Methodological Issues in Survey Research on the Inhibition of Crime

In a recent article, Grasmick and Green analyze the relationships they observe in their sample between scales for L = threat of legal punishment, S = threat of social disapproval, M = moral commitment to legal norms, Ip = self-reported past involvement in illegal behavior, and If = estimated future involvement in illegal behavior.The authors interpret their data on the basis of an assumed underlying causal model in which L, S, and M each influence I, and If.Their findings are consistent with the proposition that each of the independent variables inhibits participation in crime.Before these findings are accepted, other possible interpretations of the data must be ruled out.Grasmick and Green make no attempt to do this.Excluding correlated measurement errors, there are two rival interpretations to consider.The first is that L, S, and M are not causes of involvement in illegal behavior, but rather, are consequences of such involvement.The second is that the relationships are spurious, with nonvanishing correlations being produced by unmeasured exogenous causes of the observed variables.Each of these possibilities will be considered in turn.