Identifying Profiles of Democracies: A Cluster Analysis Based on the Democracy Matrix Dataset from 1900 to 2017

Abstract

This study examines types of democracies that result from trade-offs within the democratic quality. Recently, the existence and relevance of trade-offs has been widely discussed. The idea is that the functions associated with the quality of democracy cannot all be maximized simultaneously. Thus, trade-offs are expressed in distinct profiles of democracy. Different profiles of democracy favour certain democracy dimensions over others due to their institutional design. Conceptually, we differentiate between four different democracy profiles: a libertarian-majoritarian (high political freedom, lower political equality, and lower political and legal control values), an egalitarian-majoritarian (high equality combined with lower freedom and control values), as well as two control-focused democracy profiles (high control values either with high degrees of freedom or high degrees of equality). We apply a cluster analysis with a focus on cluster validation on the Democracy Matrix dataset—a customized version of the Varieties-of-Democracy dataset. To increase the robustness of the cluster results, this study uses several different cluster algorithms, multiple fit indices as well as data resampling techniques. Based on all democracies between 1900 and 2017, we find strong empirical evidence for these democracy profiles. Finally, we discuss the temporal development and spatial distribution of the democracy profiles globally across the three waves of democracy, as well as for individual countries.

Different types of deficient democracies: Reassessing the relevance of diminished subtypes

“… 1. We wish to underline the differences between our paper and the paper by Schlenkrich (2019), which seems to deal with the same topic. However, each paper has a different conceptual, methodological and empirical background.…”

“…However, each paper has a different conceptual, methodological and empirical background. Firstly, Schlenkrich (2019) does not elaborate on the advantages in constructing diminished subtypes compared to classical subtypes. Secondly, although both papers apply cluster analysis, these methods are not similar.…”

“…Secondly, although both papers apply cluster analysis, these methods are not similar. Schlenkrich (2019) uses a different set of cluster algorithms, while our paper applies a clustering strategy based on calibrated fit indices. Thirdly, both papers are based on the same dataset, DeMaX, but they select different variables inside this dataset: while Schlenkrich (2019) uses the trade-off indicators, our paper works with indicators from the context measurement.…”

Political Trade-Offs: Democracy and Governance in a Changing World

“…Like Hidalgo (2019), he understands democracy as a multidimensional concept whose central dimensionspolitical freedom, political equality and political and legal control-cannot all be developed comprehensively at the same time. With the help of a reformulated data set from the Varieties of Democracy project, Schlenkrich (2019) shows that trade-offs that are assumed theoretically appear in empirical findings that demonstrate corresponding profiles of democracy.…”