Don't look at the neuroimagers doing this with thousands of correlated dimensions then 😬 Do you know of published work on the reliability of distance scores, given scale reliability? I've been simulating that for another project but not sure if there's a solid answer already.

"While we lack more recent data on site costs, other data suggests they are continuing to rise: a more recent study which measured total trial costs (not just site costs) found that phase 3 trial costs had reached $55,567 per patient by 2017." 🤯 www.clinicaltrialsabundance.blog/p/to-fix-tri...

Now published in Collabra! online.ucpress.edu/collabra/art... Using test data from TwinLife (N=9,980, Culture Fair Test CFT 20-R), we show that Regularised Prediction and Poststratification yields IQ scores that differ by up to 19 IQ points (mean 4.19 pts) from those reported in the CFT 20-R manual

A pattern that surprised me when I first heard about it (from @ruben.the100.ci): in high fertility contexts, a large share of kids is born to older mothers -- it takes time to have lots of kids! Not sure how I feel about "very late fertility (ages 35+)" 😭 www.niussp.org/fertility-an...

Tired: difference scores suck! Wired: distance scores suck! I remember learning about what we called "algebraic confounding" in BA research methods, and how uninteresting results like these are... And yeah, they're everywhere.

Some corners of psych love calcuating all sorts of metric to quantify things like "psychological distance." I am quite skeptical of many of them -- for example, Euclidean distance calculated on rating scale items will be confounded by the dimensions that go into its calculation. Check this out:

Oh, to publish in a Nature brand journal about how *checks notes* the big math changes to small math.

Things that I never thought I’d hear as often as I do: “No means no!” (from the 3yo in response to the request to help clean his room)

Just received an e-mail concerning the European Summer School on Eye Movements

You know what I'm gonna give this one to her. 10/10

But personally what has really bought it all back for me is having a new arch nemesis: @dingdingpeng.the100.ci 🤣🤣🤣

Taking a look at Castro-Schilo & Grimm about "residualized change vs diff scores" -- interesting case of how spelling out estimands affects reasoning. Estimand of interest: "effect of prior cohabitation on change in relationship satisfaction pre/post marriage"> journals.sagepub.com/doi/10.1177/...

When people ask me what my article's word count is

A few things I've learnt from writing >4,000 word pieces: - They tend to be more popular than my short pieces! - They're more definitive - like reference material, not a hot take. - Not everyone will reach the end, and that's okay. - Write so that the people who do find it incredibly satisfying.

Can you recommend papers in epidemiology/biomedicine that do causal inference on observational data and thoroughly and explicitly discuss their underlying causal model and assumptions ? @dingdingpeng.the100.ci #episky

Just spoke to a colleague who couldn't get their methodologically superior study published in top journals because it got basically the same effect estimates as vastly inferior methods

MFW I see authors resubmit their paper at a different journal without any edits after they have received extensive constructive feedback during peer review.

These captchas just keep getting harder #rstats

Coincidentally, my second paper ever (where I’m first author) was just published yesterday (!) and relates to the difficulties with pre-post differences. bsky.app/profile/vilg...

A very nice to initiative where you can post your DAG: opencausal.org And because they're machine readable, they're much easier to search.

Find someone who believes in you like researchers believe in pre-post differences as valid treatment effect estimates.

This undersells the fact that you would sometimes do zoom meetings during the kayak!

Two fossils diverged in a moonlike cave, And sorry I could not take both And be one traveler, long I gave And looked down one as far as I could To where it revived in its growth;

7

Great post on the emergence of clinical trials! This section makes me wonder about "professionalization" in science in general. I do feel like many parts of the research process in my field are...surprisingly dilettantish.>

Does anyone (*taps microphone* @dingdingpeng.the100.ci) have a blog post that clearly explains the dangers of over interpreting cross-sectional cohort research. Need a reader friendly response to 'people who drank green tea had lower rates of Alzheimer's, so drinking green tea can reduce your risk."

Them: "Latent class analysis is needed" Me:

Bsky friends: if you get only one reading for a grad seminar on ethics in data analysis and reporting, what do you choose? @drandreahoward.bsky.social @kevinmking.bsky.social @foswald.bsky.social @marcuscrede.bsky.social @dingdingpeng.the100.ci @bwroberts.bsky.social @tedmond.bsky.social

Just finished my talk on causal graphs as a simple yet effective reasoning tool -- unfortunately I had to join remotely because I caught a nasty cold 😷 Still had fun talking about DAGs, caffeine during pregnancy and hunter-gatherer longevity! You can find my slides here:

I mean, it's completely fine to say there's a 95% chance that a tibble will not behave as a data frame. (you can't deny that this was an epic example, incorporating baseR-tidyverse flamewar with frequentist-CI-interpretation-controversial)

Has Charité a problem with bot PhD supervisors? Or is finding all the cars the minimal requirement for supervision?

Relevant meme from our blog post on the topic: www.the100.ci/2024/12/05/w...

Ha! The original Lancet article on the dangers of reading in bed is here: doi.org/10.1016/S014...

TIL about Flamigo. But ranks 13 to 16 also seem kinda off?

Not sure what "entirely generated" means, exactly, but I feel we need new norms around this. Here's a blog I like about "etiquette", "consent", and what should be considered "rude" when sharing AI-generated content. Disclosures after the fact do not cut it, IMO. distantprovince.by/posts/its-ru...

I really like this chart: terminology for cause vs describe vs predict.

Hannah Ritchie has built a fun little tool where you can compare energy usage of various products and activities. This is super helpful imho, because it's so hard to develop intuitions even just about the scales involved here. hannahritchie.substack.com/p/does-that-...

Just received a mysterious e-mail inquiring whether the silkie bantam rooster is still available and to my disappointment, this apparently is not some sort of code word. Looks like there's another Julia Rohrer in a different part of Germany who is selling chicken.

Schaffe Schaffe Plotsle bauen

Housing theory of everything strikes again!

Clarity in communications

Just got a request to share data underlying two articles of mine. I presume this is some meta-scientific study and I’d love to share my data, except there literally aren’t any 🤨

TIL trees are the crabs of the plant kingdom 🌲🌴🌳

Can’t wait for somebody to work out the implications of this idea in a 5,000 word post on toothfairyeconomics in the appropriate subreddit.

Finished listening to this excellent episode. Great exercise in triangulation of different lines of evidence! So should you be taking statins? Maybe! Might make sense to get your cholersterol levels checked!

Have examples of social science papers that use DAGs to justify their controls? I find these very hard to come by & would like to use for teaching.

Just came across this in the wild and I am irrationally angry. Also, be wary if the magnitude of the coefficient in your logistic regression > |4|