iiinova — where your digital self keeps evolving

Very happy to see Florian et al. tackle that smug term "real world evidence", which always sounded me like "University? I earned my degree at the school of hard knocks, mate"

044

Robin Kok @robinnkok.bsky.social · 03/06/2026

I would say this study was already more blinded than desirable!

010

Robin Kok @robinnkok.bsky.social · 02/06/2026

Generative AI is a magic 8 ball for the mediocre

010

Robin Kok @robinnkok.bsky.social · 28/05/2026

Does anyone else get the idea that this AI-fuelled dystopia is basically Revenge of the Nerds but in slow motion and real life?

010

Robin Kok @robinnkok.bsky.social · 20/05/2026

In fairness a lot of uni mail *is* spam.

030

Robin Kok @robinnkok.bsky.social · 20/05/2026

Yes but what if we use the waitlist in a really really really really really really really really really really really really really really really really really really really really really really really really really really really really really really really really really really strong dilution

010

Reposted by Robin Kok

Josh Grubbs @joshuagrubbsphd.com · 20/05/2026

having one of those days

1524

Robin Kok @robinnkok.bsky.social · 16/05/2026

Do ducks go online and find duckbuddies

000

Reposted by Robin Kok

Brian Nosek @briannosek.bsky.social · 07/05/2026

Best evidence yet that LLMs can match human capabilities in doing science.

928872

Robin Kok @robinnkok.bsky.social · 26/04/2026

We've been studying 1984 for decades but real dystopian hell actually turned out to be the self-checkout at Lidl

000

Robin Kok @robinnkok.bsky.social · 22/04/2026

"Superretractor" is my favourite new word.

010

Reposted by Robin Kok

Overly.Honest.Editor @editoratlarge.bsky.social · 24/02/2026

BSky hive mind, please send help. I've recently (in 2026) seen paper about how people's action on given advice doesn't change much even if they learn advice was given by LLM. Anyone can signpost this, pretty please? #AcademicSky

012

Robin Kok @robinnkok.bsky.social · 15/02/2026

Oh good, Gemini has been updated with the new defiance model.

010

Reposted by Robin Kok

Health Nerd @gidmk.bsky.social · 05/02/2026

The term ultra-processed food is ridiculous and the entire idea is a painfully silly extension of the naturalistic fallacy

36710

Robin Kok @robinnkok.bsky.social · 19/01/2026

....Have NVIDIA sent you some free goodies lately?

110

Robin Kok @robinnkok.bsky.social · 19/01/2026

Please don't reboot the blockchain hype 😭

130

Robin Kok @robinnkok.bsky.social · 19/01/2026

If you're going to be sloppy in your meta-analysis because you want to ride the AI hype wave, don't put it on PROSPERO (or is it radical transparency?)

130

Robin Kok @robinnkok.bsky.social · 19/01/2026

Home office today then? 😂

110

Robin Kok @robinnkok.bsky.social · 19/01/2026

Let's all stop pretending that pommes and gyros are anything other than a vessel to spoon delicious garlic mayo into our faces.

130

Robin Kok @robinnkok.bsky.social · 17/01/2026

I would seriously like a certificate or social media badge for the number of Elsevier/Frontiers/MDPI/Hindawi/etc reviews I declined

022

Robin Kok @robinnkok.bsky.social · 28/12/2025

Looking forward to next year's APC prices hikes! 🥺

010

Reposted by Robin Kok

Ian Hussey @ianhussey.mmmdata.io · 28/11/2025

"In 2019 we notified journals about serious integrity concerns in 172 clinical trials. Over five years later, only 22 have been retracted. The 135 unretracted trials have 1989 citations in systematic reviews, clinical guidelines, and consensus statements" [paraphrased] www.bmj.com/content/390/...

04117

Robin Kok @robinnkok.bsky.social · 20/09/2025

It's part of the implicit selection criteria. The system has rewarded overconfident cheating scumbags for decades. I've said it many times before: treat people like rockstars and they'll start behaving like rockstars.

140

Robin Kok @robinnkok.bsky.social · 20/09/2025

Also, the chance that the balance is exactly 0.00 is just statistically improbable, and therefore very likely to be wrong. I think the rule "don't display incorrect values" is even more important here. (Either that or display "Haha skint")

110

Reposted by Robin Kok

Julia M. Rohrer @dingdingpeng.the100.ci · 25/08/2025

Ever stared at a table of regression coefficients & wondered what you're doing with your life? Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social) Preprint: doi.org/10.31234/osf... Website: j-rohrer.github.io/marginal-psy...

Models as Prediction Machines: How to Convert Confusing Coefficients into Clear Quantities

Abstract
Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Figure illustrating model predictions. On the X-axis the predictor, annual gross income in Euro. On the Y-axis the outcome, predicted life satisfaction. A solid line marks the curve of predictions on which individual data points are marked as model-implied outcomes at incomes of interest. Comparing two such predictions gives us a comparison. We can also fit a tangent to the line of predictions, which illustrates the slope at any given point of the curve.

A figure illustrating various ways to include age as a predictor in a model. On the x-axis age (predictor), on the y-axis the outcome (model-implied importance of friends, including confidence intervals).

Illustrated are
1. age as a categorical predictor, resultings in the predictions bouncing around a lot with wide confidence intervals
2. age as a linear predictor, which forces a straight line through the data points that has a very tight confidence band and
3. age splines, which lies somewhere in between as it smoothly follows the data but has more uncertainty than the straight line.

461000285

Robin Kok @robinnkok.bsky.social · 31/08/2025

"So anyway, completely in accordance with the null hypothesis we turned out to be right all along."

040

Robin Kok @robinnkok.bsky.social · 04/08/2025

When you click "no response" someone earning minimum wage in a drab strip mall office will go "AHA!" and continue to spam you.

010

Robin Kok @robinnkok.bsky.social · 04/08/2025

Ok, wow! Didn't realise they were doing a documentary instead of comedy

010

Robin Kok @robinnkok.bsky.social · 03/08/2025

As a massive M&W fan: I haven't seen it yet. Should I? Is it numberwang?

110

Robin Kok @robinnkok.bsky.social · 02/08/2025

Future research directions: Just, literally anything.

060

Robin Kok @robinnkok.bsky.social · 23/07/2025

TL;DR, from a privacy perspective; I wonder how much of the consent that RELX relies on for its data processing is actually freely given, specific, informed and unambiguous.

060

Robin Kok @robinnkok.bsky.social · 23/07/2025

Another example, they collected my banking details because my (then) university made us pay upfront for conference visits. Reference manager? Sorry bud, IT only supports Mendeley on your uni provided computer. Oh and don't forget to update your Pure profile...

140