Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

That's generally not how things go. I don't know the details of this exact thing, but generally there will be two "very yes" responses, a "kinda yes", a "kinda no", and to "very no". You throw out all the "kinda"s and pay attention to the "very"s. 25% "very no" is a serious black eye, it's probably like 60% "kinda"s.


The breakdown is here: http://www.pewinternet.org/2019/01/16/facebook-algorithms-an...

13% "Very accurately"

46% "Somewhat accurately"

22% "Not very accurately"

5% "Not at all accurately"

(and 3% Refused, 11% Not assigned categories)


As someone who studied some formal logic and some psychology, I have to say that providing the "Not very accurately" option is both bad logic and bad research methodology.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: