If you're the type to eschew epistemic humility and present yourself as a "superstar" academic, or scientist-hero, you are probably also the type to have a relaxed attitude to good scientific practices like avoiding plagiarism. #academia
Robin Blythe
@rbly.bsky.social
Health economist dabbling in biostats and clinical informatics. https://orcid.org/0000-0002-3643-4332
It's the unseriousness of the whole thing that gets to me. What is produced as scientific research in one of the top business schools in the country. What they measure and what's being reported have nothing to do with each other. How do they not feel ashamed to circulate this report? (rhetorical)
Stephen Wild@stephenjwild.bsky.social · last wk.What
I'll never forget the first time I attended a #datascience seminar: the speaker had a huge dataset and ran a t-test on every combination of variables, pointed to each p<0.05, and said that was your starting point for a ML model or research paper
Demetri@phdemetri.bsky.social · last wk.New rule: Statistical analysis without a clearly stated research question gets you a paddlin'
Maturing is realising that most of your field's applied work is based off handwavey assumptions of 💫asymptotic equivalence💫 #econsky
The worst part is that, as you increasingly prove yourself as a capable investigator, you spend less and less time doing any actual analysis. By the time you reach a/prof, it's all grants and speaking engagements. Meat for the academic machine.
Robin Blythe@rbly.bsky.social · 2w agoWhen I made research faculty, I naively assumed I'd get to mostly do fun stuff - learning, exploring and analysing my own projects. Instead it's 95% project managing other people's grants: chasing data & IRBs, and dealing with angry primary investigators. Word of warning to any #postdoc out there!
When I made research faculty, I naively assumed I'd get to mostly do fun stuff - learning, exploring and analysing my own projects. Instead it's 95% project managing other people's grants: chasing data & IRBs, and dealing with angry primary investigators. Word of warning to any #postdoc out there!
Intense exercise in the tropics is an ongoing battle between your stomach and your heart as you try to cram as much liquid in as possible to stop your blood turning into an underseasoned ragu. At this point I should become a shareholder in Pocari Sweat with how much I've paid them
New record for me: one reviewer, 12 pages of comments. This is about twice as much as I had to address for my dissertation.
#Statsky I need your help. If you have a consulting gig or been involved in building a (stat. modeling) consulting center, would you be willing to share your business model, sample contracts/terms/conditions, your approach to recruiting clients... esp. in an academic setting, I'd appreciate it!
This is baby's first steps into the prediction-causality blood wars, but the more work I do in both, the less convinced I am there are real diffs. The last predictive model I built had a cross-val AUC = 0.96, and it was from a DAG + long discussion with clinicians about which Xs causally influence Y
Anyone have papers that they regret publishing? They might not be bad science, they could just be trite or useless (though some certainly might be bad). For the sake of the public record, it's good they're out there, but how do you handle no longer supporting your own work?
IMO the only thing holding people back from posting nuclear takes on LinkedIn is fear of getting filtered out for future jobs. Nobody in academia (afaik) is getting headhunted on LinkedIn though, so maybe we should just go for it. Mostly I'd probably just bitch about the state of science, really.
Nearly all research is, to some extent, exploratory. I don't buy that there is a clean line of separation between exploration and confirmation/hypothesis testing.
🥰 Always get warm fuzzies* when I see someone has cited my work, failing to read past the title and not realising it is directly contradicting the point they've casually tossed in the paper like a fart in a bubble bath 🥰
About once a month, for the last few years, I've been asked for a sample size in a prediction study and have been able to either point to pmsampsize or just run a quick sim myself. This package is such a blessed contribution to #rstats and #statsky @joieensor.bsky.social @richarddriley.bsky.social
pmsampsize: Sample Size for Development of a Prediction Model
Computes the minimum sample size required for the development of a new multivariable prediction model using the criteria proposed by Riley et al. (2018) <<a href="https://doi.org/10.1002%2Fsim.7992...
cran.r-project.org
IMO prereg is usually more useful heading off biased co-authors on straightforward analysis, rather than as an external quality signal. Rarely do I find myself with a clear analysis plan I can simply apply to data, making the main purpose of prereg weak to me. Lots of research is highly iterative.
Ingo Rohlfing@ingorohlfing.bsky.social · 4w agoArguing with economists: the case for preregistration #MetaScience kdoroc.substack.com/p/arguing-wi... Assuming this represents a broader picture in econ: I would have assumed that econ was more open to preregistration than other disciplines. This post suggests differently. 1/
#machinelearning people: What's the current best practice for predicting w/ longitudinal data when you've prior evidence of multilevel effects like random slopes and intercepts? An R implementation would be ideal! I dislike having to trade off testing possible interactions with ignoring slopes 🤔
Not sure I agree with the idea of trying to legislate how academic discussion should happen. Some of the best conferences I've been to were enlightening because they led to an exchange between a speaker and audience member. A better rule might just be 'don't be "that guy" when you ask questions'.
Georgia Tomova@georgiatomova.bsky.social · last mo.All conferences should adopt @epidbydesign.bsky.social ‘s question rules as standard! #SER2026
Seems as good a time as any to spruik the #RShiny app my doctoral supervisor and I made for the #rstats and #econsky Bayes-curious among you! Useful if looking to get prior distribution parameters from published literature to use in your Bayesian or decision analytic models: osf.io/preprints/os...
OSF
osf.io
Mattan S. Ben-Shachar@mattansb.msbstats.info · last mo.Apparently posteriors are sensitive to priors!* Specifying informative priors is a sign of a maturing field! I hope our new paper nudges psychology toward complex Bayesian models — based in formalized theories — and away from "default" priors that lack theoretical or practical grounding.
Gotta love it when you ask for a data dictionary, and the data dictionary is just the list of variable names you could have gotten with colnames() #rstats
As a half-Kiwi, half-Iranian native of Los Angeles, this WC match was made for me in an almost painfully poignant way. There was even a watch party at my local chelow kabab joint. Maybe the only time I've ever had FOMO from living overseas?
I suspect >95% of surprising study results and associated press releases are basically just explained by weighting/adjustment being inadequate (p < 0.05) #episky #econsky
Stephen Wild@stephenjwild.bsky.social · 2mo agoSynthetic diff-in-diff for us pendants 😉 It almost certainly has to do with how they defined their treated and control groups. I don't think the weights (either set) can properly adjust for the differences.
There's weirdly common debate in #NZ whenever policy's announced that deficit spending = bad. No consideration of whether it generates net economic value - see Labour's proposed fare cap for #publictransport. Where does this view come from? Investment in transport generates tons of societal value.
I am, once again, pleading with qualitative researchers to stop using participant quotes in your titles. Investigator bias, misrepresenting your data, and a complete lack of rigour all abound: www.tandfonline.com/doi/full/10....
Dont Care Cat
ALT: Dont Care Cat
static.klipy.com
Not sure I feel comfortable submitting to PLOS anymore either after this.
John Pickering@kiwiskinz.bsky.social · 2mo agoA while since I had a review request from PLOSOne so I looked at the Terms & Conditions. Wow they are broad! We own it all & can do what we want! Ugh. So, I'll no longer review for @plosone.org. A pity as I like their approach to emphasising good science over perceived utility. #Academia
I have seen my research with both lay and technical styles misinterpreted in citations and press releases. The main difference is that the technical pieces are just cited less overall, so I guess this kind of works? I think most people just read the title. Bad for the h-index grindset 😤 though.
William B. Fuckley@opinionhaver.bsky.social · 2mo agoAcademics could definitely write better overall, but I actually have kind of come around a little on dense technical jargon, because it serves as a useful pair of lead gloves in preventing normative preferences from irradiating your analytical clarity. Not foolproof, but some insulation.
Shanghai is a wonderful place. First time visiting China, well worth a return visit - maybe for a conference one day?
