Abstract
This article discusses and illustrates methods of computing the statistical power of hypothesis tests in regression analysis of the impacts of social policies. It first considers the importance of such calculations, both in interpreting the results of regression analysis and in designing evaluation research. The succeeding sections present technical details of the power calculations. Finally, these techniques are illustrated in the context of analysis of data from the Seattle and Denver income maintenance experiments, which were sponsored by the U.S. Department of Health, Education and Welfare in order to examine the various impacts of a negative income tax.
Get full access to this article
View all access options for this article.
