The Challenge of Empirically Assessing the Effects of Constitutions
Journal of Economic Methodology 22(1): 46-76
34 Pages Posted: 7 Mar 2017
Date Written: 2015
Mutually supporting methodologies are necessary for building a convincing case establishing a particular effect. Strengths and weaknesses of four empirical methods are discussed. Econometric methods quantify the relative importance of different factors and may assess the time frame over which constitutions matter, but have difﬁculties in dealing with nonlinear interactions among constitutional and cultural details. Cluster analysis can be a pre-requisite to other methods, and an analytic method in itself, useful for identifying the details that really matter and discovering surprising patterns in the data. I discuss the application of cluster analysis on the Comparative Constitutions Project database. Qualitative comparative analysis can reveal the hidden structure of interactions among different variables, but robustness checks are difﬁcult to perform. Case studies are useful for distinguishing between rules-in-use and rules-in-form and for discovering important informal aspects. They can deal with complex nonlinearities well, but they are often hard to generalize.
Keywords: cluster analysis, Comparative Constitutions Project, qualitative comparative analysis, case studies, analytic narratives, comparative histories
JEL Classification: B41, P50, K00, K42, O50
Suggested Citation: Suggested Citation