Reexamining and validating a multicomponent model of in-group identification for college student belonging

We examined the validity and reliability of a model of in-group identification in a new context—university students living in residential commons in the United States. We collected survey data from fall 2020 to spring 2022 and used four cross-sectional and two panel samples in the analysis. First, we used confirmatory factor analysis to determine whether a higher-order factor model or a simpler first-order factor model better fit the data. Second, we used two panel samples to test for measurement invariance in in-group identification over time. Results indicate the simpler factor structure is a better fit to the data. Also, the measure exhibits at least scalar invariance in each panel. Our results produced a simplified version of the in-group identification model, adding to the measurement tools available to assess sense of belonging. Moreover, strong evidence of measurement invariance allows a comparison of mean differences and differences in relationships between variables across time.