Psychometric validation and measurement invariance of the Arabic version of the problem-solving inventory-20: An...

This study aimed to translate and validate the Arabic PSI-20, examine its psychometric properties, test measurement invariance across key subgroups and establish an optimal cut-off score.

Problem-solving ability (PSA) is a core cognitive competence underpinning nursing students' academic success, clinical reasoning and professional readiness. Although the Problem-Solving Inventory-20 (PSI-20) is widely used internationally, no validated Arabic version is available for Arabic-speaking nursing students.

A two-phase methodological and cross-sectional validation study.

Faculty of Nursing, Mansoura University, Egypt, during the 2024-2025 academic year.

A total of 633 undergraduate nursing students participated in Phase Two.

Phase One followed Beaton's cross-cultural adaptation framework. Phase Two involved psychometric evaluation using confirmatory factor analysis (CFA), assessment of convergent and discriminant validity, internal consistency, test-retest reliability, receiver operating characteristic (ROC) curve analysis and multi-group measurement invariance across gender, GPA, family structure, living arrangements, sleep regularity and ChatGPT use.

The Arabic PSI-20 demonstrated excellent internal consistency (Cronbach's α = 0.956) and strong construct validity. CFA supported the original three-factor structure (CFI = 0.979, RMSEA = 0.055). Measurement invariance was confirmed across most subgroups, with partial invariance across ChatGPT-use categories. PSA was positively associated with self-directed learning and negatively associated with academic stress. ROC analysis identified an optimal cut-off score of ≥ 57 for detecting low PSA (AUC = 0.799).

The Arabic PSI-20 is a valid and reliable instrument for evaluating PSA.