Crime propensity, criminogenic exposure and violent scenario responses: Testing situational action theory in regression and Rasch models

Abstract

In this paper, we argue that quantitative empirical research to explain and predict criminal and related behaviour can benefit greatly from explicit theories of action linking individual and contextual factors in the causation of crime. Such theories foster a systematic selection of causal variables for data collection and hypothesis testing instead of a more indiscriminate accumulation of ‘risk factor’ correlates. Moreover, action theory encourages statistical modelling of crime causation beyond the most common linear regression. This paper illustrates both points by estimating two empirical models – a conventional logistic model and a Rasch model – on scenario response data concerning youth violence. The findings of this study show that the extent to which young people indicate a violent response to a provocation is dependent on their (law relevant) morality and ability to exercise self-control as well as the deterrent qualities (monitoring) of the setting.