STAT 110 Lecture Notes - Lecture 4: Sampling Frame, Errors And Residuals
Document Summary
There are two types of sampling errors that can occur in surveys: random sampling error- the natural error that arises when sampling from a population, nonsampling error. The margin of error in a confidence statement includes only random sampling error. Taking a larger sample reduces the margin of error but it will not eliminate random sampling error completely. Undercoverage- occurs when some groups in the population are left out of the process of choosing the sample. Nonsampling errors- errors that arise from sources not related to the act of sampling. Examples: data entry errors, nonresponse, poorly worded questions, interviewer bias, response error. Nonresponse- the failure to obtain data from an individual select for a sample. Response error- occurs when individuals do not answer truthfully. A stratified random sample arises in two steps: 1. divide the sampling frame into groups of individuals, called strata: take a srs from each stratum and combine each srs to form the complete sample.