Street Youths and the Proximate and Contingent Causes of Instrumental Crime: Untangling Anomie Theory

Abstract

Utilizing a sample of 300 homeless street youths, the research examines the individual‐level sub‐model of Baumer’s interpretation of Merton’s anomie theory. The paper explores the role the interaction between monetary goals and weak commitment to legitimate means plays in the generation of instrumental crime and the manner in which this interaction is itself moderated by blocked opportunities, monetary dissatisfaction, social modeling, cultural support, and the perceived risk of punishment. The findings reveal that a weak commitment to legitimate means, but not monetary goals, has a lower order impact on the willingness to commit instrumental crime. These two variables, however, do not interact to predict intentions to offend. Instead the findings reveal that blocked opportunities and higher levels of monetary dissatisfaction moderate the relationship between the monetary goals and weak commitment to legitimate means interaction and the willingness to offend. Findings are discussed and suggestions for further research are offered.

Keywords:

• anomie theory
• street youth
• property crime

Acknowledgments

The author would like to acknowledge the financial support of the Queen’s University Chancellor’s Research Award and the Social Sciences and Humanities Research Council of Canada. Special thanks also to Owen Gallupe for his research assistance and to Richard Tewksbury, Fiona Kay, and the anonymous reviewers for their comments.

Notes

1. Whitbeck et al. (2001) note that variations in the gender make‐up across samples might be attributable to the degree of effort made to recruit youth from across a range of locations including the street as opposed to relying strictly on agencies for contacts. The former strategy will provide a larger percentage of male participants, while the latter will produce a greater proportion of female respondents.

2. To examine if there were slope differences for the three‐way interaction effects I compared the simple slopes for high and low values (e.g., +1 SD and −1 SD) of the higher order moderators shown in Tables 4 and 5 (Dawson & Richter, 2006). These tests showed significant slope differences for the top panel of Table 4 (p < .10), the bottom panel of Table 4 (p < .05), and the bottom panel of Table 5 (p < .05).