ROLE TAKING AND RECIDIVISM: A TEST OF DIFFERENTIAL SOCIAL CONTROL THEORY

Abstract

In 1994, Karen Heimer and Ross Matsueda collaborated and developed Differential Social Control (DSC) theory. Heimer and Matsueda posit that the proximate cause of crime is role-taking which encompasses five major processes: 1) reflected appraisals of self as a rule violator, 2) anti-social attitudes, 3) anticipated disapproval of deviant acts from family and friends, 4) criminal associations, and 5) prior experience with crime and delinquency. Taking these processes together, DSC argues that the likelihood of crime and delinquency increases when an individual believes that others view him as a rule violator, holds anti-social attitudes, anticipates limited disapproval of deviance from family and friends, associates with deviant peers, and has repeatedly solved prior problematic situations using criminal or delinquent behaviors. DSC also posits that more distal factors such as role commitment and structural locations affect crime and delinquency indirectly via role-taking. Unlike other theoretical perspectives also formulated in the 1990s, DSC has received scant theoretical discussion and empirical attention. To date, DSC has only been evaluated in a handful of empirical tests. The primary aim of this dissertation is to expand the body of empirical research assessing DSC. In particular, this dissertation examines DSC's ability to explain recidivism among a sample of adult offenders released from Maryland prisons.

Overall, the results generated from this dissertation do not lend support for DSC's ability to account for recidivism. Specifically, the results revealed that only two of the five measures of role-taking, anti-social attitudes and number of prior arrests, were consistent significant predictors of recidivism. The results also indicated that measures of role-commitment were not generally related to recidivism and as a consequence, the hypothesized mediating effects of role-taking on the relationship between role commitment and recidivism by DSC were not supported. The results also showed that with the exception of age, social location measures generally were not related to recidivism and thus, definite statements on the mediating effects of DSC's central concepts on this relationship could not be drawn.