Episode 22 - Bayes Rulez, Death to P-Hacking

The overall theme of the show today is applying what we learned last time. Bayesian Probability is good. P-Value hacking - is where we get into trouble. Death to P-hacking!

Links

We start by following up on my monologue from last week about Beliefs and Probability. This leads us into an article about the Reverend Thomas Bayes himself, and how his ideas transformed our thinking on probability and belief.

And finally, are most of the scientific studies that you hear about wrong? Worry about p-hacking, and think like a Bayesian, and maybe we won't be misled as often; let’s begin to attempt to figure out how to make sense of it all.

Thomas Bayes and the Crisis in Science by David Papineau

Beware the P! A blog post against p-values by mathematician, Foursquare engineer, and friend of the Local Maximum Stephanie Yang.

Beware those scientific studies, most are wrong, by Ivan Courronne. Interestingly (though not purposefully) we have two articles today that start with the word "Beware".

Washington Post article about Twitter removing fake accounts. We just touched on this at the end, but I want to do a more full discussion of this topic on a later date. Send in your questions and comments about this topic because we are open for business!

Here's a Google Books link to a Bayes article on Fluxions. This is his defense of the foundations of Calculus. Often times in mathematics (and in life) we find the formulas that work before exploring and understanding the underlying theory, and you can see this process at work here.

Now that's going to be tough to read, but a more recent book for a more general audience is Infinitesimal: How a Dangerous Mathematical Theory Shaped the Modern World by Amir Alexander. That book also has some interesting historical coverage of mathematics, including for example the creation of the modern (Gregorian) calendar.

We mention Forrest Gump and HBO's John Adams mini-series. Interestingly (though not purposefully), Tom Hanks was involved in both.

Previous Shows

We refer back to the original Episode 0 on Bayesian Probability and to Episode 1 where we applied Bayesian logic to the news.

I mentioned Stephanie Yang as the author of the blog post about the p-value house of horrors. Her interview was episode 3.

How data analysis becomes news stories was something we explored in episode 15 with Sarah Spagnolo.

And finally, we followed up on the philosophical underpinnings of beliefs and epistemology in episode 21.