Cross-cultural Adaptation of the Activity Questionnaire for Adults and Adolescents into Turkish and Investigation of its Validity and Reliability

Abstract

Objective: The Activity Questionnaire for Adults and Adolescents (AQuAA) is used to evaluate physical activity (PA) levels in different age groups. Its validity and reliability in the Turkish language have not been studied yet. This study aims to adapt the AQuAA into Turkish and to investigate its validity and reliability. Methods: A total of 124 volunteers were included in the study. After the Turkish adaptation of AQuAA, the AQuAA-Tr version was administered to the volunteers for test-retest reliability twice, with an interval of two weeks, and the International Physical Activity Questionnaire Short Form (IPAQ-SF) for criterion validity. For construct validity, the step counts of the volunteers were followed for two weeks with the Samsung Health® smartphone pedometer application. The reliability of the AQuAA-Tr was evaluated with intra-class correlation coefficients (ICC). Spearman correlation coefficients (r) were used to analyze the relationships between continuous variables. Results: A total of 72 adolescents (51 females and 21 males, mean age 14.5 ± 0.1 years) and 52 young adults (32 females and 20 males, mean age 25.8 ± 1.3 years) participated in the test-retest reliability and criterion validity study. Thirty-four adolescents (26 female, 8 male, mean age 14.7 ± 0.2 years) and 39 young adults (27 female, 12 male, mean age 25.6 ± 1.5 years) were included in the construct validity study. The test-retest reliability of the questionnaire was in the range of strong to very strong (ICC = 0.704 to 0.982) in adolescents and moderate to strong (ICC = 0.606 to 0.851) in adults for different levels of PA. In the context of the criterion validity, although there were moderate to strong correlations (r = 0.413 to 0.768) between some PA levels of the IPAQ-SF and AQuAA-Tr in adolescents and moderate correlations (r = 0.422 to 0.525) in adults, the correlations were mostly weak or negligible. In relation to construct validity, although there were moderate correlations (r = 0.435 to 0.504) between the Samsung Health® data and some PA levels of the AQuAA-Tr in adults, the correlations were mostly weak or negligible. There were no correlations between the Samsung Health® data and AQuAA-Tr in adolescents. Conclusion: The reliability of the AQuAA-Tr was confirmed in both adolescents and adults. However, the criterion and construct validity of the AQuAA-Tr were not confirmed for either adolescents or adults. Introducing a PA questionnaire, which can provide detailed information about sedentary, light, moderate, and vigorous PA scores separately and total PA scores and allows the evaluation of PA in different categories, into our language is considered beneficial. Yet, the results of AQuAA-Tr should be interpreted carefully in the clinic.