Measurement invariance of depression symptom ratings across African American, Hispanic/Latino, and Caucasian adolescent psychiatric inpatients

Psychol Assess. 2019 Jun;31(6):833-838. doi: 10.1037/pas0000708. Epub 2019 Feb 25.

Abstract

The Beck Depression Inventory-II (BDI-II) is widely used to assess adolescent depressive symptom severity. Psychometric investigations, including factor-analytic studies, with adolescents support the reliability and validity of the BDI-II. However, a major limitation of this research is that samples have been predominantly Caucasian/White. This is critical because depressive illness is highly prevalent across race and ethnicity, and the extent to which reliability and findings generalize to non-Caucasian populations is in question. The present study recruited African American/Black (n = 96), Hispanic/Latino(a) (n = 151), and Caucasian/White (n = 97) adolescent psychiatric inpatients (Mage = 14.73) to test the measurement invariance of the BDI-II, using Osman and colleagues' two-factor solution while also assessing within-group reliability and concurrent validity by examining associations with other symptom measures. Across groups, the two-factor solution, factor loadings, and indicator thresholds were invariant. Within-group reliability estimates were adequate, and the concurrent validity was supported. This suggests BDI-II symptom comparisons between African American/Black, Hispanic/Latino(a), and Caucasian/White adolescent inpatients are valid. Critical extensions of this work may include the examination of potential invariance across depressive symptom clusters via network analysis and invariance testing of depression symptom ratings over time in ethnoracially diverse children and adolescents. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

MeSH terms

  • Adolescent
  • Black or African American / psychology*
  • Black or African American / statistics & numerical data
  • Child
  • Depressive Disorder / diagnosis*
  • Depressive Disorder / psychology*
  • Female
  • Hispanic or Latino / psychology*
  • Hispanic or Latino / statistics & numerical data
  • Humans
  • Inpatients / psychology*
  • Inpatients / statistics & numerical data
  • Male
  • Psychometrics
  • Reproducibility of Results
  • White People / psychology*
  • White People / statistics & numerical data