商务统计-C8商务统计.pdf

Chapter 8 Sampling Methods and the Centra l Lim it Theorem GOALS 1.Explain why a sample is the only feasible way to learn about a populat ion. 2.Describe methods to se lect a sample 3.Define and construct a sampling d istribut ion of the sample mean. GOALS 4.Explain the central lim it theorem. 5.Use the central lim it theorem to find probabilit ies of se lect ing possible sample means from a spec ified populat ion. Why Sample the Populat ion? · To contact the who le populat ion would be t ime consum ing. · The cost of study ing all the items in a populat ion may be prohibit ive. · The physical impossibility of checking all items in the. Why Sample the Populat ion? · The destruct ive nature of certain tests. · The sample results are adequate. Probability Sampling · A probability sample is a sample se lected such that each item or person in the populat ion being stud ied has a known like lihood of being inc luded in the sample. Methods of Probability Sampling · S imple Random Sample: A sample formulated so that each item or person in the populat ion has the same chance of being inc luded. Methods of Probability Sampling · Systemat ic Random Sampling: The items or ind iv iduals of the populat ion are arranged in some order. A random start ing po int is se lected and then every kth member of the populat ion is se lected for the sample. Methods of Probability Sampling · Strat ified Random Sampling: A populat ion is first d iv ided into subgroups, called strata, and a sample is se lected from each stratum. · C luster Sampling: A populat ion is first d iv ided into primary units then samples are se lected from the primary units. Sampling Error A sampling error is the d ifference between a sample stat ist ic and its correspond ing populat ion parameter. Example: Sampling Distribut ion of t