Dear all,

I am estimating a linear regression with an interaction effect between a continuous and a metric variable. However, the case numbers at the lowest and highest end of the metric variable are very low and this partly different between groups (group A has few cases on the lower end of the distribution and Group B at the higher end).

I just wondered if there is a way to deal with this or to rule out that this shaping my results?

All help is greatly appreciated, thank you,

Evelyn