COMM 1 Lecture Notes - Lecture 7: Simple Random Sample, Stratified Sampling, Nth Metal
Document Summary
Triangulation of measurement: use several different measures of one variable, then compare them: ex: different types of measures (facial fear, heart rate, and self-reported fear, or ex: differently phrased scales (yes/no scared vs how scared) How we select participants (or other units) for a study. Sample: a subset of the target population (who/what you want to report findings about: ex target populations: voters, fb users, married couples, juries, football fans, business owners; or tv shows, magazine ads, blog posts (content analysis) National poll n = ~1,000 +/-3: you want your margin error to be as small as possible, generally, the larger the sample size, the lower the margin of error. Simple random sampling: select elements randomly from population: listed populations: random #s table, phones: random digit dialing. Systematic random sampling: list of population, take random starting point, then select every nth element, until cycle through entire list, similar results as srs just a matter of procedural convenience.