With regards to case variety, Gary King, Robert Keohane, and Sidney Verba alert from "picking over the dependent variable". They argue for instance that scientists simply cannot make legitimate causal inferences about war outbreaks by only checking out cases in which war did happen (th