Psychometric Properties of Persian Version of “Child-Initiated Pretend Play Assessment” For Iranian Children

Background Play, particularly pretend play, has a cognitive basis and has been linked to the language and social ability. Objectives The goal of this study was to examine face and content validity, inter-rater, intra-rater and test-retest reliability of the Persian translation of the child-initiated pretend play assessment Methods Ten occupational therapists consented to be in the content validity study. Face validity was examined by five occupational therapy specialists. For reliability, 31 typically developing children aged 4 - 6 years were chosen from kindergartens of four regions of Tehran, Iran. Two weeks after the initial assessment the children were re-tested for test-retest reliability. Intra-rater and inter-rater reliability was scored from videos of the children’s play assessment. Results To be culturally appropriate for Iran, some phrases were changed and the pigs were replaced by dogs. Content validity index (CVI) and content validity ratio (CVR) were acceptable for all items. The Intra-class correlation coefficient was ICC = 0.99 for intra-rater reliability and ICC = 0.98 for inter-rater reliability. For test-retest reliability, the intra-class correlation coefficient for symbolic and combined object substitution scores and all elaborate play scores ranged from ICC = 0.69 to - 0.99. For imitated actions, the majority of children scored 0 on both test and re-test. Conclusions The Persian version of child initiated pretend play assessment has appropriate face and content validity. Inter-rater and intra-rater reliability were excellent. PEPA, combined and symbolic NOS showed good to excellent test-retest reliability. Test retest reliability for conventional NOS was moderate and NIA was not stable with more children imitating the examiner in the first test but not the re-test.