Group CFA

Often when analyzing a survey, you want to learn more about responses of particular groups or even differences between groups. There is a rich suite of tools in SEM available for comparing differences between groups, but let's start with the basics: modeling a CFA by group. Our c_sat data frame now contains a variable COUNTRY containing the location of each respondent (either USA or Canada). This is stored in your environment as the c_sat_group data frame. Compare the loadings between the two models.

lavaan and dplyr have been loaded into your environment.

Fit the defined model c_sat_model to the c_sat_group data. We will include group = "COUNTRY" to model separate CFA's for respondents in the USA and Canada.
Summarize the results. Include fit measures and standardized estimates.
Compare the average standardized estimate for each item for both groups.

Exercise

Group CFA

Instructions