Latent Group Differential Item Functioning on Depression Scale

• In Event: AERA Poster Session 14
  In Poster Session: Division D Section 1 Poster Session 1

Abstract

The mixture rating scale IRT model was applied to detect the differential item functioning on 20-item version of the Center for Epidemiological Studies Depression Scale (CES-D) using the WINMIRA computer program (von Davier, 2001). Three model fit indices (i.e., AIC, BIC and CAIC) were calculated to determine the number of latent classes. Crosstab analysis using demographic information such as race, gender, age, education level, language was performed to identify the latent classes. Item parameters and the percentages of responses on each category were also compared to understand the characteristics of latent class as well as the mean and proportion of latent classes.

Authors

• Youn-Jeng Choi, Ewha Womans University

• Giyeon Kim, Chung Ang University