Measuring Internal Migration Likelihoods across Demographic Groups: Comparing Multi-Group Confirmatory Factor Analyses and Logistic Regressions

• In Event: 546 - Section on Sociology of Population Roundtable Session (one-hour).
  In Refereed Roundtable: Table 05. Neighborhoods, Migration, and Health

Abstract

This is a study of internal migration within the United States. Its central research questions are: (1) How do migration likelihoods differ by demographic groups? and (2) How should migration researchers adjudicate between the two methods of multi-group Confirmatory Factor Analyses and logistic regressions? Using the 2011 American Community Survey, I conduct a series of multi-group Confirmatory Factor Analyses to measure respondents’ likelihood of migration, and compare these results to logistic regressions. Respondents are organized by geographic, socioeconomic, and demographic predictors. While my first research question pertains to respondents’ mobility based on their demographic groups, I also address the conversation of which statistical methods are appropriate for measuring migration likelihoods. The empirical results indicate that those with children present, minorities, and females are less likely to move. College graduates and those living in the South are more likely to move, and earning an above-median income has mixed results with migration likelihoods. While establishing adequate model fit was not an issue for my models, confirming factorial invariance was more challenging. For that reason, logistic regressions may be better-suited for studying migration likelihoods. The implications of these findings, the limitations of this project, and directions for future research are discussed.

Author

• Brock Ternes, University of Kansas