NCJ Number
183343
Journal
Crime and Delinquency Volume: 46 Issue: 3 Dated: July 2000 Pages: 330-353
Date Published
July 2000
Length
24 pages
Annotation
This article considers the ethical problems associated with random assignment and privacy in randomized field experiments, with attention to procedural, legislative, and technical approaches for reducing or avoiding the problems; examples are provided from a variety of disciplines, although the emphasis is on research in crime and delinquency.
Abstract
In the course of using randomized experiments over the past 30 years, primary ethical problems have arisen in two areas. The first concerns the random assignment feature that is common to any randomized trial. The second problem concerns individual privacy. It affects efforts to elicit information in probability sample surveys and ethnographic research, as well as in randomized experiments. This problem area also concerns efforts to link research records on individuals who participate in experiments with administrative records on the same individuals. Certain features of randomized field trials can be manipulated so as to make the experiment's design more acceptable on ethical or legal grounds. This manipulation is such that scientific standards for the fairness of a comparison can be maintained, and a legitimate statement about statistical error can be made without sacrificing compliance with professional or legal standards. This article discusses such manipulations in the following areas of the experiment's design: treatment and control conditions, units of random allocation and analysis, the level of supply of and demand for the treatment, eligibility of individuals, imbalance in random configurations, and the ratio of treated to untreated participants. In the area of privacy, a variety of approaches can be used to ensure that the research records on individuals participants remain confidential. This article discusses statistical approaches for ensuring privacy and confidentiality as well as statutory approaches for ensuring confidentiality of research records. 33 references