POL222H1 Lecture Notes - Lecture 7: Incumbent, Confounding, Dependent And Independent Variables
Document Summary
There is a number of important and interesting questions within political science which cannot be examined in experiments: in these questions, x" cannot be controlled or randomly assigned. The most feasible alternative is to conduct observational research/studies, where one actually observes x, y and z in the real world. Observational research/study: a research design in which the researcher has no control over the values of the independent variable. Therefore, in one"s observations, which concern x, y and z all at once, one should compare the average value of y across different values of x among observations with the same value of z. There exist two data types: individual survey data. Here, individuals randomly sampled from a population are studied. Here, the observations aggregated at a certain level (from city to country, school boards to electoral districts, etc. ) Example: canadian election study, 2011: over the past year, has canada"s economy : In the absence of z, there is a 17-point increase.
