Which statistical test should I use?

Photo by Emily Morter on Unsplash

Visit Choose Your Test to answer a few questions to choose the appropriate test for your specific data and question. If you are interested in my process or reasoning in developing this, read on below.

I’m scouring the web for statistical flow charts and introduction to statistics books to get a sense for what questions need to be answered to effectively determine a correct statistical test regardless of the situation. One interesting approach to this is Decision-Based Learning, developed out of Brigham Young University. They have statistical software that essentially breaks down expert decision making into concrete steps or questions for novice learners to answer. Then, if you are struggling to answer a particular question, it provides just-in-time material to fill in knowledge gaps to help you answer the question.

While I love their approach, I think it is still a little bit too complicated for what I want to do with statstest.com. StatsTest.com should be easily accessible by anyone with data wanting to do some statistical comparisons. I especially don’t want to bog down practitioners with fancy statistical jargon when they don’t need to know the statistical jargon to carry out an appropriate statistical test. I know far too many companies that are STILL relying on intuition, even in the information age, and I think it is because they have data but they don’t have the data literacy to use it in a disciplined and data-driven way.

The first purpose of statstest.com is to remove the complicated statistics barrier to enable practitioners, novice learners, and anyone to correctly choose the appropriate statistical test for their data and understand where to go next. Eventually, I think the purpose of statstest.com will be to allow people to run their statistical tests in the browser by simply uploading a spreadsheet of their data, but I’m saving that for v2.0.

Leave a Comment

Your email address will not be published. Required fields are marked *