do housework and child care than men. Based on data from married, coresidential parents, Allen and Hawkins found that mothers could be clustered into three coparenting types—active gatekeepers, intermediaries, and collaborators. Approximately 21% of mothers were active gatekeepers. Moreover, active gatekeepers engaged in 5 more hours of family work per week than intermediaries and collaborators. In the present study, we test a model of maternal gatekeping in relation to father involvement with children. We draw from the marriage literature Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1023 to examine factors that are likely to influence maternal gatekeeping. These factors include family structure, financial contributions, social capital, and ideology (Cherlin, 1992; Furstenberg, 1990). With respect to the structural hypothesis, gatekeeping may be particularly relevant to families in which the father does not reside with his children (Ihinger-Tallman, Pasley, & Buehler, 1993). The parents in these families may be divorced, separated, or never married. Numerous studies, including many that use nationally representative samples, suggest that divorced fathers, as a group, have little contact with their biological children (Furstenberg & Cherlin, 1991; Seltzer & Bianchi, 1988; Stewart, 1999). Children of divorced parents see their nonresidential fathers less frequently as time passes (Amato & Keith, 1991; Maccoby & Mnookin, 1992). The same appears to be true of never married, nonresidential fathers. These fathers are lower on all dimensions of involvement than noncustodial fathers with children born in intact marriages (Aquilino, 1994; Seltzer, 1991). Certainly, living apart from children reduces the potential for involvement with one's offspring. However, the lack of paternal involvement may be related also to mothers' restrictions on fathers (Braver & Griffin, 2000; Braver & O'Connell, 1998). Mothers may restrict father access to chil-drenbecause they are angry with their former spouse or partner orbecause they feel that the father has opted out of responsibility for his children (Greif, 1997). Divorced parents frequently report moderate to high levels of conflict with one another, and the conflict is often related to visitation issues (King & Heard, 1999) and to boundary ambiguity (Madden-Derdich, Leonard, & Christopher, 1999). Highlevelsof interparentalconflict and low levels of support may have a negative impact on the divorced parents' attachment relationship (Madden-Derdich & Arditti, 1999), which ultimately can lead to reduced contact between the nonresidential father and his children. Additionally, custodial mothers sometimes limit the father's involvement with his children because they see little value in the father's continued role (Braver & Griffin, 2000; Braver & O'Connell, 1998). Mothers also may restrict access to children if the man does not make sufficient financial contributions. Recent data show that 41% of nonresidential fathers do not pay child support; moreover, these fathers have no apparent financial reason to avoid this responsibility (Sorenson & Zibman, 2001). Nonresidential fathers' financial contributions often become a source of much conflict between parents (Amato, 2000). If the father is in arrears, the mother may prevent visitation with children even though this is illegal in many states (Greif, 2001). Taken together, these findings suggest that nonresidential fathers not only spend less time with 1024 JOURNAL OF FAMILY ISSUES / November 2003 their children than residential fathers, but some mothers also engage in considerable gatekeeping behavior toward nonresidential fathers. We also predict that fathers' financial contributions will be associated with maternal restrictions on fathers' time with children. Social capital refers to resources that adhere in the relationships between people (Coleman, 1990). In the present study, social capital is defined as fathers' positive parenting skills and competence. We were particularly interested in mothers' perceptions of father competence because of its probable linkage to maternal gatekeeping. Several researchers have suggested that fathers who have greater parenting competence may be more motivated to spend time being involved with their children (Beitel & Parke, 1998; Lamb, 1986, 1997). Others have suggested that the marital relationship mediates the association between paternal competence and father involvement (Bonney, Kelley, & Levant, 1999). According to this perspective, fathers' positive parenting abilities are linked to marital satisfaction and marital stability (Kalmijn, 1999), which in turn has a positive influence on the amount of father involvement with children (Bonney et al., 1999). Although we do not deny the potential relationship between paternal competence, marital satisfaction, and amount of father involvement, we suggest that maternal gatekeeping is the more likely process through which mothers' perceptions of father competence influence the man's actual involvement. Given the traditional role that mothers have assumed in providing primary care to their children (Coltrane, 1996), it is probable that mothers who perceive their husbands/partners as being less competent parents are also likely to restrict fathers' access to children. Maternal gatekeeping is therefore viewed as a significant mediator of the relationship between mothers' perceptions of father competence and fathers' actual involvement with their children. To facilitate the examination of parental competence and maternal gatekeeping, we included in the study a sample of families in which the fathers and father figures are likely to have lower levels of parenting competence—families receiving mandated child welfare services. Fathers were found to be the more likely perpetrators of child maltreatment (Malkin & Lamb, 1994), including shaken baby syndrome (Sinai et al., 2000) and serious physical abuse (Naidoo, 2000). It is known that maltreating mothers frequently report being socially isolated (McCurdy, 1995; Moncher, 1995), which presumably includes having a lack of father presence. In this study, we expected that mothers receiving child welfare services will report lower levels of paternal competence and higher levels of maternal gatekeeping. Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1025 Women's ideology about the role of fathers also may have an impact on paternal involvement with children (Arendell, 1996). Some have argued that fathers are unlikely to participate in their children's care without the support of their wives (Bonney et al., 1999). Women who had more liberal attitudes about the father's role in parenting tended to have husbands or partners who participated in more child care (Barnett & Baruch, 1987). Several studies revealed positive associations between mothers' beliefs about the importance of the father role to children and the level of paternal involvement with children (DeLuccie, 1995; Fagan, Newash, & Schloesser, 2000). Moreover, mothers' attitudes about the father role seem to be important predictors of father involvement even after accounting for mothers' assessments of their husbands' child care skills and interest in participating in child care (Beitel & Parke, 1998). In the present study, we explored the relationship between mothers' attitudes about the importance of the father role and amount of father involvement with children. We expected that mothers' attitudes are directly linked to father involvement, but we also explored the possibility that maternal gatekeeping mediates the relationship between mothers' attitudes about the father role and paternal involvement in child care. The heuristic model that emerges from the existing research literature is one in which fathers' parenting competence is linked to mothers' gatekeeping behavior and to amount of father involvement with children. We hypothesized a path between maternal attitudes about the importance of the father role and maternal gatekeeping and a path between maternal attitudes and amount of father involvement. We expected also that being a nonresidential father will be associated with mothers' gatekeeping behavior and fathers' actual involvement with his children. Finally, fathers' financial contributions were explored for their potential linkages to maternal gatekeeping. METHOD PARTICIPANTS AND PROCEDURES This study made use of a convenience sample of participants drawn from two sources: families receiving in-home child welfare services (n = 30) and families that were known to undergraduate students of a human behavior course (n = 72). To receive child welfare services, the local department of public welfare had to complete an investigation of the family that resulted in a founded case of child abuse or child neglect. There were 1026 JOURNAL OF FAMILY ISSUES / November 2003 two criteria applied to all families recruited for participation in this study. Families were required to have at least one child who was between the ages of 3 and 16 years and who was living with the mother or with the mother and father. There had to be a biological father, stepfather, or paramour who was involved with this child. The families receiving child welfare services were obtained from four private nonprofit agencies located in an urban community. The agency supervisors were first asked to randomly select one or two families from the caseloads of agency caseworkers. Staff members were next asked to contact the family to determine their interest in participation. The caseworkers also had a letter from the researchers that could be presented to the parents if needed. The researchers then made a follow-up telephone contact to the family to explain the purpose and procedures of the study and to determine their interest in the study. If the family and caseworker indicated interest, then the researchers arranged a time to visit the family in their home. All interviews were conducted with the mother between July and August 2000. All survey forms were read aloud to the participants and the interviewer recorded all responses. Undergraduate students attending a human behavior course collected data from 76 families between the months of October 2000 and November 2000. Each student was permitted to conduct a maximum of five interviews. They were instructed to recruitfamiliesintheirsocialnetwork. The same criteria applied for the families in this sample as for the families in the child welfare sample. The students read the survey forms aloud to the participating mothers, but respondents filled in their own responses on the questionnaire. Respondents were then instructed to deposit and seal the completed questionnaire in an envelope provided by the student, with the assurance that the student would not open the envelope. Students received approximately 1 hour of training on the administration procedures of the study, including obtaining informed consent and answering participants' questions. The total number of respondents in the study was 102. Of the mothers who participated in the study, 51% were married only once, approximately 7% had remarried, 30% were never married, 5% were separated, and 5% were divorced (see Table 1). Fathers' reported marital status was similar: Forty-nine percent were married once, 8% were remarried, 29% were never married, 7% were separated, and 7% were divorced. The average age of the female caregivers in the study was about 35 years, and the mean age of fathers was 39 years. The race or ethnic background of the mothers was 44% African American, 44% White, 8% Latino, 2% Asian/Pacific Islander, and 2% who responded as a race or ethnicity other than these listed. The racial/ethnic background of fathers Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1027 TABLE 1 Demographic Characteristics of the Study Sample Variable Mother Father Parent's marital status, n (%) Never married 31 (30.4) 30 (29.4) Separated 7 (6.9) 7 (6.9) Divorced 5 (4.9) 7 (6.9) Married only once 52 (51) 50 (49) Remarried 7 (6.8) 8 (7.8) Parent's race/ethnicity, n (%) Asian Pacific Islander 2 (2) 2 (2) Latino 8 (7.8) 10 (9.8) African American 45 (44.1) 43 (42.2) White 45 (44.1) 47 (46) Other 2 (2) 0 Parent's education (median) post-high school (technical, trade, or business school) 1 year of college Parent's employment status, n (%) Yes 68 (66.7) 91 (89) No 34 (33.3) 11 (11) Weekly work hours, M (SD) 23.64 (19.06) 37.35 (17.63) Household income (median) $30,001-$40,000 Parent's age, M (SD) 35.35 (8.66) 38.7 (10.2) Father/father figure's relationship to the child, n (%) Biological father 87 (85.3) Stepfather 10 (9.8) Mother's paramour 5 (4.9) Father/father figure resides with the child, n (%) Yes 72 (70.6) No 30 (29.4) Age of oldest child, M (SD) 10.72 (3.10) was very similar. The median education of mothers was 1-year post-high school (i.e., job training), and the median education of fathers was 1 year of college. Approximately 67% of the female caregivers were employed at the time of the study, but 89% of fathers were employed. The median household income reported by mothers ranged from $30,001 to $40,000. The vast majority of the men in the study were biological fathers (85%). A small percentage of the men were stepfathers (10%), followed by a small number of paramours (5%). The mean age of the oldest child in the household was 10.72 years. 1028 JOURNAL OF FAMILY ISSUES / November 2003 INSTRUMENTS The family background questionnaire developed by the researchers for this study included the following items pertaining to mothers: marital status, age, race/ethnicity, education, employment status, work hours during a typical week, and household income. The family background questionnaire also included items about the father or father figure. Before responding to these items, mothers were asked to identify the adult male who is most significantly involved with her oldest child (this child was also required to be between ages 3 and 16years).Mothersweretheninstructedto think about this man and his relationship to and involvement with the oldest child when completing the questionnaires. Items pertaining to fathers included relationship to the children (i.e., stepfather, biological father, mother's partner), marital status, age, race/ethnicity, employment status, work hours during a typical week, residential status, education, and average amount of money provided to the family during a typical week. The Role of the Father Questionnaire (ROFQ) was used to measure the extent that mothers believe that the father's role is important to child development (Palkovitz, 1984). The ROFQ contains 15 items that solicit parents' beliefs about the ability of men to spend quality time interacting with children while meeting their psychological needs. A sample item is "Fathers play a central role in the child's personality development." Participants indicate their level of agreement or disagreement with each item on a 5-point scale. Higher scores reflect attitudes that fathers are capable and should be involved with and sensitive to their children. Palkovitz (1984) reported good internal consistency of the measure with an alpha of .77. The reliability coefficient (Cronbach's alpha) for the present study was .61. The Inventory of Father Involvement (IFI) was used to measure mothers' perceptions of fathers' abilities to relate competently to their children (Hawkins et al., 1999). Participants were asked to consider "how good a job" the most significant male performed on nine dimensions of father involvement, including discipline and teaching responsibility, school encouragement, mother support, providing, time and talking together, praise and affection, development of talents and future concerns, reading and homework support, and attentiveness. The authors of the measure indicate that all nine dimensions can be seen as indicators of a single, global construct of father involvement. For the purposes of this study, the total IFI score (the sum of all items in the nine dimensions) was calculated and used in all analyses. Participants ranked involvement for each item on a Likert-type scale, in which response choices were 0 through 6, 0 corre- Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1029 sponding to very poor and 6 corresponding to excellent, or NA (not applicable). Sample items include disciplining children; encouraging children to do their chores; setting rules and limits for children's behavior; teaching children to be responsible for what they do; paying attention to what children read, the music they listen to, or to the TV shows they watch; and enforcing family rules. Higher scores on the IFI suggested that fathers were more competent parents. The reliability coefficient (Cronbach's alpha) for all items in the IFI was .98. A survey instrument was developed to assess the degree to which mothers restrict access of their children to the father (gatekeeping). This instrument consisted of nine items, each addressing mothers' preferences for carrying out various child care tasks rather than permitting the father/ father figure to carry out the task. Sample items included "If my child needs to be disciplined, I think that I am the one to discipline him/her, and not the father/father figure" and "If an adult needs to talk to my child about his/her behavior, I think that I am the one to do the talking, and not the father/father figure." Each item in this instrument is rated on a Likert-type scale ranging from 1 {strongly agree) to 5 (strongly disagree). The items were reversed in scoring the instrument, such that higher scores reflected more gatekeeping. The appendix presents the results of a factor analysis of the scale. The reliability coefficient for the items in this scale was .93. Amount of father involvement was assessedby measuring fathers' participation in play, eating meals, helping with homework and reading, and providing solo child care to the oldest child in the family. Solo child care was defined as watching or staying with the children when the mother is not home. Mothers were first asked to indicate if the past week (past 7 days) was a typical week with respect to the amount of time that the father was involved with his children. If the past week was atypical, then the mother was instructed to think about the last typical week. Next, the respondent was asked to indicate on a grid how many hours the father spent with his child in each type of activity during each day of the week. For each activity, the number of hours was totaled over the course of a week. RESULTS DESCRIPTIVE ANALYSES On the average, fathers spent 15.47 hours per week providing solo child care to their oldest child (see Table 2). However, there was also a wide range in the number of hours of such care provided by men. Seventeen 1030 JOURNAL OF FAMILY ISSUES / November 2003 TABLE 2 Descriptive Statistics for the Major Study Variables Variable M SD Range Father's weekly financial contribution $639.50 $782.33 $0-$4,200 Paternal competence 152.14 42.04 35-209 Maternal attitudes about the father role 62.66 5.02 47-72 Maternal gatekeeping 28.17 9.24 9-45 Father's solo child care (hours/week) 15.47 17.90 0-90 Father's play (hours/week) 11.96 12.24 0-70 Father eats meals with children (hours/week) 6.92 9.84 0-30 Father reads or helps with homework (hours/week) 3.33 5.78 0-25 fathers provided no solo child care to the child, and one father spent 90 hours watching his child while the mother was not home. Fathers spent an average of 11.96 hours per week playing with the child, 6.92 hours eating with the child, and 3.33 hours helping with homework or reading to the child. The mean paternal competence score for fathers in this study was 152.14. This score, which converts to an average item score of 4.35, suggests that mothers perceived the men to have good parenting abilities. The average score for maternal attitudes about the father role was 62.66, or 4.17 per item. This score suggests that mothers moderately agree that fathers are important to the development of their children. Finally, the mean gatekeeping score for mothers was 28.17, or 3.13 per item. On the average, mothers neither agree nor disagree that they restrict fathers' involvement with the child. PRELIMINARY ANALYSES Factor analyses were conducted on the four father involvement variables for the purpose of data reduction. Standardized scores were first calculated on all measures. Principal components analysis was used as the extraction method. Table 3 shows that all father involvement variables loaded on one distinct factor, accounting for 59.18% of the variance in the data set. Therefore, a single father involvement variable was computed by adding the standardized scores of play, eating meals, reading and helping with homework, and providing solo child care. Because of the wide range of ages of children in the study (ages 3 to 16 years), we conducted a set of preliminary analyses to determine if the relationships between major variables in the study differed by the age of Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1031 TABLE 3 Results of Factor Analysis of Father Involvement Variables Variable Factor Loading Solo child care .74 Play .81 Eating meals .85 Helping with homework/reading .67 Eigenvalue 2.37 Percentage of variance 59.18 the oldest child in the family. The sample was divided into two groups— oldest child was between ages 3 and 9 years (n = 48) and oldest child was between ages 10 and 16 years (n = 54). Table 4 presents the correlation matrix for both samples. Pearson product moment correlations for the 10-to 16-year-olds are above the diagonal and correlations for the 3- to 9-year-olds are below the diagonal. Table 4 reveals numerous associations between study variables that are similar in direction and strength for families with younger and older children. The correlations are very similar for gatekeeping and amount of father involvement, paternal competence and gatekeeping, paternal competence and father involvement, attitudes about the father role and father involvement, and nonresidential father and paternal competence. One exception was the relationship between nonresidential father and maternal gatekeeping. Note that nonresidential father was a dummy-coded variable (residential father = 0, nonresidential father = 1). There was a stronger association between being a nonresidential father and gatekeeping among families in which the oldest child was between 10 and 16 years, r(53) = .54, p < .001, than among families in which the oldest child was between 3 and 9 years, r(47) = .30,/?<. 05. Fisher's z transformation function for Pearson's r correlation coefficient was then conducted to determine whether these correlations were significantly different from each other. The Fisher's z transformation statistic was 1.44, p > .05, suggesting that the correlations were not significantly different. These findings seem to suggest that it is appropriate to combine samples in subsequent analyses. BIVARIATE ANALYSES The correlation matrix for all participants in the study is presented in Table 5. The amount of fathers' financial contributions to the family was 1032 JOURNAL OF FAMILY ISSUES / November 2003 TABLE 4 Bivariate Analyses for Independent and Dependent Variables for Families With Older Children (10-16 years) and Younger Children (3-9 years) Variable 1 2 3 4 5 6 1. Father's financial contribution 1.00 -.11 .27* .25* -.19 -.09 2. Nonresidential father -.34* 1.00 -.12 _ 57*** 54*** -.07 3. Attitudes about fathers .16 -.20 1.00 .30** -.24* .24* 4. Paternal competence .38** _ 55*** .27* 1.00 _ 54*** .35** 5. Gatekeeping -.09 .30* -.21 _ 49*** 1.00 -.24* 6. Father involvement -.19 -.14 .20 41 ** -.27* 1.00 NOTE: Statistics for families with older children (« = 54) are above the diagonal, and statistics for families with younger children (« = 48) are below the diagonal. Nonresidential father: 0 = residential father, 1 = nonresidential father. *p < .05. **p < .01. ***p < .001. TABLE 5 Bivariate Analyses for All Families Variable 1 2 3 5 1. Father's financial contribution 2. Child welfare status 3. Nonresidential father 4. Attitudes about fathers 5. Paternal competence 6. Gatekeeping 7. Father involvement .00 -.37** * — 24** -.01 33*** -. 17* - 01 1.00 .11 -.19* -.06 .21* 02 1.00 -.17* _ 5g*** 4*7*** _ 12 1.00 .31** -.26** 24* 1.00 _ 55*** 1.00 1 35* 38* 00 NOTE: Child welfare status: 0 = family does not receive child welfare services, 1 =family receives child welfare services. Nonresidential father: 0 = residential father, 1 = nonresidential father. *p < .05. **p < .01. ***p < .001. positively related to maternal perceptions of father competence. That is. the more money the father contributed to the family, the more mothers perceived fathers to be capable parents. Fathers' financial contributions were negatively associated with being a recipient of child welfare services, being a nonresidential father, and maternal gatekeeping. Residential fathers were perceived to be more competent parents than nonresidential fathers. Mothers reported significantly higher levels of gatekeeping for nonresidential fathers than they did for residential fathers. Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1033 There was no significant association between residential status of the father and the amount of father involvement. There were a number of significant associations between child welfare status and the study variables. Child welfare status was dummy coded (0 = family does not receive child welfare services; 1 = family receives child welfare services). There was a positive association between receiving child welfare services and maternal gatekeeping, and there was a negative association between child welfare status and mothers' attitudes about the importance of fathers to children. Contrary to expectation, there was no significant relationship between child welfare status and mothers' reports of paternal competence. Mothers' attitudes about the importance of fathers to children were positively related to maternal perceptions of father competence. That is, mothers who had positive attitudes about the father role also rated the father as being a more competent parent. There was also a significant and positive relationship between mothers' attitudes about the father role and amount of father involvement. There was a significant and negative relationship between mothers' attitudes and maternal gatekeeping. Mothers who had positive attitudes about the father role were less likely to restrict the father's involvement with his child. Finally, there was a strong negative association between maternal perceptions of father competence and maternal gatekeeping. Mothers who rated fathers as being competent parents engaged in less gatekeeping behavior toward the men. Paternal competence was also positively related to amount of father involvement with the child. Maternal gatekeeping was negatively related to amount of father involvement. PATH ANALYSIS The theorectical model that was derived from the literature review includes causal linkages from the four exogenous variables (father's financial contributions, attitudes about the father role, paternal competence, and nonresidential father) to maternal gatekeeping. The model also includes linkages between paternal competence and father involvement, between nonresidential father and father involvement, and between mothers' attitudes about the father role and father involvement. Finally, maternal gatekeeping was causally linked to amount of father involvement. Path analysis was used to evaluate the model. In all cases, the coefficients presented are the standardized path coefficients obtained using AMOS (Arbuckle, 1999) with maximum likelihood estimation. The 1034 JOURNAL OF FAMILY ISSUES / November 2003 model fit the data quite well, %2(2) =l.2,p = .55. The comparative fit index (CFI) (Bentler, 1990), which always lies between 0 and 1 (unity), for the tested model was 1.0 (good fit is indicated by a value that is close to 1). Browne and Cudeck (1993) also recommend the use of the population discrepancy function to assess model adequacy. One such measure is the root mean square error of approximation (RMSEA). Browne and Cudeck suggest that a RMSEA value of .08 or less is a reasonable error of approximation. The RMSEA for the tested model was .00. However, the model also contained four nonsignificant paths, including the link between financial contributions and gatekeeping (ß = .02), the link between attitudes about the father role and gatekeeping (ß = -.09), the linkbetween attitudes about the father role and amount of father involvement (ß = .11), and the link between nonresidential father and amount of father involvement (ß =. 14). Duncan (1975) suggested refining the overall model (theory trimming) to produce a better fitting model. A more parsimonious path diagram showing one possible structure of the relations among variables is illustrated in Figure 1. The causal linkages in this model are from nonresidential father to maternal gatekeeping (ß = .22, p < .05) and from paternal competence to maternal gatekeeping (ß = -.42, p < .001). The revised model included a significant linkage between paternal competence and amount of father involvement (ß = .19). Finally, there was a direct path from maternal gatekeeping to father involvement (ß = -.24, p < .05). The model appeared to fit the data quite well. The chi-square goodness-of fit index was nonsignificant, %2(5) = 6.04, p = .30. The CFI for the tested model was .999. The RMSEA for the tested model was .043, suggestinga close fit of the model in relation to the degrees of freedom. The model predicted 34% of the total variance in maternal gatekeeping and 18% of the variance in amount of father involvement. ADDITIONAL TESTS OF MEDIATION Baron and Kenny's (1986) definition of mediation was also used to -examine whether maternal gatekeeping mediates the relationship between paternal competence and father involvement. Baron and Kenny have defined a mediator as a variable that accounts for the relation between a predictor and a criterion variable. For a variable to qualify as a mediator, an independent variable must have a direct effect on the suggested mediator and the dependent variable, the mediator must have a direct effect on the dependent variable, and the effect of the mediator on the dependent variable must reduce the direct effect of the independent variable on the dependent variable. Multiple regression analyses were conducted to test for Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1035 father's financial contribution .18 attitudes about fathers .12 father involvement paternal competence -^ maternal gatekeeping lonresidentia father Figure 1: Parsimonious Path Analysis NOTE: chi-square = 6.04, p = .30, with five degrees of freedom. mediation. Paternal competence was significantly associated with gatekeeping, ß = -.55, p < .001; and father involvement, ß = .35, p < .001. Gatekeeping was significantly related to father involvement, ß = —.38,/? < .001. Gatekeeping also reduced the effect of paternal competence on father involvement from .35 to .20, p = .05. Based on Baron and Kenny's definition of mediation, these findings suggest that maternal gatekeeping mediates the relationship between paternal competence and amount of fathers' involvement with children. DISCUSSION The purpose of the present study was to develop and test a model of factors associated with maternal gatekeeping and amount of father involve- 1036 JOURNAL OF FAMILY ISSUES / November 2003 ment with children. The findings suggested a significant, negative association between maternal gatekeeping and paternal involvement. The results of the most parsimonious path analysis revealed that nonresidential status of the father was directly linked to maternal gatekeeping. Father competence was directly linked to maternal gatekeeping and to amount of father involvement. Gatekeeping was causally linked to amount of father involvement. The findings of this study were consistent with other studies (i.e., Allen & Hawkins, 1999; DeLuccie, 1995) that have found a significant relationship between maternal gatekeeping and amount of father involvement. We note, however, that the total amount of variance in the dependent variable explained by maternal gatekeeping and two additional variables—paternal competence and attitudes about the father role—was fairly small (18%). One possible explanation is that mothers' preferences to restrict fathers' involvement with children are offset by other factors, such as having a multitude of responsibilities in the home, working in paid labor, or just being tired and needing someone to share in the tasks of caring for children. Some mothers may prefer to handle child-related matters yet still rely on fathers as a matter of necessity. It is also possible that maternal gatekeeping is more closely linked to father involvement variables not measured in this study. For example, Allen and Hawkins (1999) have suggested that maternal gatekeeping may be related to nontemporal measures of father involvement, such as commitment, care, and responsibility. The path analysis confirmed our hypothesis regarding the association between mothers' perceptions of paternal competence, gatekeeping behavior, and father involvement with children. Fathers' parenting abilities appear to have both a direct and indirect influence on father involvement. The test for mediation used in this study indicated that maternal gatekeeping reduced the association between father competence and father involvement by about 40% (from .35 to .20). These findings seem to suggest that mothers play a significant role in deciding how much time fathers spend with their children, depending on how the mother perceives the competence of the father. Our findings are also consistent with those of other researchers who have suggested a direct linkage between paternal competence and the man's amount of involvement with his children. Lamb (1986, 1997) has suggested that fathers who are more competent parents are also more motivated to spend time with their children. In essence, they spend time with their children because they are "good at it" and presumably find the involvement rewarding. Thus, our findings confirm the likelihood of a complex relationship between paternal competence and father involvement. Competent fathers spend more time with Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1037 their children. At the same time, mothers tend to exclude fathers from being involved with children when the mother perceives the father to have less parenting competence. We expected that maternal gatekeeping would mediate the relationship between maternal attitudes about the father role and amount of involvement in the same manner that gatekeeping was expected to mediate paternal competency. Instead, the path analysis revealed no significant associa-tionbetween maternal attitudes about the importance of the father role and gatekeeping and no significant relationship between maternal attitudes and amount of father involvement. DeLuccie (1995), who equated mothers' attitudes about the importance of father involvement with maternal gatekeeping, found a direct, although weak, link between attitudes and amount of father involvement and a direct, stronger link with satisfaction with father involvement. Our findings suggest that these variables (attitudes about the importance of fathers and gatekeeping) are two rather dissimilar constructs. Conceivably, mothers may feel strongly that fathers should play an important role in the lives of children, but they still may restrict fathers from being involved with their children. Indeed, the findings of the present study suggest that mothers' preferences for handling child-related matters without the involvement of the father are associated more with perceptions of the father's competence than they are with attitudes about the importance of fathers to children. We raise one cautionary note with regard to our data on maternal attitudes about the importance of the father role. The moderately low reliability of the maternal attitude measure (ROFQ) may have affected the lack of association between mothers' attitudes and maternal gatekeeping. We were particularly interested in this study to examine how maternal gatekeeping was associated with the residential status of the father. We hypothesized that residential status would have both a direct and an indirect association with amount of father involvement. Contrary to expectation, the results of this study indicated that residential status of the father was causally linked to mothers' gatekeeping, and there was no direct path to amount of father involvement. The lack of association between fathers' residential status and amount of involvement is inconsistent with the extensive research literature on single-parent families. Instead, our findings are more consistent with those of researchers such as Madden-Derdich and Leonard (2000) and Braver and O'Connell (1998) who have found that mothers have a good deal of influence over divorced fathers' family involvement. Our findings also support the notion advanced by a number of scholars who have suggested that fathering is a systemic phenomenon: Mothers influence men's 1038 JOURNAL OF FAMILY ISSUES / November 2003 involvement in their children's lives in important ways, whether or not they are married (Cummings & O'Reilly, 1997; Doherty et al., 1998; Parke, 1996). We expected to find a significant associationbetween fathers' financial contributions and maternal gatekeeping in the path analysis. Instead, our findings revealed no such relationship. However, there was a significant covarying relationship between fathers' financial contributions and mothers' perceptions of paternal competence and between financial contributions and nonresidential status of the father. Whereas mothers tended to view fathers who make smaller financial contributions as being less competent parents, we note that maternal perceptions of father competence played a more significant role in predicting maternal gatekeeping. We think that the relationships between financial contributions, maternal gatekeeping, and father involvement may be very complex and in need of further study. Conceivably, fathers' financial contributions may have different meanings depending on whether the father resides with his children or not. We suggest the use of qualitative research methods to better understand these family processes. A subsample of families receiving child welfare services was included in this study to ensure that there would be a range of parenting competencies among the fathers. Although child welfare involvement was not included in the model under investigation, we still expected to find several differences between recipients of those services and other families. Contrary to expectation, mothers did not perceive the fathers in this subsample to be less competent parents than did mothers in the nonchild welfare group. However, mothers receiving child welfare service reported more negative attitudes about the importance of fathers to children and they reported more gatekeeping behavior. To a great extent, the child welfare field has focused on the mother and her responsibilities to provide an adequate level of care to her children. More needs to be known about the roles these fathers play in caring for their children and the degree to which child welfare social workers should work with these men to improve children's well-being. There were several noteworthy limitations to the present study. One limitation was that data regarding amount of father involvement were only obtained from maternal reports. This approach raises the risk of shared method variance in the measurement of family processes because mothers also reported on their own gatekeeping behavior, attitudes about fathers, and paternal competency. An alternative approach would be to obtain more objective measures of father involvement. The findings of the Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1039 present study also cannot be used to make generalizations to the larger population inasmuch as data were obtained using a convenience sample. Finally, caution should be exercised about inferring causal relationships based on correlational data. Future research should use longitudinal or experimental research designs to explore the ways in which maternal gatekeeping, mothers' attitudes about fathers, and paternal competence are causally related to amount of father involvement with children. Gatekeeping has been an understudied topic. The findings of the present study suggest that maternal gatekeeping is a significant coparenting variable that is relevant to understanding father involvement. Of particular interest is that mothers are more likely to restrict the father's involvement with children when she perceives the man to have low parenting competence. Taken together, the findings of the present study suggest the importance of thinking about fathers from a family systems perspective. Future research will need to examine more closely the processes through which maternal gatekeeping influences father involvement. Walker and McGraw (2000) have suggested that it will be important to examine maternal gatekeeping within the context of mothers' facilitation of father involvement and men's authority in families. APPENDIX Factor Analysis of the Maternal Gatekeeping Scale Item Factor Loading 1. If my child(ren) need to be disciplined, I think that I am the .82 one to discipline them, not their father (father figure). 2. If a choice has to be made about what clothing my child(ren) .62 will wear, I think that I am the one to make that decision, not their father (father figure). 3. If someone needs to talk with my child(ren)'s teacher, I am the .75 one to do it, not their father (father figure). 4. If my child(ren)'s feelings are hurt, I think that I should com- .83 fort them, not their father (father figure). 5. If my child(ren) have to go to the doctor, I think that I am the .77 one to take them, not their father (father figure). 6. If a decision has to be made about who my child(ren) will play .86 with (or spend time with), I think that I am the one to make that decision, not their father (father figure). 7. If a decision has to be made for my child(ren), I think that I am .89 the one to make it, not their father (father figure). (continued) 1040 JOURNAL OF FAMILY ISSUES / November 2003 APPENDIX (continued) Item Factor Loading 8. If an adult needs to talk to my child(ren) about their behavior, I .81 think that I am the one to do the talking, not their father (father figure). 9. If a decision has to be made about which TV shows my .80 child(ren) should watch, I think that I am the one to make that decision, not their father (father figure). NOTE: The extraction method was principal components analysis. Eigenvalue = 5.70. Percentage of variance explained by the nine items = 63.27. REFERENCES Ahrons, C. R. (1981). The continuing coparental relationship between divorced spouses. American Journal of Orthopsychiatry, 5, 415-428. Ahrons, C. R., & Miller, R. B. (1993). The effect of the postdivorce relationship on paternal involvement: A longitudinal analysis. American Journal of Orthopsychiatry, 63, 441-450. Ahrons, C. R., & Wallisch, L. (1987). The relationship between former spouses. In D. Perlman& S. Duck(Eds.), Intimate relationships: Development, dynamics, anddete-rioration (pp. 269-296). Newbury Park, CA: Sage. Allen, S. M., & Hawkins, A. J. (1999). Maternal gatekeeping: Mothers'beliefs and behaviors that inhibit greater father involvement in family work. Journal of Marriage and the Family, 61, 199-212. Amato, P. R. (2000). The consequences of divorce for adults and children. Journal of Marriage and the Family, 62, 1269-1287. Amato, P. R., & Keith, B. (1991). Parental divorce and the well-being of children: A metaanalysis. Psychological Bulletin, 110, 26-46. Aquilino, W. S. (1994). Impact of childhood family disruption on young adults' relationships with parents. Journal of Marriage and the Family, 56, 295-313. Arbuckle, J. L. (1999). Amos 4 [Computer software]. Chicago: SmallWaters. Arendell, T (1996). Co-parenting: A review of the literature (LR-CP-96-03). Philadelphia: National Center on Fathers and Families. Baber, K. M., & Monaghan, P. (1988). College women's career and motherhood expectations: New options, old dilemmas. Sex Roles, 19, 189-203. Barnett, R. C, & Baruch, G. K. (1987). Determinants of fathers' participation in family work. Journal of Marriage and the Family, 49, 29-40. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Beitel, A. H., & Parke, R. D. (1998). Paternal involvement in infancy: The role of maternal and paternal attitudes. Journal of Family Psychology, 12, 268-288. Belsky, J. (1979). The interrelation of parental and spousal behavior during infancy in traditional nuclear families: An exploratory analysis. Journal of Marriage and the Family, 41, 749-755. Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1041 Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238-246. Bonney, J. F, Kelley, M. L., & Levant, R. F. (1999). A model of fathers' behavioral involvement in child care in dual-earner families. Journal of Family Psychology, 13, 401-415. Braver, S. L., & Griffin, W.A. (2000). Engaging fathers in the post-divorce family. Marriage and Family Review, 29, 247-267. Braver, S. L., & O'Connell, D. (1998). Divorced dads: Shattering the myths. New York: Putnam. Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136-162). Newbury Park, CA: Sage. Cherlin, A. J. (1992). Marriage, divorce, and remarriage. Cambridge, MA: Harvard University Press. Cohen, R. S., & Weissman, S. H. (1984). The parenting alliance. In R. Cohen, B. Cohler, & S. Weissman (Eds.), Parenthood: A psychodynamicperspective (pp. 33-49). New York: Guilford. Coleman, J. (1990). Foundations of social theory. Cambridge, MA: Harvard University Press. Coltrane, S. (1996). Familyman: Fatherhood, housework, and gender equity. New York: Oxford University Press. Cummings, E. M., & O'Reilly, A. W. (1997). Fathers in family context: Effects of marital quality on child adjustment. In M. E. Lamb (Ed.), The role of the father in child development (Vol. 3, pp. 49-65). New York: John Wiley. DeLuccie, M. F. (1995). Mothers as gatekeepers: A model of maternal mediators of father involvement. The Journal of Genetic Psychology, 156, 115-131. Deutsch, F. M., Lussier, J. B., & Servis, L. J. (1993). Husbands athome: Predictors of parental participation in childcare and housework. Journal of Personality and Social Psychology, 65, 1154-1166. Doherty, W. J., Kouneski, E. F, & Erickson, M. F. (1998). Responsible fathering: An overview and conceptual framework. Journal of Marriage and the Family, 60, 277-292. Duncan, O. D. (1975). Introduction to structural equation models. New York: Academic Press. Fagan, J., Newash, N, & Schloesser, A. (2000). Female caregivers' perceptions of fathers' and significant adult males' involvement with their Head Start children. Families in Society, 81, 186-196. Furstenberg, F. F., Jr. (1990). Divorce andthe American family. Annual Review of Sociology, 16,379- 403. Furstenberg, F. F., & Cherlin, A. (1991). Divided families: What happens to children when parents part. Cambridge, MA: Harvard University Press. Gable, S., Belsky, J., & Crnic, K. (1995). Coparentingduringthe child's 2nd year: Adescrip-tive account. Journal of Marriage and the Family, 57, 609-616. Greif, G. L. (1997). Out of touch: When parents and children lose contact after divorce.'New York: Oxford University Press. Greif, G. L. (2001). When a divorced parent does not visit. In J. Fagan & A. J. Hawkins (Eds.), Clinical and educational interventions with fathers (pp. 67-90). Binghamton, NY: Haworth. Hawkins, A. J., Bradford, K. P., Christiansen, S. L., Palkovitz, R., Call, V R. A., & Day, R. D. (1999, November 12). The Inventory of Father Involvement: A pilot study of a new measure of father involvement. Paper presented at the National Council on Family Relations Annual Conference, Irvine, CA. 1042 JOURNAL OF FAMILY ISSUES / November 2003 Hawkins, A. J., Marshall, C. M., & Meiners, K. M. (1995). Exploring wives sense offairness about family work. Journal of Family Issues, 16, 693-721. Hochschild, A. (1989). The second shift. New York: Avon Books. Ihinger-Tallman, M., Pasley, K., & Buehler, C. (1993). Developing a middle-range theory of father involvement post-divorce. Journal of Family Issues, 14, 550-571. Jouriles, E. N, Murphy, C. M., Farris, A. M., Smith, D. A., Richters, J. E., & Waters, E. (1991). Marital adjustment, parental disagreements about child rearing, and behavior problems in boys: Increasing the specificity of the marital assessment. Child Development, 62, 1424-1433. Kalmijn, M. (1999). Father involvement in childrearing and the perceived stability of Marriage. Journal ofMarriage and the Family, 61, 409-421. King, V., & Heard, H. E. (1999). Nonresident father visitation, parental conflict, and mother's satisfaction: What's best for child well-being? Journal ofMarriage and the Family, 61, 385-396. Klein, D. M., & White, J. M. (1996). Family theories: An introduction. Thousand Oaks, CA: Sage. Kranichfeld, M. L. (1987). Rethinking family power. Journal of Family Issues, 8, 42-56. Lamb, M. E. (1986). The changing role of fathers. In M. E. Lamb (Ed.), The father's role: Applied perspectives (pp. 3-27). New York: John Wiley. Lamb, M. E. (1997). The development of father-infant relationships. In M. E. Lamb (Ed.), The role of the father in child development (3rd ed., pp. 104-120). New York: John Wiley. LaRossa, R. (1997). The modernization of fatherhood. Chicago: University of Chicago Press. Lewis, M., Feiring, C, & Weinraub, M. (1981). The father as a member of the child's social network. In M. E. Lamb (Ed.), The role of the father in child development (2nd ed., pp. 259-294). New York: John Wiley. Maccoby, E. E., & Mnookin, R. H. (1992). Dividing the child: Social and legal dilemmas of custody. Cambridge, MA: Harvard University Press. Madden-Derdich, D. A., & Arditti, J. A. (1999). The ties that bind: Attachment between former spouses. Family Relations, 48, 243-249. Madden-Derdich, D. A., & Leonard, S. A. (2000). Parental role identity and fathers' involvement in coparental interaction after divorce: Fathers' perspectives. Family Relations, 49, 311-318. Madden-Derdich, D. A., Leonard, S. A., & Christopher, F. S. (1999). Boundary ambiguity and parental conflict after divorce: An empirical test of a family systems model of the divorce process. Journal ofMarriage and the Family, 61, 588-598. Malkin, C. M.,&Lamb, M. E. (1994). Child maltreatment: A test of sociobiological theory. Journal of Comparative Family Studies, 25, 121-134. Marsiglio, W. (1995, December 11). Fathers'identity, commitment, and care: Partners mediating role. Paper presented at the Fathers and Families Roundtable on Father Care, Philadelphia, National Center on Fathers and Families. Marsiglio, W., Amato, P., Day, R. D., & Lamb, M. E. (2000). Scholarship on fatherhood in the 1900s and beyond. Journal ofMarriage and the Family, 62, 1173-1191. McBride, B. A., & Rane, TR. (1998). Parenting alliance as a predictor of father involvement: An exploratory study. Family Relations, 47, 229-236. McCurdy, K. (1995). Risk assessment in child abuse prevention programs. Social Work Research, 19, 77-88. Minuchin, S. (1974). Families and family therapy. Cambridge, MA: Harvard University Press. Fagan, Barnett / GATEKEEPING AND FATHER INVOLVEMENT 1043 Minuchin, S., & Fishman, H. C. (1981). Family therapy techniques. Cambridge, MA: Harvard University Press. Moncher, F. J. (1995). Social isolation and child-abuse risk. Families in Society, 76,421-433. Naidoo, S. (2000). A profile of the oro-facila injuries in child physical abuse at a children's hospital. Child Abuse and Neglect, 24, 521-534. Nichols, M. P., & Schwartz, R. C. (1998). Family therapy: Concepts and methods (4th ed.). Boston: Allyn & Bacon. Palkovitz, R. (1984). Parental attitudes and father's interactions with their 5-month-old infants. Developmental Psychology, 20, 1054-1060. Parke, R. D. (1996). Fatherhood. Cambridge, MA: Harvard University. Rutter, V, & Schwartz, P. (2000). Gender, marriage, and diverse possibilities for cross-sex and same-sex pairs. In D. H. Demo, K. R. Allen, & M. A. Fine (Eds.),Handbookof'family diversity (pp. 59-81). New York: Oxford University Press. Seltzer, J. A. (1991). Relationship between fathers and children who live apart: The father's role after separation. Journal ofMarriage 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. Sinai, S. H, Petree, A. R., Herman-Giddens, M., Rogers, M. K., Enand, C, & DurRant, R. H. (2000). Is race or ethnicity a predictive factor in shaken baby syndrome? Child Abuse and Neglect, 24, 1241-1246. Sorenson, E., & Zibman, C. (2001, April). Poor dads who don't pay child support: Dead-heats or disadvantaged? (Series B, No. B-30). Washington, DC: The Urban Institute. Stewart, S. D. (1999). Nonresident mothers' and fathers' social contact with children. Journal ofMarriage and the Family, 61, 894-907. Vaughn, B. E., Block, J. H, & Block, J. (1988). Parental agreement on child rearing during early childhood and the psychological characteristics of adolescents. Child Development, 59, 1020-1033. Walker, A. J., & McGraw, L. A. (2000). Who is responsible for responsible fathering? Journal of Marriage and the Family, 62, 563-569.