Use of Item Response Models in Assessing the Health Literacy Facet Understanding Health Information for Early Childhood Allergy Prevention and Prevention of COVID-19 Infections by Pregnant Women and Mothers of Infants

Abstract

Abstract. Appropriate parental health literacy (HL) is essential to preventively maintain and promote child health. Understanding health information is assumed to be fundamental in HL models. We developed N = 67 items (multiple-choice format) based on information materials on early childhood allergy prevention (ECAP) and prevention of COVID-19 infections to assess the parental HL facet Understand. N = 343 pregnant women and mothers of infants completed the items in an online assessment. Using exploratory factor analysis for ordinal data (RML estimation) and item response models (1-pl and 2-pl model), we proved the psychometric homogeneity of the item pool. 57 items assess the latent dimension Understand according to the assumptions of the 1-pl model (weighted MNSQ < 1.2; separation reliability = .855). Person parameters of the latent trait Understand correlate specifically with subjective socioeconomic status ( r = .27), school graduation ( r = .46), allergy status ( r = .11), and already infected with COVID-19 ( r = .12). The calibrated item pool provides a psychometrically sound, construct-valid assessment of the HL facet Understand Health Information in the areas of ECAP and prevention of COVID-19 infections.