2-samplet-test-NorthCarolinaStateUniversity.docVIP

ST 361 Ch8 Testing Statistical Hypotheses: Testing Hypotheses about Means (§8.2-2) : Two-Sample t Test Topics: Hypothesis testing with population means One-sample problem: Testing for a Population mean Assume population SD is known: use a z test Assume population SD is not known: use a t test Two-sample problem: : Testing for 2 population means A Special Case: the Paired t test -------------------------------------------------------------------------------------------------------------------- TWO-sample problem: Testing for 2 population means Motivating Example Is there a difference between the life of batteries made by Duracell and Eveready? Let be the mean lifetime (in days) for Duracell batteries, and be that of Eveready batteries. Perform a 5% level of test. Duracell Eveready n (batteries) 8 10 41 45 Sample SDs 18 20 Step 1: Specify the hypotheses parameters of interest = and Step 2: significance level =0.05 Step 3: test statistic ???? Step 4: p-value Step 5: conclusion: If p-value , then we reject and draw conclusion according to Otherwise do not reject , and draw conclusion according to Calculating the Test Statistics for Testing Two Means Need: Data are Normal. We will focus on the case that and are unknown (and may not be the same) Test statistic is Then this test statistic will have a t-distribution with the following df: df = (round down!!!) Note that the test statistic should be consistent with your hypotheses. That is, if your hypotheses are stated in terms of , then the corresponding test statistic should be (Back to the battery example) Step 3: test statistic -0.45 (So we cannot use the normal table) Step 4: p-value = 0.694 Step 5: conclusion: Since p-value the significance level, we don’t reject Summary of the testing procedure for two population means: Hypotheses vs. (lower-tail test) (upper-tail test) (two-sided test) Significance level Test statistic With df =