Multisite Randomized Trials in Criminology

Abstract

INTRODUCTION Relatively few randomized controlled trials (RCTs) have been conducted in criminology and criminal justice; rarer still are multisite RCTs. Yet, as the field moves toward a more evidence-based perspective, we anticipate increases in the numbers of RCTs. The results of these trials will begin to identify efficacious interventions, and more sophisticated analyses will be required to identify interventions that accomplish the desired impact. Multisite RCTs provide a mechanism for increasing our understanding of what works for whom, and why, and in what environments. Multisite studies are independent randomized experiments undertaken in two or more sites where researchers involved in the study plan and collaborate across these sites (Boruch 1997; Kraemer, 2000; Weisburd and Taxman, 2000). Some of the reasons for conducting multisite studies involve the need to replicate findings from initial single-site studies, to gain a sample size large enough to obtain sufficient power, and to discern moderate to small effect sizes. In addition to outlining further the definitional criteria of a multisite RCT, both of these justifications for conducting a multisite study are discussed in detail.