Criminal Justice and Behavior Volume: 33 Issue: 5 Dated: October 2006 Pages: 646-674
This study drew on observational research on shoplifting to explore whether implicit cultural stereotyping had an impact on offender profiling.
The results indicated that despite training and specific instructions not to use shopper demographics in the selection of probable shoplifting offenders, observers tended to rely on implicit cultural stereotypes in the selection of individuals as possible shoplifters. More specifically, even when working with a standardized data template that directed observers to sample individuals based exclusively on behavioral characteristics, observers followed a disproportionate number of shoppers who were non-White adolescent males. The findings suggest that offender stereotypes do impact profiling behaviors. The findings further suggest that attempts to train police officers and other social control agents about the dangers of implicit stereotyping would most likely be ineffective. The study evolved from a larger study that took place in a retail pharmacy/drug store located in Atlanta, GA where researchers observed customers through an array of closed-circuit television (CCTV) cameras. Observations were made over the course of 12 months and involved the use of a standardized data template that directed researchers to record the date, time, number of shoppers, ethnicity, gender, approximate age, and approximate social class. The template also allowed for the recording of shoplifting incidents, including behavioral and demographic characteristics of the shoplifter. The sampling protocol called for observations of every third shopper who was dressed in such a manner that merchandise could be concealed. It soon became evident that observers were more likely to follow shoppers who matched their idea of a probable shoplifter and thus the research was extended to detect evidence of profiling. The data analysis process involved the establishment of interrater reliability for the coding process, followed by bivariate and multivariate logistic regression analyses. Follow-up studies should use larger samples and rely on a more tightly controlled experimental design. Figures, tables, notes, references
United States of America