Measurement invariance of the Multidimensional Depression Assessment Scale (MDAS) across gender and ethnic groups of Asian, Caucasian, Black, and Hispanic

Abstract

Background: Major Depressive Disorder is a severe and highly disabling mental illness. Almost all self-reported questionnaires have overlooked the interpersonal symptoms of depression which are important across gender and culture. The Multidimensional Depression Assessment Scale (MDAS) developed by Cheung and Power (2012) entails comprehensive emotional, cognitive, somatic, and interpersonal subscales. It addresses the criticism that existing self-report depression scales might not cover sufficient phenomenological heterogeneity of depression. The current study aimed to evaluate the psychometric properties of the MDAS across gender and four major ethnic groups of Caucasian, Black, Asian and Hispanic, including reliability and concurrent validity against the Centre for Epidemiological Studies Depression Scale (CES[sbnd]D) and Patients Health Questionnaire (PHQ-9). It also aimed to establish a stable factor structure across gender and ethnic groups and test the measurement invariance to enhance its potential for clinical use. Methods: A community sample of 3499 participants from four ethnic groups were recruited via online crowdsourcing sites of Qualtrics and Amazon M Turk. Each individual completed a demographic questionnaire, the MDAS, CESD and PHQ-9. Results: There was good internal consistency (Cronbach's alpha >0.90) and concurrent reliability across gender and ethnic groups. Strict measurement invariance was established for MDAS over a four-factor factor structure corresponding to the four subscales. Conclusions: The MDAS showed good psychometric properties and measurement invariance of a four-factor structure, suggesting its potential to be used in clinical settings across gender and ethnic groups. Limitations: Participants all answered the questionnaires in English, which could hinder cultural variations in their expression of symptoms.