Department of Political Science

The Methodology of “Varieties of Democracy” (V-Dem)

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

  • Michael Coppedge, University of Notre Dame
  • ,
  • John Gerring, University of Texas, Austin, Texas
  • ,
  • Carl Henrik Knutsen, Universitetet i Oslo
  • ,
  • Joshua Krusell, Göteborgs Universitet
  • ,
  • Juraj Medzihorsky, Göteborgs Universitet
  • ,
  • Josefine Pernes, Göteborgs Universitet
  • ,
  • Svend-Erik Skaaning
  • Natalia Stepanova, Göteborgs Universitet
  • ,
  • Jan Teorell, Lunds Universitet
  • ,
  • Eitan Tzelgov, University of East Anglia
  • ,
  • Steven L. Wilson, University of Nevada, Reno
  • ,
  • Staffan I. Lindberg, Göteborgs Universitet

This article describes and discusses the new generation of methodological responses to measuring democracy and related issues generated by Varieties of Democracy (V-Dem). V-Dem is distinct in several regards in addition to its unique level of disaggregation, by the combination of: historical data extending back to 1900 and for a large selection among them to 1789 for many countries in the world; use of multiple, independent coders for each evaluative question; inter-coder reliability tests incorporated into a custom designed Bayesian item-response theory measurement model; provision of confidence bounds for all point estimates associated with expert-coded questions as well as for all indices; multiple indices reflecting varying theories of democracy; fully transparent aggregation procedures; and that all data are made freely available, including original coder-level judgments (exclusive of any personal identifying information).

Original languageEnglish
JournalBMS Bulletin of Sociological Methodology/ Bulletin de Methodologie Sociologique
Pages (from-to)107-133
Number of pages27
Publication statusPublished - 2019

    Research areas

  • Aggregation, Democracy, Democratization, Dictatorship, Expert, Index, Indicators, survey

See relations at Aarhus University Citationformats

Download statistics

No data available

ID: 164029323