Cross-cultural validation of the short form of the Physical Self Inventory (PSI-S).

Abstract

The study examined the cross-cultural validity of the short form of the Physical Self-Inventory (PSI-S) among samples of adolescents speaking French, Dutch, Turkish, Italian, and Arab. A total of 4867 adolescents (1173 Belgian Flemish, 598 French, 1222 Italian, 643 Turkish, 646 Kuwaiti, and 585 Tunisian) completed the original PSI-S version, and a revised version including a positively-worded reformulation of the three negatively-worded PSI-S items. The results supported the factor validity and reliability of revised PSI-S version across all cultural groups, and its superiority when compared to the original version. Compared to confirmatory factor analyses (CFA), relying on an exploratory structural equation modeling (ESEM) measurement model further resulted in superior solution, and more cleanly differentiated factors. PSI-S responses proved to be fully invariant across cultural groups, and presented no evidence of differential item functioning (DIF) as a function of age, gender, body mass index (BMI), and sport involvement. However, the results revealed meaningful mean level differences as a function of gender, age, sport involvement, and BMI that were mostly consistent with the results from previous studies.