iiinova — where your digital self keeps evolving

Malte Elson @malte.the100.ci · 21/09/2026

"Science is self-correcting"

Malte Elson @malte.the100.ci · 18/09/2026

No ground truth, no witnesses. Uh. Yeah. [Verse] Roll up on the dataset, don't standardize a thing, Raw units, mixed scales, let the big variable be king. Pick my k in advance 'cause I already know the story, Three clusters in the grant, so it's three, that's the glory. - Clusta Rhymes

3235

Malte Elson @malte.the100.ci · 01/08/2026

This blurb makes me wonder what the rest of the review said

240

Malte Elson @malte.the100.ci · 18/07/2026

My brain any time someone says hot deck imputation

061

Malte Elson @malte.the100.ci · 17/07/2026

The Ministry of Peace concerns itself with war, the Ministry of Truth with lies, the Ministry of Love with torture, the Ministry of Plenty with starvation, and the new KINDEROASE daycare center at TU München with---

272

Malte Elson @malte.the100.ci · 21/06/2026

5. Wording asymmetry. The Slovak gambling item used the Slovak term "hazardné hry" term plus an English "gambling" clarifier; the gaming item used only the Slovak "hranie," which the authors note is ambiguous between gaming and gambling.

100

Malte Elson @malte.the100.ci · 21/06/2026

4. Careless responding No CR checks reported (although the "gambling is gambling" item is labelled a "control item"). Slovak completion times are extreme: 6.3% finished in under 3 min (UK 0.3%), 7.0% took over an hour (UK 0.7%), slowest >11 days. Hard to square with genuine survey engagement.

120

Malte Elson @malte.the100.ci · 21/06/2026

1. Control item fails. Five items almost everyone should endorse (e.g. "offline poker for real money is gambling"). In the UK, 9.3% did not (strongly) agree with at leastone; in Slovakia, 49.8% did. 22.6% don't (strongly) agree "gambling is gambling," while 17.1% (agree "frisbee is gambling."

130

Malte Elson @malte.the100.ci · 20/06/2026

8. In Panel B of Figs 2 & 3, the bars for the "Human Average" and "LLM Average" appear larger than the individual-sample bars they're presumably computed from. We couldn't work out how that's possible, may be an error.

110

Malte Elson @malte.the100.ci · 20/06/2026

They may seem intelligent at first glance but then they get fundamentals of the universe wrong

140

Malte Elson @malte.the100.ci · 05/06/2026

The forensic report is available here: storage.courtlistener.com/recap/gov.us... Incredible stuff

44114

Malte Elson @malte.the100.ci · 25/05/2026

Well well well how the turn tables

0131

Malte Elson @malte.the100.ci · 21/05/2026

I'm sorry? Aren't you forgetting something?

150

Malte Elson @malte.the100.ci · 10/05/2026

The overconfident 8yo on the other hand, seeing gifts being exchanged

1110

Malte Elson @malte.the100.ci · 09/05/2026

Some research fields have a type of paper that basically asks: "What if a more expensive but less reliable machine can do [this] instead?"

02812

Malte Elson @malte.the100.ci · 27/04/2026

I should really not have a discretionary budget

4404

Malte Elson @malte.the100.ci · 20/03/2026

Shocking to think that telling people to imagine eating ice cream is not the same as having them eat ice cream. Don't get me wrong, it's good someone is doing these studies, but the wild claims based on vignettes never cease to amaze me.

2181

Malte Elson @malte.the100.ci · 16/03/2026

From a colleague, earlier today, during a department talk

191

Malte Elson @malte.the100.ci · 16/03/2026

Total scam. I wired Julia the advertised amount and all I got was this lousy post card

0101

Malte Elson @malte.the100.ci · 12/03/2026

Any sufficiently advanced file-drawer is indistinguishable from p-hacking

87413

Malte Elson @malte.the100.ci · 10/03/2026

When tests are used for decision-making, differences of this magnitude can lead to errors with potentially life-altering consequences (e.g., about intellectual disability)

050

Malte Elson @malte.the100.ci · 23/02/2026

We regret to inform you that your paper cannot be considered for publication, but we encourage you to submit it to our GOLD Open Access sister journal

14815

Malte Elson @malte.the100.ci · 19/02/2026

When a reviewer questions my moderated mediation model

1211

Malte Elson @malte.the100.ci · 14/02/2026

It does create awkward situations like this:

020

Malte Elson @malte.the100.ci · 03/02/2026

My wife when I catch her in our bedroom with another man

060

Malte Elson @malte.the100.ci · 03/02/2026

At low base rates, specificity works in extremely counterintuitive ways, as shown below. Generally, we think ML is going to help a lot with spotting paper mills and Scancar et al's classifier seems to work very well! BMJ Comment: www.bmj.com/content/392/... PubPeer: pubpeer.com/publications...

072

Malte Elson @malte.the100.ci · 28/01/2026

Fortunately, Conflict of Interest begins at $200,001

3233

Malte Elson @malte.the100.ci · 26/01/2026

We looked at 100 IGT studies and found abysmal reporting standards, and 244 different scoring techniques. Applied to a new dataset, the average correlation between these 244 techniques was 0. 5/15 IGT meta-analyses we looked at did not account for this hetereogenity.

1217

Malte Elson @malte.the100.ci · 25/01/2026

About the Go/No-Go task....

141

Malte Elson @malte.the100.ci · 16/01/2026

Some day, OSF's integration into my workflows as the go-to repository for research data will be seamless. But not today.

4203

Malte Elson @malte.the100.ci · 17/12/2025

While we are unable to offer publication in Nature Aging, we believe your manuscript may be suitable for our Gold Open Access sister journal.

1358

Malte Elson @malte.the100.ci · 10/12/2025

But then, why skip everything after the "might" step and go straight to asking for more funding for social psych? Clearly there never was any curiosity in learning whether LLMs can do this.

250

Malte Elson @malte.the100.ci · 02/12/2025

I think you just discovered a meme

5193