2 thoughts on “Dr. Steven Novella on Bayes’ Theorem & Pseudoscience

  1. Very hard to follow, this statistical analysis stuff. All I can say is the critiques of Bem seem to be more persuasive than the responses to the critiques. That’s not saying I know who’s right.

    My own reaction is that the tests for psi seem to be pretty piss-poor: lots and lots and lots of wrong answers but because hits seem marginally better than chance, it’s a positive? Come on. Run trails against chance but also set up control tests where the results are 100% — but subjects don’t know that — and see what happens.


  2. Bayes is perfect for this type of thing. If the prior probability is small (part of the numerator) and the probability that the hypothesis is not true is giant (part of the denominator), you end up with an extremely minute posterior probability. In other words, when the thing being proposed is “probably not true” then this analysis gives an appropriately small chance of it being true.


Commenting below. No spam or trolling, or my cats will be angry.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s