Designing and Validating a Questionnaire on Healthy Lifestyle to Reduce Depressive Symptoms among Adolescents

• Main Authors: Esra Tajik, Latiffah Abd Latiff, Chin Yit Siew, Hamidin Awang, Siti Nur' Asyura Adznam
• Format: Article
• Language: English
• Published: Tehran University of Medical Sciences 2020-02-01
• Series: Iranian Journal of Psychiatry
• Subjects: Depressive Symptom; Lifestyle; Partial Least Squares; Reliability; Validity
• Online Access: https://ijps.tums.ac.ir/index.php/ijps/article/view/1637
• ISSN: 1735-4587; 2008-2215

Objective: Most adolescents do not meet national recommendations for healthy lifestyle and reducing depressive symptom. A comprehensive educational program with its specified questionnaire is needed to improve healthy lifestyle to assess the lifestyle components. This study aimed to confirm the validity of a healthy lifestyle questionnaire to assess depressive symptoms among adolescents. Method: A descriptive predictive study using simple random sampling was performed in a secondary school in Kuala Lumpur, Malaysia, among 235 school-going adolescents (13-15 years old) with Malaysian nationality. The instrument consisted of a set of dual language (English and Malay) for both researcher-administered questionnaires (knowledge, attitude, eating behavior, and relaxation activities) and standard-validated questionnaires (Physical Activity questionnaire for Adolescents, Depression, Anxiety, and Stress Scale- 21, and Brief Copping). Data were analyzed using the kappa statistics (k) and the intraclass correlation coefficient test for reliability and Partial Least Squares (PLS) for validity. Results: The reliability of all variables was over the substantial level (ICC and/or k > 0.61). The model and construct variables were predictive of depressive symptoms by 49.4%. To analyze the validity of the questions, 11 questions were removed from the initial model for factorial loading less than 0.5. In convergent validity of theory of information-motivation-behavioral skills, AVE (average variance construct), the outer loading, was higher than 0.5. Conclusion: PLS confirmed the existence of sufficient correlations between different items of the construct. Thus, the weights of indicators appeared to be appropriate indicators for the model. The results proved that the information, motivation, and behavioral skills model was efficient for healthy lifestyle and can be a good base for further research.