Evaluating the Invariance of the Multigroup Ethnic Identity Measure Across Foreign-Born, Second-Generation and Later-Generation College Students in the United States

• Published in: Cultural diversity & ethnic minority psychology 2016-07, Vol.22 (3), p.460-465
• Main Authors: Yap, Stevie C. Y., Donnellan, M. Brent, Schwartz, Seth J., Zamboanga, Byron L., Kim, Su Yeong, Huynh, Que-Lam, Vazsonyi, Alexander T., Cano, Miguel Ángel, Hurley, Eric A., Whitbourne, Susan Krauss, Castillo, Linda G., Donovan, Roxanne A., Blozis, Shelly A., Brown, Elissa J.
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; College Students; Emigrants and Immigrants - psychology; Ethnic Groups - psychology; Ethnic Identity; Family Characteristics - ethnology; Female; Generational Differences; Group Identity; Human; Humans; Male; Measurement; Measurement Invariance; Psychometrics - statistics & numerical data; Social Identification; Students - psychology; United States - ethnology; Young Adult
• ISSN: 1099-9809; 1939-0106
• DOI: 10.1037/cdp0000068

Objectives: Past research has established that the Multigroup Ethnic Identity Measure (MEIM) exhibits measurement invariance across diverse ethnic groups. However, relatively little research has evaluated whether this measure is invariant across generational status. Thus, the present study evaluates the invariance of the MEIM across foreign-born, second-generation, and later-generation respondents. Method: A large, ethnically diverse sample of college students completed the MEIM as part of an online survey (N = 9,107; 72.8% women; mean age = 20.31 years; SD = 3.38). Results: There is evidence of configural and metric invariance, but there is little evidence of scalar invariance across generational status groups. Conclusions: This study suggests that the MEIM has an equivalent factor structure across generation groups, indicating it is appropriate to compare the magnitude of associations between the MEIM and other variables across foreign-born, second-generation, and later-generation individuals. However, the lack of scalar invariance suggests that mean-level differences across generational status should be interpreted with caution.