This preview has intentionally blurred sections. Sign up to view the full version.
Unformatted text preview: H o : μ x = μ y H a : μ x <> μ y We then proceed to calculate our test statistics z . Because in the null hypothesis μ x = μ y , μ x - μ y = 0. Thus this term goes away, leaving ( 29 ( 29 80 . 1 140 45 . 100 32 . ) 10 . 5 19 . 5 ( 2 2 = +-= z We now proceed with the formula for the two-tail test from fig.8.7 in Devore Phi(abs(z)) = Phi(1.80) = 0.964 from the z-table Then p = 2[1 – 0.964] = 0.072 = 7.2% Interpretation: the lowest possible alpha for rejecting H o is 7.2%. If the test were given at the 10% level, we would reject H o , but if it were at the 5% level, we would accept it....
View Full Document
- Fall '08
- Null hypothesis, lowest possible α-value, +Y nX nY