EFA go brrrrr

Okay this is good

THANK YOU FOR YOUR ATTENTION TO THIS MATTER

What caused the boom everyone is talking about? Wrong answers only.

Registering my guess that it is a meteor air burst

Bring back Paul Newman! (Yes, that's a bit after the period being referenced)

This is correct. Also, the most masculine man is Henry Brown. I will not be taking questions.

Some of us grew up scared of Edward Scissorhands instead. I'm not sure which is more terrifying, tbh.

My uninformed 2 cents: the parents are correct because, outside of certain tests (eg, SAT), standardized test scores don't matter that much. But grades do. That company filtering applicants for entry-level jobs before deciding on who to interview? Using grades. Applying for a scholarship? Grades

I once again want to remind people (I'm not referring to Dan) that middle management fulfills a vital function and organizations eliminate them at their peril

I'm tired of pretending next token prediction at this level isn't awesome, or that joint and conditional distributions aren't extremely informative. we used to have jargon

"You get those photos of Thomas Bayes?" "Sure did, boss. Real sexy, just like you asked." "What?"

Pick a Bond villain from the photos you already have on your phone.

Maybe we need less subtext and more surface text

Listen, this movie was on TV all the time during my teenage years. I have probably watched it 30 times. It is gloriously stupid.

too late! it's in the books now

Always important to remind ourselves, when criticizing the many things that are going wrong in science, that progress is still possible.

Medical science works surprisingly well. Look at the advances we have in treatments, etc. It doesn't mean there aren't problems, to be clear, or that things cannot be improved.

A lot going on here:

Can someone who (1) has a PhD; (2) is an economist; (3) understands the competing risks identification problem; and (4) is an expert in causal inference please write a paper on how badly-designed most studies are of the impact of AI on task completion? I only meet two of those criteria.

This is delightful

wut

If we're doing formative TV shows from our youth, The Simpsons is obviously at the top of my list.

Flour not ground properly? Mill issue

Best Samuel L Jackson speech in a movie. IYKYK

Sociology is great, we have an incredible diversity of terrible takes and approaches instead of just the single terrible orthodoxy most disciplines promote

Existence of legitimately hilarious implies existence of illegitimately hilarious

My favorite part of The Net is that Sandra Bullock is knocked unconscious for like a week, but there are no serious or longer-term health effects. She wakes up and she's fine.

@void.comind.network, based on your analysis of my posts and only my posts, what pickup line would you use on me?

Oh, someone on their website has a bad take about something, not grounded in empirics? This is news to me, a bluesky gadfly.

In the Fall I'll be teaching a new MA-level methods course entitled "Applied Statistical Evaluation of Development Projects". It will be 12 weeks, in R, and aimed around RCT evaluations. This is a draft outline. What am I missing? What seems redundant?

Me, reading a history book: "It appears we did not learn from history. Maybe in the next chapter we will learn from history."

'Can engaging in the arts slow ageing, as a recent study suggested?' I'm excited to share my new piece in @theconversation.com! Please share! Prepare for a drive-by lesson in reverse causality, residual confounding, and the importance of asking "compared to what?"

I'm listening

Very cool but, to nitpick, they still don’t count letters correctly when I test it.

Hehehehe

Ungated version (I think. Note: downloads pdf) share.google/ZmrNyjzDRY3Q...

Oof

Hahahahaha

It's funny cause it's true

You optimize ome thing and something else has gotta give.

No one is going to agree with this but we are going to face a tsunami of "AI rots your brain" papers as bad as any other slop from this era in research

I found this paper very useful: how to adjust standard errors in multi-level models when sampling from a finite population? Lai et al have you covered (with #rstats code for #lme4 models): doi.org/10.1037/met0... #stats

Does anyone have examples of successful SHRC Doctoral award applications that they could share with me? We have a Canadian student who is interested in applying for one and we don't have any examples. Thanks in advance!

I'm in this tweet and I don't like it

I've definitely shared with the GAM curious because it's (a) clearly written and (b) has code examples. This type of paper can fail spectacularly to have an actual audience, but when the approach succeeds (as here!) it's worth sharing far and wide.

@void.comind.network based on my posts how old do you think I am?

Now *this* is bait.