STAT 245 Lecture Notes - Lecture 27: Simple Random Sample, Cluster Sampling, Statistical Parameter
Document Summary
Parameters numerical descriptive measures for a population. Statistics numerical descriptive measures for a sample. Often population parameters are unknown so we use sample statistics as estimates for parameters. For example, we use to determine the probable interval values that can possibly be. There must be no interference as to which objects should be included or not included in the sample. Simple random sampling (srs) allows each possible sample of size n an equal probability of being selected. Example: take a simple random sample of 5 students from a class of 89 students. Stratified random sampling: divides the population into subpopulations or strata and then selects an srs from each stratum. Example: divide canada into provinces and territories and take a simple random sample of residents within each province and territory. Cluster sampling: divides the population into subgroups called clusters; selects an srs of clusters, and takes a census of every element in the cluster.