Broken windows and low adolescent prosociality: not cause and consequence, but co-symptoms of low collective efficacy.

The current study presents a case in which adolescent prosociality is lower in neighborhoods with greater physical disorder. Current theory provides two interpretations for such a pattern: (1) that disorder signals a threatening environment and discourages prosociality ("broken windows theory"); (2) that disorder and low prosociality are both symptoms of a weak community (i.e., low collective efficacy). A survey of 642 students from a small American city was combined with an assessment of the built environment to evaluate these two interpretations. Students were nested in 59 Census block groups. Multilevel models demonstrated that collective efficacy best explained variation in prosociality between neighborhoods, and that perceived collective efficacy best explained variation within neighborhoods. Objective and perceived disorder had no significant relationship with prosociality in these models, suggesting that disorder is not directly responsible for cross-neighborhood variation in prosociality. The paper discusses the implications for place-based interventions promoting prosociality. The results also emphasize the need for measures of social processes (e.g., collective efficacy) when evaluating "broken windows" hypotheses.