Analysis of National Youth Survey Data
Finally, given the paucity of individual level studies, we made our own brief comparison using National Youth Survey (NYS) data from wave 7 (See Table 9.2). Limited by the fact that we look at only one year of the subjects’ lives, we compared individuals who self-reported violent acts, including domestic violence acts, and compared them to those who reported theft acts but no violent acts. Simple correlations reveal that committing any violent act was negatively associated with two measures of SES (but not to income). Stealing was not associated with any measure of SES. Those who had committed violence had significantly lower SES than those who reported thefts but no violence; all three indicators (estimated income, Hollingshead index [where a higher value indicates lower SES], and the Duncan index) suggest that violent subjects have greater socioeconomic disadvantage than nonviolent-only offending subjects. When we include a group of individuals who committed both theft and violence, we find that those in the violence-only group have the lowest SES of the four groups (nonoffender, theft only, violent only, theft plus violence). In fact, in this data set, stealing is positively associated with SES—individuals who only steal have higher SES than those who report no crime. This analysis is very preliminary and is provided as just one piece of evidence to make our case, yet the results lend credence to our hypothesis.