I think obsessing about linear model-specific issues is a bad use of energy herehttps://twitter.com/toad_spotted/status/1296118945993416704?s=19 …
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Add this Tweet to your website by copying the code below. Learn more
Add this video to your website by copying the code below. Learn more
By embedding Twitter content in your website or app, you are agreeing to the Twitter Developer Agreement and Developer Policy.
| Country | Code | For customers of |
|---|---|---|
| United States | 40404 | (any) |
| Canada | 21212 | (any) |
| United Kingdom | 86444 | Vodafone, Orange, 3, O2 |
| Brazil | 40404 | Nextel, TIM |
| Haiti | 40404 | Digicel, Voila |
| Ireland | 51210 | Vodafone, O2 |
| India | 53000 | Bharti Airtel, Videocon, Reliance |
| Indonesia | 89887 | AXIS, 3, Telkomsel, Indosat, XL Axiata |
| Italy | 4880804 | Wind |
| 3424486444 | Vodafone | |
| » See SMS short codes for other countries | ||
This timeline is where you’ll spend most of your time, getting instant updates about what matters to you.
Hover over the profile pic and click the Following button to unfollow any account.
When you see a Tweet you love, tap the heart — it lets the person who wrote it know you shared the love.
The fastest way to share someone else’s Tweet with your followers is with a Retweet. Tap the icon to send it instantly.
Add your thoughts about any Tweet with a Reply. Find a topic you’re passionate about, and jump right in.
Get instant insight into what people are talking about now.
Follow more accounts to get instant updates about topics you care about.
See the latest conversations about any topic instantly.
Catch up instantly on the best stories happening as they unfold.
eigenrobot Retweeted Spotted Toad
I think obsessing about linear model-specific issues is a bad use of energy herehttps://twitter.com/toad_spotted/status/1296118945993416704?s=19 …
eigenrobot added,
Very open to it being BS, but if so I'd really like to know what problem causes the result! I imagine it's a useful story with lessons for other studies, whatever it is.
I think its failure is massively overdetermined unfortunately
if black patients have higher underlying/latent risk and white docs are assigned to higher risk patients , unless you measure that risk very accurately, white doctor/black patient outcomes will be worse, even "controlling" for risk, bc drawing from a higher risk distribution
The physician FEs should cover it if it's just white docs getting higher-risk patients in general, but if some docs see white lower-risk parents while also getting a more diverse set of higher-risk ones, that could create the interaction they found.
To me the biggest hurdles for any theory are the physician FEs, the fact the effect is specifically on the interaction, and the fact you need a theory where healthier blacks are more likely to have black docs, which is somewhat counterintuitive.
Is that counterintuitive? Wealthier people are more involved in physician selection, and it seems realistic for them to select doctors like themselves. Poorer people get whoever's around when they show up. (Whether this is empirically true is another matter).
Yeah, that's one of the most plausible explanations this endless branching thread has produced!
so at this point you've identified one reason the results are potentially borked and you havent even touched on model selection and specification, error estimation, or overfitting this is why I just dismiss this sort of thing out of hand
heh yes also data cleanliness
Need to write to Gelman and wait until it goes to the blog, heh!
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.