A composite short self-report of adolescents’ out-of-school physical activity: enhanced validity, reliability, cross-cultural applicability and identification of psychological correlates

Questionnaires of adolescents’ moderate-to-vigorous physical activity (MVPA) often capture differing timeframes and lack cross-cultural comparability, limiting their utility in large-scale international research. This study evaluated the factorial, cross-cultural, incremental, criterion validity and reliability of a composite tool that integrates three widely used Short Self-Reports (SSRs) of out-of-school MVPA (OS-MVPA). Psychological correlates of PA and accelerometer-based MVPA were used as validation criteria. In Study 1, 9,435 adolescents (ages 10–18) from seven countries (France, Greece, Italy, Portugal, Spain, Turkey, UK) completed the PACE+, a one-item weekly MVPA measure (WHO-HBSC), and four items from the Youth Activity Profile (YAP). Structural equation modelling supported a hierarchical factor model with three lower-order and one higher-order factors, demonstrating strong structural validity and cross-cultural metric invariance. The composite measure showed greater predictive power, with psychosocial variables explaining 51% of its variance–substantially more than individual SRs (35%–42%). In Study 2, involving 2,907 adolescents from four countries, the composite measure demonstrated good test–retest reliability over a six-month interval, exceeding that of individual SRs. In Study 3, 188 adolescents from three countries wore accelerometers for seven days. The composite tool exhibited strong criterion validity, explaining a higher proportion of variance in accelerometer-assessed MVPA compared to previous studies. Incremental validity for the composite tool was supported both in Study 1 and 3. Overall, findings suggest that the composite SR measure offers a valid, reliable, and culturally robust approach for assessing youth out-of-school PA in large-scale, cross-national studies, outperforming single SR instruments across multiple psychometric criteria.