Data from the 1987-92 National Crime Survey suggest that workplaces are the setting for 15 percent of the violent crimes and nearly one-fourth of the personal larcenies experienced by persons age 12 or older. Data from the 1995 National Retail Security Survey revealed that 341 major retailers estimated that they lost an average of 1.835 percent of their annual sales volume to a combination of shoplifting, employee theft, vendor theft and fraud, and bookkeeping and accounting errors. However, crimes against retail businesses are difficult to research due to the difficulty of detection, the complicated methods involved, and the lack of an audit trail to determine whether the crime was committed by an insider or an outsider. Methodological problems include the use of inventory shrinkage as a proxy measure of crime, the constantly changing sampling universe of the retail business community, low response rates, the difficulty of grouping retail chains to permit comparisons, data contamination, the interpretation of responses, and time delays before usable data are obtained. The available responses to retail crimes all have significant shortcomings. Recommended changes include improved data on crimes and losses, research to evaluate loss prevention programs, and security personnel training. 36 references