Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jul 1.
Published in final edited form as: J Community Psychol. 2020 Mar 28;48(5):1543–1563. doi: 10.1002/jcop.22346

Risk and protective family factors during childhood on youth violence among African American males: the role of mothers and nonresident fathers

Kazumi Tsuchiya 1, Daniel B Lee 2, Yiqing Qian 2, Cleopatra H Caldwell 2, Ronald B Mincy 3
PMCID: PMC7316613  NIHMSID: NIHMS1578396  PMID: 32222114

Abstract

African American male youth experience disproportionately higher levels of violence. We examined parental depression among African American mothers and nonresident fathers on parenting stress and school involvement on their adolescent sons’ school connectedness and violent behaviors. Using a longitudinal study design, parent factors were assessed when sons were nine years old on youth outcomes at age fifteen. We found that maternal depression was associated with maternal stress, and maternal stress was indirectly associated with sons’ violent behaviors through school connectedness. School involvement among nonresident fathers was positively associated with sons’ school connectedness, which was linked to less youth violent behaviors. Maternal stress and nonresident fathers’ school involvement are influential for understanding youth violence. Future interventions should incorporate a more nuanced approach when including family factors.

Keywords: African American male youth, parental mental health, parenting stress, parent-school involvement, youth school connectedness, youth violence, African American families

Introduction

Youth violence is a critical public health problem that peaks during adolescence and is the leading cause of nonfatal youth injury and death (David-Ferdon et al., 2016; NAHIC, 2007). Starting from early adolescence, violent behaviors may be a progression of behavior change from physical fighting to lethal forms of violence including violent acts involving a weapon (Dahlberg & Potter, 2001). Violence during adolescence is also a salient risk factor for violence into young adulthood (Dahlberg & Potter, 2001).

Homicide is the 3rd leading cause of death mostly due to firearms among youth 10 to 24 years old (CDC, 2016), with higher rates among males compared to females (CDC, 2014). In 2014, African American males ages 10 to 24 years old experienced the highest rates of homicide (48.2 per 100,000), compared to Hispanic (9.6 per 100,000) and non-Hispanic white males (2.6 per 100,000; CDC, 2014). These disparities highlight the need to focus on mechanisms that may exacerbate or protect against violence among African American male youth.

Youth violence is a complex and dynamic problem associated with a number of risks and protective factors (David-Ferdon et al., 2016). No one factor leads to youth violence; however, the interplay of multiple factors including individual, family, and community level factors influence violent behaviors (David-Ferdon et al., 2016; Stoddard et al., 2013). Some risk factors include prior exposure to violence, low levels of school achievement and school connection, and poor relationship quality with parents and family members (David-Ferdon et al., 2016; Zimmerman & Messner, 2013). Additionally, increasing evidence suggests the family context (e.g., parenting support, parenting stress, parent-school involvement, positive and strong parent-child relationships), youth school connectedness and positive relationships with peers and teachers may prevent or reduce violent behaviors (David-Ferdon et al., 2016; Nord et al., 1997). According to the Family Stress Model (FSM), parental stress due to economic hardship may influence youth’s emotional and behavioral outcomes (including externalizing behaviors) indirectly through the parents’ emotional well-being (e.g., depression) and parenting behaviors (e.g., involvement; Conger et al., 2002; Conger, Conger & Martin, 2010). The Social Development Model (SDM) also highlights the important role of parents and parenting behaviors on children’s outcomes (Catalano et al., 1996) and has been used to examine parent-child outcomes including school connectedness (see Kelly et al., 2012). Furthermore, expanding beyond family factors, the SDM examines influential factors for youth socialization at multiple levels (Hawkins & Weis, 1985). According to the SDM, the family, school, peers, and community are the most critical factors that influence youth socialization both directly and indirectly. However, the majority of research has examined these risk and protective factors separately rather than in tandem to investigate potential protections from the family and school for violent behaviors among minority youth, with specific attention to assessing these relationships for African American adolescent males.

Parent Depression & Youth Violent Behaviors

Parents are usually the primary socializing agents in their children’s development through parenting support and parent involvement (Cabrera, Shannon, & Tamis-LeMonda, 2007; Cabrera, Fitzgerald, Bradley, & Roggman, 2014; Lindstrom Johnson et al., 2012), both of which have salient roles in ultimately reducing youth violent behaviors. Parent depression (diagnosed or number of symptoms), however, has been linked to youth’s externalizing behaviors (Gross, Shaw, & Moilanen, 2008; Leve, Kim, & Pears, 2005; Ponnet, 2014) through a lack of parent involvement and support (Ponnet, 2014). As a robust area of research, maternal depression has been linked to poor child outcomes of internalizing and externalizing behaviors, specifically aggression (Leve, Kim, & Pears, 2005; Gross, Shaw, & Moilanen, 2008). Depressed mothers may view their parenting roles less positively and have less motivation, energy, and confidence in their involvement with their children (Lovejoy, Graczyk, O’Hare & Neuman, 2000). Most research that investigates parent influences on their children’s development has been primarily focused on mothers, with scant literature on fathers (Cabrera, et al., 2014) and nonresident African American fathers; however, one study found paternal depression among nonresident African American fathers is associated with less involvement in their children’s lives (Davis, Caldwell, Clark, & Davis, 2009). These relationships within African American families are most often assessed through pathways of financial hardship and parenting practices (Conger et al., 2002; Gutman & Eccles, 1999).

To further elaborate, paternal depression may influence parenting behaviors and family dynamics (Cabrera et al., 2014), including developing positive relationships with children and their teachers (Kohl, Lengua, & McMahon, 2000). These relational challenges may in turn influence their levels of school involvement and interactions with school personnel (Kohl et al., 2000). As supported by the FSM (Conger et al., 2002), parental depression may also invoke stress and spill over into family relations and children’s internalizing and externalizing outcomes (Cabrera et al., 2014). Shaw and colleagues (2009) found that addressing and reducing maternal depression led to reductions in internalizing and externalizing problems among their children. Thus, further assessing the links between parental mental health, family and school factors among African American mothers and fathers as mechanisms to ameliorate and prevent youth violence among their sons is critical.

African American nonresident father involvement.

The significance of nonresident African American fathers to these family dynamics is often overlooked. In 2015, about 70 percent of African American children were born to unmarried mothers, with the majority not living with their fathers (Hamilton et al., 2016). However, most African American fathers who do not live with their children (or nonresident) are involved in their children’s lives and prioritize their role as fathers (Edin, Tach, & Mincy, 2009; Carlson & McLanahan, 2010). More research is warranted considering these family structures to understand pathways between African American paternal mental health and youth violence. Therefore, this study focuses on the nonresident African American family structure.

Although previous studies have investigated individual associations of parent depression, parenting stress, parent-school involvement, youth school connectedness and violent behaviors, to our knowledge, studies examining risk and protective factors collectively for youth violence among African American families with nonresident fathers and their adolescent sons are rare. Informed by the SDM and FSM, the current study investigated the relationship between parental depression and adolescent male violent behaviors mediated by risk and protective family and school factors, including parent-school involvement and parenting stress, and youth school connectedness by parent gender. Specifically, these relationships were assessed among both African American mothers and their sons as well as among African American nonresident fathers and their sons. There are a few studies that are informed by the SDM which examine the influence of the family, across both mothers and fathers in the same study (see Drapela & Mosher, 2007; Kosterman, Haggerty, Spoth, & Redmond, 2004). Additionally, FSM has not been adequately used to examine these pathways among African American nonresident families. An important contribution of the current study is examining parallel models of these pathways for both mothers and nonresident African American fathers on outcomes for their sons at age fifteen.

Parenting Stress as a Mediator

A robust area of research using the FSM has found that parenting stress, a critical factor for relational dynamics, is associated with children’s outcomes and specifically higher externalizing behaviors (Neece et al., 2012; Puff & Renk, 2014; Mistry et al., 2002) and among African American families (Conger, et al., 2002; Mistry, et al., 2002). Typically, parenting stress is defined as stress arising from balancing multiple parenting demands including employment and financial constraints (Deater-Deckard, 2004). Patterson and colleagues (1989) posited that stressors within the family in tandem with poor parenting practices were associated with adverse outcomes among their children. Parents juggling multiple demands may be less likely to engage in effective parenting to prevent youth violence among African American adolescent boys. Conversely, though limited, a growing area of literature has found that parenting and more parental support are associated with more youth school connectedness (Kelly et al., 2011; Shochet, Smyth, & Homel, 2007). Thus, these results suggest addressing and reducing parenting stress may prevent youth violence through youth school connectedness among African American families and their children.

Parent-school Involvement as a Mediator

Parent-school involvement is protective for children and has been associated with improved student academic achievement (Barnard, 2004), fewer behavioral problems (Brookmeyer, Fanti, & Henrich, 2006), and lower rates of school dropout (McNeal, 1999). Parental involvement has been distinguished across domains at home (e.g., homework help) and at school, including participating in school activities and attending parent-teacher conferences (Kim, 2009; Kohl et al., 2000). Parent-school involvement in particular is critical for their children’s schooling as they may be able to advocate for their children within the school (Barnard, 2004; McNeal, 1999). Thus, the visibility of parents in school is an important role in their children’s academic outcomes, even after controlling for their involvement in academics at home (Kim, 2009).

Despite the known positive effects of parent-school involvement and child outcomes, this area has been underdeveloped among African American families (Howard & Reynolds, 2008). Previous research suggests that African American parents encounter numerous barriers to school involvement (Koonce & Harper, 2005; Williams & Sanchez, 2013) or only became involved when there were problems at school (Howard & Reynolds, 2008; Lindstrom Johnson et al. 2012; Murray et al., 2014). These parents may possess fewer resources (e.g., financial, time constraints), demanding work schedules and juggling child care responsibilities while also navigating challenging school environments (e.g., negative interactions with school administrators), which in turn influences their ability to invest in the education and development of their children (Kohl, et al. 2000; Koonce & Harper, 2005; Murray et al., 2014). In sum, these findings suggest that multiple demands may reduce the time and resources African American parents have to engage within the school environment.

Furthermore, the majority of parent-school involvement research has examined the potential positive role of parents among mothers and young children with limited studies on nonresident fathers (Cabrera, et al., 2014) and their adolescent child. Findings suggest that nonresident African American fathers became more involved when their children were engaged in delinquent behaviors and that their involvement was associated with lower suspension and expulsion among secondary school children (Coley & Medeiros, 2007; Nord et al., 1997). Taken together, these results suggest nonresident African American fathers’ involvement may be protective for violent behaviors among their adolescent children.

Youth School Connectedness as a Mediator

Emerging research suggests that youth school connectedness may be protective for youth risky behaviors (Brookmeyer et al., 2006; Chapman, Buckley, Sheehan, & Shochet, 2013; Foster et al., 2017). School connectedness has been defined as the level to which the youth feel that they are valued and a part of the school community at large and that the adults and peers care for their well-being (CDC, 2009). A more limited focus in the literature has been on school connectedness as a mediator. Some evidence shows youth who feel more connected at school are more likely to do better in school, have better emotional health (Kidger, et al., 2012), and are less likely to engage in violence (Brookmeyer et al., 2006; Foster et al., 2017). The majority of studies on school connectedness among youth focus on understanding the role of school (e.g., teachers, environment) and peer factors (see Chapman et al., 2013; Kidger et al., 2012; Voisin et al., 2005).

Parent and school factors are associated with youth school connectedness (Foster et al., 2017; Kelly et al., 2012; Shochet, Smyth, & Homel, 2007). Some scholars argue youth school connectedness may be linked to their socialization experiences within their families (Barber & Olson, 1997). Foster and colleagues (2017) found youth who felt more connected with their parents reported lower depressive symptoms and conduct problems and youth who reported higher school connectedness reported more positive outcomes. Specifically, a growing area of research suggests that parent support through high-quality parent child relationships are linked with greater school connectedness for youth (Kelly et al., 2012; Shochet, Smith, & Homel, 2007). These findings indicate that connectedness in these critical contexts may be foundational for youth development. The majority of prior studies investigated school connectedness in a broad sample of youth and with limited examination of specific sub-groups of youth and social contexts (e.g., neighborhoods with constrained resources). African American adolescent males continue to remain underrepresented in studies that have investigated the influence of school and family environments on youth violence (see Bernat & Resnick, 2009; Nation et al., 2003).

The Current Study

Based on previous findings, limited studies have assessed both the role of mothers’ and fathers’ mental health as risk or protective factors for violent behaviors among African American male youth. Across the parenting and family literature, fathers and nonresident fathers have been typically excluded as an area of interest. From a developmental perspective, late childhood is a time where families still exert a great influence on youth decision-making (Lindstrom Johnson et al. 2012). At this stage, peer influences are less salient than during the adolescent period. Thus, we focus on parent factors when the youth were nine years old and parental influence remains critical to predict youth outcomes at age fifteen. Given these critical gaps, our study aims to understand how African American mothers’ and nonresident fathers’ mental health during childhood is linked to violent behaviors among African American adolescent males at age fifteen. We also consider mediating mechanisms of parenting stress and parent-school involvement during childhood and adolescent male school connectedness at age fifteen as critical determinants.

Both the Family Stress Model (FSM) and the Social Development Model (SDM) provide theoretical grounding for understanding the direct and indirect pathways for how family and school contexts affect violence among African American adolescent boys. Previous research using the FSM have assessed these pathways in samples of African American mothers (Conger et al., 2002; Guttman & Eccles, 1999), with limited assessment among fathers, especially nonresident African American fathers. Cabrera and colleagues (2014) elucidated an updated framework for fathers which includes pathways of father functioning for parenting and involvement which provide further support in the assessment of these pathways. The SDM framework is also useful in understanding youth outcomes of school connectedness and violence as it highlights the importance of parenting and parent involvement and the reciprocal nature of parent and child behaviors. Consistent with SDM, parenting and parent support has a direct and positive relationship on youth school connectedness, academic outcomes, and risky health behaviors (Kelly et al., 2012; Shochet, Smyth, & Homel, 2007). To further understand the role of family and school contexts on youth outcomes, the current study contributes to the parenting literature for understanding these pathways among African American mothers and nonresident fathers on outcomes of school connectedness and violent behaviors among adolescent boys.

Hypotheses

Based on previous research, 1) we expect that depression among mothers and nonresident fathers during childhood will be positively associated with youth violent behaviors at age fifteen. To test potential mediators or indirect relationships between parental depression and youth violent behaviors, we also hypothesize that 2a) parenting stress and school involvement for both mothers and nonresident fathers will mediate the relationship between parental depression during childhood and youth violent behaviors. Specifically, parental depression will be associated with more parenting stress and lower school involvement. Parenting stress will be associated with less school connectedness and conversely parent-school involvement will be associated with more school connectedness for their sons at age fifteen. These combined factors contribute to their sons’ violent behaviors at age fifteen. We also expect that 2b) adolescent males’ school connectedness will mediate the relationship between parenting stress and parent-school involvement for mothers and nonresident fathers on their violent behaviors. That is, more school connectedness will be linked to less violent behaviors at age fifteen.

Method

Data and Sample

The study sample used data from a subsample from the Fragile Families and Child Wellbeing Study (FFCWS), a longitudinal birth-cohort study. The baseline sample includes 4,898 births, between 1998 and 2000, from 20 US cities. The FFCWS oversampled children who were born to unmarried couples (3:1 ratio), with the majority of the children not living with their biological fathers at birth. Follow-up surveys via telephone were conducted across multiple years when the focal child was age one (W2; follow-up 1), three (W3; FU 2), five (W4; FU 3), nine (W5; FU 4), and fifteen (W6; FU 5). For our study, we focused on parenting at age nine, when the child was still attached to the household with fewer competing influences from their peers predicting to child outcomes during high school at age fifteen.

Our study focused on African American families and their sons, specifically among families where fathers do not live with their sons, the family structure for 70% of African American youth (Hamilton et al., 2016). Thus, the analytic sample included two samples: 1) African American mothers and their sons (N=481) and 2) African American nonresident fathers and their sons (N= 267). For both samples, we included those who were not married to each other or were not romantically involved (e.g., friends, divorced/separated) or did not live with their child’s mother or father (see Fagan, Palkovitz, Roy, & Farrie, 2009). Previous research has found that African American families are less likely to cohabit even when they are romantically involved, compared to whites or Hispanics (Carlson & McLanahan, 2010). For Wave 5 (follow-up 4), the sample of nonresident fathers (N = 267) either had a romantic relationship with the mother but did not live with the mother/no visitation (3.4%), were separated/divorced (10.9%), friends (47.9%), or had no relationship (37.8%). In this longitudinal study, the analytic sample used parent measures when the son was nine years old (W5) and on sons’ outcome measures when they were fifteen years old (W6). The mean age for mothers was 32.72 years (SD = 5.33) and 35.12 years (SD= 6.30) for fathers. The mean level of education was in between a high school diploma and some college for both mothers (M=2.49, SD=0.88) and fathers (M=2.33, SD= 0.89).

Measures

Parent Depression.

Depression was assessed using questions from the Composite International Diagnostic Interview (CIDI)-Short Form (Kessler, Andrews, Mroczek, Utsun, & Wittchen, 1998; Bendheim-Thoman Center for Research on Child Wellbeing, 2013). The CIDI is a standardized measure for assessing mental disorders including clinical depression. Items included 1) losing interest, 2) feeling tired, 3) change in weight, 4) trouble sleeping, 5) trouble concentrating, 6) feeling worthless, and 7) thinking about death. Depression diagnosis using the CIDI conservative measure was used as a dichotomous measure for both mothers and fathers from W5 of the FFCW study to indicate whether they had depression.

Parent-school Involvement.

School involvement was measured using four-items (from W5) regarding levels of parent involvement in school activities. Respondents responded to items asking, since start of the school year have you: 1) attended an open house or back-to-school night, 2) attended meetings of parent-teacher organization, 3) gone to regularly scheduled parent-teacher conferences, and 4) attended a school event where your child participated. Categories ranged from 0 (not in this/last school year), 1 (once in this/last school year), and 2 (more than once in this/ last school year). Responses were averaged for this scale with higher scores indicating more school involvement. The reliability of this scale demonstrated acceptable internal reliability (α = 0.67 for mothers and α = 0.85 for fathers). Inter-class correlations across these items ranged between 0.38 to 0.57 for mothers and 0.67 to 0.81 for fathers.

Parenting Stress.

Parenting stress included three items (W5) adapted from the JOBS Child Outcomes Study (Bendheim-Thoman Center for Research on Child Wellbeing, 2013). This scale examines the parents’ aggravation or perceptions of stress in taking care of their children with responses ranging from 1 (Strongly Agree) to 4 (Strongly Disagree). Items included were “being a parent is harder than I thought it would be,” “taking care of my child is much more work than pleasure,” and “I often feel tired, worn out, exhausted from raising a family.” Items were reverse coded and averaged across these three-items. Higher scores are indicative of higher stress. In this study, the parenting stress scale demonstrated acceptable reliability (α = 0.62 for mothers and α = 0.60 for fathers). Additionally, the inter-correlations across items ranged from 0.39 to 0.50 for mothers and 0.32 to 0.50 for fathers.

Sons’ School Connectedness.

Sons’ school connectedness (from W6) was measured by four items adapted from the Panel Study of Income Dynamics Child Development Supplement (PSID-CDS-III) regarding their perceptions of the degree of inclusiveness, closeness, and safety at school (Bendheim-Thoman Center for Research on Child Wellbeing, 2015). Items included if they feel close to people at school, feel like they are a part of the school, are happy to be at school, and feel safe at school. Responses ranged from 1 (Strongly Agree) to 4 (Strongly Disagree). Response categories were reverse coded and averaged across four-items, with higher scores indicative of greater school connectedness. African American adolescent boys’ response on this measure had good reliability for the sample with mothers (N = 481; α = 0.71) and fathers (N = 267; α = 0.73). Additionally, the inter-correlations across items ranged from 0.34 to 0.63 for the sample with mothers and sons and from 0.38 to 0.60 for the sample with fathers and sons.

Sons’ Violent Behaviors.

Sons’ violent behaviors (W6) were examined using measures adapted from both the Add Health Wave I and Wave II In-Home survey, asking adolescents whether they engaged in violent behaviors (Bendheim-Thoman Center for Research on Child Wellbeing, 2015). Responses ranged from never, 1 or 2 times, 3 or 4 times, or 5 or more times, and included whether the respondent had gotten into a serious physical fight, taken part in a group fight, or hurt someone badly enough to need bandages or medical care. Responses were averaged across these three-items for violence. Adolescent responses on this measure yielded acceptable scale reliability for the sample with mothers (α = 0.67) and sample with fathers (α = 0.58). The inter-correlations ranged from 0.47 to 0.72 across items for the matched mother-son sample and from 0.29 to 0.71 for the matched father-son sample.

Some scales in the study had low reliability (parenting stress, violence), potentially due to limited number of items included for the scale. However, given the current study estimated these models using structural equation modeling which adjusts for measurement error (Kline, 2011), these scales were acceptable for use in the study.

Covariates.

Parents’ age, parents’ education, mother and fathers’ relationship quality, and parents’ closeness with their son were included as controls in the analysis, as previous research has shown that these factors are associated with the key predictors and outcomes (Coleman, 1966; Coley & Medeiros, 2007; Davis et al., 2009; Nord, Brimhall, & West, 1997).

Analytic Approach

Structural equation modeling (SEM) was used to test direct and indirect pathways between parental depression to violent behaviors among African American males using samples of parent-child dyads, after adjusting for measurement errors. All study analyses were conducted in Mplus 8.2 (Muthén & Muthén, 2018) and models were estimated within two samples – (1) mothers and their sons and (2) nonresident fathers and their sons. As a preliminary step, we examined descriptive statistics and bivariate correlations between study variables. We used confirmatory factor analyses (CFAs) to assess the construct validity of parenting stress (W5), parent-school involvement (W5), youth school connectedness (W6), and youth violent behaviors (W6). Perceptions of school connectedness and violent behaviors were measured in the subsequent follow-up (W6) after the parental outcomes (e.g., maternal depression) were measured (W5). After estimating measurement models, we fit structural models to examine whether mothers’ and nonresident fathers’ depression uniquely shapes parent-school involvement and parenting stress, as well as youth school connectedness and violent behaviors. Of note, we fit one structural model for the mother sample and another structural model the nonresident father sample to test multiple pathways from parental depression to youth violent behaviors. Moreover, the two structural models were identical (see Figures 1 and 2) with exception to the covariates. That is, in the mother sample, covariates were related to the mother (e.g., mother’s age), whereas the covariates were related to nonresident fathers (e.g., nonresident father’s age) in the nonresident father sample. To assess indirect effects (i.e., mediation) from parental depression to youth violent behavior in MPlus, we implemented MacKinnon’s and colleagues’ (1995) product method to test the indirect effect of parental depression on violent behaviors. That is, a significant indirect effect would indicate a mediation effect. The fit of our measurement and structural models was considered adequate if the RMSEA value was less than .05, CFI value was greater than .90, and SRMR was less than .05 (MacCallum et al., 1996). All measurement and structural models were estimated using Weighted Least Square estimation (WLSMV) since categorical outcomes (e.g., Likert scales) depart from multivariate normality.

Missing data.

Missing data was a nontrivial issue for each variable in the full data set. Missing data patterns across both mothers, nonresident fathers, and youth factors indicate that missing data rates were below 20% across all variables included in the study. For mothers, missing data ranged from 1–3% for all the predictors and covariates. For nonresident fathers, missing data ranged from 1–4% for all covariates and depression, and for parenting stress and school involvement the missing data ranged between 12–13%. For youth reported outcomes (school connectedness and youth violence) missing data was about 18% in both mothers’ and nonresident fathers’ samples. While FIML is a common method for accommodating missing data in structural equation models, this method is not available for SEMs estimated with WLSMV (i.e., default is pairwise deletion). We, therefore, preprocessed the data implementing Bayesian imputation. Specifically, each variable with missing data was treated as an outcome to impute all missing values (i.e., rectangularize the data), and multiple datasets were iteratively produced (i.e., 100 datasets). Our SEMs of interest were then estimated using WLSMV on each copy of the imputed data, and the model parameters are pooled using Rubin’s (1976) formula. While approximately 5 imputations have been recommended in previous studies (Rubin, 1976), we imputed 100 datasets to ensure that the pooled parameter estimates were as accurate as possible. The Bayesian estimation was conducted with two unique chains, a minimum of 25,000 iterations per chain, and the potential scale reduction cut-off (i.e., convergence) was set to .01 (Mplus default is .05). Plot trace lines for both chains were examined to ascertain convergence. Lastly, we conducted a sensitivity analysis by comparing model coefficients from the imputed datasets (i.e., pooled) and the raw data (i.e., pairwise deletion; default for WLSMV).

Results

Descriptive Statistics

Mothers’ sample.

Table 1 provides descriptive statistics and Pearson’s correlations of the key variables and covariates for the mothers’ sample (N = 481) in the study. Of the mothers’ in the sample, 14.4% were depressed (N = 68). Mothers’ reported levels of school involvement was 1.30 (SD = 0.56) and 2.33 for maternal stress (SD = 0.83). Sons’ average levels of school connectedness was 3.40 (SD = 0.61) and 1.33 for violence (SD = 0.50). Mothers’ reported relationship quality with the father was 4.16 (SD = 1.52) and 3.78 for closeness with their son (SD = 0.50).

Table 1.

Descriptive statistics and Pearson’s correlations for mothers (N=481)

Variables 1 2 3 4 5 6 7 8 9
1 Maternal depression 1.00
2 Maternal school involvement −.08@ 1.00
3 Maternal stress .13** −.03 1.00
4 Sons’ school connectedness −.03 .06 −.13* 1.00
5 Sons’ youth violence .03 −.08@ .13** −.28*** 1.00
6 Mothers’ age −.02 .01 −.07 −.04 −.06 1.00
7 Mothers’ education −.06 .10* −.11* .04 −.07 .08@ 1.00
8 Relationship quality .05 −.05 .09* −.04 −.01 −.08 .00 1.00
9 Closeness with son −.01 .13** −.09* .02 .07 .07 −.07 −.17*** 1.00
Mean / % (for depression) 14.4 1.30 2.33 3.40 1.33 32.72 2.49 4.16 3.78
SD/ N (for depression) 68 .56 .83 .61 .50 5.33 .88 1.52 .50
Range 0–2 1–4 1–4 1–4 23–50 1–4 1–6 1–4
Open in a new tab

Note:

@

p < .1;

*

p < .05;

**

p < .01;

***

p < .001.

Nonresident fathers’ sample.

Table 2 provides nonresident fathers’ descriptive statistics and Pearson’s correlations of the key variables and covariates in the study (N = 267). Fourteen percent of fathers were depressed (N = 37). Fathers’ reported scores of school involvement was 0.61 (SD = 0.68), and 2.07 for paternal stress (SD = 0.80). The average sons’ school connectedness was 3.40 (SD = 0.62) and 1.32 for violent behaviors (SD = 0.45). Nonresident fathers’ reported mean scores of relationship quality with their sons’ mother was 3.46 (SD = 1.44) and 3.27 for closeness with their sons (SD = 0.95).

Table 2.

Descriptive statistics and Pearson’s correlations for nonresident fathers (N=267)

Variables 1 2 3 4 5 6 7 8 9
1 Paternal depression 1.00
2 Paternal school involvement −.05 1.00
3 Paternal stress .10 −.04 1.00
4 Sons’ school connectedness −.05 .29*** −.07 1.00
5 Sons’ youth violence .01 −.09 .02 −.21** 1.00
6 Fathers’ age −.01 .02 −.18** −.02 −.02 1.00
7 Fathers’ education .09 .15* −.04 .07 −.05 .15* 1.00
8 Relationship quality .11@ −.34*** .12@ −.15* .00 −.06 .01 1.00
9 Closeness with son −.09 .26*** −.21** .06 .01 .03 .03 −.42** 1.00
Mean / % (for depression) 14.0 .61 2.07 3.40 1.32 35.12 2.33 3.46 3.27
SD/ N (for depression) 37 .68 .80 .62 .45 6.30 .89 1.44 .95
Range 0–2 1–4 1–4 1–4 25–54 1–4 1–6 1–4
Open in a new tab

Note:

@

p < .1;

*

p < .05;

**

p < .01;

***

p < .001.

Correlations

Inter-correlations between the study variables were not the same across mothers and fathers. Mothers’ parenting stress was positively associated with their depression, sons’ violent behaviors and relationship quality with the father, and negatively associated with closeness with their son. Sons’ school connectedness was negatively associated with their violent behaviors and mothers’ parenting stress. Mothers’ education was positively associated with their school involvement and negatively associated with their parenting stress. Closeness with their son was positively associated with their school involvement and negatively associated with their relationship quality with the father.

Nonresident fathers’ school involvement was negatively associated with their relationship quality with their son’s mother and was positively associated with their son’s school connectedness. Sons’ school connectedness was negatively associated with their violent behaviors. Their education (positive), quality of their relationship with their mother (negative), and closeness with their son (positive) were all associated with nonresident fathers’ school involvement. Fathers’ age and closeness with their son were negatively associated with their parenting stress, while fathers’ age and their education were positively associated. Closeness with their son was negatively associated with their quality of relationship with the mother.

Measurement Models for the Maternal Sample

Chi-square-based fit statistics were unproducible for maternal stress and youth violent behaviors as the three-item latent factors were “just identified” (i.e., saturated). Item-level factor loadings ranged between .61 to .75, and .62 to .97, respectively. Measurement models for maternal school involvement (i.e., χ2 (2) = 2.70, p = .26; RMSEA = .03; CFI = .99) and youth school connectedness (i.e., χ2 (2) = 0.95, p = .62; RMSEA = .00; CFI = .99) fit the data well, with factor loadings that ranged between .63 to .84, and .58 to .82, respectively.

Measurement Models for the Nonresident Father Sample

Our unidimensional measures of parenting stress and youth violent behaviors were “just identified” (i.e., saturated); therefore, chi-square-based model fit statistics were not able to be estimated. Unstandardized factor loadings for paternal stress ranged from .51 to .75, suggesting that all items contributed to the measurement of parenting stress. Unstandardized item-level factor loadings for youth violent behaviors ranged from .47 to 1.26. We observed good model fit for our measure of paternal school involvement (i.e., χ2 (2) = 0.25, p = .88; RMSEA = .00; CFI = 1.00) and youth school connectedness (i.e., χ2 (2) = 0.95, p = .62; RMSEA = .00; CFI = 0.99), and observed factor loadings that ranged from .85 to .94, and .58 to .82, respectively (results not shown).

Structural Model Linking Maternal Depression and Youth Violent Behaviors

A structural model was fit to examine pathways bridging maternal depression to youth violent behaviors. Specifically, maternal depression was linked with maternal stress and maternal school involvement, and maternal stress and school involvement was linked with youth school connectedness. Additionally, youth school connectedness was linked with youth violent behaviors (see Figure 1). Covariates included mother’s age, educational attainment, relationship with the child, and relationship with the nonresident father (see Table 3). Among mothers and sons, the structural model fit the data well (i.e., χ2 (121) = 141.27, p = .10; RMSEA = .02; CFI = 0.99). While the direct effect between maternal depression and violent behaviors were not observed, we found that maternal depression was associated with higher levels of parenting stress (b = .47; Figure 1), maternal stress was negatively associated with youth school connectedness (b = −.21), and youth school connectedness was associated with less youth violent behaviors (b = −.38). The indirect effect from maternal stress to youth violent behaviors through youth school connectedness was significant (b = .08, p = .02), with more maternal stress associated with less school connectedness, and in turn, higher violence. Mothers with higher educational attainment (b = .15) and mothers who felt close to their son (b = .29) were more likely to be involved in their son’s school. Maternal educational attainment, on the other hand, was associated with less maternal stress (b = −.16).

Figure 1.

Figure 1

Open in a new tab

Structural model of maternal depression and sons’ violent behaviors

Table 3.

Structural Model for Mothers and Sons

Youth Violent Behavior (Wave 6) b s.e. p
 Youth School Connectedness −.38 .09 < .01
 Maternal School Involvement −.10 .08 .24
 Maternal Stress .13 .09 .15
 Maternal Depression .02 .20 .19
 Mother’s Age −.02 .01 .14
 Mother’s Educational Attainment −.05 .09 .57
 Mother-Son Relationship Quality .21 .16 .20
 Relationship Quality with Child’s Father −.03 .04 .50
Youth School Connectedness (Wave 6) b s.e. p
 Maternal School Involvement .09 .07 .21
 Maternal Stress −.21 .08 .01
 Maternal Depression −.01 .19 .97
 Mother’s Age −.02 .01 .14
 Mother’s Educational Attainment .01 .08 .86
 Mother-Son Relationship Quality −.01 .12 .91
 Relationship Quality with Child’s Father −.02 .04 .65
Maternal School Involvement b s.e. p
 Maternal Depression −.26 .16 .10
 Mother’s Age −.00 .01 .81
 Mother’s Educational Attainment .15 .07 .03
 Mother-Son Relationship Quality .29 .10 < .01
 Relationship Quality with Child’s Father −.02 .04 .65
Maternal Stress b s.e. p
 Maternal Depression .47 .18 .01
 Mother’s Age −.01 .01 .27
 Mother’s Educational Attainment −.16 .07 .02
 Mother-Son Relationship Quality −.21 .12 .08
 Relationship Quality with Child’s Father .06 .04 .13
Mediating Pathways Indirect effects
 Maternal stress -> school connectedness -> violence b = .08, s.e. = .04, p = .02
Open in a new tab

Structural Model Linking Nonresident Paternal Depression and Youth Violent Behaviors

A structural model was also fit to examine pathways linking nonresident paternal depression to youth violent behaviors. As in the maternal depression model, paternal stress and paternal school involvement mediated the effect of nonresident paternal depression on youth school connectedness. Moreover, youth school connectedness was linked with youth violent behaviors (see Figure 2). Covariates included nonresident father’s age, educational attainment, relationship with the child, and relationship with the nonresident father (see Table 4). Our hypothesized structural model fit the data well (i.e., χ2 (121) = 141.67, p > .05; RMSEA = .03; CFI = .98). While our hypothesized direct and indirect effects were not observed, we observed a positive association between school involvement and youth school connectedness (b = .40; Figure 2) and an inverse association between youth school connectedness and violent behavior (b = −.31). Moreover, the indirect effect from paternal school involvement to youth violent behaviors through youth school connectedness was significant (b = −.12, p = .03), with more school involvement associated with less school connectedness, and in turn, higher violence. In addition, fathers reporting a positive relationship with their sons were more likely to endorse less paternal stress (b = −.29) and higher levels of school involvement (b = .27). In addition, age was inversely associated with paternal stress (b = −.04). Lastly, fathers’ educational attainment (b = .31) and perceptions of closeness with the mother of their son (b = −.29) also influenced their school involvement.

Figure 2.

Figure 2

Open in a new tab

Structural model of paternal depression and sons’ violent behaviors

Table 4.

Structural Model for Nonresident Fathers and Sons

Youth Violent Behavior (Wave 6) b s.e. p
 Youth School Connectedness −.31 .11 < .01
 Paternal School Involvement .04 .12 .76
 Paternal Stress .01 .11 .93
 Paternal Depression .00 .25 .99
 Father’s Age −.01 .02 .51
 Father’s Educational Attainment −.09 .11 .44
 Father-Son Relationship Quality .04 .11 .74
 Relationship Quality with Child’s Mother −.03 .08 .72
Youth School Connectedness (Wave 6) b s.e. p
 Paternal School Involvement .40 .12 < .01
 Paternal Stress −.08 .13 .51
 Paternal Depression −.04 .26 .89
 Father’s Age −.02 .02 .30
 Father’s Educational Attainment .00 .11 .99
 Father-Son Relationship Quality −.11 .12 .27
 Relationship Quality with Child’s Mother .00 .08 .99
Paternal School Involvement b s.e. p
 Paternal Depression −.13 .23 .57
 Father’s Age .00 .01 .71
 Father’s Educational Attainment .31 .10 < .01
 Father-Son Relationship Quality .27 .12 .02
 Relationship Quality with Child’s Mother −.29 .07 < .01
Paternal Stress b s.e. p
 Paternal Depression .41 .26 .12
 Father’s Age −.04 .02 .01
 Father’s Educational Attainment −.06 .10 .55
 Father-Son Relationship Quality −.29 .11 .01
 Relationship Quality with Child’s Mother .06 .07 .42
 Mediating Pathways Indirect Effects
 Paternal school involvement -> school connectedness -> violence b = −.12, s.e. = .06, p = .03
Open in a new tab

Sensitivity Analysis

Measurement Model.

The model fit and factor loadings of our latent variables (i.e., parenting stress, parent-school involvement, youth school connectedness, and youth violent behaviors) were near-identical between the measurement models generated from our imputed and raw data (results not shown). Measurement models for parent-school involvement and youth school connectedness fit the data well regardless of how missing data was handled. The absolute differences in the factor loadings between the imputed and raw data, on average, was .016. When taken together, the measurement models generated from the imputed and raw data were not substantively different.

Structural Model.

The structural models for mother and nonresident fathers fit the imputed and raw data well (results not shown). The mean absolute difference in path coefficients was .044 within the nonresident father model, and the absolute differences ranged from 0 to .08. For the mothers’ model, the mean absolute difference in path coefficients was .016, and ranged from 0 to .05. Importantly, substantive differences were not observed between structural models generated from the imputed and raw data.

Discussion

Understanding factors that heighten or protect against violent behaviors among African American male youth continues to be a pressing issue. Given the robust area of research and theoretical mechanisms of the Family Stress Model (FSM) for maternal depression and their children’s outcomes (Conger et al., 2010; Gross, Shaw, & Moilanen, 2008; Shaw, et al., 2009), this study has sought to understand whether depression among mothers and nonresident fathers during childhood is indirectly linked to violent behaviors at age fifteen, an area with scarce research. Additionally, parenting stress (Puff & Renk, 2014), parent-school involvement (Brookmeyer et al., 2006), and youth school connectedness (Foster et al., 2017) as mediators may further explain youth violent behaviors among African American male youth. Thus, this study investigated these pathways collectively within African American families across both mothers and nonresident fathers. Our results suggest that both family and school factors are influential for violent behaviors among African American male youth with pathways differing by parent gender and residential status.

We did not find a direct and indirect relationship between maternal depression and male youth violent behaviors in the current study. That is, maternal stress did not mediate the relationship between maternal depression during childhood and youth violent behaviors. Thus, our hypotheses were not supported. However, maternal depression was associated with higher maternal stress and we also found that maternal stress was indirectly associated with their sons’ violent behaviors at age fifteen through male youth school connectedness. These results support previous literature regarding maternal depression and parenting stress as suggested by the FSM (Conger et al., 2002; Conger et al., 2010; Puff & Renk, 2014). As posited by Conger and colleagues (1994), parents who are stressed are more likely to be managing multiple demands and may be less involved in their children’s lives. That is, stress may affect their parenting practices and in turn negatively impact their children (Conger et al., 2010; Cabrera et al., 2014). Our results similarly suggest that maternal stress has implications for their sons’ overall experiences of feeling disconnected to school, which in turn, increases their violent behaviors. Scholars have previously found school connectedness to be associated with violent behaviors (Brookmeyer et al., 2006; Foster et al., 2017); however, empirical research has not elucidated factors that may influence school connectedness beyond the peer (e.g., friends) and school context, including school climate and classroom size (see McNeely et al., 2002; Thompson et al., 2006). Thus, this study highlights the critical role of maternal stress for African American male adolescents’ school connectedness and violent behaviors.

Regarding nonresident fathers, contrary to our hypotheses, we did not find a direct nor an indirect association for the relationship between paternal depression and their sons’ violent behaviors. Given the scarcity of research in this area, a plausible explanation may be that since nonresident fathers do not live with their sons, nonresident fathers’ mental health may not be a critical factor for their sons’ violent behaviors. Nonresident fathers’ mental health but may be linked to other outcomes for their children (e.g., internalizing behaviors). The residing parent (i.e., mother) may have a salient effect on children’s outcomes through parenting and family dynamics as supported by our results from this study and previous research (Conger et al., 2010; Cabrera et al., 2014). A review of the literature found that paternal depression was linked to parenting among young children, which may be due to the increased demands of taking care of infants and toddlers and among fathers who live with their children (Wilson & Durbin, 2010). Of the few studies among African American fathers, one study found that nonresident African American fathers who were depressed were less likely to be involved with their sons (Davis, Caldwell, Clark, & Davis, 2009). We did not find that paternal depression was linked to their school involvement, which is different from general involvement (e.g., levels of contact) as school activities may occur less frequently compared to activities at home.

Interestingly, we did find that nonresident fathers’ school involvement is associated with higher levels of sons’ school connectedness and in turn, their school connectedness is negatively associated with their violent behaviors. Nord and colleagues (1997) found that nonresident fathers’ school involvement was negatively associated with their children’s likelihood to be suspended or expelled. This finding highlights the importance of nonresident fathers’ school involvement for positive child outcomes. However, the Nord and colleagues’ (1997) study did not examine other factors including depression and stress as mediators of this relationship. Another study among a sample of youth living in adverse circumstances (e.g., poverty, high crime rates, unemployment) found that both parent and school connectedness were associated with better adolescent adjustment (Foster et al., 2017). Supporting these findings, the results of the current study also suggest that nonresident fathers’ involvement in school may increase their sons’ positive perceptions of the school environment and feelings of connectedness to their school.

With regards to understanding the school context, previous empirical research has focused on characteristics of the students (e.g., household structure, extracurricular participation) and school environment (e.g., school financial resources, classroom climate) on youth school connectedness (McNeely et al., 2002; Thompson et al., 2006). However, the results of our study expand the previous literature in assessing the role of family on school connectedness, and particularly among nonresident African American fathers. Based on our findings, these nonresident fathers may have become involved if their son was having either academic or behavioral issues as teachers may reach out to parents if their child is underperforming or having behavioral problems (Howard & Reynolds, 2008; Lindstrom Johnson, et al. 2012; Murray et al., 2014). Coley and Medeiros (2007) found that African American fathers became involved once they found out that their child was engaged in delinquent behaviors; however, research has not previously examined the role of nonresident fathers in schools for youth male violence in tandem with their mental health and parenting stress. Thus, our findings provide a unique contribution to the literature.

Implications

This study can be used to inform future programs and interventions that focus on preventing or reducing violence among African American male youth through family and school factors. The results of this study suggest the pathways for African American mothers and nonresident fathers may differentially influence their sons’ outcomes, which should be considered in future intervention designs. Given our findings of a positive relationship between maternal depression and stress and extant research in this area (Conger et al., 2010; Gross, Shaw, & Moilanen, 2008; Puff & Renk, 2014; Shaw et al., 2009), future programs should consider addressing maternal depression and stress simultaneously as they relate to school connectedness and violent behaviors among their sons. In addition, providing strategies to manage stress and relationship dynamics with their children may potentially be effective for reducing adverse outcomes among African American adolescent boys. Previous parenting programs have largely focused on family dynamics and strengthening relationships to address outcomes for their children (Kaminski, Valle, Filene, & Boyle, 2008). Our findings hold promise for future family interventions that aim to prevent or reduce youth violent behaviors to consider maternal mental health and parenting stress for African American families.

Our results suggest that nonresident fathers who attended school meetings (e.g., teachers, school events) was associated with increased levels of sons’ school connectedness and in turn, contributed to less violence. This warrants attention to involving nonresident fathers to improve African American adolescent boys’ outcomes. Previous research on school involvement has found that ethnic minority parents would also like to know when their child is doing well in school (Smrekar & Cohen-Vogel, 2001), and not only when their child is experiencing disciplinary issues in school settings. Therefore, schools and administrators could consider ways to recognize student achievement in multiple ways and on multiple occasions inviting father participation to increase school connectedness and violence prevention among African American male youth.

African American parents from disadvantaged backgrounds in particular face numerous barriers in being involved in their children’s schooling (William & Sanchez, 2013). These include power dynamics due to their lower educational levels (Barton, Drake, Perez, St. Louis, & George, 2004) and lack of time due to demanding work schedules and unpaid leave (Murray et al., 2014). Further, teachers’ negative perceptions toward the efficacy and capacity of minority parents (Hoover-Dempsey, Walker, Jones, & Reed, 2002), an unwelcoming school environment, and poor communication with parents (Murray, et al., 2014) were adversely associated with parent involvement. Given these findings, it is clear that strengthening family-school relationships is warranted for African American families, with a particular emphasis on involving nonresident fathers, and with ample opportunities to build relationships across parents, teachers, and school administrators. Tailored efforts would need to be considered when designing programs and interventions to involve nonresident fathers. Strategies may include training teachers to be more effective in working with fathers who do not live with their children (e.g., conducive meeting schedules for fathers’, providing a welcoming environment), and increased and timely efforts in communicating with the father (e.g., less jargon, informal communication), as shown in previous research among minority parents (Kim, 2009). Other studies have found that school policies (e.g., homework completion, academic achievements) and reaching out to parents to participate in school may increase ethnic minority parent involvement (Kohl et al., 2000; Murray et al., 2014), and may also be effective when working with nonresident fathers. The majority of fatherhood programs have focused on parenting, child support payments, and employment opportunities (Caldwell, Tsuchiya, Assari, & Thomas, 2019), with a limited focus on school involvement. As suggested by our results, fathers can be a protective socializing influence even though they may not live with their sons, as supported by other studies (Cabrera et al., 2014; Caldwell et al., 2019).

Although not the main focus of our study, we did find that the quality of relationship between parents specifically mothers and nonresident fathers, was significantly associated with fathers’ school involvement, as previously supported in the literature (Paulson et al., 2011). Our findings suggest that the quality of the relationship between parents may be salient for fathers’ involvement beyond other factors as indicated by a strong co-parenting literature. Another study found that among African American nonresident fathers, interpersonal stress was associated with their depressive symptoms (Tsuchiya et al., 2018). The quality of the relationship between parents may be reflective of their levels of communication, where a stronger relationship may reflect more communication. For example, fathers may be more aware of what is occurring in their sons’ lives with regards to their sons academic and behavioral outcomes, which may in turn impact nonresident fathers’ levels of school involvement. The findings of our study call for increased efforts to involve African American nonresident fathers in schools to ultimately influence positive outcomes for their sons. Furthermore, we can no longer ignore the role of nonresident fathers if we are looking to prevent and reduce African American male youth violence.

Limitations

There are several limitations to be noted for this study. The sample was comprised of African American mothers and nonresident fathers and their sons. Thus, our findings are not generalizable for female youth, African American fathers who live with their children, and across all nonresident African American fathers. The analytic sample was comprised from W5 of FFWS; thus, the sample was not nationally representative. Additionally, a little less than half (44.5%) of the sons from the mothers’ matched sample overlapped with the sons matched with the nonresident fathers’ sample. For the purpose of our study, we treated both mothers and nonresident fathers as separate samples and thus, we did not control for the effects of the other parent. Our purpose was to assess family and school factors by parent gender and residential status and thus we did not examine community, teacher, and peer factors on youths’ school connectedness and violent behaviors. Future studies may consider investigating these factors in tandem with community, school, and peer factors to examine the accumulation of risk and protective factors or whether specific domains (see Kim, Gilman, Hill, & Hawkins, 2016) are associated with violent behaviors among adolescent African American boys.

Previous research supports the mechanisms examined in this study regarding parental mental health, parenting stress, and parent-school involvement on youth outcomes of school connectedness and violence among African American adolescent boys (Conger et al. 2010; David-Ferdon et al., 2016). Both school connectedness and violent behaviors were assessed at the same wave. Nevertheless, it is plausible that violent behaviors could influence school connectedness; however, previous research support the direction of our findings (Brookmeyer et al., 2006; Foster et al., 2017). While we found results consistent with theory, youth violence can be shaped by other contextual factors including community and peer factors. We demonstrate the importance of parent characteristics and school connectedness; however, future research should examine multiple contexts/factors simultaneously to unify our understanding of youth violence. There also could be variations in mother and nonresident father involvement at home; however, we were unable to include a reliable measure for both mother and nonresident father involvement in school activities at home in our study.

A strength of the study was the longitudinal design where we assessed parent factors during childhood (W5) to predict youth outcomes at age fifteen (W6). Additionally, due to the low frequency of parent depression in our data, an indirect relationship may not have been observed for parental depression and youth violent behaviors due to inadequate power to detect differences. Furthermore, parent depression and stress measures were collected when the youth were nine years old on outcomes when these youth were fifteen years old, with a six-year gap in time between these time points; thus, the significance between mental health and stress on violence may have been diminished through youth school connectedness when they could be more influenced by peers. Future studies should consider assessing these relationships in a larger sample of African American parents experiencing depression and include stronger measures for parenting stress (with more items) while simultaneously understanding levels of nonresident father involvement.

Conclusion

Youth violence is the leading cause of death for African American male youth. Thus, it is vital to consider multiple approaches to reduce violence in this population. Our findings suggest that both African American mothers and nonresident fathers may mitigate youth violence among African American adolescent males by enhancing their school connectedness. Broadly, this study emphasizes the importance of considering both risk and protective factors of family and school involvement across African American families with nonresident fathers as potential avenues for addressing violence among African American adolescent males. Using this approach to inform future interventions on violent behaviors among African American male youth will have lasting impacts into adulthood.

Acknowledgements

This research was funded by the Rackham Graduate School at the University of Michigan and the Eunice Kennedy Shriver National Institute of Child Health and Human Development under award numbers T32HD095134-01A1 and P2CHD041023.

The authors would like to thank the Fragile Families Wellbeing Study data administrators and Kirsten Herold from the University of Michigan for her feedback and support for the manuscript.

Footnotes

Conflict of Interest: The authors declare that they have no conflict of interest.

References

  1. Barber BK, & Olsen JA (1997). Socialization in context: Connection, regulation, and autonomy in the family, school, and neighborhood, and with peers. Journal of adolescent research, 12(2), 287–315. [Google Scholar]
  2. Barnard WM (2004). Parent involvement in elementary school and educational attainment. Children and youth services review, 26(1), 39–62. [Google Scholar]
  3. Barton AC, Drake C, Perez JG St. Louis K, & George M (2004). Ecologies of parental engagement in urban education. Educational Researcher, 33(4), 3–12. [Google Scholar]
  4. Bendheim-Thoman Center for Research on Child Wellbeing. (2013). Scales documentation and question sources for the nine year wave of the Fragile Families and Child Wellbeing Study.
  5. Bendheim-Thoman Center for Research on Child Wellbeing. (2015). Data user’s guide for the fifteen-year follow-up wave of the Fragile Families and Child Wellbeing Study.
  6. Bernat DH, & Resnick MD (2009). Connectedness in the lives of adolescents In DiClemente RJ, Santelli JS, & Crosby RA (Eds.), Adolescent health: Understanding and preventing risk behaviors (pp. 375–389). San Francisco, CA: Jossey-Bass. [Google Scholar]
  7. Brookmeyer KA, Fanti KA, & Henrich CC (2006). Schools, parents, and youth violence: A multilevel, ecological analysis. Journal of Clinical Child and Adolescent Psychology, 35(4), 504–514. [DOI] [PubMed] [Google Scholar]
  8. Cabrera NJ, Fitzgerald HE, Bradley RH, & Roggman L (2014). The ecology of father‐child relationships: An expanded model. Journal of Family Theory & Review, 6(4), 336–354. [Google Scholar]
  9. Cabrera NJ, Shannon JD, & Tamis-LeMonda C (2007). Fathers’ influence on their children’s cognitive and emotional development: From toddlers to pre-K. Applied Development Science, 11(4), 208–213. [Google Scholar]
  10. Caldwell CH, Tsuchiya K, Assari S, & Thomas A (2019). Fatherhood as a strategy to reduce men’s health disparities: Lessons learned from the Fathers and Sons Program. In Griffith DM, Marino A, & Thorpe RJ (Eds.), Men’s Health Equity: A Handbook. [Google Scholar]
  11. Carlson MJ, & McLanahan SS (2010). Fathers in fragile families. The role of the father in child development, 5, 241–269. [Google Scholar]
  12. Catalano RF, Kosterman R, Hawkins JD, & Newcomb MD (1996). Modeling the etiology of adolescent substance use: A test of the social development model. Journal of Drug Issues, 26(2), 429–455. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Centers for Disease Control and Prevention (2014). National Center for Injury Prevention and Control. Web based Injury Statistics Query and Reporting System (WISQARS). Retrieved from www.cdc.gov/injury.
  14. Centers for Disease Control and Prevention. (2016). Youth online: High school YRBS. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion; Retrieved from http://www.cdc.gov/healthyyouth/data/yrbs/. [Google Scholar]
  15. Centers for Disease Control and Prevention (2009). School Connectedness: Strategies for increasing protective factors among youth. Atlanta, GA: U.S. Department of Health and Human Services. [Google Scholar]
  16. Chapman RL, Buckley L, Sheehan M, & Shochet I (2013). School-based programs for increasing connectedness and reducing risk behavior: A systematic review. Educational Psychology Review, 25(1), 95–114. [Google Scholar]
  17. Coleman JS, Campbell EQ, Hobson CJ, McPartland J, Mood AM, Weinfeld FD, & York RL (1966). Equality of educational opportunity. Washington, DC: U.S. Government Printing Office. [Google Scholar]
  18. Coley RL, & Medeiros BL (2007). Reciprocal longitudinal relations between nonresident father involvement and adolescent delinquency. Child Development, 78(1), 132–147. [DOI] [PubMed] [Google Scholar]
  19. Conger RD, Elder GH, Lorenz FO, Simons RL, & Whitbeck LB (1994). Families in troubled times: adapting to change in rural America. Walter de Gruyter Inc, New York. [Google Scholar]
  20. Conger R, Conger K, & Martin M (2010). Socioeconomic status, family processes, and individual development. Journal of marriage and the family, 72, 685–704. doi: 10.1111/j.1741-3737.2010.00725.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Conger RD, Wallace LE, Sun Y, Simons RL, McLoyd VC, & Brody GH (2002). Economic pressure in African American families: a replication and extension of the family stress model. Developmental psychology, 38(2), 179. [PubMed] [Google Scholar]
  22. Dahlberg LL, & Potter LB (2001). Youth violence: Developmental pathways and prevention challenges. American Journal of Preventive Medicine, 20(1), 3–14. [DOI] [PubMed] [Google Scholar]
  23. David-Ferdon C, Vivolo-Kantor AM, Dahlberg LL, Marshall KJ, Rainford N & Hall JE (2016). A Comprehensive Technical Package for the Prevention of Youth Violence and Associated Risk Behaviors. Atlanta, GA: National Center for Injury Prevention and Control, CDC. [Google Scholar]
  24. Davis RN, Caldwell CH, Clark SJ, & Davis MM (2009). Depressive symptoms in nonresident African American fathers and involvement with their sons. Pediatrics, 124(6), 1611–1618. [DOI] [PubMed] [Google Scholar]
  25. Deater-Deckard K (2004). Parenting stress. Yale University Press, New Haven. [Google Scholar]
  26. Drapela LA, & Mosher C (2007). The conditional effect of parental drug use on parental attachment and adolescent drug use: Social control and social development model perspectives. Journal of Child & Adolescent Substance Abuse, 16(3), 63–87. [Google Scholar]
  27. Edin K, Tach L, & Mincy R (2009). Claiming fatherhood: Race and the dynamics of paternal involvement among unmarried men. The Annals of the American Academy of Political and Social Science, 621, 149–177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Fagan J, Palkovitz R, Roy K, & Farrie D (2009). Pathways to paternal engagement: Longitudinal effects of risk and resilience on nonresident fathers. Developmental psychology, 45(5), 1389. [DOI] [PubMed] [Google Scholar]
  29. Foster CE, et al. (2017). Connectedness to family, school, peers, and community in socially vulnerable adolescents. Children and youth services review, 81, 321–331 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Gross HE, Shaw DS, & Moilanen KL (2008). Reciprocal associations between boys’ externalizing problems and mothers’ depressive symptoms. Journal of abnormal child psychology, 36(5), 693–709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Guttman LM, & Eccles JS (1999). Financial strain, parenting behaviors, and adolescents’ achievement: Testing model equivalence between African American and European single- and two-parent families. Child Development, 70, 1464–1476. [DOI] [PubMed] [Google Scholar]
  32. Hamilton BE, Martin JA, & Osterman MJ (2016). Births: Preliminary data for 2015. National vital statistics reports: from the Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System, 65, 1–15. [PubMed] [Google Scholar]
  33. Hawkins D, & Weis JG (1985). The social development model: an integrated approach to delinquency prevention. Journal of Primary Prevention, 6(2), 73–97. [DOI] [PubMed] [Google Scholar]
  34. Hoover-Dempsey KV, Walker JM, Jones KP, & Reed RP (2002). Teachers involving parents (TIP): Results of an in-service teacher education program for enhancing parental involvement. Teaching and Teacher Education, 18, 843–867. [Google Scholar]
  35. Howard TC, & Reynolds R (2008). Examining parent involvement in reversing the underachievement of African American students in middle-class schools. Educational Foundations, 22, 79–98. [Google Scholar]
  36. Kaminski JW, Valle LA, Filene JH, & Boyle CL (2008). A meta-analytic review of components associated with parent training program effectiveness. Journal of Abnormal Child Psychology, 36(4), 567–589. [DOI] [PubMed] [Google Scholar]
  37. Kelly AB, O’Flaherty M, Toumbourou JW, Homel R, Patton GC, White A, & Williams J (2012). The influence of families on early adolescent school connectedness: Evidence that this association varies with adolescent involvement in peer drinking networks. Journal of abnormal child psychology, 40(3), 437–447. [DOI] [PubMed] [Google Scholar]
  38. Kessler RC, Andrews G, Mroczek D, Utsun B, & Wittchen H (1998). The World Health Organization Composite International Diagnostic Interview Short-Form (CIDI-SF). International Journal of Methods in Psychiatric Research, 7, 171–185. [Google Scholar]
  39. Kidger J, Araya R, Donovan J, & Gunnell D (2012). The effect of the school environment on the emotional health of adolescents: a systematic review. Pediatrics, 129(5), 925–949. [DOI] [PubMed] [Google Scholar]
  40. Kim BE, Gilman AB, Hill KG, & Hawkins JD (2016). Examining protective factors against violence among high-risk youth: Findings from the Seattle Social Development Project. Journal of criminal justice, 45, 19–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Kim Y (2009). Minority parental involvement and school barriers: Moving the focus away from deficiencies of parents. Educational Research Review, 4(2), 80–102. [Google Scholar]
  42. Kline RB (2011). Principles and practice of structural equation modeling. New York: The Guilford press. [Google Scholar]
  43. Kohl GO, Lengua LJ, & McMahon RJ (2000). Parent involvement in school conceptualizing multiple dimensions and their relations with family and demographic risk factors. Journal of school psychology, 38(6), 501–523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Koonce DA, & Harper JW (2005). Engaging African American parents in the schools: A community-based consultation model. Journal of Educational and Psychological Consultation, 16, 55–74. [Google Scholar]
  45. Kosterman R, Haggerty KP, Spoth R, & Redmond C (2004). Unique influence of mothers and fathers on their children’s antisocial behavior. Journal of Marriage and Family, 66(3), 762–778. [Google Scholar]
  46. Leve LD, Kim HK, & Pears KC (2005). Childhood temperament and family environment as predictors of internalizing and externalizing trajectories from ages 5 to 17. Journal of abnormal child psychology, 33(5), 505–520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Lovejoy MC, Graczyk PA, O’Hare E, & Neuman G (2000). Maternal depression and parenting behavior: A meta-analytic review. Clinical psychology review, 20(5), 561–592. [DOI] [PubMed] [Google Scholar]
  48. Johnson SL, Finigan N, Bradshaw C, Haynie D, & Cheng T (2012). Urban African American parents’ messages about violence. Journal of adolescent research, 28(5), 511–534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. MacCallum RC, Browne MW, & Sugawara HM (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological methods, 1(2), 130. [Google Scholar]
  50. MacKinnon DP, Warsi G, & Dwyer JH (1995). A simulation study of mediated effect measures. Multivariate Behavioral Research, 30, 41–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. McNeal RB (1999). Parental involvement as social capital: Differential effectiveness on science achievement, truancy, and dropping out. Social Forces, 78, 117–144. [Google Scholar]
  52. McNeely CA, Nonnemaker JM, & Blum RW (2002). Promoting school connectedness: Evidence from the national longitudinal study of adolescent health. Journal of School Health, 72, 138–146. [DOI] [PubMed] [Google Scholar]
  53. Mistry RS, Vandewater EA, Huston AC, & McLoyd VC (2002). Economic well‐being and children’s social adjustment: The role of family process in an ethnically diverse low‐income sample. Child development, 73(3), 935–951. [DOI] [PubMed] [Google Scholar]
  54. Murray KW, Finigan-Carr N, Jones V, Copeland-Linder N, Haynie DL, & Cheng TL (2014). Barriers and facilitators to school-based parent involvement for parents of urban public middle school students. SAGE open, 4(4), 2158244014558030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Muthén LK (2018). Mplus Users Guide Los Angeles, CA: Muthén & Muthén; 2010; Computer software and manual. [Google Scholar]
  56. Nation M, Crusto C, Wandersman A, Kumpfer KL, Seybolt D, Morrissey-Kane E, & Davino K (2003). What works in prevention: Principles of effective prevention programs. American Psychologist, 58(6–7), 449–456. [DOI] [PubMed] [Google Scholar]
  57. National Adolescent Health Information Center. (2007). Fact sheet on violence: Adolescents and young adults. San Francisco: National Adolescent Health Information Center. [Google Scholar]
  58. Neece CL, Green SA, & Baker BL (2012). Parenting stress and child behavior problems: A transactional relationship across time. American journal on intellectual and developmental disabilities, 117(1), 48–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Nord CW, Brimhall D, & West J (1997). Fathers’ involvement in their children’s schools (NCES 98–091). Washington, DC: U.S. Department of Education, National Center for Educational Statistics. [Google Scholar]
  60. Patterson G, DeBarsyshe B, & Ramsey E (1989). A developmental perspective on antisocial behavior. Am Psychol, 44, 329–335. [DOI] [PubMed] [Google Scholar]
  61. Paulson JF, Dauber SE, & Leiferman JA (2011). Parental depression, relationship quality, and nonresident father involvement with their infants. Journal of Family Issues, 32, 528–549. [Google Scholar]
  62. Ponnet K (2014). Financial stress, parent functioning and adolescent problem behavior: An actor–partner interdependence approach to family stress processes in low-, middle-, and high-income families. Journal of youth and adolescence, 43(10), 1752–1769. [DOI] [PubMed] [Google Scholar]
  63. Puff J, & Renk K (2014). Relationships among parents’ economic stress, parenting, and young children’s behavior problems. Child Psychiatry & Human Development, 45(6), 712–727. [DOI] [PubMed] [Google Scholar]
  64. Rubin DB (1976). Inference and missing data. Biometrika, 63(3), 581–592. [Google Scholar]
  65. Shaw DS, Connell A, Dishion TJ, Wilson MN, & Gardner F (2009). Improvements in maternal depression as a mediator of intervention effects on early childhood problem behavior. Development and psychopathology, 21(2), 417–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Shochet IM, Smyth T, & Homel R (2007). The impact of parental attachment on adolescent perception of the school environment and school connectedness. Australian and New Zealand Journal of Family Therapy, 28(2), 109–118. [Google Scholar]
  67. Smrekar C, & Cohen-Vogel L (2001). The voices of parents: Rethinking the intersection of family and school. Peabody journal of education, 76(2), 75–100. [Google Scholar]
  68. Stoddard SA, Whiteside L, Zimmerman MA, Cunningham RM, Chermack ST, & Walton MA (2013). The relationship between cumulative risk and promotive factors and violent behavior among urban adolescents. American journal of community psychology, 51(1–2), 57–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Thompson DR, Iachan R, Overpeck M, Ross JG, & Gross LA (2006). School connectedness in the health behavior in school-aged children study: The role of student, school, and school neighborhood characteristics. Journal of School Health, 76, 379–386. [DOI] [PubMed] [Google Scholar]
  70. Tsuchiya K, Qian Y, Thomas A, De Loney EH, & Caldwell CH (2018). The effects of multiple stressors and social support on depressive symptoms among nonresident Black fathers. American Journal of Community Psychology, 62(3–4), 464–475. [DOI] [PubMed] [Google Scholar]
  71. Voisin DR, Salazar LF, Crosby R, Diclemente RJ, Yarber WL, & Staples-Horne M (2005). Teacher connectedness and health-related outcomes among detained adolescents. Journal of Adolescent Health, 37(4), 337–e17. [DOI] [PubMed] [Google Scholar]
  72. Wilson S, & Durbin CE (2010). Effects of paternal depression on fathers’ parenting behaviors: A meta-analytic review. Clinical psychology review, 30(2), 167–180. [DOI] [PubMed] [Google Scholar]
  73. Williams TT, & Sanchez B (2013). Identifying and decreasing barriers to parent involvement for inner-city parents. Youth & Society, 45, 54–74.c [Google Scholar]
  74. Zimmerman GM, & Messner SF (2013). Individual, family background, and contextual explanations of racial and ethnic disparities in youths’ exposure to violence. American journal of public health, 103(3), 435–442. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES