Comparing CFA models using ANOVA

So far throughout the course we have been loading manifest variables to one and only one factor. This is not, however, a requirement of CFA. A manifest variable can, for example, load to two factors. The previous lessons provided you with several ways to compare model fits. Now let's look at a special case of model comparison -- the nested model.

A nested model uses the same variables and patterns as another model but specifies at least one other parameter. In this special case, we can use analysis of variance (ANOVA) to test for a statistically significant difference in model fit. Let's create a nested model in lavaan and compare models using anova().

lavaan has been loaded into your environment.

Print this model definition c_sat_model_a passing it to the cat() function.
Define a second model c_sat_model_b. This will be the same as c_sat_model_a except we will reflect the item EU1 on both the CSU and EU models.

Exercise

Comparing CFA models using ANOVA

Instructions 1/3