HWcorrections

Hypothesis Test for comparing 2 independent proportions

First, go to Stat, Basic Statistics, 2 proportions .
The following screen will appear.

Click on "Summarized Data".
Enter n in the column for trials and the number of "succcessess" under success.
Click on Options. The following box will appear.

To Compute a Hypothesis Test

Don't worry about the confidence level box.
Fill in the test mean in the box. Most likely this number will be equal to zero.
Chose the alternative that matches your alternative hypothesis.
Click on "Use pooled estimate of p for test".

To Compute a Confidence Interval or Bound.

Enter your level of confidence.
Do not worry about the blank for test mean
Next pick the alternative that matches your alternative hypothesis.

If you want a two sided test, select "not equal" as your alternative.

Click on "Use pooled estimate of p for test".

Note: Your screen may look a little different; however, you should have the same boxes to fill in for Confidence and Alternative.

Test and CI for Two Proportions

Sample      X      N Sample p
1           4     45 0.088889
2           4     45 0.088889

Estimate for p(1) - p(2): 0
95% CI for p(1) - p(2): (-0.117589, 0.117589)
Test for p(1) - p(2) = 0 (vs not = 0): Z = 0.00 P-Value = 1.000

* NOTE * The normal approximation may be inaccurate for small samples.