Hacker News .hnnew | past | comments | ask | show | jobs | submitlogin

I found this to be a great example of the flaws with "evidence based" reporting rather than "science based"

Science would suggest that this result is preliminary, and results should be studied with Bayesian statistics. This article makes headlines, sure, but shouldn't on its own dictate any behavior.

PS. I found the false dichotomy between things like "moving for a person with crutches" and "being good at your job" annoying. This was likely more a product of reporting than the study though.



It seems more like the difference between "reporting" and "science" to me. If you're interested in the science, consider reading the linked research paper instead[0]. It may adopt something closer to the tone you're looking for.

FWIW, I agree that Bayesian statistics strikes me as regrettably underused in modern social science. But it's my impression that scientists of many stripes continue to use classical statistics instead, so calling something that doesn't use Bayes "not science" seems a bit off.

[0] http://www.socialemotions.org/page5/files/Condon.etal.2013.p...


The article made no pretense at being journalistic reporting. It's published in the Op-Ed section, after all.




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

Search: