Statistical Sciences 2244A/B Chapter Notes - Chapter 6: Confounding, Observational Study, Time Point
Document Summary
Observation versus experiment: observational study: observes people but doesn"t attempt to influence responses, purpose is the compare existing groups, experiment: imposes treatment on people to record responses, purpose to see if treatment caused change in response. Lurking variables: the variable in the experiment not record and affected the response: observational studies may fail because of the explanatory and lurking variable are confounded (mixed) Sampling: population: information wanted in a group of people. Sample: part of population used to collect information: used to draw conclusions about population. Sampling design: how a sample is picked from a population: what population you want to describe, what do we want to record, not all studies are drawn directly from the population of interest. Everyone has an equal chance of getting selected: sample entire population. Cohorts and case-control studies: case-control: (random sample) group of individuals with a case is compared to a group of individuals with a control, applies to observational studies.