Disagreement over the Evaluation of a Controlled Experiment
Hans Zeisel
Abstract
Hans Zeisel
Abstract
Under the leadership of the U.S. Department of Labor, the states of Texas and Georgia engaged in what was perhaps the most careful large-scale randomized controlled experiment ever conducted in a natural setting. It was designed to test the widely supported theory that the recidivism rate of ex-convicts could be reduced or retarded by providing them for a time with a monthly monetary allowance. During the evaluation of the experiment, a radical difference of opinion over the correct interpretation of its results arose between the evaluators (Peter H. Rossi, Richard A. Berk, and Kenneth J. Lenihan) and Hans Zeisel, a member of the advisory board. This article, together with the comment and reply that follow it, explicates this difference and attempts to resolve it.
OpenAlex reports 39 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Under the leadership of the U.S. Department of Labor, the states of Texas and Georgia engaged in what was perhaps the most careful large-scale randomized controlled experiment ever conducted in a natural setting. It was designed to test the widely supported theory that the recidivism rate of ex-convicts could be reduced or retarded by providing them for a time with a monthly monetary allowance. During the evaluation of the experiment, a radical difference of opinion over the correct interpretation of its results arose between the evaluators (Peter H. Rossi, Richard A. Berk, and Kenneth J. Lenihan) and Hans Zeisel, a member of the advisory board. This article, together with the comment and reply that follow it, explicates this difference and attempts to resolve it.
Key concepts: Allowance (engineering), Natural experiment, Recidivism, Randomized experiment, Test (biology), Difference in differences, Psychology, Interpretation (philosophy)