The assumptions of a two-way ANOVA are:
- Normality assumption: The distribution of the population must be normal, which means that it is symmetrical and bell-shaped. This assumption is often violated in many real-life situations like when the sample size is small. In such cases, transformations can be done to make the distribution normal again.
- Homogeneity assumption: The variances for both groups must be equal. If this assumption does not hold true, then we can perform an F test to see whether there is a significant difference between the two groups.