I'm using lme4 for mixed effects linear regression, and I'd like to compare two models using anova(). However, I'm perplexed by the Df column. If I have two nested models, where the second has more independent variables than the first, how can the second model have more degrees of freedom?
One example can be found at this question: How to interpret ANOVA output when comparing two nested mixed-effect models?
Another example (sorry about the formatting)
Df AIC BIC logLik deviance Chisq Chi Df Pr(>Chisq)
aov_models$base 7 11862 11907 -5924.2 11848
aov_models$augmented 13 11786 11869 -5880.1 11760 88.113 6 < 2.2e-16 ***
