Socially desirable response bias in criminology: an example of its effect in testing the effects of self-control

Over the past two decades, there was a proliferation of research in criminology on the role of self-control in explaining criminal conduct. A meta-analysis of the criminological literature indicates that self-control is a significant predictor of both criminal and analogous behaviors (Pratt and Cullen, 2000). Much of this literature relies on cognitive measures of self-control, such as Grasmick et al.’s scale (1993), that require respondents to engage in a self-appraisal of the types of people they are, an activity that would be subject to a range of intentional or unintentional response biases. Hirschi and Gottfredson (1993) caution criminologists against using cognitive measures of self-control. Their line of reasoning is supported by research indicating that self-report criminal behavior is underreported among more serious delinquents (Hindelang et al., 1981). By extension, then, those with low self-control are either unwilling or unable to participate in surveys, thus attenuating the effects self-control on crime. Research has found that one’s self-control influences survey responses, suggesting that external behavioral measures recorded from neutral observers may be more accurate for measuring self-control and predicting its influence on criminal behavior (Piquero et al., 2000).