Testing for approximate measurement invariance of instructional quality in the Teaching and Learning International Survey (TALIS) 2018

This study investigates how consistently instructional quality is measured among secondary school teachers across 47 countries and economies using the TALIS 2018 teacher data (N = 127,607). We use two methods: the conventional approach (multiple group confirmatory factor analysis) and alignment optimization to test the comparability of instructional quality in an international large-scale survey. The results show that while basic (configural) and structural (metric) consistency is present, full comparability (scalar invariance) is not achieved using the conventional method. However, the alignment optimization method reveals partial comparability across countries. These findings imply that while comparisons of instructional quality can be made across most countries, caution is needed when interpreting results in certain cases. This study highlights the importance of testing measurement invariance in cross-national education research.