Checking for influential observations in logistic regression is the same as for MLR (Section 5.21). Fit the model, plot the Cook’s distances and DFBetas, and, if there are observations with extreme values, conduct a sensitivity analysis to see if their removal impacts your conclusions (Section 5.24).
::influenceIndexPlot(fit.ex6.3.int, vars = "Cook", carid=F, main = "Cook's distance")
# Compute DFBETAS <- dfbetas(fit.ex6.3.int) DFBETAS # To see the spelling of the terms # colnames(DFBETAS) # Index plot for each predictor # (results only shown for one plot) plot(DFBETAS[, "alc_agefirst"], ylab="AlcAge") abline(h = c(-0.2, 0.2), lty = 2)
Conclusion: There appear to be a few potentially influential points based on Cook’s Distance, and a few points that are influential for the main effect of age of first alcohol use.