« Back to Results

Achievement Tests II: On the Validity of Comparisons Across SES and Place

Paper Session

Saturday, Jan. 4, 2020 10:15 AM - 12:15 PM (PDT)

Marriott Marquis, Rancho Santa Fe 2
Hosted By: American Economic Association
  • Chair: John B. Klopfer, University of Hong Kong and United States Naval Academy

A Rosetta Stone for Human Capital

Dev Patel
Harvard University
Justin Sandefur
Center for Global Development


International comparisons of human capital figure prominently in economists' explanations of poverty, economic growth, and trade patterns. But how can we accurately measure the global distribution of skills when countries do not take the same tests? We develop a new methodology to non-parametrically link scores from distinct populations. By administering an exam combining items from different assessments to 2,300 primary students in India, we estimate conversion functions among four of the world's largest standardized tests spanning 80 countries. Armed with this learning "Rosetta Stone", we revisit various well-known results, showing, inter alia, that learning differences between most- and least-developed countries are larger than existing estimates suggest. Applying our translations to microdata, we match pupils' socio-economic status to moments of the global income distribution and document several novel facts: (i) students with the same household income score significantly higher if they live in richer countries; (ii) the income-test score gradient is steeper in countries with greater income inequality; (iii) girls read better than boys at all incomes but only outperform them in mathematics at the lowest deciles of the global income distribution; and (iv) the test-score gap between public and private schools increases with inequality, partially due to a rise in socio-economic sorting across school types.


Eric A. Hanushek
Stanford University
Annika B. Bergbauer
Ifo Institute
Ludger Woessmann
Ifo Institute and CESifo


The significant expansion of varying forms of student testing, while controversial in many countries, has not been generally linked to educational performance. Different testing regimes provide varying information to parents, teachers, and decision makers. We investigate how different types of information relate to student achievement. Our empirical analysis exploits data from over two million students in 59 countries observed across six waves of the international PISA test 2000-2015. Relying on the country panel feature of the data, we investigate how testing reforms relate to country performance on PISA tests over time, taking out country and year fixed effects. Expansion of standardized testing with external comparison, both school-based and student-based, is associated with improvements in student achievement. This effect is stronger in low-performing countries. By contrast, reforms to solely internal testing without external comparison and internal teacher monitoring including inspectorates are not related to changes in student achievement.
Jong-Wha Lee
Korea University
Jeffrey Livingston
Bentley University
JEL Classifications
  • I2 - Education and Research Institutions
  • C1 - Econometric and Statistical Methods and Methodology: General