what you described is prevalent in social sciences. Ton of biases and causation/... | Hacker News

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

curiousjorge on July 16, 2015 | parent | context | favorite | on: How to learn data science

what you described is prevalent in social sciences. Ton of biases and causation/correlation error and putting blind trust in some arcane statistical analysis without knowing what they really mean. Conclusion: statistically significant is the magic word peppered throughout academic literature.

achompas on July 16, 2015 | [–]

Agreed. Fortunately, excellent social scientists really care about this -- see Andy Gelman's blog for many rants on this topic.

Balgair on July 16, 2015 | [–]

I'll echo for biology and sports sciences (think moneyball).

http://www.wired.com/2009/09/fmrisalmon/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182327/

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