Multilevel Research Designs: Case Selection, Levels of Analysis, and Scope Conditions

Subnational research in comparative politics has been growing steadily in the last two decades. However, methodological advances have been rather limited. This article builds upon Snyder’s (Stud Comp Int Dev 36(1):93–110, 2001) subnational comparative method and extends its logic to the comparison of subnational units from different countries. It proposes a novel typology of multilevel research designs that focuses particularly on cross-national small-N analysis (CSNA). This research design offers three different logics of qualitative case selection to achieve a sound trade-off between internal and external validity. This article analyzes the advantages and limitations of the underlying logics of CSNA and illustrates their use with recent empirical research from Latin American countries. It concludes by highlighting its versatility and offers a series of best practices in order to produce more generalizable findings than the majority of single-country subnational comparisons.