Testing the need for disaggregation of the dependent variable in criminology

Many criminological theories about individual violation can be broadly classified into two groups: (1) those that attempt to explain violation of personally held conventional norms (normative violations) and (2) those that attempt to explain the following of deviant norms (cultural deviance). I argue that each set of theories focuses on a unique dependent variable, and that social learning theory is one example of a theory that explicitly integrates these dependent variables under one explanatory heading. When testing theories, students of crime/deviance have largely ignored this distinction, often relying on undifferentiated dependent variables designated as simply “crime.” This research attempts to ascertain if differentiating dependent variables with respect to normative belief improves the explanatory power of certain theoretically based independent variables (derived from social control and subcultural theories); in addition, it attempts to identify some independent variables that differentially affect normative violations and cultural deviance (perceived certainty and strain); it also examines the implication that comparisons of theories with different dependent variables to one another when using an undifferentiated dependent variable of “crime” are inaccurate. Hypotheses are tested using data from three different surveys. Results were mixed. On one hand, overall the explained variance of the theoretical models was increased when the moral belief of actors was taken into account. However, when the separate theories were tested using variables specific to their focus either weak or contrary results were found. Despite these conditional findings, the fact that some were favorable to the argument set forth should encourage more detailed examination of the possibility that crime/deviance students need to more carefully specify their dependent variable according to the focus of specific theories.