# Comparing two proportions in XLSTAT

An Excel sheet with both the data and the results can be downloaded by clicking here.

We are going to compare the proportion of male and female Public Relation (PR) manager per earning category in the US (Source: US Census Bureau, Census 2000 special tabulation).

As can be seen in the dataset the distribution is very different between male and female.

The earning strata that seems to be the most similar between gender is a yearly salary of \$50,000 to \$74,999: 29.32% of the male and 25.00% of the female PR manager are earning this salary.

We are now going to test the following null hypothesis H0: There is no gender difference in the proportion of people earning \$50,000 to \$74,999.
The alternative hypothesis will be Ha: There is a bigger proportion of male PR manager earning \$50,000 to \$74,999 than female PR manager.

Go to the menu Parametric tests / Tests for two proportions.

Once the dialog box is opened, fill in the following information:

1. Frequency 1: 8990 (Female PR managers earning \$50,000 to \$74,999)
2. Sample size 1: 35955 (Total Female PR managers)
3. Frequency 2: 6619 (Male PR managers earning \$50,000 to \$74,999)
4. Sample size 2: 27340 (Total Male PR managers)

The Data format is Frequencies.

We select a z test rather than a Monte Carlo method as we do not need to use Monte Carlo simulation.

Leave the default value Sheet to display the results in a new sheet of the current workbook.

Move on to the next tab: Options.

For the Alternative hypothesis set up the option Proportion 1 –Proportion 2 < D where D is 0, as the proportion of female PR manager earning \$50,000 to \$74,999 is inferior to the same proportion for male.

Keep the significance level to 5%.

There are two diferent formula to compute the variance we will use the first one.

Click on OK.

## Interpreting the results of a test comparing two proportions

In the result sheet look at the value of the p-value. It is inferior to the limit of 0.05 therefore we should reject the hypothesis and conclude that the proportion of female PR manager earning \$50,000 to \$74,999 is significantly less than the proportion of male PR manager.

Click here for other tutorials.

Kovach Computing Services (KCS) was founded in 1993 by Dr. Warren Kovach. The company specializes in the development and marketing of inexpensive and easy-to-use statistical software for scientists, as well as in data analysis consulting.

Like us on  Facebook

### Get in Touch

• Email:
sales@kovcomp.com