NCJ Number
100389
Journal
American Behavioral Scientist Volume: 27 Issue: 4 Dated: (March-April 1984) Pages: 423-452
Date Published
1984
Length
30 pages
Annotation
This article examines types and uses of computerized, systematic data searching, particularly in law enforcement, and discusses some of the limitations of such techniques.
Abstract
Systematic data searching permits the analysis of complex transactional data that can reveal discrepancies, contradictions, and irregularities indicative of many types of low-visibility, white-collar crime. It can permit discovery of patterns of offenses and yield profiles of likely offenders. Two types of commonly used techniques are matching and profiling. Matching involves the comparison of information from two or more distinct data sources. It may be used to discover impersonation and false representation, such as that involved in double dipping or false titling of stolen vehicles. Profiling permits investigators to correlate data items to assess how close a person or event comes to a predetermined characterization or model of infraction. It can be used to discover such activities as welfare or insurance fraud, cheating on tests, and employee drug use. While these techniques have enhanced discovery, they involve two major costs. Such techniques are prone to at least five sources of error: erroneously reported or entered data, time lags, hardware and software problems, the acontextual nature of the decision process, and the probabilistic nature of profiling. Use of such techniques also may involve infringement of privacy, search and seizure, and burden of proof guarantees. 33 notes and 39 references.