TY - JOUR

T1 - A critical evaluation of alignment optimization for improving cross- national comparability in international large-scale assessments

AU - Sandoval-Hernández, Andres

AU - Carrasco, Diego

AU - Eryilmaz, Nurullah

N1 - Publisher Copyright: © 2025 The Authors

PY - 2025/12

Y1 - 2025/12

N2 - This study critically examines the use of alignment optimization to improve cross-national comparability of teacher and principal scales from the Teaching and Learning International Survey (TALIS) 2018. By investigating key psychometric properties, including dimensionality, reliability, and measurement invariance, the study highlights critical challenges in international large-scale assessments. While unidimensionality and high internal consistency were established for all scales, traditional multiple-group confirmatory factor analysis (MGCFA) suggested that scalar invariance could not be fully established for most scales, raising concerns about the robustness of cross-national comparisons under strict invariance assumptions. In contrast, alignment optimization emerged as a flexible and robust method, significantly enhancing the comparability of principal scales, all of which met alignment criteria. However, persistent challenges were identified for many teacher scales, which fell below alignment thresholds, emphasizing unresolved methodological complexities. This study demonstrates the transformative potential of alignment optimization for advancing psychometric rigor in global educational research and underscores the need for innovative approaches to address lingering comparability issues in international assessments.

AB - This study critically examines the use of alignment optimization to improve cross-national comparability of teacher and principal scales from the Teaching and Learning International Survey (TALIS) 2018. By investigating key psychometric properties, including dimensionality, reliability, and measurement invariance, the study highlights critical challenges in international large-scale assessments. While unidimensionality and high internal consistency were established for all scales, traditional multiple-group confirmatory factor analysis (MGCFA) suggested that scalar invariance could not be fully established for most scales, raising concerns about the robustness of cross-national comparisons under strict invariance assumptions. In contrast, alignment optimization emerged as a flexible and robust method, significantly enhancing the comparability of principal scales, all of which met alignment criteria. However, persistent challenges were identified for many teacher scales, which fell below alignment thresholds, emphasizing unresolved methodological complexities. This study demonstrates the transformative potential of alignment optimization for advancing psychometric rigor in global educational research and underscores the need for innovative approaches to address lingering comparability issues in international assessments.

KW - Alignment optimization

KW - Cross-national comparability

KW - Educational measurement

KW - International large-scale assessments (ILSAs)

KW - Measurement invariance

KW - Multiple-group confirmatory factor analysis (MGCFA)

KW - Principal scales

KW - TALIS 2018

KW - Teacher scales

KW - Educational science

UR - http://www.scopus.com/inward/record.url?scp=105016638555&partnerID=8YFLogxK

U2 - 10.1016/j.stueduc.2025.101519

DO - 10.1016/j.stueduc.2025.101519

M3 - Journal articles

AN - SCOPUS:105016638555

VL - 87

JO - Studies in Educational Evaluation

JF - Studies in Educational Evaluation

SN - 0191-491X

M1 - 101519

ER -