Item response theory and differential test functioning analysis of the HBSC-Symptom-Checklist across 46 countries

Area of Science:

• Psychometrics
• Adolescent Health
• Cross-cultural Psychology

Background:

• The Health Behaviour in School-aged Children (HBSC) Symptom Checklist (SCL) is a widely used, non-clinical measure for psychosomatic complaints in adolescents.
• Previous research has explored its psychometric properties, but dimensionality, Graded Response Model (GRM) adequacy, and cross-country measurement invariance require further clarification.

Purpose of the Study:

• To investigate the dimensionality of the HBSC-SCL.
• To assess the adequacy of the Graded Response Model (GRM) for describing the HBSC-SCL data.
• To examine measurement invariance of the HBSC-SCL across different countries.

Main Methods:

• Analysis of data from 229,906 adolescents (aged 11, 13, 15) across 46 countries from the 2018 HBSC survey.
• Exploratory graph analysis for dimensionality assessment.
• Multigroup invariance analyses and alignment analysis to evaluate cross-country measurement properties and GRM fit.

Main Results:

• Exploratory graph analysis indicated a one-dimensional structure for the HBSC-SCL in 16 countries; deviation from unidimensionality was minimal in most countries.
• Configural and metric invariance were supported across 32 countries, but scalar invariance was not.
• Alignment analysis identified non-invariance for specific items, including irritability and feeling nervous/bad temper.

Conclusions:

• The HBSC-SCL demonstrates consistent and reliable unidimensionality across most participating countries.
• These findings support the use of the HBSC-SCL as a valid instrument for population health analyses and early identification of mental health status in adolescents.