The researchers posited that studies of offender decision-making have often simplified the analysis into the decision to offend or not offend. This study explored a range of alternatives within the "not offending" category using a framework derived from the concept of crime displacement. Decision trees were employed to analyze the multi-staged decision-making processes of criminals who were blocked from offending due to a situational crime control or prevention measure. The researchers were primarily interested in how offenders evaluated displacement options as available alternatives, and posited that a better understanding of how criminals respond to crime control and prevention efforts, beyond simple desistance, could help to expand offender decision-making theory and provide insight into the efficacy of crime prevention practices.
All data were collected through face-to-face offender interviews (n=200). The researchers employed mixed-methods design in that the interview format was designed to allow both open-ended questions as well as hypothetical scenario methods. Each interview involved three parts: (1) offender experiences; (2) a crime control measures survey; and (3) situational crime vignettes. Subjects were first asked about their experiences involving situations in which they wanted to commit a crime but chose not to do so due to a crime control or prevention measure. Next, subjects were asked to assess the effect of a standard list of 10 to 17 control/prevention measures for their particular crime type and to explain why they thought the measure did or did not have an effect. Finally, subjects were given a series of situational vignettes, each describing a prevented crime situation, followed by five displacement options (spatial, temporal, target, persist/tactical, and functional) and a desistance option. Qualitative data from the subjects' explanations and experiences was also obtained to provide a more in-depth understanding of their decisions.
Semi-structured interviews were conducted with 200 adult offenders, either in jail or on probation under the authority of the Texas Department of Criminal Justice, from 14 counties. To be included in the sample, an offender had to have a minimum of three convictions for predatory property or street crime (auto theft, vehicle burglary, residential/commercial burglary, shoplifting, or street/commercial robbery).
Adult offenders, either in jail or on probation under the authority of the Texas Department of criminal Justice, having had a minimum of three convictions for predatory property or street crime (i.e., auto theft, vehicle burglary, residential/commercial burglary, shoplifting, or street/commercial robbery).
Variables include basic information about the offenders who participated in the study. The variables in this data set include a unique offender/respondent identifier, which can be used as a key variable when relating this to other datasets from this study (PartID). Apart from basic demographics (i.e., gender, age at the time of the interview, race/ethnicity), information is available about the offender's age at first arrest, and also about the offender's preferred crime type. Finally, two nominal variables indicate whether data are available for the offender form the crime control measures survey and from the situational vignettes.
The variables in this dataset relate to the data obtained from the crime prevention/control measures survey. The variables in this data set include a unique offender/respondent identifier, which can be used as a key variable when relating this to other datasets from this study (PartID). Additional variables include the crime type preferred by the offender, the crime control/prevention measures, the effect the measures had on the offender's behavior, the rank assigned based on the impact of the offender's behavior, and the offender's response to the crime control/prevention measure.
Variables included relate to data collected using the situational vignettes. The variables in this data set include a unique offender/respondent identifier, which can be used as a key variable when relating this to other datasets from this study (PartID). Other variables indicate the crime type and control/prevention measure the vignettes depicted, the vignette number, the behavioral responses selected by the offender and the order of preference, as well as a free text field outlining the reasons for their choice. The data in this dataset is clustered both on the individual vignette and the individual offender. Additional variables include crime type offender would displace to when engaging in functional displacement, and the effort, risk, and reward associated with the chosen behavioral response of the offender.