To Örebro University

Objectives: Assessing the predictive accuracy of using prior crime, place attributes, ambient population, community structural, and social characteristics, in isolation and combined when forecasting different violent and property crimes.

Method: Using multilevel negative binomial regression, crime is forecasted into the subsequent year, in 50-m grid-cells. Incidence rate ratio (IRR), Prediction Accuracy Index (PAI), and Prediction Efficacy Index (PEI*) are interpreted for all combined crime generators and community characteristics. This study is partially a test of a crude version of the Risk Terrain Modeling technique.

Results: Where crime has been in the past, the risk for future crime is higher. Where characteristics conducive to crime congregate, the risk for crime is higher. Community structural characteristics and ambient population are important for some crime types. Combining variables increases the accuracy for most crime types, looking at the IRR. Taking the geographical area into account, crime history in combination with both place- and neighborhood characteristics reaches similar accuracy as crime history alone for most crime types and most hotspot cutoffs.

Conclusions: Crime history, place-, and neighborhood-level attributes are all important when trying to accurately forecast crime, long-term at the micro-place. Only counting past crimes, however, still does a really good job.

The current study examines the relative importance of fear of crime and its three aspects—affective, behavioral, and cognitive—in relation to frailty (daily activities, health problems, psychosocial functioning), news consumption, trust in neighborhood, and perceptions of rising crime among older adults in a mid-sized Swedish municipality. Using logistic regression, we analyzed data from a subsample of 336 participants from a Swedish cross-sectional study of adults aged 65 and older (mean age = 76.62; age range = 64 to 106 years). Contrary to the common portrayal of older adults as highly fearful of crime, our findings reveal that their primary worries are more closely linked to health problems, social isola-tion, and other personal vulnerabilities than to fear of crime. Trust in the neighborhood is associated with lower levels of the affective and behavioral aspects of fear of crime. Conversely, perceptions of rising crime are positively related to the affective and behavioral aspects of fear. Health problems and problems in psychosocial functioning are positively linked to the affective aspect. Further, problems in psychosocial functioning are positively related to the behavioral aspect. Overall, while crime is a worry, it is overshadowed by more immediate health and social issues.

Decades of criminological research have highlighted the principle that "place matters" when analysing crime dynamics. While acknowledging this, our study emphasises that place matters at specific times of the day and days of the week. This paper explores spatio-temporal patterns of assaults in public spaces in Stockholm, Sweden, using a time-space modeling approach. By segmenting annual crime registry data into time-specific models (e.g., weekday mornings and weekend nights), sixteen models were created covering two years. Using Risk Terrain Modeling, spatio-temporal cells were analysed with Spearman's Rank Correlations and Predictive Accuracy and Efficiency measures to explore and compare time-specific models to a more generalised annual model. Results suggest that time-specific models perform better in smaller geographical areas compared to a yearly model. However, correlations and GIS mapping show that micro-grid crime hotspots fluctuate within larger hotspots and between years. These findings underscore the need for dynamic hotspot monitoring within crime analysis and for law enforcement to adapt resource allocation at both hotspots and within hot times. By shifting from static annual models to more nuanced temporal models, authorities can enhance crime prevention strategies and optimise interventions at high-risk locations during critical temporal windows.

In environmental criminology, various methods exist to forecast unsafety. Some are more complex than others. To determine their practicality, we must compare the accuracy of simple, transparent, and functional methods with slightly more complex methods and those requiring more data collection.

The overall aim of the current dissertation was to examine the relationship between crime history, environmental and neighborhood characteristics in forecasting unsafety, both crime and fear of crime, in various geographical locations. Study I compared the predictive accuracy of two methods using historical crime exposure and different crime-time-periods for violent and property crimes. Study II compared the predictive accuracy of prior crime, place attributes, ambient population, and community structural and social characteristics for various crime types. Study III examined the relationship between violent and property crime, as well as community structural and social characteristics, and different types of fear of crime.

The findings of the current dissertation suggest that, overall, a one-size-fits-all approach is not effective. Simpler methods are generally comparable to more complex ones in long-term crime forecasting at the micro-level. However, at the neighborhood level, social integration plays a significant role in determining levels of perceived safety and fear of crime.

Perceived unsafety, fear of crime, and avoidance were studied in relation to different types of crime, crime in different time perspectives, concentrated disadvantage, collective efficacy, urbanity, age structure, and neighborhood disorder. Four data sources were used on a large Swedish city; a community survey from 2012 and 2015 among residents, census data on socio-demographics, police data on reported violent (assault and robbery in the public environment), and property crimes (arson, property damage, theft, vehicle theft, and residential burglary) and geographical information on local bus stops and annual passengers visiting these bus stops. Collective efficacy primarily, but also concentrated disadvantage, was strongly related to perceived unsafety, across 102 neighborhoods. Collective efficacy was strongly related to fear of crime. It was not viable to relate the neighborhood variables with avoidance, however. Fear of specific violent crimes was different from fear of specific property crimes and should for future reference be examined separately. Crime, visible disorder, urbanity, and age structure do not seem as important.

There are multiple geographical crime prediction techniques to use and comparing different prediction techniques therefore becomes important. In the current study we compared the accuracy (Predictive Accuracy Index) and precision (Recapture Rate Index) of simply counting crimes: Simple Count with Kernel Density Estimation in the prediction of where people are reported to commit violent crimes (assault and robbery) and property crimes (residential burglary, property damage, theft, vehicle theft and arson), geographically. These predictions were done using a different number of years into the future and based on a different number of years combined to do the crime prediction, in a large Swedish municipality. The Simple Count technique performed quite well in comparison to simple Kernel Density Estimation no matter what crime was being predicted, making us conclude that it may not be necessary to use the more complex method of Kernel Density Estimation to predict where people are reported to commit crime geographically.

Uniformed presence is commonly thought to create feelings of safety in people.However, do differently uniformed people contribute to an equal amount of safety and arethere situation-dependent differences? The present study examined the association betweenvarious types of uniformed presence and people’s feelings of safety through a questionnaireamong 352 respondents (18–86 years) (49.1 % women). The questionnaire contained picturesof situations perceived as relatively safe and unsafe with or without uniformed presence. Therespondents estimated how safe they thought they would feel in these situations with nouniformed presence, two police officers, six police officers, a police vehicle, two securityguards, or two police volunteers. Results showed that uniformed presence did not increasefeelings of safety in a situation perceived as relatively safe, making patrol unnecessary. Insituations perceived as relatively unsafe however, all types of uniformed presence increasedfeelings of safety. Foot patrolling police contributed to the greatest increase in feelings ofsafety. Security guards and police volunteers created similar amounts of feelings of safetymaking police volunteers a cost-effective alternative. All types of foot patrol were better thanvehicle patrol, making non-police groups an alternative to vehicle patrol. Some situational,gender, and age differences were found.