Objectives: To examine the extent to which different operationalizations of target, displacement, and comparison sites affect the interpretation of a CCTV impact evaluation. Methods: Different operationalizations are used to examine change in property and violent crime before and after CCTV was introduced in a residential neighbourhood in Malmö, Sweden, using paired t tests, odds ratio effect sizes and weighted displacement quotients. Results: Despite most results being non-significant as a result of low statistical power, different operationalizations appear to produce varying results. This issue is most prominent for target sites and property crime while it is less so for catchment and comparison sites, and violent crime outcomes. Conclusion: Valid operationalizations of research areas are highly important for CCTV impact evaluations and may affect the study outcome. Future research with access to larger data sets is encouraged to elaborate on these findings and further analyse the impact of measurement choices on experimental outcomes.

Objectives: To explore the extent to which unsafe locations are concentrated to micro-places within the city of Malmö, Sweden, and whether there is a temporal stability in these micro-places over time.

Methods: Information on unsafe locations is obtained from an open-ended item across three waves of a random sample community survey. Reported unsafe locations are geocoded as polygon, polyline, and point features and merged with a 200 by 200-m grid-cell network using both unadjusted and weighted counts.

Results: The results suggest that unsafe locations are concentrated to a small share of grid-cells using different metrics. There are also signs of spatial clustering and a temporal stability of unsafe locations over time.

Conclusions: As unsafe locations are concentrated to a small share of micro-places the results have important implications for both theory and practice. However, further research exploring unsafety and fear of crime at micro-places is highly warranted.

Objectives: To explore how locations perceived as unsafe by city inhabitants are associated with spatial risk factors, neighbourhood collective efficacy and neighbourhood disorder. Methods: Information on unsafe locations is obtained from two open-ended items in a random sample community survey. Reported unsafe locations in the respondents’ own neighbourhood (Q18) and in other partsof the city (Q19) are geocoded and merged with a 200 by 200-metre grid-cellnetwork. The data are analysed using multilevel models with grid-cells nested in neighbourhoods. Results: A number of spatial risk factors are correlated with unsafe locations, but these associations seem to vary between unsafe locations in one’s own neighbourhood and in other parts in the city respectively. While the results indicate that there is substantial between-neighbourhood variance inunsafe locations, the association between neighbourhood-level collective efficacy and disorder is generally weak and non-significant, although this differs depending on the outcome. Conclusions: Using a place-based approach to the fear of crime, this study contributes to the current body of research by studying specific unsafe locations and their characteristics. The findings may guide practitioners to better understand why some micro-level places are perceived as unsafe and how public perceptions of safety may be improved.

The introduction of experience research using smartphone applications has enabled researchers to explore previously uncharted territories within the social science landscape. The use of experience research methodologies may both avoid some of the biases associated with conventional survey methods and enable researchers to gather information on situational dimensions of different outcomes. While existing smartphone-based experience research is found in several scientific disciplines, one outcome that has rarely been studied using this approach is fear of crime. Although fear of crime here only serves as an example of an outcome that may benefit from being examined in situ, the main focus of this article is directed at assessing the feasibility of gathering momentary and daily assessments on fear of crime using an experience sampling research design. A sample of 191 university students participated in a non-incentivized research study by downloading the smartphone application STUNDA, completing a baseline survey and repeatedly answering signal-contingent surveys, event-contingent surveys, and daily assessments across a self-defined study period. The results indicate that it is feasible to conduct research on fear of crime using a smartphone application. However, variations in participation across groups indicate that the methodology is perhaps best suited to dedicated groups of participants with a special interest in the studied outcome. Methodological considerations and implications for future research are discussed.