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 datasets. Conclusion: Given its superior coverage and comparability, the SWIID remains the optimum source of data for broadly cross-national research on income inequality.

Important Figure

Figure 7: k-fold By-Observation Cross-Validation Results

BibTeX Citation

@article{Solt2020,
    author = {Solt, Frederick},
    journal = {Social Science Quarterly},
    number = {3},
    pages = {1183-1199},
    title = {Measuring Income Inequality Across Countries and Over Time: The Standardized World Income Inequality Database},
    volume = {101},
    year = {2020}}