Cross-ethnic measurement invariance of the SCARED and CES-D in a youth sample.

ABSTRACT

This study evaluated the cross-ethnic measurement invariance of 2 common screening measures of anxiety and depressive symptoms in youth. The measurement invariance of the Screen for Childhood Anxiety and Related Emotional Disorders (SCARED) and the Center for Epidemiologic Studies Depression Scale (CES-D) was tested across 881 African American (Black; n = 396), Hispanic (n = 185), non-Hispanic White (White; n = 166), and Asian/Indian (n = 134) youth in the 7th grade. The measures were administered as part of a grade-wide screening to identify youth with elevated anxiety and depressive symptoms. The 5-factor model of the SCARED and the 4-factor model of the CES-D best represented the data for all ethnic groups. Results provided support for strong invariance of the SCARED across all 4 ethnic groups. Results provided support for strong invariance of the CES-D across Black, White, and Asian/Indian youth, and partial strong invariance for Hispanic youth. Overall, results suggest that factor means and total scores can be compared across groups. Factor mean differences across groups were identified for both measures. In particular, Hispanic youth reported greater levels of anxiety and depression in certain domains than White, Black, and Asian youth. Our findings support the use of the SCARED and CES-D as tools for measuring anxiety and depressive symptoms in ethnically diverse youth in the United States. Results are discussed with respect to the importance of establishing measurement invariance for screening measures of anxiety and depressive symptoms prior to comparing symptom levels across ethnic groups.