Sample selectivity and the validity of international student achievement tests in economic research

Larger rates of exclusion, non-response, and age-specific enrollment are related to better country average scores on international student achievement tests. But accounting for sample selectivity does not alter existing evidence that academic achievement enters importantly in economic growth regressions.

Research Highlights► Sampling affects international student achievement tests used in cross-country analyses. ► Sampling issues include larger rates of exclusion, non-response, and age-specific enrollment rates. ► These issues will affect country average scores. ► They do not affect estimates of academic achievement on economic growth.