African-American fathering of young children in violent neighborhoods: paternal protective strategies and their predictors.

Using an ecological framework, this study investigated African-American fathers' use of five protective strategies to keep their preschool children safe from community violence. Father, child, and contextual predictors of fathers' protective strategies were also examined. In-depth interviews with 61 African-American Head Start fathers and father figures revealed that participants were most likely to adopt the strategy of monitoring and teaching personal safety, followed by teaching neighborhood survival tactics, reducing exposure to violent media, engaging in community activism, and instructing children to fight back. Overall, parenting practices, social support, and psychological functioning were the best predictors of these strategies, with one exception. Child's gender was the best predictor of the strategy "reduce exposure to violent media," with fathers of sons more likely to limit such exposure. Implications of these findings for researchers and practitioners are discussed.

Key Words: African-American fathers, community violence, preschool children

**********

Community violence is a national epidemic that is exposing growing numbers of families to crime, drug activity, and homicide (U.S. Department of Health and Human Services, 2000). Parenting in violent neighborhoods is an especially challenging task as parents cannot rely on many of the child rearing practices used in safer contexts. Allowing children to play on the playground or in their yard, walking children to school, or encouraging children to explore their environment may be dangerous and place children at risk of harm. The constant threat of community violence forces many parents to find alternative parenting strategies that will ensure their children's safety at all times.

Over the past decade, researchers have begun to investigate the strategies mothers and female caregivers use to protect their preschool and school-age children from violence exposure (Garbarino, Kostelny, & Dubrow, 1991; Hill, Hawkins, Raposo, & Cart, 1995; Holland, Koblinsky, & Anderson, 1995; Jarrett, Jefferson, & Roach, 2000; Mohr, Fantuzzo, & Abdul-Kabir, 2001; Randolph, Koblinsky, & Roberts, 1996). These studies reveal several common behavioral coping strategies used by mothers, including keeping children physically close, providing constant supervision/chaperonage, teaching practical household safety skills (e.g., not sitting by windows), and restricting neighborhood activities (e.g., the use of community playgrounds). Some mothers also report the use of spiritual or cognitive strategies (e.g., prayer and positive thinking) and community-based strategies (e.g., reliance on informal neighborhood leaders and local institutions) to keep their children safe.

Most of the research on maternal strategies to protect children living in violent neighborhoods is qualitative in nature and focuses on low-income African-American families (e.g., Hill et al., 1995; Jarrett et al., 2000; Randolph, Koblinsky, & Roberts, 1998). African-American families are 10 times more likely than European-American families to live in neighborhoods where at least 30 percent of residents are poor (Duncan, Brooks-Gunn, & Klebanov, 1994) and where there are high levels of joblessness (Chase-Lansdale & Gordon, 1996). African-American families are also disproportionately represented in neighborhoods characterized by high violence, crime, and drug activity (Sampson, Raudenbush, & Earls, 1997).

Father absence appears widespread among low-income African-American families, as many fathers have never married or lived in the same household as their child's mother (Ventura & Bachrach, 2000). However, recent evidence from the Fragile Families study has begun to challenge the "absent father myth," suggesting that--at least early on--many unwed fathers are involved in their children's lives regardless of their residential status (Carlson & McLanahan, 2002). In another study of urban African-American families, approximately half of nonresidential fathers had regular contact and provided some financial support during their children's preschool years (Coley & Chase-Landsdale, 1999).

Qualitative studies of low-income African-American fathers reveal the complexities of these men's roles in family life (Anderson, Kohler, & Letiecq, 2002; Hamer, 1998; Jarrett, Roy, & Burton, 2002), suggesting that definitions of fatherhood comprise both economic and relational aspects (such as visiting children and spending "quality time") and that many fathers endeavor to fulfill their roles "in the face of tremendous sociocultural barriers" (Nelson, Clampet-Lundquist, & Edin, 2002, p. 552). While a majority of African-American fathers are nonresidential, approximately 40% live in the same home as their children, including 4% who are the sole parent (U.S. Census Bureau, 2000). Such evidence suggests that many fathers are present in young children's lives in some capacity; yet little is known about the parenting practices of these men--especially in violent neighborhoods. Thus, the purpose of this study was to investigate African-American fathers' use of various strategies to keep their young children safe from community violence. A second purpose was to examine the extent to which selected father, child, and contextual factors predicted the use of fathers' protective strategies.

THEORETICAL FRAMEWORK

A conceptual framework that is particularly relevant to the study of African-American fathering in violent neighborhoods is the cultural ecology model (Bronfenbrenner, 1986; Hamer & Marchioro, 2002). This model stresses the need to examine patterns of socialization and parenting competencies based on cultural contexts that are central to the attitudes, skills, and values of parents within a specific culture or subculture (Ogbu, 1981). The ecological model influenced our work by expanding our conceptualization of fathering beyond the individual and family levels to also include larger contexts or ecologies that affect African-American family functioning and well-being. For example, we recognize that African-American men operate as members of kin networks and communities and that these larger systems influence and are influenced by men's ability to nurture, provide for, and protect their families (Allen & Connor, 1997; McAdoo, 1993).

Consistent with the ecological framework, many researchers have begun to investigate the complex array of factors related to fathering, including individual "father" factors such as psychological well-being, parenting knowledge and skills, and residential status (Anderson, Kohler, & Letiecq, in press; Doherty, Kouneski, & Erickson, 1998). Doherty et al., in their model of influences on fatherhood, also consider child factors (e.g., gender, age), mother factors (e.g., attitude toward father), co-parental factors (e.g., marital status, custodial arrangement), and contextual factors (e.g., economic opportunity, social support). When exploring factors likely to influence fathering in violent neighborhoods, we chose to examine fathers' psychological well-being, fathers' parenting practices, fathers' personal history of violence exposure, their child's gender, and one contextual factor--social support. Although there are undoubtedly many other factors that may influence fathering in violent neighborhoods, this exploratory study is a first attempt to shed light on some of the relationships that may emerge when fathers attempt to protect their children from community violence. Following is a brief review of our selected variables and speculations as to how they may relate to fathers' use of protective strategies.

FATHER FACTORS

Psychological Well-Being. Studies examining psychological adjustment and parenting quality consistently show a positive relationship between parent psychological well-being and parenting attitudes and skills (Andrews-Cameron, 1998; Brody, McBride Murry, Kim, & Brown, 2002). Research on depression has shown that maternal depression may have adverse consequences for children, depleting the energy mothers have to nurture their children's cognitive and social-emotional development (Brody et al., 2002; Taylor, Zuckerman, Harik, & Groves, 1994). These findings appear consistent across racial and ethnic groups (Mistry, Vandewater, Huston, & McLoyd, 2002). Although there is a paucity of research investigating the psychological well-being of low-income African-American fathers, Anderson et al. (in press) recently examined depressive symptomatology in low-income, nonresidential African-American fathers in the Washington, D.C., metropolitan area. Findings revealed that 56% of the study fathers reported depressive symptoms indicating cause for clinical concern. One can speculate that a number of Head Start fathers living in low-income, high violence neighborhoods are likely to experience depressive symptoms that diminish the psychological resources they have available to protect their children from community violence.

Parenting Practices. Parenting attitudes and practices also seem likely to affect the ways in which fathers equip their young children to deal with community violence. Baumrind (1967) identified three parenting typologies related to child behavior, including (1) authoritative parenting, which includes, the demonstration of warmth, nurturance, consistency, and verbal reasoning in raising a child; (2) authoritarian parenting, which involves the use of control, coercive tactics, physical punishment, and rule-making without consulting the child, and (3) permissive parenting, which is characterized by lack of punishment and failure to follow through with child discipline. One can speculate that fathers who adopt authoritative parenting practices may make greater efforts to establish and explain rules for personal, home, and neighborhood safety than fathers who use other parenting styles. African-American fathers who use authoritarian practices may restrict their children's exposure to the neighborhood in order to have greater control over their behavior and to protect them from harm. In contrast, permissive parents may monitor theft children less frequently, make fewer attempts to teach personal safety, and be less likely to limit children's neighborhood activity than parents who adopt other parenting styles.

History of Violence Exposure. Researchers investigating violence exposure have recently noted two problems with previous studies in this area, including the failure to study at-risk groups and the failure to disentangle the effects of experiencing, witnessing, and initiating violent behavior (Langinrichsen-Rohling & Neidig, 1995). Some violence theories suggest that individuals who are victimized by violence adopt aggression as a personal coping strategy (e.g., Widom, 1989). Such a strategy may, in turn, be transmitted to children, who model parental behavior or are directly taught to adopt an aggressive posture as a protective technique. Likewise, fathers who have witnessed violence, been victimized by violence, or themselves use physical aggression as a strategy for conflict resolution may be more likely to encourage children's use of aggressive behavior in peer conflict situations than fathers with little history of violence exposure or use. It is also plausible that fathers who have witnessed or experienced violence in their lives may be more aware of community dangers and may employ more hypervigilant monitoring strategies than fathers with little history of violence exposure.

CHILD FACTOR

Child's Gender. We were also interested in examining how fathers' protective strategies might differ as a function of the gender of their preschool child. Previous studies involving mainly White children suggest that parents of preschoolers are more likely to use physical or power-assertive styles of punishment with sons than with daughters, which may encourage sons to adopt more aggressive behavior (Maccoby & Jacklin, 1974; Smetana, 1988). Fathers are less likely to interpret fighting as aggressive behavior in their sons, so they may ignore it more than they do with daughters (Perry, Perry, & Weiss, 1989). Finally, fathers generally have more rigid ideas about gender roles than mothers and enforce them more strongly with young children (Wood, 1994). Such factors suggest that African-American fathers may employ different strategies in preparing preschool sons and daughters to be safe from community violence.

CONTEXTUAL FACTOR

Social Support. A contextual variable that may affect African-American fathers' protective strategies is the social support they receive from family, friends, co-workers, church members, teachers, and other professionals. Social support has been defined as emotional, instrumental, material, or informational assistance offered by members of a person's informal or formal (community agency, institution) networks (Dunst & Trivette, 1990). Support from extended networks has been found to bolster self-esteem (Taylor, Chatters, Tucker, & Lewis, 1990), enhance parent-child relationships (Crnic, Greenberg, Ragozin, Robinson, & Basman, 1983), and strengthen one's ability to deal with social problems (Sarason, Sarason, & Shearin, 1986). Thus, fathers who experience a high level of social support may be more proactive in developing and teaching safety procedures than fathers who lack such support. Fathers with limited social support may feel that the major way they can keep children safe in violent neighborhoods is to confine children to their homes or severely restrict their outdoor play. Moreover, fathers may be less likely to engage in community activism (e.g., neighborhood watch patrols, local clean-up activities) to reduce pervasive neighborhood violence without the aid of informal or formal support networks.

Although one can speculate about possible relationships between father, child, and contextual factors and the strategies fathers use to protect their children from community violence, there are no previous studies on this topic. To shed light on these relationships and inform the extant literature, the current study investigated two research questions: (1) To what extent do African-American fathers employ various strategies to protect children from community violence? and (2) What father, child, and contextual factors best predict or account for the variance in these paternal protective strategies?

METHOD

SAMPLE AND COMMUNITY CHARACTERISTICS

For this study, 61 African-American Head Start biological and social fathers volunteered to participate in in-depth interviews. Table 1 presents demographic characteristics of the sample. Most participants were the child's biological father (67.2%) and reported living in the same household with their preschooler (77.0%). Other participants identified themselves as the child's uncle, grandfather, or stepfather. Four participants had other kinship or social ties to the child--mother's boyfriend, child's cousin, child's great uncle, and close family friend. Social fathers were included in this study to reflect the "fictive" father presence common in African-American communities, where role flexibility and concern for children regardless of biological connection has been a strong tradition (Billingsley, 1968; Jarrett et al., 2002). All fathers lived in the same low-income neighborhood as their child or in close proximity. Although we did not ask fathers specifically about their personal income due to the sensitive nature of the topic, we chose to use the enrollment of the targeted child in Head Start (a national education program designed to enhance the school readiness of economically disadvantaged children) and father's residence in a low-income neighborhood as proxies for low-income status.

This study took place in southeast Washington, D.C., and a Maryland county adjoining the District of Columbia. Southeast Washington, D.C., is an area that has experienced high levels of community violence according to the Uniform Crime Report and Violent Crime Index (Federal Bureau of Investigation, n.d.). In comparison to the 50 states, the District of Columbia has had the highest teen violent death rate and the highest child death rate due to homicide since 1985 (Annie E. Casey Foundation, 1999). The Maryland county adjoining the District of Columbia also had high rates of community violence, including the fifth highest death rate due to homicide, suicide, and violent deaths of all 24 Maryland counties in 1998. Between 1990 and 1998, the county's juvenile violent crime arrest rate increased 25% (Advocates for Children and Youth, 2000). Targeted neighborhoods in this county had been identified as violent "hot spots" based on county police data measuring murder/negligent manslaughter, rape, robbery, and aggravated assault.

MEASURES

The following measures were included in a larger interview schedule administered orally to participants. This schedule included both closed-ended and open-ended questions; however, for the purpose of this study, only closed-ended items were included in the analyses. Each of the measures was chosen for its sound psychometric properties and previous use with African-American parents (e.g., Anderson et al., in press; Letiecq, Anderson, & Koblinsky, 1998; Randolph et al., 1998).

INDEPENDENT VARIABLES: FATHER, CHILD, AND CONTEXTUAL FACTORS

Psychological Well-Being. Father's psychological well-being was measured using the Center for Epidemiological Studies Depression Scale (CES-D; Radloff, 1977). The scale is a short, 20-item self-report measure specifically constructed to study depressive symptomatology in the general population. Respondents were asked to indicate how often they felt certain ways during the past week using a four-point scale ranging from 0 = rarely or none of the time (less than 1 day) to 3 = most or all of the time (5-7 days). Summing the 20 items yielded total scores ranging from 0 to 60, with a cut-off score of 16 indicating cause for clinical concern. The current study found the CES-D to have internal consistency with a Cronbach's alpha of .73.

Parenting Practices. The Parenting Practices Questionnaire (PPQ; Robinson, Mandleco, Olsen, & Hart, 1995) is a 62-item measure developed for use with mothers and fathers of preschool and school-age children. The PPQ assesses global parenting typologies consistent with Baumrind's (1967) typologies, with respondents obtaining separate scores for the authoritative, authoritarian, and permissive subscales. Each respondent was asked how often an item described him as a father of a preschool child using a 5-point scale anchored by 1 = never and 5 = always. The PPQ was scored by summing the subscale items and dividing by the total number of items within each subscale. Cronbach's alphas established the internal consistency of the subscales: authoritative, [alpha] = .93; authoritarian, [alpha] = .84; and permissive, [alpha] = .70.

History of Violence Exposure. Father's history of violence exposure was assessed using a modified version of the Conflict Tactics Scale (CTS; Langhinrichsen-Rohling & Neidig, 1995). Fathers were asked to report on the frequency of nine conflict behaviors (curse, threaten to hurt, push, slap, kick, hit with fist, hit with object, threaten with a knife or gun, use a knife or gun) experienced anywhere in their environment using a 7-point scale. The nine conflict tactics were presented in three sections addressing witnessing of violence, victimization (i.e., violent acts fathers have personally experienced), and personal use of aggressive/violent behavior. Response options included: 0 = never; 1 = once; 2 = twice; 3 = 3-5 times; 4 = 6-10 times; 5 = 11-20 times; and 6 = more than 20 times. Subscale scores were computed by summing the nine items of each subscale. Cronbach coefficient alphas for each subscale were: witnessing violence, [alpha] = .92; victimization, [alpha] = .82; and personal use, [alpha] = .83.

Gender of Target Child. Fathers were also administered a Demographic Questionnaire specifically designed for this study. This measure ascertained information about the target Head Start child, including the child's gender, age, and date of birth. Other demographic data collected for the study are presented in Table 1.

Social Support. Social support was assessed using a modified version of the Family Support Scale (FSS; Dunst, Jenkins, & Trivette, 1984). The 18-item FSS measures the degree to which different sources of support were helpful to families in raising young children during the previous six months. The FSS was modified by adding four items to the original scale: the helpfulness of the father's current partner, her parents, relatives, and friends (if different from the child's biological mother). Respondents were asked to rate the helpfulness of various support sources using a five-point scale anchored by 4 = extremely helpful and 0 = not at all helpful. Three subscales of support were analyzed: familial supports, including parents, partner, and own children; extra-familial supports, including friends, co-workers, social groups, and church members; and professional supports, including teachers, doctors, and social service workers. Indices of helpfulness were computed by summing the items within each subscale and dividing by the number of subscale items. Cronbach's coefficient alphas for the subscales established acceptable reliability: familial support, [alpha] = .80; extra-familial support, [alpha] = .79; and professional support, [alpha] = .72.

DEPENDENT VARIABLE

Parenting in Violent Neighborhoods. Last, this study employed a new quantitative measure of protective strategies used to keep children safe from community violence, the Parenting in Violent Neighborhoods Scale (PVNS). (1) The 47-item PVNS was constructed using data from three focus groups of fathers (Letiecq & Koblinsky, to press) and previous community violence studies examining maternal protection strategies (e.g., Hill et al., 1995; Holland et al., 1995; Randolph et al., 1998). Using correlation matrices and confirmatory factor analyses with Varimax rotation (Comrey & Lee, 1992), five subscales emerged: (1) monitoring and teaching personal safety (e.g., "I permit my child to play on playgrounds only when directly supervised by an adult," "I teach my child to tell the teacher if another child picks on her/him"); (2) teaching neighborhood survival tactics (e.g., "I talk to my child about safe routes for walking in the neighborhood," "I tell my child to avoid drug dealers or troublemakers in the neighborhood"); (3) reducing media violence exposure (e.g., "I keep my preschool child from playing video games that have a lot of violence," "I do not allow my preschool child to watch TV or movies that have violent scenes"); (4) engaging in community activism (e.g., "I participate in neighborhood watch or other groups that try to reduce neighborhood violence," "I call the police when I hear gunshots"); and (5) fighting back (e.g., "I tell my preschool child to fight back in order to be safe;" "I carry a weapon, like mace or a knife, in case I need to protect myself or my child"). Item response options were anchored by 0 = never and 4 = always. Cronbach's coefficient alphas for the five subscales were monitoring and teaching personal safety (16 items), [alpha] = .91; teaching neighborhood survival (10 items), [alpha] = .80; reducing media violence exposure (4 items), [alpha] = .56; community activism (7 items), [alpha] = .81; and fighting back (10 items), [alpha] = .84.

PROCEDURE

We began this study by establishing an advisory committee made up of Head Start fathers, teachers, community leaders, and experts in the field of African-American family life to assist with recruitment of fathers, selection of culturally relevant measures, development of the PVNS, and interpretation of findings. After conducting a pilot study to confirm the reliability and cultural sensitivity of selected measures, we enlisted the help of Head Start teachers and staff to identify and recruit fathers and father figures of Head Start children to participate in the study. We also employed snowball-sampling techniques, asking interested fathers to help us identify other men involved with Head Start children. In total, 61 fathers/father figures consented to participate in one-on-one interviews conducted by trained African-American male graduate and undergraduate student interviewers. Following the interview schedule developed for this study, interviewers read aloud all items and recorded fathers' responses. Interviews took place at the father's home or Head Start center and lasted approximately 1.5 to 2 hours. All participants received a $25 stipend for their time and effort. Collected data were checked for errors, cleaned, and entered into SPSS for Windows.

RESULTS

PRELIMINARY ANALYSES

Given that this study included both biological and social fathers, we first compared the demographic profiles of these two groups of men. Using independent t tests and chi square analyses, we found few significant differences; however, biological fathers were younger (M = 33.9, SD = 6.5) than social fathers (M = 40.8, SD = 15.3; t (59) = 2.47, p < .05), more likely to live with the child in the same household (85.4%) than social fathers (60.0%; [chi square] (1, N = 61) = 4.89, p < .05), and involved in the Head Start child's life for more years (M = 4.1, SD = 0.8) than social fathers (M = 3.6, SD = 1.1; t (59) = 2.13, p < .05). We also examined the demographic profiles of fathers as a function of their marital status (married, not married) and residential status (residential, nonresidential), but no significant differences emerged. Next, we examined the independent variables as a function of father's relationship to the target child (biological, social), father's marital status, residential status, and child's gender using independent t tests and chi-square analyses. These preliminary analyses of the independent variables revealed no significant child relationship, marital, residential, or gender differences, which allowed us to aggregate the data for all subsequent analyses. Table 2 presents the means, standard deviations, and bivariate correlation coefficients for the independent variables.

PARENTING IN VIOLENT NEIGHBORHOODS

Also presented in Table 2 are five protective parenting strategies fathers used to keep their children safe. Overall, fathers reported monitoring and teaching personal safety "very often" (M = 2.8, SD = 0.7), followed closely by teaching neighborhood survival tactics (M = 2.7, SD = 0.8). Fathers reported reducing exposure to violent media slightly more than "half the time" (M = 2.5, SD = 0.8) and engaging in community activism a little less than "half the time" (M = 1.8, SD = 0.8). Participating fathers were least likely to use fighting back to protect children from violence, reporting this strategy only "once in a while" (M = 1.4, SD = 0.7).

PREDICTORS OF FATHERS' PROTECTIVE STRATEGIES

To examine father, child, and contextual factors that best predict paternal protective strategies, five regression models were run (see Table 3). Predictor variables included father's psychological well-being (depression), parenting practices (authoritative, authoritarian, permissive), father's history of violence exposure (victimization, witness), child's gender, and social support (familial, extra-familial, professional). Note that because the variable "personal use of violence" was not significantly intercorrelated with any protective strategy, it was omitted from the regression analyses. The rationale for selecting the "best" regression model (i.e., the equation that maximizes [R.sup.2]) was based on constructing the model with backward elimination (Pedhazur, 1982). After the first step, in which all variables were entered into the model, the variable with the smallest partial correlation coefficient was examined, and, if the probability of its F was greater than the criterion value of .05, the variable was removed. This procedure was repeated until the "best" model was constructed.

As shown in Table 3, five variables were significant predictors of monitoring and teaching personal safety, accounting for 58% of the variance: authoritative parenting, permissive parenting, paternal depression, family support, and extra-family support. Three parenting variables were significant predictors of teaching neighborhood survival tactics and explained 37% of the variance: authoritative parenting, authoritarian parenting, and permissive parenting. Child gender was the only significant predictor of reducing exposure to violent media, accounting for 9% of the variance; fathers of sons were more likely to limit violent media exposure than fathers of daughters. Extra-familial support and authoritative parenting were significant predictors of engaging in community activism, explaining 25% of the variance. Finally, three variables were significant predictors of fighting back and accounted for 37% of the variance: paternal depression, family support, and extra-family support.

DISCUSSION

AFRICAN-AMERICAN FATHERS' PROTECTIVE STRATEGIES

One major goal of this study was to examine the extent to which African-American fathers employed five strategies to protect their children from community violence. Findings revealed that fathers were most likely to adopt the strategy of monitoring and teaching personal safety by confining young children to their homes, supervising their children closely on streets and playgrounds, and teaching their children personal safety skills, such as telling the teacher or the parent if "picked on" by another child, and resolving peer disputes peacefully with calm words or by walking away. Fathers in this study reported actively teaching their preschoolers prosocial skills that foster empathy and reduce peer conflict, a finding that may reflect the fathers' history of involvement with the Head Start program. Fathers' hypervigilant monitoring of preschoolers' contact with peers and the larger community is a strategy also adopted by African-American mothers of preschool and elementary school children in poor inner-city neighborhoods (Hill et al., 1995; Jarrett et al., 2000; Mohr et al., 2001; Randolph et al., 1998). While such close supervision and confinement may be critical to keeping children physically safe, such restriction may hinder young children's ability to explore their environment, cultivate social relationships, master motor learning skills, and achieve other developmental milestones (Holland et al., 1995).

Another common protective strategy reported by fathers was teaching neighborhood survival tactics, including instruction about safe routes, how to respond to sounds of gunfire, how to dial 911, and how to avoid drug dealers and neighborhood "troublemakers." This strategy also included teaching children about the "real-life pain that comes from violence, such as bleeding or dying when you get shot." Despite fathers' positive intentions, some of this information may be too cognitively complex or frightening for three-, four-, and five-year-olds. Preschoolers are unlikely to be able to identify drug dealers, and therefore unable to avoid them. Teaching preschoolers about the pain, death, and violence associated with drug selling may also increase children's fears and anxieties, undermining positive psychological and adaptive functioning (Garbarino, Dubrow, Kostelny, & Pardo, 1992). When considering the challenges of parenting in violent neighborhoods, fathers (and mothers) face difficult choices about how to promote healthy physical, cognitive, and social-emotional development while simultaneously ensuring child safety. Clearly, in violent contexts, these two parental functions may be at odds with one another.

The growing isolationism in low-income, high-violence neighborhoods, such as those found in the current study, contributes to the erosion of traditional strengths in African-American community life. Restricting children's neighborhood contact--literally keeping children out of sight of one's friends and neighbors--runs counter to African-American values of child-centeredness (Hill, 1993) and collectivism (Nobles & Goddard, 1993), where children are parented by the entire neighborhood and represent the continuity and well-being of the community. Several fathers noted that there are no longer grandparents, "big mamas," and other surrogate parents on the street to nurture, teach, and discipline their children. Although some fathers attempted to counterbalance neighborhood dangers by taking their young children to safer environments--parks, malls, and recreation centers in nearby suburbs--such strategies do not build the social support networks that formerly characterized their inner-city blocks and neighborhoods. Moreover, many fathers reported lacking the resources (money, transportation) and free time to access these safer places on a routine basis.

A third strategy adopted by African-American fathers was to reduce children's exposure to violent media. Such efforts may help to reduce aggressiveness and desensitization to violence in young children, as well as increase their perceptions of safety in the world around them (Smith & Donnerstein, 1998). However, as parents keep their children indoors to avoid potential violence in the community, this strategy may be difficult to employ consistently, especially with older children present. Many fathers related how they kept their child occupied in the home by watching television or playing video games--both forms of media containing high levels of violent imagery and therefore requiring close parental supervision (Villani, 2001).

A fourth protective strategy, used occasionally by African-American fathers of preschoolers, was engaging in community activism. It is notable that the inner-city fathers in this investigation appeared more likely to adopt this strategy than mothers in previous studies, who often feared retribution for their activism from gangs or drug dealers (Hill et al., 1995; Jarrett et al., 2000; Mohr et al., 2001). Many fathers reported participation in some community-level activities, such as Neighborhood Watch or church-based projects, to monitor potential troublemakers, clean up a block, or remove drug markets. It seems likely that some fathers believe they have the physical strength and support of other males to confront neighborhood problems, while mothers often lack these resources. Moreover, fathers may feel it is their duty as men to engage in community action that will improve their family's safety and well-being.

A final strategy, used infrequently by fathers, was teaching both sons and daughters to confront potential danger by fighting back. Some fathers also modeled this strategy by carrying weapons for protection. The importance of posturing, standing up for oneself, and earning respect (particularly from dealers and gang members) has been noted in other studies of African-American men in low-income neighborhoods in the Baltimore/Washington, D.C., area (Whitehead, 1997). However, the finding that fathers used this practice only "once in a while" may reflect their knowledge of the real dangers of conflict in violent neighborhoods, where confrontation can quickly escalate to injury and death--even among the youngest of children.

PREDICTORS OF PATERNAL STRATEGIES

Beyond identifying the relative use of child safety strategies by African-American fathers residing in violent neighborhoods, this study examined father, child, and contextual predictors of these protective strategies. Overall, the most salient predictors included fathers' parenting practices, their social support, and their psychological well-being.

With regard to parenting practices, fathers who employed more authoritative parenting styles were more likely to monitor and teach personal safety, teach neighborhood survival practices, and engage in community activism. In essence, fathers who relied on nurturance, consistency, verbal reasoning, and problem-solving in parenting used similar tactics to ensure their child's safety. Conversely, fathers who adopted permissive styles of parenting were less likely to supervise children and prepare them with personal or neighborhood-level safety skills. It is possible that some permissive fathers underestimated neighborhood dangers, living under an "illusion of invulnerability" where one has enhanced feelings of control and low levels of fear and anxiety (Perloff, 1983). While this adaptive coping mechanism may promote fathers' sense of safety, it appears likely to increase children's exposure to community danger and limit their ability to protect themselves. Interestingly, fathers who favored authoritarian practices--who were more controlling and rigid about family rules--were also more involved in trying to control their child's neighborhood behavior by teaching them detailed, often complex strategies for identifying dangerous neighborhood individuals and places and handling crisis situations (e.g., dialing 911). Although it has been noted that some of these tasks may challenge preschoolers' developmental skills, fathers believed that strict enforcement of safety rules and routine rehearsal of safety drills (e.g., ducking when you hear gunfire) were essential to protecting children in volatile neighborhoods.

A second significant predictor of fathers' protective strategies was social support. Our findings suggest that fathers with more social support, especially familial support, were more likely to monitor and teach children personal safety. A strong family support network may help to buffer families from the stress of violence and provide fathers with additional time to develop children's personal safety and peer relationship skills. Greater extra-familial support from friends, neighbors, church members, and coworkers was a strong predictor of engaging in community activism. This finding suggests that when fathers were able to identify a critical mass of caring adults, they were more willing to involve themselves in interventions to establish informal social control and improve public safety in their neighborhoods. Interestingly, fathers with more extra-familial support reported lower levels of engagement in child monitoring and personal safety instruction, and teaching children to handle peer aggression by fighting back. It is possible that some of these fathers relied on members of their extra-familial support network to provide childcare for their preschoolers, and expected these caregivers to share responsibility for helping to protect their children from community violence. With greater support from friends, neighbors, the church, and coworkers, fathers may have felt less need to teach children aggressive strategies like fighting back because they were more confident of the network's ability to maintain safety within their child's home, school, and neighborhood environment. More socially isolated fathers may have sensed a greater need to teach their offspring "to stick up for themselves." Surprisingly, higher levels of family support were not only associated with teaching more personal safety, but also with teaching children to fight back. Possibly some members of the fathers' extended families believed that children should learn to stand up for themselves and passed these views on to fathers, resulting in children being taught a wider repertoire of personal safety and defensive skills.

In addition to parenting practices and social support, fathers' psychological well-being emerged as another factor that predicted paternal protective strategies. Fathers who reported having more depressive symptoms were more likely to monitor and teach personal safety and to instruct children to fight back than those with fewer depressive symptoms. Although a somewhat perplexing relationship, one can speculate that fathers with more emotional distress were more likely to perceive a dangerous environment for their children, and responded by employing a wider range of protective strategies. Thus, fathers who felt more helpless and powerless in their own lives were more likely to carry weapons and to teach their children aggressive skills to protect themselves in a threatening environment. It should be noted, however, that less than 10% (5) of fathers in this study scored in the clinical range for depressive symptomatology. Therefore, it is possible that our measure did not adequately tap into fathers' psychological health or that the measure is a better proxy for some other characteristic of fathers' well-being, such as their willingness to share personal feelings. Fathers who are more comfortable with their feelings may be more likely to talk with children about how it feels to be bullied or afraid, or how children can protect themselves from potential peer aggression. Clearly, more research is needed to explore the relationship between paternal mental health and protective parenting strategies.

Although gender of the child was not predictive of the majority of protective strategies, it was a significant predictor of father's attempts to reduce their preschoolers' exposure to violent media. Fathers of sons were more likely to limit their child's exposure to media violence on TV and in video games than fathers of daughters. Fathers may have had special concerns about sons encountering violent images of African-American men and experienced fears that sons would internalize the images or imitate the behavior. Fathers also may recognize that as boys grow older, they are more likely than girls to confront life-threatening aspects of neighborhood life. The extremely high rates of homicide and serious violent crime victimization for African-American male youth (Federal Interagency Forum on Child and Family Statistics, 1999) reinforce these fears. Since most media perpetrators of violence are male, fathers of young daughters may have had less concern about their child's exposure to media violence. However, caution should be exercised in interpreting this finding because child's gender explained a relatively small percentage (9%) of the variance, and the subscale for reducing media violence exposure had only marginal reliability.

Finally, our measure of father's history of violence exposure did not significantly predict any of the paternal protective strategies. Regardless of fathers' prior experiences with violence, virtually all were aware of neighborhood threats to young children's safety, so this "father factor" may have been less salient in predicting paternal safety behaviors. It is also possible that a social desirability bias resulted in fathers giving more socially acceptable answers to items dealing with personal use of violent behavior, particularly because many interviews were administered at a Head Start center. Obtaining accurate information about a sensitive topic like violence exposure and use may be better achieved using an anonymous questionnaire than a personal interview.

IMPLICATIONS FOR PRACTITIONERS

Overall, the African-American fathers of Head Start children in this study were actively involved in attempting to protect their children from community violence. Findings suggest that other fathers in violent neighborhoods might benefit from intervention programs that promote authoritative parenting styles and enhance fathers' social support networks. Fathers of both sons and daughters may be motivated to join initiatives that focus on working together to eliminate violence in the community, rather than programs more narrowly focused on parenting. However, these interventions should include efforts to develop authoritative parenting practices that foster nurturance, verbal reasoning, problem-solving, and promotion of children's prosocial skills and discourage more permissive styles of parenting. Such program components may not only be important in keeping children safe, but may also contribute to a young child's development of security, trust, and empathy (Garbarino et al., 1992). Father or parent initiatives involving African-American families should draw on Africenttic principles (e.g., communalism, spirituality, harmony) to help fathers restore some of the cohesion and mutual aid that has long sustained African-American neighborhoods despite economic hardship (e.g., Billingsley, 1968; Nobles & Goddard, 1993; Randolph, Damond, & Washington, 1995). Parenting initiatives might also establish social support networks of low-income fathers and work to enhance their coping strategies, interpersonal communication, and sense of empowerment (Fagan & Stevenson, 2002). Such programs may provide participants with an extended social family who can share the task of tackling neighborhood problems, as well as the rewards and challenges of being the father of a young child.

Early childhood educators and family practitioners must also recognize that before fathers can focus on building social support and enhancing parenting practices, they may first need help in dealing with mental health issues, such as depression. Although this study did not find fathers with high levels of depressive symptomatology, other studies suggest that parents living in violent neighborhoods may be struggling with depression, post-traumatic stress disorder, anxiety, anger, fear, isolation, and guilt (Anderson et al., in press; Garbarino et al., 1991; Lotion & Saltzman, 1993). Such parents may find themselves preoccupied, distracted, and unable to provide consistent, effective parenting (Pinderhughes, Nix, Foster, & Jones, 2001), and their feelings of distress may contribute to coping strategies such as arming themselves or teaching their children to respond aggressively in conflict situations. Family practitioners must reach out to families living in violent communities to provide the comprehensive support and mental health services that will facilitate parental functioning and promote child well-being.

LIMITATIONS AND FUTURE RESEARCH

This exploratory study is unique in shedding light on African-American fathers' protective strategies and several ecological variables that best predict those strategies; however, it is not without limitations. Despite use of numerous tactics to identify and recruit fathers for participation, the study is characterized by its small, nonrandom, convenience sample of volunteer biological and social fathers and is therefore limited in the generalizability of its outcomes. The majority of study fathers were highly involved with their Head Start child, regardless of their biological relationship to the child, marital status, or residential status. Almost twice as many fathers (77%) lived with their child as African-American fathers in the population at large (40%). While current findings may appear to contradict some studies suggesting that biology and marriage matter (for review, see Hofferth, Pleck, Stueve, Bianchi, & Sayer, 2002), researchers have noted that African-American fatherhood is an active, flexible relationship where families depend on both biological and social fathers to rear their children (Hamer, 1998; Jarrett et al., 2002; Letiecq & Koblinsky, in press). The complex ways in which African-American fathers define their roles and responsibilities in children's lives have yet to be fully understood.

To extend generalizability, future research should attempt to replicate this study's findings using larger, more diverse samples of fathers. A broader sampling of fathers of children living in violent neighborhoods may reveal that less involved fathers use other strategies or vary in their frequency of strategy use. Efforts should also be made to consider the socioeconomic status of fathers in low-income African-American neighborhoods because there is often great variability in the backgrounds of residents of these urban areas. Moreover, future research should include other predictor variables, such as father's relationship with his child's mother, his work schedule, and the number and ages of children, since it is possible that these variables influence fathers' protective strategies as well as their general parenting styles (Hamer, 1998).

It is clear that our ability to quantitatively measure fathers' history of violence exposure and parenting in violent neighborhoods is in its infancy. This study utilized a new measure--the Parenting in Violent Neighborhoods Scale--to assess the strategies fathers adopt to keep young children safe from community dangers. Although we found this measure to be psychometrically reliable and culturally sensitive for use with African-American fathers (with the exception of the "reduce violent media exposure" subscale, which had marginal reliability), continued research is needed to refine the measure and establish its construct and content validity. Further research is also needed to examine the measure's utility in assessing the protective strategies of African-American mothers and to establish its appropriateness for use with parents from other cultural groups.

Table 1
Demographic Characteristics of the Sample (N = 61)

Demographic Characteristic               M (SD) or     Range
                                         N (%)

Father's Characteristics
Age in years                             36.2 (10.6)   18 to 70 years
Education                                12.7 (2.1)    8 to 17 years
  8-11th grade completed                 10 (16.4%)
  High school diploma or GED             31 (50.8%)
  13+ years of school completed          20 (32.8%)

Marital status
  Single, not living with partner        33 (54.1%)
  Living with partner                    28 (45.9%)

Employment status
  Employed (Yes)                         50 (82.0%)
  Number of hours worked weekly          42.6 (13.5)   10 to 80 hours

Father's family and household characteristics

  Age at birth of first child            23.4 (5.0)    15 to 34 years
  Total number of biological children     3.1 (2.3)    1 to 12 children
  Number of adults living in household    2.0 (0.6)    1 to 4 adults
  Number of children living in            2.0 (1.4)    0 to 6 children
    household

Target Head Start Child's Characteristics
Child's age                               4.0 (0.8)    3 to 6 years
Child's gender
  Male                                   32 (52.5%)
  Female                                 29 (47.5%)

Participant relationship to target child
  Biological father                      41 (67.2%)
  Stepfather                              4 (6.6%)
  Grandfather                             5 (8.2%)
  Uncle                                   7 (11.5%)
  Other                                   4 (6.6%)

Participant father living in same
  household with HS child (Yes)          47 (77.0%)
Years of participant father
  involvement in HS child's life          3.9 (0.9)    0.5 to 6 years


Table 2
Means, Standard Deviations, and Correlation Coefficients of Independent
and Dependent Variables

                        1         2         3         4         5

Psychological well-being:
1.  Depression          --        .28 *     .39 **    .28 *     .30 *

Parenting practices:
2.  Authoritative                 --       -.23       .11       .13
3.  Authoritarian                          --         .46 **    .11
4.  Permissive                                       --         .11

Violence exposure:
5.  Victim of                                                   --
    violence
6.  Witness of
    violence
7.  Personal use of
    violence

Child's gender:
8.  Gender (0 =
    girl, 1 = boy)

Social support:
9.  Familial support
10. Extra-familial
    support
11. Professional
    support

Protective strategies:
12. Monitor and
    teach personal
    safety
13. Teach
    neighborhood
    survival tactics
14. Reduce violent
    media exposure
15. Engage in
    community
    activism
16. Fight back

                        6         7         8         9        10

Psychological well-being:
1.  Depression          .25       .29 *     .01       .03      -.13

Parenting practices:
2.  Authoritative       .31 *     .05      -.16       .24      -.06
3.  Authoritarian      -.09       .11       .19       .18       .11
4.  Permissive         -.07       .02       .06       .31 *     .03

Violence exposure:
5.  Victim of           .62 **    .73 **    .14       .01       .07
    violence
6.  Witness of          --        .54 **    .10      -.01      -.10
    violence
7.  Personal use of               --       -.11      -.02      -.00
    violence

Child's gender:
8.  Gender (0 =                            --        -.02       .02
    girl, 1 = boy)

Social support:
9.  Familial support                                 --         .54 **
10. Extra-familial                                             --
    support
11. Professional
    support

Protective strategies:
12. Monitor and
    teach personal
    safety
13. Teach
    neighborhood
    survival tactics
14. Reduce violent
    media exposure
15. Engage in
    community
    activism
16. Fight back

                        11        12        13        14        15

Psychological well-being:
1.  Depression         -.26 *     .38 **    .31 *     .02       .16

Parenting practices:
2.  Authoritative       .19       .69 **    .50 **    .03       .36 **
3.  Authoritarian       .01      -.15      -.05      -.09       .08
4.  Permissive          .16      -.17      -.30 *    -.15      -.05

Violence exposure:
5.  Victim of           .05       .19       .21      -.09       .15
    violence
6.  Witness of          .08       .29 *     .28 *    -.05       .16
    violence
7.  Personal use of    -.05       .03       .06      -.19       .07
    violence

Child's gender:
8.  Gender (0 =        -.12      -.16       .03       .30 *     .01
    girl, 1 = boy)

Social support:
9.  Familial support    .45 **    .21       .15      -.07       .35 **
10. Extra-familial      .50 **   -.18       .06      -.01       .33 *
    support
11. Professional        --        .01       .02      -.23       .27 *
    support

Protective strategies:
12. Monitor and                   --        .53 **    .13       .35 **
    teach personal
    safety
13. Teach                                   --        .24       .54 **
    neighborhood
    survival tactics
14. Reduce violent                                    --        .09
    media exposure
15. Engage in                                                   --
    community
    activism
16. Fight back

                        16        M         (SD)

Psychological well-being:
1.  Depression          .44 **    8.43      (5.77)

Parenting practices:
2.  Authoritative       .16       3.77      (0.62)
3.  Authoritarian       .33 **    2.29      (0.52)
4.  Permissive          .23       2.19      (0.47)

Violence exposure:
5.  Victim of           .30 *     6.05      (6.57)
    violence
6.  Witness of          .26 *    14.77     (11.44)
    violence
7.  Personal use of     .24       4.90      (6.33)
    violence

Child's gender:
8.  Gender (0 =         .04       0.53      (0.50)
    girl, 1 = boy)

Social support:
9.  Familial support    .35 **    1.86      (0.71)
10. Extra-familial     -.07       1.29      (0.71)
    support
11. Professional       -.03       1.48      (0.64)
    support

Protective strategies:
12. Monitor and         .22       2.80      (0.70)
    teach personal
    safety
13. Teach               .16       2.67      (0.78)
    neighborhood
    survival tactics
14. Reduce violent     -.13       2.46      (0.76)
    media exposure
15. Engage in           .09       1.76      (0.82)
    community
    activism
16. Fight back          --        1.42      (0.74)

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


Table 3
Regression Analyses Examining Predictors of Five Paternal Protective
Strategies

Strategy          Predictor                   Slope            Beta
                  Variables                [+ or -] SE

1. Monitor and
   teach
   personal
   safety

                  Authoritative         0.02 [+ or -] 0.01      .520
                  Paternal              0.03 [+ or -] 0.01      .235
                    depression
                  Permissiveness       -0.02 [+ or -] 0.01     -.220
                  Family support        0.26 [+ or -] 0.11      .265
                  Extra-familial       -0.25 [+ or -] 0.11     -.250
                    support

                  Predictor                    p <           [R.sup.2]
                  Variables                                   Change

                  Authoritative               .001              .48
                  Paternal                    .017              .04
                    depression
                  Permissiveness              .042              .03
                  Family support              .027              .02
                  Extra-familial              .026              .01
                    support

                  F (5, 54) = 15.18, p < .001; [R.sup.2] = .58

Strategy          Predictor                   Slope            Beta
                  Variables                [+ or -] SE

2. Teach
   neighborhood
   survival
   tactics

                  Authoritative         0.02 [+ or -] 0.01      .526
                  Permissiveness       -0.05 [+ or -] 0.01     -.410
                  Authoritarian         0.02 [+ or -] 0.01      .312

                  Predictor                    p <           [R.sup.2]
                  Variables                                   Change

                  Authoritative               .001              .25
                  Permissiveness              .002              .06
                  Authoritarian               .021              .06

                  F (3,56) = 11.11, p < .001; [R.sup.2] = .37

Strategy          Predictor                   Slope            Beta
                  Variables                [+ or -] SE

3. Reduce
   exposure to
   violent
   media

                  Child's gender (a)   0.45 [+ or -] 0.19      .296

                  Predictor                    p <           [R.sup.2]
                  Variables                                   Change

                  Child's gender (a)          .021              .09

                  F (1, 58) = 5.58, p < .05; [R.sup.2] = .09

Strategy          Predictor                   Slope            Beta
                  Variables                [+ or -] SE

4. Engage in
   community
   activism

                  Extra-familial       0.41 [+ or -] 0.03      .378
                    support
                  Authoritative        0.02 [+ or -] 0.01      .352

                  Predictor                    p <           [R.sup.2]
                  Variables                                   Change

                  Extra-familial              .002              .11
                    support
                  Authoritative               .003              .14

                  F (2, 57) = 9.52, p < .001; [R.sup.2] = .25

Strategy          Predictor                   Slope            Beta
                  Variables                [+ or -] SE

5. Fight back

                  Family support        0.52 [+ or -] 0.13      .498
                  Paternal              0.05 [+ or -] 0.01      .384
                    depression
                  Extra-familial       -0.30 [+ or -] 0.14     -.289
                    support

                  Predictor                    p <           [R.sup.2]
                  Variables                                   Change

                  Family support              .000              .12
                  Paternal                    .001              .19
                    depression
                  Extra-familial              .030              .06
                    support

                  F (3, 56) - 10.72, p < 0.001; [R.sup.2] = .37

This research was supported by a Head Start Research Scholars Grant from the United States Department of Health and Human Services/Administration for Children and Families, and the National Council on Family Relations Ruth Jewson Dissertation Award. The authors would like to thank the fathers who participated in this study, James Shird and the Significant Male Task Force, the Head Start teachers and staff who recruited fathers, and our interview team. This study benefited from the helpful comments provided by Suzanne Randolph, Elaine Anderson, Deborah Roberts, Tony Whitehead, and Aria Crump.

NOTE

1. A full copy of the PVNS is available from the first author.

REFERENCES

Advocates for Children and Youth (2000). Maryland kids count factbook. Baltimore, MD: Author.

Allen, W., & Connor, M. (1997). An African American perspective on generative fathering. In A J. Hawkins & D.C. Dollahite (Eds.), Generative fathering: Beyond deficit perspectives (pp. 52-70). Thousand Oaks, CA: Sage.

Anderson, E., Kohler, J., & Letiecq, B. (in press). Predictors of depression among low-income, non-residential fathers. Journal of Family Issues.

Anderson, E., Kohler, J., & Letiecq, B. (2002). Low-income fathers and "responsible fatherhood" programs: A qualitative investigation of participants' fathering experiences and perceptions of program efficacy. Family Relations, 51, 148-155.

Andrews-Cameron, T. (1998). African American, urban preschool children's internal working models of attachment: Associations with adolescent mothers' parenting behavior and psychological characteristics. Dissertation Abstracts International, 58(12-B), 6839.

Annie E. Casey Foundation. (1999). Kids count data book: State profiles of child well-being. Baltimore, MD: Author.

Baumrind, D. (1967). Child care practices anteceding three patterns of preschool behavior. Genetic Psychology Monographs, 75, 43-88.

Billingsley, A. (1968). Black families in white America. Englewood Cliffs, NJ: Prentice-Hall.

Brody, G., McBride Murry, V., Kim, S., & Brown, A. (2002). Longitudinal pathways to competence and psychological adjustment among African American children living in rural single parent households. Child Development, 73, 1505-1516.

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development: Research perspectives. Developmental Psychology, 20, 1074-1081.

Carlson, M., & McLanahan, S. (2002). Fragile families, father involvement, and public policy. In C.S. Tamis-LeMonda & N. Cabrera (Eds.), Handbook of father involvement: Multidisciplinary perspectives (pp. 461-488). Mahwah, N J: Lawrence Earlbaum.

Chase-Lansdale, P.L., & Gordon, R. (1996). Economic hardship and the development of five- and six-year-olds: Neighborhood and regional perspectives. Child Development, 67, 3338-3367.

Coley, R., & Chase-Lansdale, P.L. (1999). Stability and change in paternal involvement among urban African American fathers. Journal of Family Psychology, 13, 1-20.

Comrey, A., & Lee, H. (1992). A first course in factor analysis. Hillsdale, NJ: Lawrence Erlbaum.

Crnic, K., Greenberg, M., Ragozin, A., Robinson, N., & Bashman, R. (1983). Effects of stress and social support on mothers and premature and full-term infants. Child Development, 52, 857-865.

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

Duncan, G., Brooks-Gunn, J., & Klebanov, P. (1994). Economic deprivation and early childhood development. Child Development, 65, 296-318.

Dunst, C., Jenkins, V., & Trivette, C. (1984). The Family Support Scale: Reliability and validity. Journal of Individual, Family, and Community Wellness, 1, 45-52.

Dunst, C., & Trivette, C. (1990). Assessment of social support in early intervention programs. In S. J. Meisels & J.P. Shonkoff (Eds.), Handbook of early childhood intervention (pp. 326-349). Cambridge, England: Cambridge University Press.

Fagan, J., & Stevenson, H. (2002). An experimental study of an empowerment-based intervention for African American Head Start fathers. Family Relations, 51, 191-198.

Federal Bureau of Investigation (n.d). Uniform crime reports: Crime in the United States--1997. Retrieved December 8, 2002, from http://www.fbi.gov/ucr/97cius.htm

Federal Interagency Forum on Child and Family Statistics. (1999). America's children: Key national indicators of well-being, 1999. Federal Interagency Forum on Child and Family Statistics. Washington, DC: U.S. Government Printing Office.

Garbarino, J., Dubrow, N., Kostelny, M., & Pardo, C. (1992). Children in danger: Coping with the consequences of community violence. San Francisco: Jossey Bass.

Garbarino, J., Kostelny, K., & Dubrow, N. (1991). No place to be a child: Growing up in a war zone. Lexington, MA: Lexington Books.

Hamer, J. (1998). What African American noncustodial fathers say inhibits and enhances their involvement with children. The Western Journal of Black Studies, 22, 117-127.

Hamer, J., & Marchioro, K. (2002). Becoming custodial dads: Exploring parenting among low-income and working-class African American fathers. Journal of Marriage and the Family, 64, 116-129.

Hill, H., Hawkins, S., Raposo, M., &Carr. P. (1995). Relationship between multiple exposures to violence and coping strategies among African-American mothers. Violence and Victims, 10(1), 55-71.

Hill, R. (1993). Research on the African-American family: A holistic perspective. Westport, CT: Auburn House.

Hofferth, S., Pleck, J., Stueve, J., Bianchi, S., & Sayer, L. (2002). The demography of fathers: What fathers do. In C.S. Tamis-LeMonda & N. Cabrera (Eds.), Handbook of father involvement: Multidisciplinary perspectives (pp. 63-90). Mahwah, NJ: Lawrence Earlbaum.

Holland, C., Koblinsky, S., & Anderson, E. A. (1995, June). Maternal strategies for protecting Head Start children from community violence: Implications for family-focused violence education programs. Paper presented at the annual meeting of the National Head Start Association, Washington, DC.

Jarrett, R., Jefferson, S., & Roach, A. (2000, June). Family and parenting strategies in high risk African American neighborhoods. Paper presented at the National Head Start Association Annual Meeting, Washington, DC.

Jarrett, R., Roy, K., & Burton, L. (2002). Fathers in the "hood": Insights from the qualitative research on low-income African-American men. In C.S. Tamis-LeMonda & N. Cabrera reds.), Handbook of father involvement: Multidisciplinary perspectives (pp. 211-248). Mahwah, N J: Lawrence Earlbaum.

Langhinrichsen-Rohling, J., & Neidig, P. (1995). Violent backgrounds of economically disadvantaged youth: Risk factors for perpetuating violence? Journal of Family Violence, 10, 379-397.

Letiecq, B., Anderson, E., & Koblinsky, S. (1998). Social support of homeless and housed mothers: A comparison of temporary and permanent housing arrangements. Family Relations, 47, 415-422.

Letiecq, B., & Koblinsky, S. (in press). Parenting in violent neighborhoods: African American fathers share strategies for keeping young children safe. Journal of Family Issues.

Lorion, R., & Saltzman, W. (1993). Children's exposure to community violence: Following a path from concern to research to action. In D. Reiss, J. E. Richters, M. Radke-Yarrow, & D. Scharff (Eds.), Children and Violence (pp. 55-65). New York: Guilford.

Maccoby, E., & Jacklin, C. (1974). The psychology of sex differences. Stanford, CA: Stanford University Press.

McAdoo, J. (1993). The roles of African American fathers: An ecological perspective. Families in Society, 74(1), 28-35.

Mistry, R., Vandewater, E., Huston, A., & McLoyd, V. (2002). Economic well-being and children's social adjustment: The role of family process in an ethnically diverse low-income sample. Child Development, 72, 935-951.

Mohr, W., Fantuzzo, J., & Abdul-Kabir, S. (2001). Safeguarding themselves and their children: Mothers share their strategies. Journal of Family Violence, 16, 75-92.

Nelson, T., Clampet-Lundquist, S., & Edin, K. (2002). Sustaining fragile fatherhood: Father involvement among low-income, noncustodial African American fathers in Philadelphia. In C.S. Tamis-LeMonda & N. Cabrera (Eds.), Handbook of father involvement: Multidisciplinary perspectives (pp. 525-553). Mahwah, NJ: Lawrence Earlbaum.

Nobles, W., & Goddard, L. (1993). Toward an African-centered model of prevention for African American youth at high risk. In L.L. Goddard (Ed.), An African-centered model of prevention for African American youth at high risk (pp. 115-129). Rockville, MD: USDHHS/PHS/SAMHSA.

Ogbu, J. (1981). Origins of human competence: A cultural-ecological perspective. Child Development, 52, 413-429.

Pedhazur, E. (1982). Multiple regression in behavioral research: Explanation and prediction. Fort Worth, TX: Harcourt Brace College Publishers.

Perloff, L. (1983). Perceptions of vulnerability to victimization. Journal of Social Issues, 39(2), 41-61.

Perry, D., Perry, L., & Weiss, R. (1989). Sex differences in the consequences that children anticipate for aggression. Developmental Psychology, 25, 312-319.

Pinderhughes, E., Nix, R., Foster, E., & Jones, D. (2001). Parenting in context: Impact of neighborhood poverty, residential stability, public services, social networks, and danger on parental behaviors. Journal of Marriage and the Family, 63, 941-953.

Radloff, L. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1,385-401.

Randolph, S., Damond, M., & Washington, D. (1995). African American youth and HIV prevention. In Quakenbush, M., Clark, K., & Nelson, M. (Eds.), The HIV challenge: Prevention education for young people (2nd ed., pp.347-366). Santa Cruz, CA: ETR Associates.

Randolph, S.M., Koblinsky, S.A., & Roberts, D.D. (1996). Studying the role of family and school in the development of African-American preschoolers in violent neighborhoods. Journal of Negro Education, 65(3), 282-295.

Randolph, S.M., Koblinsky, S.A., & Roberts, D.D. (1998). Parenting practices of African American mothers of preschoolers in high and low violent neighborhoods. Paper presented at the Annual Meeting of the American Orthopsychiatric Association, Arlington, VA.

Robinson, C., Mandleco, B., Olsen, S., & Hart, C. (1995). Authoritative, authoritarian, and permissive parenting practices: Development of a new measure. Psychological Reports, 77, 819-830.

Sampson, R., Raudenbush, S., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918-924.

Sarason, I., Sarason, B., & Shearin, E. (1986). Social support as an individual difference variable: Its stability, origins, and relational aspects. Journal of Personality and Social Psychology, 50, 845-855.

Smetana, J. (1988). Concepts of self and social convention: Adolescents' and parents' reasoning about hypothetical and actual family conflicts. In M. Gunnar & W.A. Collins (Eds.), Minnesota symposia on child psychology (pp.79-122). Hillsdale, NJ: Erlbaum.

Smith, S., & Donnerstein, E. (1998). Harmful effects of exposure to media violence: Learning of aggression, emotional desensitization, and fear. In R.G. Geen & E. Donnerstein (Eds.), Human aggression: Theories, research, and implications for social policy (pp. 167-202). San Diego, CA: Academic Press.

Taylor, L., Zuckerman, B., Harik, V., & Groves, B. (1994). Witnessing violence by young children and their mothers. Developmental and Behavioral Pediatrics, 15(2), 120-123.

Taylor, R., Chatters, L., Tucker, M., & Lewis, E. (1990). Developments in research on Black families: A decade review. Journal of Marriage and the Family, 52, 993-1014.

U.S. Census Bureau. (2000). Statistical abstract of the United States: 2000 (120th ed.). Washington, DC: U.S. Government Printing Office.

U.S. Department of Health and Human Services. (2000). Healthy people 2010: Understanding and improving health (2nd ed.). Washington, DC: U.S. Government Printing Office.

Ventura, S., & Bachrach, C. (2000). Nonmarital childbearing in the United States, 1940-99. National Vital Statistics Report, 48(16). Hyattsville, MD: National Center for Health Statistics.

Villani, S. (2001). Impact of media on children and adolescents: A 10-year review of research. Journal of the American Academy of Child and Adolescent Psychiatry, 40, 392-400.

Whitehead, T. (1997). Urban low-income African American men, HIV/AIDS, and gender identity. Medical Anthropology Quarterly, 11, 411-447.

Widom, C. (1989). The cycle of violence. Science, 244, 160-166.

Wood, J. (1994). Gendered lives: Communication, gender, and culture. Belmont, CA: International Thomson Publishing.

Correspondence concerning this article should be addressed to Bethany Letiecq, Department of Health and Human Development, 318 Herrick Hall, Montana State University, Bozeman, MT 59717. Electronic mail: [e-mail]

BETHANY L. LETIECQ Montana State University, Bozeman

SALLY A. KOBLINSKY University of Maryland, College Park