Measurement invariance of the higher-order model of Preschool Anxiety Scale (PAS) across child age, gender, parental anxiety, and pandemic period in England

The Preschool Anxiety Scale (PAS) is a parent-report scale measuring young children’s anxiety symptoms involving five specific anxiety symptoms (separation anxiety, physical injury fears, social phobia, obsessive-compulsive disorder, generalised anxiety) that load on a higher-order factor representing general anxiety shared by all specific anxiety symptom subtypes. Although the PAS has been widely used to assess anxiety symptoms in young children, few studies have tested its measurement invariance for group comparisons. Using data from a sample of 2,221 children and their parents/carers in England, this study investigated the measurement invariance of the higher-order model of the PAS across child age (4-6 years vs. 6-7 years), gender (girls vs. boys), parental anxiety (low vs. high level), and children’s living circumstances (before vs. after the removal of COVID-19 restrictions). Our findings demonstrated the good factor structure, internal consistency, and convergent validity of the higher-order model of the PAS in all subgroups and supported its configural, metric, scalar invariance across these subgroups. Therefore, the findings suggest that the PAS is a reliable and valid instrument for assessing specific anxiety symptoms and general anxiety among young children in England, and that comparisons can be made between the subgroups under examination. Keywords: Anxiety symptoms, Preschool anxiety scale, Measurement invariance, Group comparisons, Early childhood