POLI 210 Lecture Notes - Lecture 1: Spurious Relationship, Design Of Experiments, Random Assignment

• School: McGill University
• Department: Political Science
• Course: POLI 210
• Professor: Dietlind Stolle

Document Summary

Research design lecture september 20,2016 time order. Experimental design so advantageous to test causality: other influences/factors can be excluded random assignment prevents self-selection into treatment. Correlation is a minimum requirement for establishing causation. Problems in causal inference: observed relationships between a and b, assumption/theory: causality a b, spurious relationship: c a/b. The experimental conditions are called treatments and represents the independent variable of the study. Addressing the causality issue in empirical research: using the experimental logic for observational studies (quasi-experiment, using a variety of experimental designs, not just classic experiments.