BUSS1020 Lecture Notes - Lecture 7: Confidence Interval, Statistic, Stratified Sampling
Document Summary
Analysis easier+ more practical than for whole pop. The list of individuals in a population; e. g. maps, databases. If sections of pop are excluded in sampling frame, results may be biased. Judgement sample: refer each other experts based on skills: e. g. Items are chosen randomly using known probabilities that closely match those in a population: Simple random sample (srs): individuals are chosen from a pool where every indv has equal prob of being selected, can be selected with/without replacement, e. g. drawing names from a hat. Stratified sampling: so effective against bias (note: most effective method: cons: costly+ long sampling process because complicated. Cluster sampling: pros: cost effective, cons: need prior info on population+ less efficient+ cluster opinions may shift with time. Sampling distributions: distribution of pop selected from pop several times. Assumes that the sample is normally distributed. Used when too complicated to assess entire population but want to assess.