Cooks distance glm in r is a measure of the influence of each observation on the fit of a generalized linear model (glm). It is calculated as the change in the deviance of the model when the observation is omitted, divided by the residual degrees of freedom. Cooks distance can be used to identify influential observations that may be affecting the fit of the model.
Cooks distance is a useful tool for identifying influential observations in a glm. However, it is important to note that it is not a measure of the importance of an observation. An influential observation may not be important, and vice versa.