Abstract
This article presents a structured framework for conducting and reporting cross-cultural psychometric research, aimed at addressing practical challenges and enhancing methodological coherence. Cross-cultural psychometrics involves the evaluation and adaptation of psychological measurement instruments across diverse cultural contexts to ensure conceptual equivalence, validity, and reliability. The proposed framework consists of six core stages. First, instrument selection based on theoretical foundations ensures alignment between the instrument’s conceptual model and the target culture; if misaligned, the development of a new instrument is recommended. Second, cross-cultural adaptation follows Beaton et al.’s (2000) five-step process: initial translation by two bilingual translators (one informed, one blind), synthesis of translations, back-translation, expert committee review for semantic and conceptual equivalence, and cognitive debriefing or pilot testing. Third, sampling and implementation involves determining sample size using either a subject-to-variable (STV) ratio of 10:1 or a minimum of 200 participants, with ethical recruitment and confidentiality safeguards. Fourth, item analysis includes statistical indicators such as mean, standard deviation, skewness, kurtosis, item discrimination, and item-total correlation (≥0.30). Fifth, reliability evaluation employs Cronbach’s alpha (≥0.70), McDonald’s omega, and intraclass correlation coefficient (ICC), with ICC values interpreted from poor (0.50) to excellent (0.90). Sixth, validity assessment encompasses face and content validity (qualitative and quantitative), criterion, convergent, discriminant, factorial, and known-group validity. Confirmatory factor analysis (CFA) is recommended for factorial validity, with acceptable model fit indicated by CFI and TLI > 0.95, RMSEA < 0.06, and SRMR < 0.08. Known-group validity is supported when the instrument significantly differentiates between groups with varying levels of the underlying construct.