Reliability and validity of parent and child versions of the multidimensional anxiety scale for children in community samples.

OBJECTIVE: To evaluate the psychometric properties of the Multidimensional Anxiety Scale for Children (MASC), which measures physical symptoms, harm avoidance, social anxiety, and separation/panic.

METHOD: Using a community sample of Australian children (8-13 years old), the internal stability, factor structure, and convergent validity of child- and parent-report versions of the MASC were examined. A total of 499 children participated at time 1, which represented 18% of children invited to participate; 91% were retained at 12-month follow-up.

RESULTS: Moderate to strong internal reliability was exhibited across all MASC subscales, and support was found for the original four-factor structure of the measure. A higher-order anxiety factor was found to account for the intercorrelation between the four factors. Low parent-child agreement was found on the MASC, whereas correlations between the MASC and the Spence Children's Anxiety Scales demonstrated good convergent validity within reporter (excepting the harm avoidance scale). Scores over a 12-month period showed greater stability for parent reports than for child reports.

CONCLUSIONS: This study is the first to demonstrate that the factor structure of the MASC holds for both child and parent reports of anxiety. The MASC is useful for directly comparing symptom reporting across children and parents during clinical assessment.