Does it — the policy, the program, the clinical intervention — work? As Hill and Lynn (2001) note, this is the “big question” that policy makers and researchers are drawn to answer. Of lower priority in the policy research pecking order are questions concerning why a policy, program or treatment works (or does not work). Yet answers to such questions are essential if a successful intervention is to be replicated, scaled up, or adapted to different contexts. Thus, the big question for those interested in governance and public management is: What do organization, management, and other aspects of context contribute to the policy and program outcomes we observe?
A good example of the “Does it work?” fixation is found in the field of substance abuse treatment. Gerstein and Harwood (1990) observe that the question most frequently and emphatically asked of substance abuse treatment researchers and other experts is: Does treatment work? The brief and pointed answer, they submit, based on the available clinical and research data, is “It varies.” “Drug treatment is not a single entity but a variety of different approaches to different populations and goals” (1990:132).
A clinical research focus has dominated studies of substance abuse treatment processes and outcomes. These studies have produced some clear findings about aspects of substance abuse treatment that correlate with positive post-treatment outcomes. For example, there are consistent associations between program effectiveness and retention in treatment, counseling intensity, methadone dosage levels, regular participation in 12-step self-help groups, and provision of psychiatric and vocational services. Yet there is considerable variation in the administration and implementation of substance abuse treatment programs, including those practices known to be effective, of which we understand comparatively little. In their recent