The V-Dem Measurement Model: Latent Variable Analysis for Cross-National and Cross-Temporal Expert-Coded Data​

34 Pages Posted: 30 May 2019

See all articles by Daniel Pemstein

Daniel Pemstein

North Dakota State University

Kyle L. Marquardt

National Research University Higher School of Economics - School of Political Science

Eitan Tzelgov

Göteborg University - V-Dem Institute; University of Gothenburg

Yi-ting Wang

National Cheng Kung University

Juraj Medzihorsky

Göteborg University - School of Business, Economics and Law

Joshua Krusell

Göteborg University - Department of Political Science

Farhad Miri

Göteborg University - V-Dem Institute

Johannes Roemer

Göteborg University - V-Dem Institute

Date Written: March 2019

Abstract

The Varieties of Democracy (V–Dem) project relies on country experts who code a host of ordinal variables, providing subjective ratings of latent - that is, not directly observable - regime characteristics over time. Sets of around five experts rate each case (country-year observation), and each of these raters works independently. Since raters may diverge in their coding because of either differences of opinion or mistakes, we require systematic tools with which to model these patterns of disagreement. These tools allow us to aggregate ratings into point estimates of latent concepts and quantify our uncertainty around these point estimates. In this paper we describe item response theory models that can that account and adjust for differential item functioning (i.e. differences in how experts apply ordinal scales to cases) and variation in rater reliability (i.e. random error). We also discuss key challenges specific to applying item response theory to expert-coded cross-national panel data, explain the approaches that we use to address these challenges, highlight potential problems with our current framework, and describe long-term plans for improving our models and estimates. Finally, we provide an overview of the different forms in which we present model output.

Suggested Citation

Pemstein, Daniel and Marquardt, Kyle L. and Tzelgov, Eitan and Wang, Yi-ting and Medzihorsky, Juraj and Krusell, Joshua and Miri, Farhad and Roemer, Johannes, The V-Dem Measurement Model: Latent Variable Analysis for Cross-National and Cross-Temporal Expert-Coded Data​ (March 2019). V-Dem Working Paper 2019:21. Available at SSRN: https://ssrn.com/abstract=3395892 or http://dx.doi.org/10.2139/ssrn.3395892

Daniel Pemstein (Contact Author)

North Dakota State University ( email )

Fargo, ND 58105
United States

Kyle L. Marquardt

National Research University Higher School of Economics - School of Political Science ( email )

Moscow
Russia

Eitan Tzelgov

Göteborg University - V-Dem Institute ( email )

United States

University of Gothenburg ( email )

Viktoriagatan 30
Göteborg, 405 30
Sweden

Yi-ting Wang

National Cheng Kung University ( email )

No.1, University Road
Tainan
Taiwan

Juraj Medzihorsky

Göteborg University - School of Business, Economics and Law ( email )

Vasagatan 1
Goteborg, 40530
Sweden

Joshua Krusell

Göteborg University - Department of Political Science ( email )

Box 711
Göteborg, S-405 30
Sweden

Farhad Miri

Göteborg University - V-Dem Institute ( email )

United States

Johannes Roemer

Göteborg University - V-Dem Institute ( email )

United States

Register to save articles to
your library

Register

Paper statistics

Downloads
10
Abstract Views
123
PlumX Metrics