| Violations of the equidispersion assumption can be of two kinds: overdispersion or underdispersion. |
| The primary Poisson assumption is equidispersion, i.e. the equality of the mean and the variance functions. |
| In single-level settings, a negative binomial model is often preferred as it relaxes the restrictive equidispersion assumption. |
| Named equidispersion, it should be investigated in the data exploration and disclosed in the model report. |
| To see whether there are violations to equidispersion assumption, we check whether the conditional mean is smaller than the conditional variance. |
| The generalized Poisson model is another model used to account for deviation from equidispersion in data to be analyzed. |
