Also have heard this called "Box-hacking", which is when someone deploys the Box quote to defend their questionable model because they don't want to change it or think about it

And you actually *should* care about the safety and autonomy of porn actors, even though I know they get the least compassion On Here.

Suppression and censorship of sex work is always, always dangerous for sex workers. They’re pros trying to do their job safely and deserve worker’s rights.

Real talk: our lab is at the cutting edge of tracking climate change-related deaths, and we have exactly $0 of external funding to do it. We're getting $0 from NIH and NSF for the next few years. If you're able to materially support our work, it keeps projects like HAL alive.

Aliquoting the feeling of Cuomo losing into a dozen little bottles to be opened at a later date when I need an energy boost

Either or!

Anyone aware of a good paper that shows confounding by indication in a DAG?

Thanks this is great!

Ok question for both of you as this is a problem I am actively thinking about for a kind of DAG experiment study I am working on - is there a good paper out there that lists all of the limitations of DAGs and what they can't show?

Does anyone have a good lay way of saying/effective communicating "data cut" when talking about updated health admin data

I feel like the politer version of this (but still rude) would be to use the "there are no routine statistical questions, only questionable statistical routines" quote

This has devolved into the nerdiest convo about modification

But at that point are you even doing science or just concepts of some science

This part I am less sure about predicting what scale should be impacted I feel like multiplicative is compounding risk at an indv lvl vs additive is about excess risk generated at at the pop level. Even though I think this vibes based idea is not correct since both scales are measured at pop level

So I was thinking the same thing re scale and actually think there is a possibility to actually strengthen your argument by hypothesizing the correct scale of effect modification especially if you ground it in theory a priori with a plausible mechanism

I mean methods arguments can get a little intense 🤷🏽‍♀️

Adding this to the list! On a quick scan it looks great and also wow what a flex to start a paper with the 42 Hitchhiker's quote *chefs kiss*

What about stratifying? Correct me if I am wrong but if iirc you can stratify by the moderator and look at unstrat est & comapre to stratified ests and can interpret 1) est in strat same as unstrat = no conf no mod 2) est is diff but same in strat = conf 3) est is diff & est diff within strat = mod

*spins*

*Lands on "Prassad"*

Fuck!

What are people's favourite tips and resources for writing a good peer-review!

a man in a suit and bow tie says " the problem ... is me " Alt: a man in a suit and bow tie says " the problem ... is me "

This is helpful but the issue the problem/bottle neck is me 🥲

#episky #academicsky does anyone have any good tips on wrapping up papers as first author and getting them submitted?

@amjepi.bsky.social @madhupai.bsky.social @raulpachecovega.bsky.social @colincarlson.bsky.social

I want to talk about epi in this space - but also shitpost, let's see how this goes