The current longitudinal study examined the protective effects of parenting processes on measures of adolescent adjustment (health-compromising and violent behaviors) in a sample of poor African American youth (N ϭ 2,867
A R T I C L E
INTRODUCTIONDespite recent decreasing national crime trends, youth violence remains a serious public health concern in the United States (Hamburg, 1998;Snyder & Sickmund, 1999). Rates of violence perpetration and victimization by and of youth remain near-catastrophic (Brener, Simon, Krug, & Lowry, 1999), especially in comparison to other Western nations (Krug, Mercy, Dahlberg, & Powell, 1998;. Elliott, Huizinga, and Ageton (1985) identified that among adolescents, the ones most at-risk for violence perpetration and victimization are poor, urban male youth. In fact, growing up in a community where poverty, frustration, and hopelessness are endemic is highly predictive of later violence perpetration (Osofsky, 1995), both for male and female adolescents.Studying and further understanding violence and its etiology is paramount if we seek to address the problem and potentially offer solutions to it (Tolan, 2001;Tolan & Guerra, 1994). Because studies have established that violence perpetration frequently co-occurs with health-compromising behaviors (Basen-Engquist, Edmundson, & Parcel, 1995;Farrell, Kung, White, & Valois, 2000;Harris & Ryan, 2002;Orpinas, Basen-Engquist, Grunbaum, & Parcel, 1995), we need to also examine the etiology of such behaviors that may in fact follow the same etiological path as violence itself. Indeed, a number of investigations have provided consistent evidence that the same risk factors responsible for youth violence are also responsible for health-compromising behaviors (e.g., Brook, Balka, & Whiteman, 1999;Fitzpatrick, 1997;Gottfredson & Hirschi, 1990;Hawkins, Catalano, & Miller, 1992;Hindelang, Hirschi, & Weis, 1981;Jessor, 1992;Loeber, Farrington, Stouthamer-Loeber, & Van Kammen, 1998b;Rutter, Giller, & Hagell, 1998).In a review of the literature, Tolan (2001) concluded that an epidemiological approach (i.e., large-scale, population-based samples) is necessary to further study and understand the etiology of violence and analogous behaviors. Furthermore, he suggested that given variability in risk factors depending on developmental timing (early, middle, or late adolescence) and depending on the sex of the adolescent (Reese, Vera, Thompson, & Reyes, 2001), we need to study the etiology for early, middle, and late adolescents, as well as for both male and female youth. Therefore, the current investigation examined the etiology of adolescent health-compromising and violent behaviors employing a multicohort, longitudinal sample of inner-city youth from high-crime neighborhoods in the Mobile, Alabama metropolitan area. Specifically, the study investigated the protective function of parenting processes in this high-risk developmental milieu for both health-compromising and violent behaviors over the course of 3 years-in male and female as well as early, midd...