MSIT 3000 Lecture Notes - Lecture 2: Confidence Interval, Null Hypothesis, Statistical Hypothesis Testing

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

Document Summary

Create a confidence interval for p phat +- z square root of pq/n. Sample size calculations n= (z/me) pq use . 5 for p if no value is given. Set up the hypothesis for a test (use p, not p hat) Calculate the z test stat, or p-value (given the test stat) Given p-value and alpha, reach the correct conclusion (know the hypothesis) if p-value < alpha reject ho. Sufficient evidence for ha if p-value > alpha ftr ho. Pvalue = . 03, alpha =. 05. reject the null hypothesis. There is sufficient evidence to conclude the population proportion exceeds . 70. Confidence intervals for m. mean +-t(se) or mean +-t (s/square root of n) N=140, confidence level = 90% tcalc has df n-1 = 139. P(x> blank) = . 05 1-. 9/2=. 05 so t* = 1. 656. Hypothesis testing: set up hypothesis (use m, not mean) calculate the t test stat or the p-value (given the t test stat) example: