STAT 30100 Lecture Notes - Lecture 2: Influenza Vaccine, Statistical Unit, Confounding
Document Summary
How to generate an srs: collect data, anecdotal data: comes from stories about cases that do not necessarily represent a large group. Example: a friend tells you that she revived a flu shot and then got the flu. Can you conclude that flu shots don"t work: available data: were produced for some other purpose but many help answer the question of interest. How to get data: observational study: (survey) we observe individuals and measure variables of interest but do not attempt to influence the responses, designed experiment: we impose some experiment on individuals and observe the responses. Things to be careful about: confounding variables: occurs when the effect of two or more variables are related or mixed up. How to avoid: causation: the changes in the factors are the only cause of the changes in the response variables (outcomes). A good experiment should be able to minimize the effects of lurking variables.