Abstract
BACKGROUND: This paper takes on 'Farrington's challenge' 'to bridge the gap between risk factor research and more complex explanatory theories' by offering an explanation for the unexplained statistical phenomenon of cumulative risk. We argue that cumulative risk primarily reflects the social contexts which crime relevant causal processes operate in and draw upon for their content and efficacy. AIMS: This paper tests if the immediate causes of crime according to Situational Action Theory (crime propensity and criminogenic exposure) can account for the relationship between cumulative risk (reflecting key features of family, neighbourhood, school and peer contexts) and crime involvement. METHODS: The paper uses data from the Peterborough Adolescent and Young Adult Development Study (PADS+) reflecting the social lives and criminal behaviour of a randomly sampled UK age cohort from ages 12 to 24 (2003-2016). Data used are drawn from parent and participant questionnaires, space-time budgets, community surveys, the UK Census and land use databases. RESULTS: Cumulative risk statistically accounts for 7% and 8% of the variance in crime prevalence and frequency, respectively, whereas crime propensity and criminogenic exposure account for 52% and 58%, respectively. Moreover, and importantly, measures of crime propensity and criminogenic exposure fully account for (statistically mediate) the association between cumulative risk and both crime prevalence and crime frequency. CONCLUSIONS: Cumulative risk does not represent cumulative causation. The phenomenon of cumulative risk is best understood as representing the social context. Future research should focus on identifying features of social contexts that provide relevant content to and impact the efficacy of key action (and developmental) processes in crime causation.