STAT W21 Lecture Notes - Lecture 28: Randomized Experiment, Observational Study, Blind Experiment
Document Summary
How individuals come to be in the treatment group is important and distinguishes. Most reliable way to determine whether a treatment has an effect is through a controlled experiments from observational studies randomized experiment. Differences between the control group and the treatment group other than the treatment can be responsible for observed differences in outcome between the two groups. With historical controls, the control and treatment groups tend to differ in more did not receive the treatment ways than just the treatment, and confounding tends to be a problem. Cross-sectional comparison: comparing individuals of different ages at a single moment in time. Longitudinal comparison: following individuals over time as they age. What is true for the parts is not necessarily true for the whole. Experiment: if an investigator gets to decide who will receive the treatment. Need not use the method of comparison to isolate the effect of treatment using controls. If they do, they are controlled experiments.