Studying implementation beyond deficit analysis: The top-down view reconsidered

Contemporary researchers of policy implementation make a plea for explaining variation in policy outputs. At the same time, still much implementation research, dispersed across the social sciences, entails studies of single cases in which a perceived gap between the intentions and the results of a public policy is analysed. In this article, a case is made for the lasting relevance of studying single policy processes, seen ‘from the top’, provided that the multi-dimensional character of these processes is taken into account. Empirical material from a study of educational inclusion policy in the United Kingdom shows how public policies may refer to different values (normative dimension) and imply ongoing policy formation between a variety of actors, each with particular stakes (political dimension), while policy goals seldom speak for themselves (practical dimension). By consequence, in implementation research the issues of, respectively, what needs explanation (the explanandum), locus specification and the appropriate unit of observation and analysis need attention.