Structural validity and measurement invariance of the scale of oral health outcomes for 5‐year‐old children

Abstract

Background: Important psychometric approaches (structural validity, measurement invariance) remain underdeveloped inmeasuring oral health-related quality of life, particularly for preschool children across diverse contexts.Aim: This study aimed to evaluate the structural validity of the child's self-reported version of the Scale of Oral Health Outcomesfor 5-year-old children (SOHO-5) and test the measurement invariance from cultural and clinical/non-clinical comparisonperspectives.Design: Three datasets were analysed: two from Brazil and one from the United Kingdom (UK). One Brazilian dataset wasderived from clinical data collection (nbr- cl. = 193), while the others were from non-clinical epidemiological school-based studies(nbr-ncl. = 768, nuk-ncl. = 296). Dimensionality was tested through parallel analysis and confirmed by unidimensional indexes.Measurement invariance across datasets was tested via multi-group Confirmatory Factor Analysis (CFA).Results: Unidimensionality was empirically confirmed for all three datasets. The multi-group CFA tests reached partial scalarinvariance threshold between the Brazilian and UK non-clinical datasets. However, there was no scalar equivalence when com-paring non-clinical with clinical datasets, neither within Brazil nor between countries.Conclusion: The child's self-reported version of the SOHO-5 is a unidimensional oral health-related quality-of-life measure thatis psychometrically comparable across different cultures (partial scalar invariance), but not between clinical and non- clinical groups.