COMM 1225 Lecture Notes - Lecture 4: Summary Statistics, Sampling Error, Escadrille 103
27 Jan 2017
School
Department
Course
Professor
Document Summary
3 requirements to establish causation: covaration, temporal precedence, nonspuriousness, covariation. When variables co-vary, they change together, there is an existing relationship. Correlation does not imply causation, covariation with strong correlations can occur just by chance: temporal precedence. The cause (x) precedes the effect (y) in time. Ruling out possibility of reverse causation (y actually affects x) Watching tv doesn"t cause weight gain, people who are already overweight have no interesting in working out and therefore they watch tv: nonspuriousness. No plausible alternative explanation for relationship observered between. Ruling out potential confounders (a. k. a there could be a third variable that is accounting for changes in x or y) Objective produce comparable information/data by asking the right people the right questions at the right time. Margin of error chance: trickier to estimate/discover than sampling error, error due to some systematic (non-random) factors: i) ii) iii) Population the entire group of individuals that we want information about.
