|Home | About | Journals | Submit | Contact Us | Français|
Using multi-informant data from 134 two-parent African American families, the goals of this study were to (a) describe parent – adolescent warmth and shared time as a function of parent and youth gender and (b) assess links between these indices of relationship quality and adolescent adjustment. Mixed-model ANCOVAs revealed that mothers reported warmer relationships with adolescents than fathers, and both parents reported warmer relationships with younger versus older offspring. Interparental differences in time spent with sons and daughters and older and younger siblings were also found. Tests of multilevel models indicated that greater maternal warmth was associated with fewer depressive symptoms and less risky behavior for sons, and more paternal warmth and shared time with fathers were associated with less risky behavior in youth. Discussion highlights the utility of cultural ecological and family systems perspectives for understanding parent-adolescent relationships and youth adjustment in African American families.
Maintaining strong family bonds during the renegotiation of parent – child relationships in adolescence is paramount for youth adjustment (Steinberg, 2001). Parent – adolescent relationships characterized by high warmth and involvement may protect youth from adjustment problems (Carlton-Ford, Paikoff, Oakley, & Brooks-Gunn, 2008; Galambos, Barker, & Almeida, 2003). Whereas most research on this topic is based on between-family comparisons of youth, family systems theory calls attention to within-family variability. From a systems perspective, families comprise interdependent individuals who make unique contributions to family dynamics and who have different experiences in and perceptions of their shared family context (e.g., Minuchin, 1985). Although a systems perspective holds that mothers and fathers will make unique contributions to their sons’ and daughters’ development, empirical research on the roles of gender for parent – adolescent relationships is sparse, and we know little about the differential experiences of boys and girls who grow up in the same family (Galambos, Berenbaum, & McHale, 2009).
Even more limited is research that extends a family systems approach to ethnic minority families. Given the significance of cultural values and experiences in shaping family dynamics, findings from the literature on White families may not generalize to other sociocultural groups (Garcia Coll et al., 1996). A cultural ecological perspective also directs attention to the substantial variability in family dynamics and youth adjustment that exists within a particular ethnic group and emphasizes the importance of ethnic homogeneous designs wherein the processes underlying within group variability can be illuminated.
Researchers have generally ignored the substantial degree of variability in the circumstances of African Americans, instead focusing primarily on families in challenging circumstances. For instance, although nearly 40% of African American youth live with both parents (Kreider & Ellis, 2009), most research includes only single-mother households. When African American fathers are included, the focus has primarily been on the implications of fathers’ residential status for youth development. Although fathers’ residency contributes to adolescent adjustment, residing with a father does not necessarily imply strong father – child bonds (Salem, Zimmerman, & Notaro, 1998; Thomas, Krampe, & Newton, 2008), and more research is needed examining the implications of father – adolescent relationship quality for youth outcomes.
Building on family systems and cultural ecological tenets, the goals of this study were to (a) describe parent – adolescent warmth and shared time as a function of parent and youth gender and (b) assess links between these indices of relationship quality and adolescent internalizing and externalizing behavior, including the moderating role of parent and adolescent gender and birth order. In an effort to understand the potentially unique roles of mothers and fathers in their children’s adjustment, we used data from a sample of two-parent African American families. Much of the prior literature on this topic has been subject to mono-reporter biases, which may be particularly problematic during adolescence (Dekovic & Buist, 2005). Thus, in addition to its conceptual contributions, this study improves on prior study designs by incorporating information from four family members. Finally, this study’s focus on the role of potentially modifiable family dynamics in youth adjustment may have important applied implications for youth and family prevention and intervention programs (Cauce, Cruz, Corona, & Conger, 2011).
Research on White samples indicates that, during adolescence, youth begin to identify more strongly with their same-gender parent, and that the quality of this relationship has implications for their adjustment (Crouter, Manke, & McHale, 1995). Thus, researchers have called for greater attention to the role of parent and child gender in research on family dynamics (McHale, Crouter, & Whiteman, 2003). Issues of gender may be particularly relevant for African American families: A popularized notion holds that African American mothers “raise” their daughters by providing rules and discipline, but “love” their sons, by displaying warmth and permissiveness (Hill & Zimmerman, 1995; Mandara, Murray, & Joyner, 2005). Hill (2001) suggested that gendered differential treatment may occur because mothers’ heightened concerns for their sons’ safety and the barriers to success experienced by African American men manifest themselves in higher expectations for daughters and greater tolerance for sons’ misbehavior.
Systematic data on African American parents’ gendered differential treatment of their children, however, are rare. Consistent with parenting in other racial/ethnic groups (Smetana & Daddis, 2002), there is evidence that African American parents engage in more monitoring of daughters than sons (Kerr & Stattin, 2000). In a study by Hill and Zimmerman (1995), African American mothers with chronically ill sons were found to treat their children more favorably than mothers with chronically ill daughters. Although these findings cannot be generalized to a healthy sample, the authors suggested that small size and physical disability were viewed by mothers as more debilitating for boys than girls in the context of greater constraints on the adult roles of African American men. Finally, Mandara, Varner, and Richman (2010) found evidence of an interaction between youth gender and birth order for mothers’ behaviors; compared to second-born daughters, second-born sons had fewer responsibilities, less cognitive stimulation, and more mother – child conflict.
In order to advance understanding of gendered differential treatment in African American families, two methodological constraints of previous work in this area must be addressed. First, though it has been suggested that African American mothers are warmer toward sons than daughters but that fathers act in the opposite way (Mandara, Murray, & Joyner, 2005), few studies have actually examined differential treatment in two-parent African American families. Second, empirical investigations into gender differences in African American parents’ treatment of offspring have largely relied on between-family comparisons of parents with sons versus parents with daughters. As family scholars have noted, within-family comparisons of mothers’ versus fathers’ warmth toward their sons versus their daughters may better illuminate family gender dynamics (McHale et al., 2003). Accordingly, given limitations in existing evidence, we chose not to advance a specific hypothesis regarding African American parents’ warmth toward sons versus daughters in two’parent families. While gendered differential treatment may be evident in these families, it also may be that similar to European American families, African American parents’ treatment of boys and girls does not differ significantly (Lytton & Romney, 1991). On the other hand, consistent with past research on European Americans (McKinney & Renk, 2008) and the role of mothers as kin-keepers across multiple ethnic groups (Turner, Young, & Black, 1996), we predicted that mothers would report warmer relationships with their children than fathers.
In addition to parental warmth, we also examined gender differences in parents’ shared time with adolescents. Evidence from White families revealed that, in the face of mothers’ overall higher level of involvement (Phares, Fields, & Kamboukos, 2009), parents spend more time with same-gender compared to opposite-gender adolescent offspring (McHale, Crouter, & Tucker, 1999). Our knowledge of African American parents’ time with offspring comes almost exclusively from studies of families with infants and young children (Hossain & Rooparnaine, 1993; Roopnarine, Fouts, Lamb, & Lewis-Elligan, 2005). Results from this work suggest that African American parents’ involvement in child care largely mirrors that of their White counterpartsl although fathers were involved with their children, they were less involved than mothers. Building on this work, we used within-and between-family comparisons to examine African American parents’ shared time with adolescents. Given findings about parents’ time with adolescents from White samples, in conjunction with evidence that African American parents’ resemble White parents in the amount of time they spend with infants and young children, we tested the predictions that (a) mothers would spend more time with adolescent offspring than fathers, and (b) parents would spend more time with same-gender offspring.
Research based primarily on White samples revealed clear links between parent – adolescent relationship quality, including parental warmth and involvement, and adolescent adjustment (Carlton-Ford, et al. 2008; Galambos et al., 2003). Although these associations are receiving more attention among African American parents and children, much remains unknown about the adolescent adjustment implications of parent – adolescent relationships, particularly father – adolescent relationships, for racial/ethnic minority youth (Coley, 2001). Furthermore, when racial/ethnic minority families have been considered, there have been mixed findings with regard to gender (Bowman, Prelow, & Weaver, 2007), with some work suggesting that associations between parent – adolescent relationship quality and adolescent adjustment differ for mothers and fathers and their sons and daughters and some suggesting uniformity across parent and child gender.
Research on parent – adolescent relationship qualities and internalizing behaviors in African American families has largely been constrained to examination of one parent or to youth of one gender. For example, in a study of African American mothers, Bynum and Kotchick (2006) found that youths’ reports of maternal relationship quality were positively related to their self-esteem and negatively related to their depressive symptoms. Focusing on African American fathers and daughters, Cooper (2009) determined that relationship quality was positively related to self-esteem. Using data from one primary caregiver (89% mothers) and adolescents, Sagrestano, Paikoff, Holmbeck, and Fendrich (2003) provided compelling longitudinal evidence that positive parent – adolescent relationships were related to fewer depressive symptoms in African American youth. Finally, in a multiracial study of the effects of fathers’ residence, Booth, Scott, and King (2010) found that close relationships with fathers were more influential for adolescents’ internalizing behavior than was family structure. Maternal relationship quality was controlled for but otherwise was not a focus of the research. Taken together, these findings led us to predict that mother – and father – adolescent relationship warmth and shared time would be negatively related to internalizing symptoms.
Fathers’ encouragement of autonomy, in conjunction with their more peer-like relationships with their offspring (relative to mothers’), are thought to facilitate the development of adolescents’ adjustment in the world beyond the family (Parke & Buriel, 1998). Evidence supporting this theory in African American samples can be gleaned from work on the effects of father involvement. For instance, the greater risk of externalizing behavior seen in African American boys relative to girls is considerably lessened when fathers are present (Mandara & Murray, 2006; Mandara et al., 2005). Others have found father – adolescent relationship quality in African American samples is protective against adolescents’ alcohol use (Booth et al., 2010; Caldwell, Sellers, Bernat, & Zimmerman, 2005). The role of mother – adolescent relationship quality in youth externalizing behaviors is less clear (Bowman et al., 2007). Given this pattern of findings, we tested the prediction that paternal warmth and shared time would be negatively associated with youths’ risky behavior and that these associations would be more consistent than those for mother – adolescent relationships and youths’ risky behavior.
The data came from mothers, fathers, and two adolescent-age siblings in 134 families who participated in a study of African American family relationships (McHale et al., 2006). Families that self-identified as African American or Black, included a mother and father figure, and who lived together with at least two offspring were targeted for recruitment. Two strategies were used to generate the sample in two urban centers in the mid-Atlantic region: First, African American community members recruited about half of the sample by advertising the study in their communities. Second, families who had been identified using a purchased marketing list were sent letters describing the study and were asked to respond by postcard or toll-free number if they were eligible and interested.
Given our interest in adolescent internalizing and externalizing outcomes, we focused on wave three of the study when relevant measures were collected. Of the original 202 families participating in the first phase of the larger study, 10 families in which one parent was not African American, 8 coparents who were not in a couple relationship, and 22 families in which parents divorced across the study period were omitted; data were also unavailable for 9 families that withdrew and 10 fathers who did not participate. Additionally, we omitted nine families with siblings age 20 or older. Of the 134 families in the present analyses, mean ages of mothers and fathers were 43.08 (SD = 5.66) and 45.63 (SD = 6.86), respectively. The majority of families (>80%) included two or three children. Mothers’ education averaged 14.79 (SD = 1.88) years and fathers’ education, 14.35 (SD = 2.39) years, with a score of 12 signifying high school graduate, 14 signifying some college, and 16 signifying a bachelor’s degree. Given the high correlation between mothers’ and fathers’ education, r = .45, the mean was included as a control. Most parents were employed (80% of mothers and 89% of fathers) and worked full-time hours (mothers, M = 32.04, SD = 18.02; fathers, M = 42.20, SD = 19.23). Family income (mothers’ plus fathers’) averaged $99,931.86 (SD = $81,568.24), with two families reporting no income for the past year and eight reporting incomes over $200,000. With respect to couple relationships, 14 couples had been cohabiting for at least two years (M = 9.29, SD = 6.39), and the rest had been married an average of 16.48 years (SD = 6.75).
Two adolescent siblings from each family also participated. Older adolescents averaged 15.87 years (SD = 1.69) and younger adolescents averaged 12.48 years (SD = 1.11) of age, the majority were biologically related to mothers (92 % of older and 96 % of younger adolescents) and fathers (76% of older and 82% of younger adolescents), and the sample was approximately equally divided by gender (51% girls) and dyad gender constellation (47% same gender).
Two procedures were used for data collection. First, family members were interviewed individually in their homes by a team of two interviewers, almost all of whom were African American. Family members reported on relationship experiences and personal characteristics during the past year. Interviews lasted about 2 hours for parents and 1 hour for adolescents. Following the completion of interviews, families received a $200 honorarium.
The second data collection procedure was used to obtain information about adolescents’ daily activities. In the month following the home interview, seven telephone interviews were conducted (five on weekday evenings, two on weekend evenings). During these calls, youth reported on their daily activities outside of regular school hours; the type of activity, how long the activity lasted, and with whom they had engaged in each activity. Calls were scheduled in the evening.
Parent – adolescent relationship warmth was reported by mothers and fathers using an 8-item (e.g., “I am a person who makes my child feel better after talking over his/her worries with me”) 5-point rating scale (1 = not at all to 5 = very much) from the parents’ version of the Child’s Report of Parental Behavior Inventory (CRPBI; Schaefer, 1965). Higher scores represented greater warmth, and alphas ranged from .87 to .89. Parent – adolescent shared time was measured using adolescent reports from the seven nightly telephone interviews. We aggregated youth reports across all activities and all seven calls to create a measure of the time (in minutes) that adolescents spent with their mothers and with their fathers.
Depressive symptoms were measured using the Children’s Depression Inventory (CDI; Kovacs, 1985). For 26 items (given institutional review board issues, the item on suicide ideation was dropped), adolescents were asked to choose the statement that best described them over the past week (e.g., “I am sad once in a while,” “I am sad many times,” “I am sad most times”). Items were summed, with higher scores indicating more depressive symptoms. Alphas were .78 for older and .72 for younger adolescents. Risky behaviors were assessed using the 18-item Risky Behavior Scale (Eccles & Barber, 1990). Older and younger adolescents reported on their involvement in risky behaviors (e.g., “Drink alcohol without your parents’ permission,”) using a scale of 1 (never) to 4 (more than 10 times in the past year). Items were summed, with higher scores indicating more risky behavior. Alphas were .86 for older and .78 for younger adolescents.
The results are organized by our aims: (1) to describe parent – adolescent warmth and shared time as a function of parent and youth gender and (b) to assess links between these indices of relationship quality and adolescent adjustment, including the moderating role of parent and adolescent gender and birth order. Parent – adolescent warmth, shared time, depressive symptoms, and risky behavior were square-root transformed in all models due to skewness. Given that youth age, parents’ education, and biological relatedness may be confounded with the independent variables, they were controlled in the analyses (Amato & Booth, 1996).
Parent – adolescent relationship quality was assessed using 2 (older sibling gender) × 2 (younger sibling gender) × 2 (parent gender) × 2 (birth order) mixed model ANCOVAs with parental warmth and parent – youth shared time as dependent variables (in separate models). Parent gender and sibling were within-groups factors, sibling genders were between-groups factors, and parents’ education and adolescents’ age were entered as covariates. Biological status of children was tested as a covariate but was never significant and thus was removed from the final models. Means and standard deviations for study variables are presented in Table 1.
Untransformed means suggested that parental warmth was quite high, averaging greater than 4 on a 5-point scale in all cases. Analyses revealed no significant differences in parent –adolescent warmth as a function of older siblings’ gender, younger siblings’ gender, and gender constellation of the sibling dyad. The sibling effect trended toward significance, F(1, 124) = 3.13, p < .10, and there was a significant effect of parent gender, F(1, 124) = 4.13, p < .05. As Table 2 Illustrates, parents tended to report warmer relationships with younger relative to older offspring and mothers reported warmer relationships with adolescents than did fathers.
Untransformed means for parent – adolescent shared time revealed that mothers spent about 8 to 10 hours a week with adolescents and fathers spent about 6 to 7 hours a week with adolescents. Parent – adolescent shared time did not differ as a function of older siblings’ gender, younger siblings’ gender, or gender constellation of the sibling dyad, and here the within-family effects of parent and birth order also were nonsignificant. Two significant interactions emerged, however, between parent and older sibling gender, F(1, 108) = 4.11, p < .05, and between parent and younger sibling gender, F(1, 108) = 7.58, p < .01. To follow up these interactions, two t tests were conducted to compare the difference between mothers’ and fathers’ time with older and with younger siblings as a function of adolescents’ gender. These revealed that the difference between mothers’ and fathers’ time with youth was greater if the older sibling was girl (M = 7.45, SD = 6.92) compared to a boy (M = 1.17, SD = 8.30), t(116) = 4.45, p < .001. The mother – father difference was also greater when the younger sibling was a girl (M = 8.56, SD = 7.89) versus a boy (M = 2.13, SD = 7.99), t(125) = 4.56, p < .001. Positively signed means indicate that mothers spent more time with children than fathers, on average, but that this difference was pronounced when the family included a daughter and is suggestive of mothers’ gender-typed differential treatment of offspring (Table 3).
Differences between mothers’ and fathers’ time with sons and daughters were further qualified by 3-way Parent × Birth Order × Older Sibling Gender, F(1, 109) = 20.02, p < .001, and Parent × Birth Order × Younger Sibling Gender, F(1, 109) = 17.95, p < .001, interactions. To test the gender component of these interactions we conducted two t tests, comparing mother – father differences in time spent with older and younger siblings as a function of older and younger adolescents’ gender (see Figure 1). When the older sibling was a boy, the difference between mothers’ time with the younger compared to the older sibling (favoring the younger) was greater than the difference between fathers’ time with the younger versus the older sibling. Further, when the younger sibling was a girl, the difference between mothers’ time with the younger compared to the older sibling (favoring the younger) was greater than the difference between fathers’ time with the younger versus the older sibling. In both cases, mothers’ as compared to fathers’ tendency to be less involved with older sons and more involved with younger daughters indicates interparental (gendered) differental treatment of children.
To address our second aim—exploring links between parent – adolescent relationship qualities and adolescent depressive symptoms and risky behavior—we used a multilevel modeling (MLM) strategy. This approach extends multiple regression to account for dependencies in the data (i.e., between siblings and between parents in the same family). In addition, an MLM framework provides for the use of cases with missing data (Raudenbush & Byrk, 2002). We tested separate two-level random intercept models with family as the unit of analysis for the outcomes of depressive symptoms and risky behavior. The Level 1 model accounts for dependencies between members of the same family; here we included within-family predictors that differed for older and younger siblings (i.e., mother – adolescent warmth, father – adolescent warmth, mother – adolescent shared time, and father – adolescent share time, and gender of sibling). Level 2 estimates, including parents’ education and adolescents’ age, are characteristics shared by youth in the same family but differ across families. By including mothers, fathers, and older and younger sons and daughters in the same analyses we were able to test the moderating roles of parent and adolescent gender and birth order on relationship-adjustment linkages.
In the case of significant interactions by gender, to test the slopes, we reran the same models treating girls and older siblings as the reference group (i.e., dummy coding instead of effect coding). Parent education and offspring age were included as controls in all models (see Table 4).
Beginning with depressive symptoms, main effects indicated that girls trended toward higher depressive symptoms than boys, but that there was no effect of birth order. This analysis also revealed that maternal warmth was negatively associated with adolescent reports of depressive symptoms. Neither fathers’ warmth nor mothers’ or fathers’ shared time with adolescents was related to adolescents’ depressive symptoms.
Turning to risky behavior, boys engaged in risky behavior more frequently than girls. Additionally, adolescent age was positively related to risky behavior. Findings also showed that paternal warmth and shared time were negatively related to adolescents’ risky behavior. Maternal warmth was significantly and negatively related to risky behavior, but this effect was qualified by gender. Follow-up tests revealed that maternal warmth was unrelated to daughters’ risky behavior, β = − .88, SE = 3.56, t = − .25, ns, but negatively related to sons’, β = − 10.42, SE = 3.78, t = − 2.76, p < .05. Mothers’ shared time with adolescents was unrelated to their risky behavior, and there were no interactions involving gender or birth order.
Responding to the call for research on normative processes in African American families, this study explored parent – adolescent relationship qualities and their associations with adolescent adjustment in two-parent families. The design of this study was in keeping with a key tenet from the cultural ecological perspective on the importance of illuminating factors that explain the substantial within-group variability in minority youth’s adjustment. Finally, our research addressed the popularized notion that African American mothers are warmer with sons (Hill & Zimmerman, 1995; McLoyd, 1990), and that in two-parent families, fathers work to balance this pattern (Mandara et al., 2005).
Regarding differences in parental warmth, results revealed no evidence that either mothers or fathers differed in their levels of warmth toward sons versus daughters. Although these findings suggest that African American parents’ “love” did not differ based on adolescent gender, two issues deserve consideration. First, this study did not investigate rules or expectations about responsibilities or behavior. Stricter parenting of daughters—evidence of “raising daughters”—may be apparent even if parents experience equally warm relationships with children of both genders. Second, more research is needed to explore the circumstances under which differential affection toward sons and daughters may occur. For instance, some research has shown that stressful family circumstances contribute to higher levels of parents’ differential treatment of offspring (Crouter, McHale, & Tucker, 1999).
Although adolescent gender was unrelated, our analyses revealed a significant effect of parent gender on parental warmth. Consistent with findings on White families, mothers reported warmer relationships with their adolescents than did fathers. This pattern may emerge because women tend to be more emotionally expressive in close relationships than men (Sprecher & Sedikides, 1993). Because parents’ reports of parent – adolescent warmth may be biased by their inclination to respond in socially desirable ways, however, future work should examine youths’ relationship perceptions. Our data also revealed that, controlling for youth age, parents reported warmer relationships with younger versus older siblings. This pattern may reflect parents’ learning from their experiences with earlier born offspring to be more effective in their roles with later-born offspring (Shanahan, McHale, Osgood, & Crouter, 2007).
Findings pertaining to parent – adolescent time were consistent with the idea that adolescence is a period of intensified gender socialization, particularly by mothers (Crouter et al., 1995). Within family comparisons revealed mother – father differences in gendered differential treatment of adolescent-aged siblings: When the older sibling was a boy, mothers invested more time in the younger sibling, and when the younger sibling was a girl, mothers spent relatively less time with the older sibling. On the other hand, though fathers spent more time with younger siblings, neither older nor younger sibling gender moderated this effect. Further research is needed to fully understand this pattern of findings. It may be that during adolescence, African American mothers, but not fathers, are focused on gendered socialization of their offspring and thus devote more time to daughters. Given past work indicating that African American parents of infants spent relatively equal amount of time with boys and girls (Hossain & Roopnarine, 1993), an important area for longitudinal research is to explore patterns of change in African American mothers’ and fathers’ time with sons and daughters across development. It is notable that in this sample, mothers did not spend significantly more time with youth than fathers, a pattern that may reflect the more flexible parenting roles that are often discussed as a major strength of African American families (Hill, 1999).
This study also contributes to the literature documenting links between parent – adolescent relationship qualities and adjustment of African American youth. In line with a family systems approach, we relied on reports from both parents (parent – adolescent warmth) and youth (shared-time, depressive symptoms, and risky behavior). Findings revealed that mothers’ and fathers’ relationships with their sons and daughters functioned in unique ways. For instance, maternal warmth was linked to fewer depressive symptoms in youth. In contrast, mothers’ warmth was negatively related to sons’ but not daughters’—risky behavior. In the case of fathers, warmth and shared time were associated with lower levels of youth risky behavior. This latter pattern is consistent with the idea that fathers’ socialization may be more important for offspring’s experiences in the world beyond the family (Parke & Buriel, 1998). For instance, fathers have been shown to spend a greater proportion of their time with adolescents engaged in leisure and recreational activities than mothers (Montemayor & Brownlee, 1987); the negative link between fathers’ time and risky behavior may emerge because of fathers’ supervision, guidance, or companionship in youths’ activities outside the home.
In addition to its substantive contributions, this study’s methodological strengths advance understanding of African American family dynamics. First, by focusing on a largely working- and middle-class community sample, this research moves beyond a deficit perspective wherein African American families may be implicitly pathologized (Garcia Coll et al., 1996). Further, as much research on African American families has focused on single mothers or families experiencing economic hardship, it has been hard to disentangle the impact of sociodemographic factors from normative family processes. Second, by relying on data from four family members this research avoided a mono-reporter bias. Indeed, fathers’ involvement in this study is noteworthy given that African American men are rarely represented in social science research, and past work has often relied on mothers’ reports of fathering (Coley, 2001). Additionally, by including mothers and fathers in the same analyses we were able to illuminate the unique effects of each parent on their children’s adjustment. Third, directly comparing mothers and fathers from the same families in their relationships with sons and daughters provided a deeper understanding of the roles of parent and child gender in African American family dynamics.
This study also has applied implications. Findings revealed that variability in parent –adolescent relationship quality in two-parent families has implications for youth problem behaviors and suggests the importance of strengthening these family dynamics in an effort to support youths’ positive adjustment. Poor mental health outcomes have been identified as an area in which African American youth fare considerably worse than their counterparts from other races/ethnicities (Cauce et al., 2011). Considering the present findings, maternal warmth may be particularly important for promoting mental health well-being among African American youth. With regard to risky behavior, our findings revealed that even among residential fathers and controlling for mother – adolescent relationship quality, the warmer fathers were with their adolescent-aged children and the more time they spent with them, the less likely sons and daughters were to participate in delinquent activities. Thus, targeting both domains of relationship quality in interventions may help to reduce negative youth outcomes.
In the face of its strengths, several factors limited this study’s conclusions. First, our sample, including its small size, the geographically circumscribed location from which it was gathered, and the relatively high levels of income and education of participants, limited the generalizability of our findings. Second, the cross-sectional design did not allow for conclusions about direction of effect. Parent – child influences are bidirectional (Kerr, Stattin, & Ozdemir, 2012), and youths’ adjustment may have implications for parent – adolescent relationships as well as the other way around. Third, though we measured two components of relationship quality, there are many other aspects of parent – adolescent relationships that may be linked to gender and have distinct implications for youth outcomes. Finally, our relatively small sample size limited power to detect interactions between parent gender, youth gender, and youth birth order
Nonetheless this study provided important insights into family processes in the neglected population of two-parent working-and middle-class African American families. The research design, which allowed for within-family comparisons, illuminated family gender dynamics that would not be evident in between-family comparisons: Rather than supporting the contention that African American parents demonstrate gendered differential treatment toward sons and daughters, our findings reflect the multidimensional nature of gender (Ruble & Martin, 1998). Adolescent gender was not a factor in parental warmth, but mothers reported warmer relationships than fathers, and parents’ shared time differed as a function of their offspring’s gender. This work also contributed to a small body of literature showing the uniquely important contributions of African American fathers to their children’s adjustment: Beyond mother – adolescent relationship quality, higher paternal warmth and parent – youth shared time were associated with lower levels of adolescent risky behavior. Given the relatively high rates of risky behavior among African American youth (Morenoff, 2005), this finding underscores the importance of directing interventions toward African American fathers.
This research was supported by grant R01-HD32336-02 from the National Institute of Child Health and Human Development to Ann Crouter and Susan McHale (PIs).
Christine E. Stanik, The Pennsylvania State University.
Elizabeth M. Riina, Columbia University*
Susan M. McHale, The Pennsylvania State University**