I designed this program to perform statistical significance testing (2-prop t-tests with equal variance) against 2 to 50 groups and is ideal for anyone who needs to complete large amounts of testing.

If you have been using a spreadsheet to calculate significance across multiple groups, one against each other, this will save you considerable time. Consider having to compare 20 groups each against the other; if you are using a program that only compares 2 at a time, then you will need to repeat the calculations 190 times (19+18+...+1). This program does the work for you.

I hope you find it useful and please feel free to email me with any comments or feedback.

Stat Testing Program Example Video



Frequently Asked Questions

What significance test is used in the calculations?
The calculations are t-tests for proportions and assume equal variance. This article in Wikipedia does a nice job of explaining the test, assumptions, and calculations.

I am only able to do 15 groups at a time instead of the 50 that you said. Why is this?
Unfortunately, you need to have the new version of Microsoft Excel (2007) to be able to run more than 15 groups at a time. If you do have Excel 2007, then what you need to do is download the file, and then go to save as - macro enabled workbook (.xlsm) then close out of excel and after you re-open it, then it should work for up to 50 groups. The reason for this is because Excel 2003 only has 256 possible columns while 2007 has 16,634, and the way I designed the program calls for lots of columns the more groups you have.

I think a problem occurred with either the setup or the equations.
Most of my experience with individuals encountering issues with the macro is because their regional language setting is not set to US English. For some reason other language settings often present problems, even UK English. I do not know of an easy way around this, but I would recommend downloading the pre-setup files above because those should work. Please contact me if you do encounter a problem, and I will certainly try to help.

The output does not match up to a set of tabs or another calculation program I have, what's going on?
Small differences are normal but if you notice big differences, please send me an email (will@wrightw.com). From my experience the reason for the small differences is because tab programs usually use tests that match the data better by doing such things as actually calculating variance instead of assuming equality. I would always recommend using tab significance in the event of a discrepancy.

How can I contact you?
Please email me at will@wrightw.com. I try to respond to all emails within one business day.