Investigation of the structural invariance of the Ethnic Identity Scale with Native American youth.

This study examined the structural invariance of the 17-item Ethnic Identity Scale (EIS; Umaña-Taylor, Yazedjian, & Bámaca-Gomez, 2004) with a Native American sample. Participants were 144 students (55% female) enrolled in a Native American boarding school in the south-central U.S. Participants ranged in age from 13 to 19 years (M = 15.9, SD = 1.49). Confirmatory factor analysis was performed using maximum likelihood estimation. The original model was a poor fit for the data set. Removal of Items 1 and 2 and modeling the covariation between Items 6 and 11 resulted in an overall chi square value of Χ²(86, N = 144) = 152.21, p < .0001 (CFI = .936; RMSEA = .073; 90% CI on RMSEA = .054, .092), suggesting an acceptable fit. Substantial factor loadings were evident for the remaining 15 items. Analyses for the most part replicated Umaña-Taylor, Yazedjian, and Bámaca-Gomez's (2004) original 3-factor structure. We hypothesize that the minor differences between the present structure and that of Umaña-Taylor et al. (2004) may be an artifact of this particular sample.