Measuring Income Inequality Across Countries and Over Time: The Standardized World Income Inequality Database

Abstract

Objective: This article documents wide-ranging revisions to the Standardized World Income Inequality Database (SWIID), which seeks to maximize the comparability of income inequality estimates for the broadest possible coverage of countries and years. Methods: Two k-fold cross-validations, by observation and by country, are used to evaluate the SWIID's success in predicting the Luxembourg Income Study (LIS), recognized in the field as setting the standard for comparability. Results: The cross-validations indicate that the new SWIID's estimates and their uncertainty are even more accurate than previous versions, extending its advantage in comparability over alternate income inequality data sets. Conclusion: Given its superior coverage and comparability, the SWIID remains the optimum source of data for broadly cross-national research on income inequality.