Harmonization of SDQ and ASEBA Phenotypes: Measurement Variance Across Cohorts

Area of Science:

• Psychometrics
• Cross-cultural psychology
• Child and adolescent mental health

Background:

• Harmonizing scores from different instruments is crucial for meta-analysis.
• Measurement invariance across populations is a key step in score harmonization.
• The Child Behaviour Check List (CBCL) and Strength and Difficulties Questionnaire (SDQ) are widely used for assessing child mental health.

Purpose of the Study:

• To examine measurement invariance of harmonized anxiety/depression and ADHD scores from the CBCL and SDQ.
• To assess invariance across different countries (Australia, Netherlands, UK) and age groups.
• To validate a Bayesian method for modeling measurement non-invariance in polytomous items and small group numbers.

Main Methods:

• Utilized Bayesian methods for modeling measurement non-invariance, adapted for polytomous items and small cohorts.
• Analyzed data from three large cohorts: Australia (10-11.5 years), Netherlands (11-13.5 years), and UK (14-19 years).
• Applied the Verhagen and Fox (2013a) approach to assess cross-cohort invariance.

Main Results:

• Found minimal differential functioning of harmonized anxiety/depression and ADHD scores between the CBCL and SDQ across the studied cohorts.
• Demonstrated that a single harmonization model is applicable across Australian, UK, and Dutch child cohorts.
• Confirmed the robustness of the harmonization model across diverse national and age samples.

Conclusions:

• Harmonized scores for anxiety/depression and ADHD using CBCL and SDQ exhibit strong measurement invariance across different countries and age groups.
• The findings support the use of combined analyses of CBCL and SDQ data in international comparative research.
• The adapted Bayesian modeling approach is effective for assessing measurement invariance in multi-cohort studies.