Deviant Modeling, Peer Gender, and Theft: An Experimental Test

• In Event: Advances in Research on Juvenile Delinquency and Crime

Abstract

We present an experiment in which we isolated the effect of specific aspects of deviant peer modeling on theft. Participants were randomized into one of 12 groups (4x3 factorial design) and given a chance to steal a gift card (current n=157). There were four conditions in which the two confederates acted in distinct ways: 1) Control; 2) Verbal-only (confederates talked about stealing during an opportunity to do so); 3) Behavioral-only (confederates stole without talking); 4) Verbal-plus-behavioral (confederates both talked about and stole a card). The groups were further divided into gender-similar (both confederates were the same gender as the participant), gender-different (both confederates were a different gender than the participant), and mixed groups (both male and female confederates present). Preliminary results demonstrate that a larger percentage of participants in the behavioral (16%) and verbal-plus-behavioral groups (9%) committed theft than in the control (6%) or verbal-only (2%) groups. Gender composition also appeared to play a role. Theft was relatively rare when participant gender was the same as both confederates (7%) or different from both (5%); it was more common when both male and female confederates were present (13%). Results will be discussed in terms of theoretical implications.

Authors

• Owen Gallupe, University of Waterloo

• Patrick C. Lalonde, University of Waterloo

• Holly Nguyen, Pennsylvania State University

• Jennifer L. Schulenberg, University of Waterloo