Wednesday, August 09, 2017

Selection Bias

Psychiatrist Scott Alexander over at Slate Star Codex, in plugging Frederick DeBoer's essay on the subject, makes the following important observation:
If an educational program shows amazing results, and there’s any possible way it’s selection bias – then it’s selection bias.
I recommend both articles. 

2 comments:

RichardJohnson said...

In a similar vein, courtesy of Maggie's Farm: Is it fair to say that most social programmes don’t work?

Assistant Village Idiot said...

My oft-repeated cliche is that the first pass of most social programs picks off the low-hanging fruit and is probably effective. After that, it's more dollars, effort, and aggravation chasing less and less result.

Great article. A hard-nosed approach to effective altruism, recognising the difficulties and not being swayed by pictures of puppies and kitties, would be valuable.