A myriad of options: Validity and comparability of alternative international education variables

Education is a key variable in survey-based research, impacting various outcomes and serving as a cornerstone in social stratification research. However, measuring education in cross-national surveys is challenging due to the complexity and variability of educational systems across countries. For comparative research, the education variable does not only need to be valid, but also comparable across countries, which implies a similar degree of validity in different countries. Simultaneously, education measures need to be parsimonious, especially for self-administered surveys. This study, expanding on previous research, evaluates the validity and comparability of 16 alternative education coding specifications using nearly 400 dependent variables to identify an optimal solution that could serve as a standard. We use European Social Survey (ESS) rounds 5 to 10 data, which provide a highly detailed international education variable serving as a benchmark. Results suggest that, while no alternative variable performs as well as the benchmark, a variable comprehensively covering educational quality ranks second in terms of validity, while the three-digit version of ISCED 2011 with a reduced quality indicator and coverage of the general/vocational divide, yields better comparability across countries than the other alternatives. These findings offer insights for comparative surveys to select an optimal international education variable for different survey modes and for researchers analysing survey data.