STA 261 Lecture Notes - Lecture 1: Random Assignment, Confounding, Statistical Unit
Document Summary
Observational studies can only reveal association or correlations. Retrospective study- conducting an analysis on the past. Identify subjects with a specific characteristic and then look into their history to find things that may be related. Ex: case control study (used in medical research) Prospective study- identifying subject"s inn advance and then collecting data as events unfold ex: cohort study. Sample- cross section of a popn at a given time also called cross-sectional study. Designed experiments- researcher"s impost treatments and controls. Also called randomized, comparative experiments requires a random assignment of subjects to treatments must identify at least 1 explanatory variable (factor) to manipulate and at least 1 response variable to measure key vocab. Subjects, participants, or experimental units: on who or what we experimented. Levels of a factor- the specific values the experimenter chooses for a factor. Treatment- a combo of specific levels from all the factors that an experimental unit receives.
