POL SCI 3 Lecture Notes - Lecture 6: Milgram Experiment, Panel Data, Randomized Experiment
16 Sep 2016
School
Department
Course
Professor
Document Summary
Hypothesis x y (1) show x and y are correlated (2) convince not fluke (3) convince cause precedes effect (rules out y x) (4) convince not spurious. Remaining questions (1) causal mechanism (2) causal heterogeneity. Need to go through x w and w y to confirm. X gaining unexpected money (financial winfall) w tax preferences (want taxes to be lower) y favorability for republican party. X y, p the moderating/ interaction variable has a arrow pointing toward the causal arrow. P: designates subgroups of cases for cases for which you think the effect of x on y differs. For the correlation between x education and y income sex is p, the correlation is stronger for men and weaker for women. Effect can be positive for one group and negative for another group. Medicine can be positive and negative for different groups. X (watch debate or not) y (rating on trump)
