POLI 30 Lecture Notes - Lecture 6: Causal Inference, Confounding, Dependent And Independent Variables
Document Summary
From our theory we"ve derived a hypothesis. We look at the data to see if there"s a pattern. Neyman/ rubin: a causal effect is the difference between what happened and what would have happened. We can"t prove causality without a time machine. This is also about making sure that x causes y, not the other. Students decide to study or not before they get their final grade. Gender and socialization happen before attitudes about gun control. Two variables are correlated when changes in one variable occur together with. Example: if x goes up, y goes up. Parliamentary regimes last longer than presidential regimes. You have to be able to tell a plausible story that connects the iv to the dv. Not everyone agrees that this is necessary. Parliamentary systems are less likely to fail because executive-legislative conflict will cause immediate elections. Democracies fight less because of audience costs.