Analysing implementation through cross case comparisons

Understanding implementation processes requires a research design that allows researchers to effectively engage with highly contingent processes.  This has led to a reliance on detailed examinations of single cases.... [ view full abstract ]

Understanding implementation processes requires a research design that allows researchers to effectively engage with highly contingent processes.  This has led to a reliance on detailed examinations of single cases.  However, the strength of the case study approach is also its biggest weakness in that this sort of analysis provides little or no guidance as to which of the processes uncovered by the analysis are the most relevant or important.  Well-tested theory can overcome this problem, but in an interdisciplinary research area such as implementation, "there is a tendency to cite theory, but not apply it" (van der Heijden & Kuhlmann 2017: 549), or when particular theoretical approaches are applied, concept proliferation or concept stretching occurs as the existing theory is unable to full explain empirical observations (van der Heijden & Kuhlmann 2017: 546-547).  For policy practitioners, theoretical critiques of a particular policy are only of interest if they provide guidance on what needs to be done to increase the effectiveness of existing policies or programs.  

The aim of this paper is to identify a new methodological approach to studying implementation that has the potential to increase cross-case comparability whilst retaining the capacity to provide a thick description of process.  The paper begins with a discussion of case study research and its place in the social sciences.  Attention then turns to why cross case comparisons are useful for academics and policy practitioners.  Section three discusses methodological issues facing implementation researchers, while section four examines existing approaches to comparative research using idiographic case studies or a combination of idiographic and nomethic case studies.  Section fiver reviews a sample of empirical studies to determine: 1) the type of data used in each study, 2) whether the data focuses on processes or outcomes or both, 3) the ways in which different types of data are combined, and 4) what researchers can achieve by combining data in these ways.  The paper concludes with a discussion of the extent to which the methodological approaches discussed in sections four and five may be useful in researching issues of concern to policy practitioners. 

References 

van der Heijden and J. Kuhlmann 2017, 'Studying Incremental Institutional Change', Policy Studies Journal, 45 (3).