4 ms·

Btw, in areas where it's OK to apply math (like medicine or spam filtering) we have a science which tells us how to determine posterior probability given prior

by gleb 11y ago

Btw, in areas where it's OK to apply math (like medicine or spam filtering) we have a science which tells us how to determine posterior probability given prior and new information. It's called statistics and Bayes rule.

bonobo3000 11y ago
Hahaha yes, absolutely :) This is another thing that makes stereotypes tricky - it takes a little bit of statistical insight to see how they can be both right in some sense but completely wrong in the other, but statistics is very unintuitive, and math doesn't grab headlines. So that kind of argument is difficult to use with non-nerds, but ultimately i think its the only accurate one. That brings up another good point in that its really a perfect opportunity for media outlets that like to turn controversy into clicks. Its very controversial and very broadly applicable.
j_m_b 11y ago
Stats can also be used to obfuscate and mislead. P-hacking (mentioned in the article) is the most well known.
jeremyjh 11y ago
A challenge - I'm sure you know - is that a study found that doctors seldom apply Bayes theorem more accurately than they'd do by rolling dice. http://slatestarcodex.com/2013/12/17/statistical-literacy-among-doctors-now-lower-than-chance/ http://slatestarcodex.com/2013/12/17/statistical-literacy-am...
yummyfajitas 11y ago
The more academic paper I linked to (by the author being discussed in this article) suggests that people use stereotypes a lot like a Bayesian prior. Once they have knowledge about a specific individual, that knowledge usually drowns out the stereotype. http://www.rci.rutgers.edu/~jussim/unbearable%20accuracy%20of%20stereotypes.pdf http://www.rci.rutgers.edu/~jussim/unbearable%20accuracy%20o...
- gleb 11y ago
  The part that I found particularly interesting is towards the end of this article. That using stereotypes [as priors] makes people more accurate. And ignoring stereotypes makes people less accurate.

by gleb 11y ago

bonobo3000 11y ago

Hahaha yes, absolutely :) This is another thing that makes stereotypes tricky - it takes a little bit of statistical insight to see how they can be both right in some sense but completely wrong in the other, but statistics is very unintuitive, and math doesn't grab headlines. So that kind of argument is difficult to use with non-nerds, but ultimately i think its the only accurate one. That brings up another good point in that its really a perfect opportunity for media outlets that like to turn controversy into clicks. Its very controversial and very broadly applicable.

j_m_b 11y ago

Stats can also be used to obfuscate and mislead. P-hacking (mentioned in the article) is the most well known.

jeremyjh 11y ago

A challenge - I'm sure you know - is that a study found that doctors seldom apply Bayes theorem more accurately than they'd do by rolling dice. http://slatestarcodex.com/2013/12/17/statistical-literacy-among-doctors-now-lower-than-chance/ http://slatestarcodex.com/2013/12/17/statistical-literacy-am...

yummyfajitas 11y ago

The more academic paper I linked to (by the author being discussed in this article) suggests that people use stereotypes a lot like a Bayesian prior. Once they have knowledge about a specific individual, that knowledge usually drowns out the stereotype. http://www.rci.rutgers.edu/~jussim/unbearable%20accuracy%20of%20stereotypes.pdf http://www.rci.rutgers.edu/~jussim/unbearable%20accuracy%20o...

gleb 11y ago
The part that I found particularly interesting is towards the end of this article. That using stereotypes [as priors] makes people more accurate. And ignoring stereotypes makes people less accurate.