MGMA01H3 Lecture Notes - Lecture 10: Cluster Sampling, Simple Random Sample, Stratified Sampling
Document Summary
It takes out some of the guess work involved in making marketing decision. It can supplement (not replace) good sense. Stage 2: develop plan and research design. How people react to a newspaper"s topic selection and space allocation. Choice of research design depends upon research objectives. Primary: data gathered for the specific problem at hand. Secondary: data that have been previously gathered for some purpose. Relatively inexpensive and fast, but inaccurate and ill-fitting the problem. Best suited for exploratory research e. g. , people meters, checkout scanners. Design of survey is critical - wording, ordering, response format. Determine content, wording, response format for each question. Simple random sampling: each element of population has an equal probability of being selected. Stratified sampling: entire population is divided into strata, and do simple random sampling from each stratum. Cluster sampling: entire population is divided into clusters each representing the population, and do random sampling of clusters. Information accuracy: sample representativeness, respondent error, administrative error.