statistician409

statistician409 on Hacker News

Women Are Outnumbered in Online Debates

Posted by statistician409 on Tuesday January 06, 2015 @01:37PM
from the kristof.blogs.nytimes.com dept.

http://kristof.blogs.nytimes.com/2015/01/06/how-to-get-more-women-to-join-the-debate/

( Read More... | no comments | on HN )

Response to the Hacker News Comments on the Sexual Assault Model

Posted by statistician409 on Friday December 26, 2014 @09:34AM
from the obsessionwithregression.blogspot.com dept.

http://obsessionwithregression.blogspot.com/2014/12/it-is-hard-to-write-about-sexual.html

( Read More... | no comments | 4 points | on HN )

Re:How Likely Is Someone Accused of Assault by Multiple People to Be I... on Tuesday December 23, 2014 @04:10PM (#8790240)

(I'm the author of the post) the fact that accusations may not be independent is of course important (although I'd question the aptness of the above commenter's comparison to witch trials); I'd welcome your suggestions for incorporating correlations into the model. This is one of the reasons the tool discussed in the NYT's article could be powerful -- because it allows us to be more confident that accusations were levied independently, since people submit to the tool without knowing about other accusers.