POL222H1 Lecture Notes - Lecture 7: Democratic Peace Theory, Justin Trudeau, Confounding
Document Summary
Many interesting and important political science questions cannot be examined through experiments. The feasible second best solution is to use the actual observation of x, y, and z in the real world. Research design in which the researcher does not have control over the values of the independent variable. Divide observations according to the values of x and z. Compare the average value of y across observations with different values of x. Set value of z - account for confounding variable. Individual"s perception of economy --> individual"s approval of pm. Can test same theory using individual survey data. If an individual"s perception of the state of national economy improves, this voter is more likely to approve or favour the incumbent pm or vote for the incumbent pm. Surveys of randomly sampled data from canadian population. Compare average feelings across individuals with different perception of national economy and with or without college education.