Suchergebnisse

Social scientists tend to accept lower standards of rigor in cross-national surveys than in national surveys, leading to heroic conclusions about differences between nations on the basis of obviously faulty data. Arising perhaps from a commendable wish to respect different cultural norms, even some of the most conscientious cross-national studies make the mistake of permitting considerable variations by country in the type and quality of the methods they deploy. Meanwhile, analysts of cross-national data frequently abandon offering explanations and interpretations in favor of league tables of distributions showing merely "gee whiz" national differences. This article acknowledges the formidable obstacles in the way of achieving rigor in large-scale comparative studies and offers 10 possible rules to mitigate the difficulties. It suggests that bigger is usually worse and recommends routine cross-national collaboration in analysis and interpretation—not just in design, development, and execution—among scholars in each of the nations under the microscope.

This paper is a systematic analysis of the comparative method. Its emphasis is on both the limitations of the method and the ways in which, despite these limitations, it can be used to maximum advantage.The comparative method is defined and analyzed in terms of its similarities and differences vis-à-vis the experimental and statistical methods. The principal difficulty facing the comparative method is that it must generalize on the basis of relatively few empirical cases. Four specific ways in which this difficulty may be resolved are discussed and illustrated: (1) increasing the number of cases as much as possible by means of longitudinal extension and a global range of analysis, (2) reducing the property space of the analysis, (3) focusing the comparative analysis on "comparable" cases (e.g., by means of area, diachronic, or intranation comparisons), and (4) focusing on the key variables.It is argued that the case study method is closely related to the comparative method. Six types of case studies (the atheoretical, interpretative, hypothesis-generating, theory-confirming, theory-infirming, and deviant case analyses) are distinguished, and their theoretical value is analyzed.