Caroline Mellgren

---

What's neighbourhood got to do with it?

The overarching aim of this thesis is to contribute to an increased understanding of how the neighbourhood context acts to influence individual reactions to crime. The general framework is that the social and physical make-up of residential neighbourhoods influences individuals, over and above individual background characteristics. Disorder is an important neighbourhood-level factor and its presence is more or less pronounced in different neighbourhoods. It acts as a sign of a general urban unease and has potential negative consequences for the individual as well as for the community at large. Four studies have been conducted each with its own specific objective. The first study reviews the Swedish crime survey literature in order to assess the national evidence for neighbourhood effects, paying special attention to methodological issues. Overall, the current literature provides mixed evidence for neighbourhood effects. Methodological issues were identified as obstacles to drawing general conclusions and specific areas that need improvement were identified. The second study examines the origins of disorder at the neighbourhood level and the relationship between disorder and crime. Two theory-driven models of the relationship between population density, disorder, and crime are tested alongside an examination of whether these models are equally applicable to data collected in two cities, Antwerp in Belgium and Malmö in Sweden. The results found some support for direct effects of disorder on crime in both settings, independent of structural variables. Some differences between the two settings were observed suggesting that the disorder-crime link may vary by setting. To further examine the influence of neighbourhood context, the role played by neighbourhood level disorder in relation to worry about criminal victimization has been tested in a multilevel model in the third study. Overall the hypotheses of the influence of both neighbourhood level and individually perceived disorder, in shaping individual worry were supported. Individual background explains most of the variance but neighbourhood context has independent effects on worry. Individual level perceived disorder mediated the effect of neighbourhood disorder on worry suggesting that the effect of context is indirect through its effect on individual perception. The fourth study investigates whether it is possible to identify any unique neighbourhood effects on the extent to which residents apply crime preventive strategies. Initially some of the total variance in the dependent variables was found to be situated between neighbourhoods. This indicates that the neighbourhood context may influence individuals’ willingness to take crime preventive action. As expected, individual characteristics explained a majority of this between-neighbourhood variance. An important finding is that the contextual variables appear to have different effects on different activities, highlighting the need to study different actions separately.