Premium
Testing the Measurement Invariance of Nativism *
Author(s) -
Zhao Yikai
Publication year - 2019
Publication title -
social science quarterly
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.482
H-Index - 90
eISSN - 1540-6237
pISSN - 0038-4941
DOI - 10.1111/ssqu.12594
Subject(s) - psychological nativism , measurement invariance , confirmatory factor analysis , scale (ratio) , psychology , construct (python library) , social psychology , econometrics , structural equation modeling , political science , immigration , geography , mathematics , computer science , statistics , law , cartography , programming language
Objective Survey questions regarding immigration have been frequently used to compare how mass publics think about nativism, but it remains unclear whether the resulting nativism measures are comparable across countries. In this article, we empirically examine whether nativism achieves scalar invariance—an important measurement feature for valid instrumentation—across very different country contexts. Method We examine measurement invariance of a nativism scale using survey data from 22 countries representing 16,096 individual respondents. To test for invariance, we utilize multigroup confirmatory factor analysis (MG‐CFA). Results While the nativism scale fails to achieve full scalar invariance, we are able to establish partial scalar invariance, indicating the nativism scale is valid across many national contexts. Though highly correlated, the means of the latent nativist variable also provide different rankings than the means of an additive scale. Conclusions With a few limitations, nativist sentiments can be meaningfully measured across political contexts. Across 19 of the 22 countries, respondents answered survey items on nativism in ways that suggest their answers reflect the same underlying latent construct.