Global Dataset on Education Quality: A Review and Update (2000-2017)

37 Pages Posted: 28 Sep 2018 Last revised: 29 Sep 2018

See all articles by Harry A. Patrinos

Harry A. Patrinos

World Bank

Noam Angrist

University of Oxford; Youth Impact

Date Written: September 27, 2018


This paper presents the most expansive and current cross-country dataset on education quality. The database includes 164 countries/territories covering over 98 percent of the global population from 2000-2017. Of these countries, 111 are low or middleincome economies. Harmonized learning outcomes are produced using a conversion factor to compare international and regional standardized achievement tests. These tests include PISA, TIMSS, PIRLS, SACMEQ, LLECE and PASEC. In addition, this paper includes the Early Grade Reading Assessment (EGRA) for the first time. This enables extension of the database substantially. This methodological update paves the way to include a series of assessments that are increasingly common in developing countries that are often excluded from large international assessments. The database includes mean scores and disaggregates the data by subject, level, and gender. This paper further presents a series of methodological improvements including measures of uncertainty, fixed conversion factors for greater comparability over time and year-by-year data.

Keywords: Education for Development (superceded), Education For All, Educational Populations, Educational Sciences, International Trade and Trade Rules, Economics of Education

Suggested Citation

Patrinos, Harry Anthony and Angrist, Noam and Angrist, Noam, Global Dataset on Education Quality: A Review and Update (2000-2017) (September 27, 2018). World Bank Policy Research Working Paper No. 8592, Available at SSRN:

Harry Anthony Patrinos (Contact Author)

World Bank ( email )

1818 H Street NW
Washington, DC 20433
United States

Noam Angrist

University of Oxford ( email )

Mansfield Road
Oxford, Oxfordshire OX1 4AU
United Kingdom

Youth Impact ( email )


Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
PlumX Metrics