Description:

Can I use a t-test if my data are non-normal? What if the sample size is small? What's an exact test? And how many tests are there, anyway?

People working with real data are often confused about hypothesis testing and paralyzed by the number of tests and their requirements. But for many common problems, you don't need specialized tests at all.

In this talk, I present a framework for using simple simulations to estimate p-values. This framework works well for a wide range of problems, which is why I say that there's only one test.

You don't need prior knowledge of statistics (and you might be better off without it). I show programming examples in Python, but the talk is accessible to anyone with basic programming skills.

About Allen B. Downey

Allen B. Downey, author of Think Stats. Allen is a Professor of Computer Science at Olin College of Engineering. He has taught at Wellesley College, Colby College and U.C. Berkeley. He has a Ph.D. in Computer Science from U.C. Berkeley and Master's and Bachelor's degrees from MIT.