The validity of incivility measures in public housing*

Abstract

Both criminologists and policy makers have argued that abandoned cars, buildings, and apartment units, public drunkenness, graffiti, and other assorted unpleasantries communicate to many persons, the absence of both formal and informal social control. Wilson and Kelling hypothesized that these incivilities, or signs of disorder, lead to a higher rate of crime, victimization, and residents' perceptionof the fear of crime. In his recent review of the incivilities thesis, Taylor identified four types of measurement strategies for incivilities including residents' perceptions and on-site assessments. Using data from two public housing developments in inner-city Denver, this paper measures incivilities using both of these strategies. Exploratory and confirmatory factor analysis reveal that these two strategies appear to index different factors (between-measurement heterogeneity), and that variation also exists within on-site assessments (within-measurement heterogeneity). Thus the analysis fails to find evidence of multimethod convergent validity. Implications for future research are addressed.