### Chi Square

Using the Chi Square Function

The chi square function of EZAnalyze allows you to conduct a "non-parametric" statistical hypothesis test to see if observed frequencies differ from expected values significantly. To keep the EZAnalyze functions consistent, the chi square function requires that the data be "raw" data for analysis. This is different than how chi square is typically used, where you create a chi square table first. The good news is that EZAnalyze creates the chi square table for you from your raw data, so you can interpret the results exactly the same as you would with any other chi square calculator.

To conduct a chi square analysis, you need to identify two categorical variables. For example, if you wanted to know if the number of male and female nurses is significantly different than the number of male and female police officers, your categorical variables would be "gender" and "occupation," with each containing two categories (male/female, and nurse/police officer). This would create a "two by two" chi square table, and tell you if the observed differences in your data are statistically significant. If you asked the same question, but also included firefighters as another occupation, you would then have a two by three chi square table. In EZAnalyze, there is technically no limit to the number of rows and columns your chi square table can contain, but larger tables are difficult to interpret and often not useful unless you have a very large sample.

To conduct a chi square analysis, select "Advanced" from the EZAnalyze menu in Excel, and then choose "Chi square".

When the "Chi square " dialogue box appears, you will be presented with two lists of the variables in your data sheet. Select a variable from the list under "Rows variable" and a different variable from "Columns variable" to create your chi square table. It is usually better to have more rows than columns, but it is really a matter of personal preference!

When you click OK, a results report will be printed on a separate sheet for your review. (click on "results report" for information on how to interpret this analysis)

Go to Index of Help Topics