Abstract
The literature on perceptions of police is growing, yet the enthusiasm is outpacing methodological rigor. This study (a) examined the factor structure of items assessing procedural justice and legitimacy, (b) tested whether the factors were uniquely associated with youth self-reported offending (SRO), and (c) identified whether effects on subsequent SRO operated through legitimacy. Using data derived from the 1,216 youth in the Crossroads Study, as well as supplemental models with Pathways to Desistance data, factor analyses established a factor structure, negative binomial regressions examined associations with SRO, and indirect effects analysis within a structural equation model framework identified whether associations on SRO operated through legitimacy. A five-factor solution emerged: Voice, Neutrality/Impartiality, Distributive Justice/Bias, Respect, and Legitimacy. In the adjusted model, only Distributive Justice/Bias and Legitimacy were directly associated with concurrent SRO. However, all procedural justice scales had indirect effects on subsequent offending through legitimacy. Implications for methodology and procedural justice theory are discussed.