Session Ready
Exercise

Model comparison with ANOVA

Comparing models can be difficult and many methods exist that are beyond the scope of this course such as model selection (e.g., AIC). For example, including too many predictors can cause models to be overfit.

One tool you are likely familiar with, Analysis of Variance (ANOVA), can be used to compare two different lmer models and test if one model explains more variability than the other model. Specifically, ANOVA can be used to test the amount of variability explained by lmer models.

If you wanted to see if Year is important for predicting Crime in Maryland, we can build a null model with only County as a random-effect and a year model that includes Year. You can then compare the two models using the anova() function.

If Year explains a significant amount of variability, then the P-value will be less than your pre-specified threshold (usually 0.05).

Instructions 1/2
undefined XP
  • 1
  • 2
  • Build a model that only includes County as a random-effect.
  • Build a model that includes Year2 as a random-effect slope.
  • Compare null_model (first) to year_model (second) using anova().