Generalized Linear Mixed Models can also handle within-subject tests.

GLMM is called mixed because we are mixing fixed and random effects.

• Random Effects: factors whose levels were sampled randomly from a large population, but whose specific level values we don’t care about. E.g. Subject is a random effect.
• Fixed Effects: factors of interest that we manipulate in a study.

As said, subject is a random effect, which, if included, makes the model a Mixed linear model. Including them allows us to correlate measures across the same subjects across different rows in the metatable.

Advantages

• we can handle missing data points
• better handle unbalanced designs
• no need for sphericity test one-way-repeated-measures-anova

Disadvantages

• computationally more intensive
• they retain larger DFres (Denominator degrees of freedom)

A nested effect comes into play when the levels of a factor shouldn’t be pooled just by their label alone. E.g. subjects is nested in the Keyboard factor and Posture factor.

References

• one-way-repeated-measures-anova
• anova-assumptions
• example-analysis-of-an-ab-test
• factorial-anova
• one-way-anova
• analysis-of-a-linear-mixed-model