Learning Through Controlled Experiments: Community Service and Heroin Prescription in Switzerland

Abstract

Europe, over the past two decades, has seen many innovations in the field of corrections, particularly new sanctions that are becoming increasingly popular as alternatives to imprisonment, such as community service. Innovative approaches have also been tested in the field of drug treatment, including large-scale drug-substitution programs. Usually, such programs have been evaluated, if at all, under the form of before-after studies. Thus, little is known about treatment effects, particularly in the longer run and/or compared to alternative approaches. Two controlled experiments conducted recently in Switzerland involving community service and heroin prescription to addicts may indicate a shift to more rigorous evaluations. They both illustrate the potentials of controlled experiments for progress in knowledge as well as some problems in methodological, legal, ethical, and practical respects. Whereas controlled experiments are necessary to learn in some areas, more conventional before-after studies may be valid under particular circumstances.