I am pretty sure this tweet is wrong on two or three levels. Can you find them all? But maybe I am wrong. Thoughts? https://twitter.com/davidshor/status/849316701862862851 …
-
This Tweet is unavailable.
-
Replying to @eigenrobot
Problems off the top of my head: 1. No causal identification. No causal identification. No causal identification. No c
1 reply 0 retweets 3 likes -
Replying to @eigenrobot
2. Even allowing causal interpretation for every coefficient, Female Educational Attainment's effect size is ~2.5 times larger than unions'
2 replies 0 retweets 1 like -
Replying to @eigenrobot
3. Permitting crimes against statistics and science and caring about this: collective bargaining's p > 0.05
1 reply 0 retweets 1 like -
Replying to @eigenrobot
4. I don't see a charitable interpretation of this travesty outside of me completely misunderstand that table.
1 reply 0 retweets 0 likes -
Replying to @eigenrobot
5. If not that, the author is either (i) grossly incompetent, or (ii) a hack of the worst kind, blackening the good name of Data. Ugh. /fin
2 replies 0 retweets 1 like -
Replying to @eigenrobot
yeah bad tweet. also, trying to imagine a story where higher top marginal rates causally increase gender gap. not easy.
1 reply 0 retweets 1 like -
Replying to @nickchk @eigenrobot
I wouldn't even be surprised if the story turned out to be true, given that unions tend to flatten wage dists. But that's not evidence.
1 reply 0 retweets 1 like -
Replying to @nickchk @eigenrobot
also it would require unions to not have a gender-segregating effect between occupations
1 reply 0 retweets 0 likes -
impression I've been getting in data sci interviews is econometrics emphasis on interpretation + modeling, even implicit, is a huge asset
1 reply 0 retweets 0 likes
this, especially for avoiding stupid pitfalls like the ones above, though also very much in feature engineering
-
-
Replying to @eigenrobot
Certainly makes sense! It's not like our stats curriculum is cutting-edge in other aspects
0 replies 0 retweets 1 likeThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.