Article Text

Download PDFPDF
48 Assessing safety in urban environments: a pilot randomised controlled study
  1. Bernadette Hohl1,
  2. Charles Branas2,
  3. Douglas Wiebe2
  1. 1Rutgers University, USA
  2. 2University of Pennsylvania, USA


Statement of purpose Crime is distributed in patterns of time and space, it is not random. Moreover, an individuals’ risk of victimisation varies with individual characteristics. Perceptions of safety are often misaligned with risk yet these perceptions guide our everyday behaviour. The purpose of this study was to determine whether it is feasible to explore individuals’ perception of safety in urban neighbourhoods using 360 degree photographs of urban environments. Furthermore, to determine if these perceptions were accurate based on actual crime for each location

Methods/approach Using a randomised controlled experimental design, participants were assigned to view 360 degree photos of urban spaces where an adolescent fatal injury had occurred or a matched space where an adolescent spent time but no injury occurred.

Results Of those who agreed to participate (n = 21), 73% were female, 76% white and 57% grew up in a very large city or suburbs of a very large city. Cases and controls were not significantly different on any demographic indicator measured at baseline. There was a nearly significant difference in mean safety scores for cases compared to controls (0.81, p = 0.09) but not in the expected direction. A significant difference in mean safety scores between cases and controls that grew up in a large city or suburb (1.39, p = 0.04) yet again, not in the expected direction. Spearman’s correlations for safety ratings and actual crimes show significant but very weak associations, except robbery, theft and vandalism.

Conclusions Consistent with existing literature, individual’s perception of safety in urban environments is not congruent with risk based on objective measures. No significant difference was found between safety measures for cases and controls suggesting environmental markers, such as disorder, are not sufficient indicators of safety.

Significance and contributions This study demonstrates unique Methods for further exploration of perceptions of disorder, physical environment and safety.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.