BUSS1020 Lecture Notes - Lecture 9: Statistical Significance, Non-Sampling Error, Central Limit Theorem

• School: University of Sydney
• Department: Business Studies
• Course: BUSS1020
• Professor: All

Document Summary

Buss1020 lecture 9 hypothesis testing: one sample. A claim, often about a population parameter (population mean, population proportion) States a default or status quo claim or assertion (preselected option) E. g. average diameter of a manufactured bolt is 30mm h0: = 30. Always about a population parameter, not a sample statistic. Tests usually begin by assuming null hypothesis is true: similar to notion of innocent until (cid:862)p(cid:396)o(cid:448)e(cid:374)(cid:863) guilt(cid:455) Refe(cid:396) to (cid:862)status (cid:395)uo(cid:863) / histo(cid:396)ical (cid:448)alue / a (cid:396)ele(cid:448)a(cid:374)t (cid:448)alue to the test. May / may not be rejected in the test. Cannot be proven (but can be rejected) by the test. E. g. average diameter of a manufactured bolt is not equal to 30mm (h1: 30) The hypothesis that researcher is trying to find evidence for / against / most interested in. Then sample the population & find sample mean. Suppose sample mean age was (cid:454) = 20. Much lower than claimed mean population age of 50.