groups

the groups to be compared as a list of estimated responses. The list must have 2 or more elements, and each group must have two or more observations.