POL222H1 Chapter Notes - Chapter 3: External Validity, Internal Validity, Field Experiment
Document Summary
4. 1: comparison as the key to establishing causal. Use different research designs to conclude true / false. Experiment: research design in which researcher both controls and randomly assigns values of independent variables to participants. Ensures that comparison is as pure as possible. Internal validity: research design = high confidence in causal levels. Random assignment: participants assign randomly to x values. Random sampling: how to select cases for inclusion in study (everyone in population has equal chance of being selected) Survey experiment: random assignment and sample --> internal and external validity. Natural: arise as if random assignment has occurred. Cannot control every individual variable (especially in social sciences) Observational studies are not experimental but emulate them. Cross-sectional observation: variation across spatial units at a single time. Time series observation: variation within single spatial unit over many time units. Spatial and time units that define data sets - date set dimensions.
