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

The phrase "measured on a widely used and scientifically validated personality inventory" jumps out at me. What does that really mean?

Myers-Briggs is 'widely used' and proponents would probably argue it has been 'empirically validated' (by this one study no one can reproduce...). Myers-Briggs still lacks the "retest reliability" necessary for this sort of analysis.

In fact, in most personality tests, you'd expect wild swings in personality without drugs, even by just waiting a day.

Stop right here. Go watch Ben Goldacre on "Bad Science:" http://www.ted.com/talks/ben_goldacre_battling_bad_science.h...

Now, stop giving any attention to campy press release results. I don't even really disagree with the thesis here, but without a full description of the methods, and probably a metanalysis of previous work in this field, the 'canned results' buzzers should be ringing loudly in all of our heads.





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

Search: