Hacker News new | past | comments | ask | show | jobs | submit login

At least give me a diff-in-diff, guys.

To be fair, they acknowledge their limitations and the fact that they can't make a causal interpretation. (Bad hospitals -> nurses leave?) But why the hell design the study this way?




How would you design it otherwise? An interventional study would presumably be ideal, but probably impossible to achieve.




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

Search: