PSYCHOMETRIC CHALLENGES IN MULTILINGUAL AND MULTICULTURAL EDUCATIONAL ASSESSMENTS OF UNIVERSITY SCIENCE EDUCATION STUDENTS

This study examined the psychometric challenges of multilingual and multicultural educational assessments among university science education students in Nigeria. The purpose was to investigate how linguistic and cultural diversity influences the reliability, validity, and fairness of assessments, and to identify strategies for improving accuracy. Guided by three research questions, the study employed a descriptive survey design with a sample of 400 undergraduate science education students selected through multistage sampling across four universities. Two instruments were used: the Multilingual and Multicultural Assessment Experience Questionnaire (MMAEQ) and a standardized Science Achievement Test (SAT). Data were analyzed using descriptive statistics, multi-group confirmatory factor analysis (CFA), differential item functioning (DIF), and multivariate analysis of variance (MANOVA). Findings revealed that students perceived significant psychometric challenges, including language barriers, cultural bias in test content, lack of standardized translations, and differential item functioning. Linguistic background was shown to influence test performance, with English first-language students achieving higher reliability and validity coefficients than peers from indigenous or mixed linguistic backgrounds. Respondents strongly endorsed strategies such as culturally adapted test items, language accommodations, pretesting across cultural groups, and the use of DIF analysis. These findings underscore the need for culturally responsive and linguistically sensitive assessment practices. The study concludes that traditional psychometric approaches are insufficient in multilingual and multicultural contexts. Assessment bodies, universities, and policymakers must integrate inclusive psychometric practices to ensure fairness, reduce bias, and enhance validity in science education assessments. The study provides evidence-based recommendations and highlights the importance of capacity building among test developers to sustain improvement in higher education assessment systems.