XLSTAT: One Proportion Test in Excel

 

Dataset for Example

The data are the results of the launch of a coin 30 times. We would like to assess if the coin is biased or not so we want to compare the results to the proportion 0.5 so half of the time a tail, half of the time a head.

 

Setting Up

• Open XLSTAT.

• Click on the Parametric test menu and select the Tests for one proportion feature:

• In the General tab, select the Test Z option and fill in the following information:

 

In the Options tab:

• Choose the alternative hypothesis: Proportion - Test proportion different from D, since 19 is greater than the expected frequency of 15.

• Leave the default significance level at 5%.

• Set the Variance (confidence interval) option to Test proportion, as the goal is to compute the confidence interval around the expected proportion of 0.5 (this option appears only when using the Wald confidence interval).

• Click on the OK button to launch the computations.

• The results are displayed on a new Excel sheet.

 

Interpreting the Results

The result is that as the p-value for this test is 0.072 which is superior to the 0.05 limit, the null hypothesis should be accepted and the difference between the proportions is equal to 0. This means that the coin is not biased.