Cross-Cultural Comparison of the Strengths and Difficulties Self-Report Questionnaire in Twelve Asian and European Countries

• Published in: Journal of the American Academy of Child and Adolescent Psychiatry
• Main Authors: Sourander, Andre, Westerlund, Minja, Kaneko, Hitoshi, Heinonen, Emmi, Klomek, Anat Brunstein, How Ong, Say, Fossum, Sturla, Kolaitis, Gerasimos, Lesinskiene, Sigita, Li, Liping, Nguyen, Mai Huong, Kumar Praharaj, Samir, Wiguna, Tjhin, Zamani, Zahra, Gilbert, Sonja
• Format: Journal Article
• Language: English
• Published: United States 08.10.2024
• Subjects: cross-cultural research; measurement invariance; Strengths and difficulties questionnaire (SDQ); adolescent mental health; factor structure
• ISSN: 1527-5418

The self-report SDQ is widely used globally, hence, the validity of the inter-group comparisons is essential. We examined the structure of the self-report SDQ in a large multinational adolescent sample, tested its measurement invariance across genders and countries and compared youth mental health in 12 European and Asian countries. This study is part of a cross-cultural research on child and adolescent wellbeing and mental health in 12 Asian and European countries EACMHS. The sample (N = 26,306) came from a cross-sectional school-based survey of adolescents. We used confirmatory factor analysis to assess a common measurement model for the self-report SDQ and the measurement invariance of the model across gender and country. Fit indices in the total sample, in each gender and each of the 12 countries separately supported the use of the first order three-factor model (without the reverse-coded items) as a common measurement model for the self-report SDQ. Measurement invariance analyses provided good support for configural, metric and scalar invariance across gender, however, metric invariance across countries was not supported. There were significant gender main effects for all SDQ subscales, except for hyperactivity/inattention. Culture had significant main effects and moderated the magnitude of gender differences in all subscales. The findings support the use of the correlated three-factor model comprising the positive dimension of prosocial behavior and two broad groupings of internalizing and externalizing problems, without the reverse-coded problem items, as a common measurement model for the self-report SDQ internationally.