La regresión múltiple en la investigación macro-comparativa: límites y alternativas

Abstract

This paper offers criticisms of, and alternatives to, the use of multiple regression in cross-national comparison of welfare states and comparative political economy generally. A conspicuous feature of the methodological ferment in comparative social policy and political economy is its confirmation of the hegemony of MR. This paper reviews and extends well-known criticisms in principle of MR, but its main focus is on practical examples showing the value of alternative methods. Some of the arguments arguments are of a general nature, but the emphasis is on macro-comparative research. Two principal conclusions will emerge. First, even though technical means are available to deal with many of the limitations of MR, these solutions are either unconvincing or else require such a high technical investment that they offer questionable returns of scholarly investment. Second, dissatisfaction with MR does not necessarily mandate radical qualitative alternatives such as QCA or complete abandonment of numerical data. Lowtech forms of analysis (tables and graphical methods) and multivariate statistical techniques other than MR (such as cluster analysis and multi-dimensional scaling) constitute viable and useful alternatives.