SOC221H5 Lecture Notes - Lecture 7: Interrupted Time Series, Design Of Experiments, Natural Experiment
Document Summary
Experiment is common in natural science, but different in social sciences. To demonstrate causation, can be established through a controlled setting like a laboratory. Experiment is one of the few methods that can establish causality where the researcher can observe and measure causation. Easiest method for meeting the three conditions of causality: Internal validity: if there is sufficient evidence that a causal relationship exists between two or more variables. Refers to asking whether or not an experiment is solid to the point that it measures the actual causality (lower external validity) External validity: if the results of a study can be generalized beyond the specific research context in which they were generated. Laboratory experiment - high degree of control over the setting, artificial setting (lower external validity because it is a controlled environment) Research that takes place in an artificial setting over. Field experiment - natural setting; less control (higher external validity)
