Hypothesis Testing The z-test for Proportions
Hypothesis Testing The z-test for Proportions
Hypothesis Testing The z-test for Proportions
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
2. Collect the data<br />
3. Compute the <strong>test</strong> statistic<br />
z =<br />
4. Make the Decision<br />
• Accept H0 if: − z ≤ z ≤ z<br />
• Reject H 0 if:<br />
x − μ0 x − μ<br />
n ≈ n<br />
σ s<br />
α / 2<br />
α / 2<br />
z < −z<br />
z > z<br />
0<br />
α / 2 or<br />
<strong>The</strong> one tailed <strong>test</strong><br />
α / 2<br />
To <strong>test</strong><br />
H0 : μ ≤ μ0<br />
(some specified value of μ)<br />
against<br />
HA : μ > μ0<br />
1. Use the <strong>test</strong> statistic<br />
z =<br />
x − μ0 x − μ<br />
n ≈ n<br />
σ s<br />
2. Use as the Acceptance and Critical Region<br />
• Accept H0 if: z ≤ zα<br />
/ 2<br />
• Reject H0 if: z > z<br />
Accept H 0<br />
0<br />
α / 2<br />
zα<br />
0<br />
α<br />
Reject H 0<br />
z<br />
<strong>The</strong> one tailed <strong>test</strong> – other direction<br />
To <strong>test</strong><br />
H0 : (some specified value of μ)<br />
against<br />
HA : 0 μ<br />
μ ≥ μ0<br />
μ <<br />
Acceptance and Critical Region<br />
• Accept H 0 if:<br />
• Reject H 0 if:<br />
Test Statistic<br />
α<br />
Reject H 0<br />
z =<br />
z ≥ −z<br />
z < −z<br />
α / 2<br />
α / 2<br />
x − μ0 x − μ<br />
n ≈ n<br />
σ s<br />
Expect z to be negative if H 0 is false<br />
Example:<br />
<strong>The</strong> Acceptance and Critical region:<br />
− zα<br />
Accept H 0<br />
We are interested in measuring the concentration of<br />
lead in water and we want to know if it exceeds the<br />
threshold level μ 0 = 10.0<br />
We take n = 40 one-litre samples measuring the<br />
concentration of lead.<br />
Statistical results:<br />
x<br />
= 12 . 1 and s = 1.<br />
2<br />
0<br />
0<br />
z<br />
8