Violent Behavior of Juveniles in a Multiethnic Society: Effects of Personal Characteristics, Urban Areas, and Immigrants' Peer Networks

Abstract

Using data from a large school survey based on 14,301 cases, we examined several risk factors, such as low self-control, culture of honor, and social bonds, for juvenile violence among immigrant and native-born Germans. Our empirical analyses indicate that violent peer networks and low self-control are more relevant for the explanation of juvenile violence than culture of honor. Moreover, these risk factors are also a consequence of the spatial concentration of Turkish and other immigrants living in large German cities. In these cities, culture of honor as well as low self-control (high temper) are more prevalent. Hence, residence in large cities could contribute to an explanation of why some immigrant groups exhibit higher risk than natives. Therefore, further research should focus more on social environments like networks and the residential segregation of juveniles from immigrant background living in larger cities.

KEYTERMS:

• Violent behavior
• peer networks
• ethnic origin
• immigrants
• zero-inflated count regression

Notes

1. Classes have been sampled in different sampling points: in urban areas of Dortmund, Kassel, Munich, Oldenburg, and Stuttgart; in smaller cities of Schwaebisch-Gmuend and Lehrte; and in the counties of Peine and Soltau-Fallingbostel. At the level of the sample point it is not a random sample, but it covers a broad range of different areas and different socioeconomic contexts (see Table 1). In eastern Germany, a random sample of classes has been drawn from the federal district of Thuringia. Due to the fact that the proportion of immigrants is very low in East Germany and due to the different sampling procedure, students from the district of Thueringen have been excluded from the analysis.

2. The categories shown in Tables 1 and 2 are derived from an adapted form of the conflict tactic scales (see Appendix 1 and also Wetzels et al., 2001, p. 230). Only parental violence experienced during childhood can be modeled as a causal predictor of violence in adolescence, because parental violence experienced during early adolescence could also be a consequence of delinquent behavior of juveniles.