STAT200 Lecture Notes - Lecture 8: Paired Difference Test, Null Hypothesis, Decision Table
4 Aug 2018
School
Department
Course
Professor
Document Summary
Independent random samples; large sample sizes (n1 and n2 >50 when p or q > . 10); ho: d. = 0 standard normal for comparisons with z; Pool the estimate of p based on the null hypothesis. Independent random samples; large sample sizes (n1 and n2 >50 when p or q > . 10); ho: d! 0 standard normal for comparisons with z: decision table for two proportions, use this table to help in the difference of proportions test. In the case where the variances are assumed to be equal: we calculate the pooled proportion based on information from two samples, and then we use that pooled proportion in the calculation of the, paired difference test. Standard 1 error: when you have a situation where we record a pre and post test for the same individual, we cannot treat the samples as independent. In these cases we can do a paired difference test.
