Abstract
Many pronouncements have been made in the process of site selection for new prisons. A review of the research literature on the effect of prisons on communities reveals that most of this information is unsupported by good research design. This essay presents a critique of the methodological and research design problems of prison impact studies. Various threats to validity are explored and two complementary methodologies, a single-site time series and multisite time series, are proposed for prison impact studies. Strict attention to internal-validity concerns will provide better answers to causal mechanisms and outcomes than is currently the case.
Get full access to this article
View all access options for this article.
