Spatial analysis of the distribution of crime and disorder often depart from the concept of neighborhood as geographical unit of analysis. Collective efficacy, the combination of working trust and expectations of a willingness to act for the common good, has been shown to tap into important social mechanisms that can explain the differences between neighborhoods. How such social mechanisms function in a neighborhood and how it impacts on spatial distribution of disorder within rather than between neighborhoods is rarely studied. In a case study of four Swedish neighborhoods differences within and between neighborhoods have been studied. Social networks, collective efficacy and indirectly the social construction of neighborhoods have been discussed in focus groups with residents of the neighborhoods and in key informant interviews with property managers and active members of local associations. Social boundaries within the neighborhoods emerge clearly from respondents’ stories with corresponding intra-neighborhood differences in respondents’ perceptions of collective efficacy, safety and disorder. Few respondents express any neighborhood-wide feelings of social cohesion which points toward the possibility that neighborhoods can be considered aggregates of smaller socio-spatial units. To study whether the socio-spatial units identified through the interviews differ in levels of physical disorder a systematic social observation (SSO) has been performed during 2011. The data have been analyzed both through aggregation based on social boundaries expressed in the interviews and through the use of density- and hotspot based methods where no pre-defined geographical units are needed. Results from the analysis of spatial distributionof physical disorder show that the social boundaries identified within neighborhoods matter in determining how disorder is distributed within the neighborhood. A preliminary conclusion is that the social mechanism of collective efficacy is best understood on much smaller units of analysis than typical neighborhoods.