MSIT 3000 Lecture Notes - Lecture 1: Simple Random Sample, Stratified Sampling, Cluster Sampling

• School: University of Georgia
• Department: Management Sciences and Information Technology
• Course: MSIT 3000
• Professor: Steve Morris

Document Summary

Generalizing from the data at hand to the world at large is something that market researchers, investors, and pollsters do every day. To do it wisely, they need three fundamental ideas. We"ll discuss each idea separately, but the ideas are: examine a part of the whole, randomize, the sample size is what matters. Part of the whole (population, sample, bias, random selection) population: all individuals you wish to study sample: subset of the population idea: use the sample to make conclusions about the entire population. This can be done as long as bias is avoided. The characteristics of a biased sample do not match the characteristics of the entire population. Best way to reduce bias is randomization. Sampling variability and sampling error are pretty much the same. Draw a sample of 50 students and you find the average gpa is 3. 04. The ratio of the sample size to the population size doesn"t matter.