«Abstract This study examined the stability of belonging to a gang in early adolescence, the behaviour proﬁles, family characteristics, and ...»
The Road to Gang Membership:
Characteristics of Male Gang and Nongang
Members from Ages 10 to 14
Wendy M. Craig, Queen’s University, Ontario, Frank Vitaro,
Claude Gagnon and Richard E. Tremblay,
University of Montréal, Canada
This study examined the stability of belonging to a gang in early adolescence, the
behaviour proﬁles, family characteristics, and friendships of nongang and gang
members. The subjects in the present study were originally part of a larger sample of boys. One hundred and forty-two boys who had a complete data set at ages 11, 12, 13, and 14 were selected for the present study. Loglinear analyses indicated that gang membership was stable from ages 13 to 14, but not at earlier ages. Boys were divided into three groups: stable gang members (children who belonged to a gang at ages 13 and 14); unstable gang members (children who belonged to a gang at either age 13 or 14) and nongang members. Repeated analyses of variance indicated that stable gang members had signiﬁcantly higher scores than nongang members on teacher ratings of ﬁghting behaviour, hyperactivity, inattention and oppositional behaviour, and self-reported delinquent activities (drug and alcohol use, stealing and vandalism). Peers rated gang members as more aggressive than nongang members. The results are discussed from a developmental perspective.
Keywords: gangs, delinquency, development, behaviour problems Many researchers studying gangs have focused on adolescence when gang members’ involvement in crime is particularly salient and perhaps most easily studied (e.g., Spergel, 1990). Consequently, the majority of research has neglected to provide a developmental perspective and understanding of the factors that begin before adolescence which may lead to gang membership. The purpose of this paper was to examine the frequency and stability of gang membership, family characteristics, concurrent behaviours, friendships, and school attitudes of stable and unstable gang members and nongang members in childhood and early adolescence.
Understanding the mechanism of or the catalyst for the initiation, continuation, or discontinuation of gang membership is an important issue, because youth gang This research was supported by grants from Conseil Québecois de la recherche sociale, Fonds de la recherche en sante du Québec, and the Social Science and Humanities Research Council. We would also like to thank Lise Desmarais-Gervais, Hélène Beaumont, Hélène Beauchesne, and Maria Rosa for their participation. Reprint or information requests should be sent to the ﬁrst author, Psychology Department, Queen’s University, Kingston, Ontario, Canada, K7L 3N6, phone (613) 533-6014.
© Blackwell Publishers Ltd 2002. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA.
54 Wendy M. Craig, Frank Vitaro, Claude Gagnon and Richard E. Tremblay members engage in more violent crimes and have more police contact than nongang members (Klien & Maxson, 1989). Furthermore, research suggests that gang members with arrest records are responsible for a disproportionate amount of crime (Bobrowski, 1988). Finally, gang criminal activity extends into adulthood (Spergel, 1989, 1987;
Robin, 1967; Hagedorn, 1988). The cost to individuals (i.e., involvement in delinquent activities) and to society (i.e., serious and violent offenses, homicides, sale and distribution of drugs) necessitates further empirical investigation of the individual characteristics and processes associated with gang membership (Klien & Maxson, 1989).
Thornberry, Krohn, Lizotte, and Chard-Wierschem (1993) have tested three models to understand the relationship between gang membership and delinquency: a selection model, a social facilitation model, and an enhancement model. A selection model is consistent with social control theory (Hirshi, 1990) and the propensity theory of crime (Gottfredson & Hirschi, 1990). According to this model, gangs recruit individuals who are already delinquent or have a propensity toward delinquency. Consequently, belonging to a gang would not impact signiﬁcantly on delinquent activities since the propensity for that behaviour would already exist before belonging to a gang. In a social facilitation model individuals become delinquent when they are active members of a gang. Thus, gang membership results in increased antisocial behaviour due to the normative nature of rule breaking behaviour in gangs and the group processes of the gang.
The enhancement model combines both the selection and social facilitation models.
According to this model, gang members recruit individuals who have a history of aggressive and antisocial behaviour problems and these problem behaviours become augmented since gangs will continue to support the values of, and provide structure for engaging in deviant acts. This pattern of afﬁliation may also result in the escalation of delinquent acts.
The results of the Thornberry et al. (1993) study supported the social facilitation model. They found that gang members entering the gangs did not have higher rates of delinquency or drug use before entering the gang than nongang members. Once these individuals became gang members, their rates of delinquency and drug use signiﬁcantly increased. But, once these gang members left the gang there was a signiﬁcant reduction in these activities. While the results have important theoretical signiﬁcance, they do not explain the process or causal mechanisms leading to gang membership.
There is also no attempt made to identify family, peer, and school factors which may distinguish stable gang members, unstable gang members, and nongang members.
Finally, the results are based on self-report information. The present research, using a multi-informant approach, examines family, peer, and school factors which may differentiate stable gang members from unstable and nongang members.
Although most research concurs that gang members have dysfunctional families, the family characteristics of gang and nongang members have not been well documented (Klien & Maxson, 1989). Based on adolescent self-reports, Hoyd (1985) found that alienation from parents was the initial cause of adolescents turning toward delinquent and gang activity. Other studies have related gang membership to harsh discipline (Friedman, Mann, & Friedman, 1975), a lack of a male role model in the home, unstable economic situations, and rejection by the family (Vigil, 1988). These family characteristics may result in children feeling alienated and isolated and consequently they may seek a peer group, such as a gang, that will be supportive of their behaviours. This process may be more intensiﬁed during preadolescence when Social Development, 11, 1, 2002 © Blackwell Publishers Ltd 2002 Adolescent Gang Members 55 individuals are increasing in their autonomy and independence from the family environment and spending more time with peers. The present study investigates family characteristics such as the amount of communication, punishment, rules, parental monitoring, and other socio-family characteristics of gang and nongang members.
Hartup (1983) found that gang membership was predicted by not getting along with others. Hirschi (1969) suggested that gang members lack the ability to form effective, fulﬁlling relationships with others. Research on the friends of gang members indicates that gangs do not typically encourage active members to have friends, especially close ones outside the gang (Horowitz, 1983; Miller, 1977; Spergel, 1989). Gang members may be socially isolated from their age appropriate and/or a socially competent peer group. Gang members may tend to associate with individuals who will accept them and are like themselves, in terms of behaviours, values, and goals, as in the case of aggressive children (Hymel, Wagner, & Butler, 1990; Cairns & Cairns, 1991). As a result, their socialization experiences are imbalanced in the direction of negative interactions and they are at continued risk for learning aggressive behaviours. This process may be particularly heightened during early adolescence when individuals spend more time with their friends that is unsupervised by an adult (Cairns & Cairns, 1997). Children with aggressive behaviour tendencies during childhood, may be more likely than prosocial children to seek out a social context, such as a gang, where their antisocial behaviour is supported. Thus, we hypothesize that the stable gang members will have more behavioural problems than the transient or nongang members. In summary, the research on both family and friendships of gang members suggests that they experience some difﬁculties in their social relationships.
Participant-observer studies indicate that gang members are typically behind in their studies, or are school dropouts (Klien & Maxson, 1989). Horowitz (1983) suggested that the high dropout rate among gang members is due to the fact that school is a context in which individual inadequacies may become apparent. In this sense, gang members may be alienated from school. In addition, gang members do not believe that school is helpful to them (Klien & Maxson, 1989). One of the objectives of the present study is to examine early school behaviour. It is possible that children who are experiencing problems in school may engage in disruptive behaviors and be at risk of becoming gang members.
A variety of theoretical and methodological problems have hindered the development of adequate knowledge about gangs. For example, researchers have tended to employ small nonrandom samples, usually without comparison groups (Spergel, 1990), or have implemented participant observer methodology. Another methodological issue in gang research concerns the over-reliance on self-report data. The paucity of reliable, empirical research on preadolescent and early adolescent gang members may be a function of these methodological difﬁculties. The present research contributes to our understanding of individuals who belong to a gang by: the consideration of self, peer, parent and teacher reports; including a comparison group (nongang members); providing a developmental perspective; and examining behaviours in preadolescence and early adolescence, as well as friendships. Given that gang behaviour is primarily a male phenomenon (Bowker, Gross, & Klien, 1980), the present study focuses on preadolescent male gang members.
The speciﬁc objectives of the present research are: to describe the frequency of gang membership during the preadolescent and early adolescent periods (at ages 11, 12, 13, and 14); to examine the stability of belonging to a gang at ages 11, 12, 13, and 14; to Social Development, 11, 1, 2002 © Blackwell Publishers Ltd 2002 56 Wendy M. Craig, Frank Vitaro, Claude Gagnon and Richard E. Tremblay compare the family, behavioral, peer and school proﬁles of boys who belong to a gang and boys who do not belong to a gang, at ages 11, 12, 13, and 14. It is expected that boys who belong to a gang will have less supervision from their parents, poorer communication in the home, and fewer rules than those who do not belong to a gang. It is predicted that teachers and parents will report more externalizing behaviour problems for gang members than nongang members. Gang members will report more delinquency and alcohol use than nongang members. Another purpose of this study was to examine the peer relations of boys who belong to a gang and those who do not belong to a gang. It is expected that gang members will be more aggressive and less popular than nongang members. In addition, friends of gang members will be engaged in delinquent activities. Finally, there will be a higher frequency of gang members than nongang members who are behind in school.
Method Participants The subjects were originally part of a larger sample of 1,034 boys studied in kindergarten in 1984 in a longitudinal study. This study took place in a low socioeconomic area of a large metropolitan city in the province of Quebec. To control for culture, boys were only included in the sample if both parents were born in Canada and the parents’ mother tongue was French. From this original sample, depending on the year and the informant source, there was gang membership information on between 640 and 1013 boys. These subjects were used in the loglinear analyses examining the stability of gang membership.
From that sample, a subsample of 142 boys was selected for the present study. These boys had complete data sets from parents, self, teachers, and peer ratings from ages 11 through 14.
Based on self-report information, boys were divided into three groups:
stable gang members (children who belonged to a gang at ages 13 and 14); unstable gang members (children who belonged to a gang at either age 13 or 14) and nongang members. Due to the disproportionate number of nongang subjects a random sample of 66 boys was selected for the group comparison tests. There were 25, 51, and 66 boys in the stable, unstable, and comparison groups, respectively. On kindergarten teacher ratings, there was no difference between those who were selected for the study and those who were not.
Instruments (i) Social Behaviour Questionnaire (SBQ) (Tremblay, Vitaro, Gagnon, Piché, & Royer, 1991). The SBQ comprise 28 items from the Preschool Behavior Questionnaire (Behar & Springﬁeld, 1974) which is an adaptation of the Children’s Behaviour Questionnaire (Rutter, 1967) and 10 items from the Prosocial Behavior Questionnaire (Weir, Stevenson, & Graham, 1980). This questionnaire measures appropriate and inappropriate behaviours of children. It is designed to identify maladjusted children.
This scale examined the following: ﬁghting (intimidates, ﬁghts, hits), oppositional behaviour (does not share, irritable, blames others, inconsiderate), inattention (unable to concentrate, easily distracted, is in the moon, gives up easily), hyperactivity (ﬁdgets, restless), anxiety-withdrawal (sad, cries easily, worries, is alone, is afraid), and prosocial behaviour (comforts others, helps others who are sick, tries to help others who Social Development, 11, 1, 2002 © Blackwell Publishers Ltd 2002 Adolescent Gang Members 57 are hurt, sympathetic to others, offers help to others who are in trouble, invites others to play, and helps clean up). The alpha reliability coefﬁcients for each of the subscales ranged from.74 to.91, indicating high internal consistency. All items are scored on a 3-point scale: does not apply, sometimes, and frequently.