This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 11.2 Inferences About Two Means, Independent Samples 1) Two sample Z test (comparison of two means) When testing if two population means are equal ( 1 = 2 ) we use the two sample Z statistic: Z = 2 2 2 1 2 1 2 1 2 1 ) ( ) ( n n x x + note: both 1 and 2 are known Note: 1 2 is the hypothesized difference between the means, which is assumed to be zero for the TI 2sample Z test. This statistic has a standard normal distribution. TI: 2SampZTest PHStat: Two sample test for the mean both 1 and 2 known Example: 1 x = 84 2 x = 91 n 1 = 16 n 2 = 16 1 = 8 2 = 4 test: Ho: 1 = 2 Ha: 1 < 2 = .05 The above values yield a z = 3.1305 ( 1 x 2 x ) And a pval = 0.000873 2) Two Sample t test For the two sample t we use the statistic: t = 2 2 2 1 2 1 2 1 2 1 ) ( ) ( n s n s x x + This will have the standard t distribution if 1 = 2 Example: 1 x = 84 2 x = 91 n 1 = 16 n 2 = 16 s 1 = 8 s 2 = 4 test: Ho: 1 = 2 Ha: 1 < 2 = .05 TI: 2SamptTest...
View Full
Document
 Fall '10
 RaySievers
 Statistics

Click to edit the document details