SOCIAL CLASS AND VIOLENT BEHAVIOR (DELINQUENCY).

Abstract

The purpose of this study is to specify more precisely the relationship between social class and crime or delinquency by focusing on a particular offense, violent behavior. Police records and survey data from four studies are analyzed. These data sets are derived from samples from around the United States and span nearly a twenty-year period. The empirical analysis begins by examining the relationship between parental status and violent behavior. One of the principal conclusions drawn from this analysis is that the magnitude of the relationship between social class and violent behavior is contingent upon the way class or status is defined or measured. Depending upon the way class is operationalized, its relationship with violent behavior is nonexistent, moderate, or relatively strong. To account for this social distribution of violent behavior, four major theories of crime and delinquency are tested. Structural and cultural theories in general fail to provide an adequate explanation. For example, consistent with research on general delinquent behavior, there is almost no support for the propositions of strain theory. In contrast, processual and psychological theories of crime and delinquency receive considerable support. Social learning theory measures of imitation are among the strongest correlates of violent behavior. In one data set, imitation measures completely account for the class distribution of violence. Social control theory measures, such as attachment to others, are also strongly correlated with violent behavior. A series of crucial experiments comparing control theory propositions with predictions of other theories yield results more consistent with control theory. The relationship between adolescent status and violent behavior is also examined. Parallel to the analysis of parental status, the results vary depending on the measure of adolescent status used. These results suggest that there are distinct dimensions of adolescent status which must be assessed separately.