iiinova — where your digital self keeps evolving

Reposted by Hauke Roggenkamp

Sharon Machlis @smachlis.bsky.social · 17/09/2026

Quarto 2 will be 100 times faster than the current edition. You can try an early beta, but as of now it only does HTML output and can't run code yet. It'll be "about a year of work" before Posit advises users to move to Q2, @cscheid.net told #PositConf2026. github.com/quarto-dev/q2 #QuartoPub

github.com

GitHub - quarto-dev/q2: q2 is the experimental implementation of Quarto 2 in Rust.

q2 is the experimental implementation of Quarto 2 in Rust. - quarto-dev/q2

1274

Reposted by Hauke Roggenkamp

Institute for Replication @i4replication.bsky.social · 08/09/2026

1/ What is the prevalence of personally identifiable information (PII) in author-collected data publicly shared in the social sciences? We audited 327 replication packages from 11 leading journals in economics, political science, and psychology Answer: surprisingly high 🧵

25537

Reposted by Hauke Roggenkamp

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 04/09/2026

It's a systemic problem in science reporting that these summaries exaggerate the evidence. In the actual candy study, the participants were all students, status was manipulated in the lab by telling some students they were better than others, and "nearly twice" means 1 piece vs 2 pieces.

Piff and his colleagues found that upper-class drivers were more likely to cut other cars off at junctions and ignore pedestrians at crossings. Wealthier people also appeared more willing to lie at work. In one experiment, volunteers were left alone with a jar of sweets ostensibly meant for children. When the researchers returned, upper-class people had snaffled nearly twice as many candies as lower-class people.

1329166

Reposted by Hauke Roggenkamp

Ana Martinovici @anamartinovici.bsky.social · 31/08/2026

I would say anyone who is a co-author, regardless of how Ariely was involved in the project.

011

Hauke Roggenkamp @hroggenkamp.bsky.social · 01/09/2026

Two years ago @mvreiss.bsky.social and I reproduced a paper for @i4replication.bsky.social's mass reproduction project. It reproduced cleanly, and we still came away seeing how fragile these pipelines can be: a hard-coded path, an updated package, an undocumented manual step. Hence this Comment.

1105

Reposted by Hauke Roggenkamp

Andrew Gelman et al. @statmodeling.bsky.social · 31/08/2026

I am shocked, shocked to learn that this 2002 paper by Dan Ariely is based on dubious data and doesn’t replicate. statmodeling.stat.columbia.edu/2026/08/31/s...

statmodeling.stat.columbia.edu

I am shocked, shocked to learn that this 2002 paper by Dan Ariely is based on dubious data and doesn’t replicate. | Statistical Modeling, Causal Inference, and Social Science

26421

Reposted by Hauke Roggenkamp

Beth Clarke @bethclarke.bsky.social · 21/08/2026

I’m hiring a PhD student to work with me on metascience in psychology at @zpid.bsky.social! The position is very open - could cover a wide range of topics/themes 🤸🏻‍♀️✨ If you have questions, please reach out! Applications are open until October 1st. leibniz-psychology.onlyfy.jobs/en/job/fyjwt...

leibniz-psychology.onlyfy.jobs

PhD in Psychological Meta-Science (m/f/d)

Become part of the open science movement The Leibniz Institute for Psychology Information (ZPID) in Trier (Germany) is an internationally recognized research-based psychology service center located in...

36157

Reposted by Hauke Roggenkamp

Colin Camerer @cfcamerer.bsky.social · 15/06/2026

NEW great progress from @i4replication.bsky.social in automating reproducibility "The Engine executes code, flags numerical mismatches, raises possible implementation issues, and proposes robustness checks, but leaves the substantive judgments to people" www.i4replication.org/blog/the-ai-...

graph showing that LLMs in a replication Engine work well

041

Reposted by Hauke Roggenkamp

Quentin André @quentinandre.bsky.social · 31/05/2026

New blog post: Make Your Papers Legible to Machines (They’re Already Reading) quentinandre.net/posts/legibl...

061

Reposted by Hauke Roggenkamp

Ian Hussey @ianhussey.mmmdata.io · 12/03/2026

I can't stop thinking about this. Imagine: - 2200 studies, as they say >2000 - 2 studies per article given piloting/multi-study articles - No p-hacking (lol); 1 test per study (lol), 5% False Positive Rate That's 55 significant studies that are false positives Actual publications: 64

33810

Reposted by Hauke Roggenkamp

Josh Grubbs @joshuagrubbsphd.com · 12/03/2026

I can’t decide if the file drawer or the absolute poverty of methodological effort it takes to run 2,000 MTurk studies is more concerning to me

2212

Reposted by Hauke Roggenkamp

Ana Martinovici @anamartinovici.bsky.social · 10/03/2026

“We wanted to create a way to celebrate those who have the courage to admit when they’ve made a mistake [...] Correcting the record shouldn’t be seen as a failure, but as a vital part of the scientific process. This award is about recognizing the integrity it takes to prioritize truth over ego”

022

Reposted by Hauke Roggenkamp

Julia M. Rohrer @dingdingpeng.the100.ci · 03/06/2025

Academia will form these little pockets -- people whose theorizing is outrageous & supported by methods outdated since the 90s -- but once it reaches a critical size those people just review each others papers & grants, form societies, hand out awards etc, like a self-contained parallel society.

2645989

Reposted by Hauke Roggenkamp

Dean Eckles @eckles.bsky.social · 21/03/2025

Mediation analysis or "three correlations in a trench coat"

0244

Reposted by Hauke Roggenkamp

Andrew Heiss @andrew.heiss.phd · 20/03/2025

I’ve long used FiveThirtyEight’s interactive “Hack Your Way To Scientific Glory” to illustrate the idea of p-hacking when I teach statistics. But ABC/Disney killed the site earlier this month :( So I made my own with #rstats and Observable and #QuartoPub ! stats.andrewheiss.com/hack-your-way/

Screenshot of the linked Quarto website, with input checkboxes to change different conditions for a regression model that predicts economic performance based on US political party, with a reported p-value

571468434