There are two purposes to the present study. Our methodological purpose is to develop and test a procedure and instrument for assessing crime- and fear-related features of the urban residential environment. We examine three classes of cues: symbols of social and physical disorder, territorial functioning, and architectural 'defensible space' features. Past research examining the physical environment correlates of fear of crime has relied almost exclusively on subjective perceptions of the environment rather than on independent and objective measures thereof. Our theoretical purpose is to test the 'disorder' thesis of Skogan, and Wilson and Kelling, that actual physical incivilities erode resident's confidence in their neighborhood and lead them to infer that serious local problems, unrelated to the physical environment, are serious. We conducted environmental assessments and resident interviews (n = 412) on 50 blocks in 50 Baltimore neighborhoods. The assessments demonstrated high levels of inter-rater reliability and concurrent validity, controlling for social class. Regression analyses showed that physical incivilities were independently linked to perceptions of social and crime-related problems. The results show that reliable and valid assessment of crime- and fear-related environmental features can be conducted. They also support the central kernel of the Wilson and Kelling, and Skogan thesis, that the actual presence of disorder-related cues engender perceptions of social and crime problems.
