Reposted by Matt WilliamsLeonid Schneider @forbetterscience.bsky.social · 1h"At the University of Calgary, Griep’s handling of data also allegedly led to problems. So serious that he had to leave under a non-disclosure agreement." Deported into a Dutch professorship. Academia is harsh to fraudsters! 021
Reposted by Matt WilliamsPeer Community In @peercommunityin.bsky.social · 28/09/2026A PCI recommendation can be a valuable part of a publication strategy: · The recommendation and peer-reviews are citable · You participate in open science initiatives · You can publish in the journal of your choice 👉 Explore your publishing options with PCI: peercommunityin.org/submit-a-pre...peercommunityin.orgSubmit a preprint to PCI - Peer Community InOnce an article has been recommended by a PCI, the authors can opt to leave it on a preprint server, to publish it in PCI Journal, to submit it to a PCI-friendly journal or to any other journal 076
Matt Williams @matthewmatix.bsky.social · 27/09/2026The study also doesn't account for insincere responses. Our recent paper showed most participants who endorsed two contradictory theories showed signs of insincerity (such as agreeing that the Canadian military are preparing a force of intelligent giant killer raccoons to invade nearby countries...) 130
Matt Williams @matthewmatix.bsky.social · 27/09/2026That 91% figure isn't something observed in the study, it's an predicted probability of incoherent endorsements for participants sitting at the ceiling of the conspiracy scale. Which may have been few or no people in the sample.... academic.oup.com/pnasnexus/ar...academic.oup.comStill very much dead and alive: Incoherence in conspiracism as a departure from Bayesian rationalityAbstract. Incoherence has long been considered a hallmark of belief in conspiracy theories, though recent research has cast doubt on this claim. We identif 130
Reposted by Matt WilliamsJack Wilkinson @jdwilko.bsky.social · 24/09/2026The integrity of evidence synthesis is threatened by problematic randomised controlled trials. These may be fraudulent, or subject to critical errors. INSPECT-SR is a tool to assess trustworthiness of RCTs - out today: www.bmj.com/content/394/...bmj.comINSPECT-SR tool for assessing trustworthiness of randomised controlled trialsThe integrity of evidence synthesis is threatened by problematic randomised controlled trials, where there are serious concerns about the trustworthiness of the data or findings. Such concerns could b... 12415
Matt Williams @matthewmatix.bsky.social · 25/09/2026On behalf of the psychology delegation I apologise and ask for guidance on the proper term. "Just a linear model, dammit"? 120
Reposted by Matt WilliamsKristoffer Magnusson @rpsychologist.com · 25/06/2026New interactive blog! "Why Adjusted Regression Coefficients Are Less Descriptive Than They Look" rpsychologist.com/descriptive-... 1520769
Reposted by Matt WilliamsThe Cochrane Collaboration @cochrane.org · 24/09/2026INSPECT-SR – a tool for assessing the trustworthiness of randomized controlled trials – has been published today in @bmj.com, giving researchers a structured way to identify problematic studies before they're included in systematic reviews. 🔗 www.cochrane.org/about-us/new... #ResearchIntegrity 03124
Matt Williams @matthewmatix.bsky.social · 24/09/2026I don't have a paid subscription so have only tested it with small things but I have not managed to get a false positive or false negative out of it yet. Seems pretty good. 030
Matt Williams @matthewmatix.bsky.social · 24/09/2026Lots of examples of this: scholar.google.com/scholar?hl=e...scholar.google.com 010
Matt Williams @matthewmatix.bsky.social · 24/09/2026This paper looks excellent! I've been looking at a related issue with causal inference in qual papers. It seems common for qual papers to make causal claims based on participants saying they were affected by something.I'm not sure if there exists any epistemological stance for which that makes sense 220
Matt Williams @matthewmatix.bsky.social · 23/09/2026I cite that paper in tons of reviews and decision letters too! Sorry if this makes the world mad at you both 010
Reposted by Matt WilliamsJeff Sharlet @jeffsharlet.bsky.social · 21/09/2026The provost of Dartmouth, at which I teach, has been using AI to write both academic papers & national op-eds, according to brave reporting by @thedartmouth.bsky.social. Many students are already bitter about AI. This leaves us as faculty no ground to stand on. www.thedartmouth.com/article/2026...thedartmouth.comOp-eds, academic articles by Provost Santiago Schnell flagged as ‘AI-written’ by ‘near-zero error’ AI detector PangramSchnell said he used artificial intelligence for “making writing processes more efficient,” but that AI was not a “substitute” for his own “intellectual judgment or authorship.” 701480490
Matt Williams @matthewmatix.bsky.social · 18/09/2026Thanks, but I think that's a paid product that journals can add to their workflows. I'm trying to hunt down something free that a rando like me can just use as needed! 110
Matt Williams @matthewmatix.bsky.social · 18/09/2026Forensic meta-science folks: What is the most efficient way to check an unpublished paper for tortured phrases (e.g., during peer review)? The Problematic Paper Screener website lists a number of software vendors but I'm struggling to find anything that I can use myself! 232
Matt Williams @matthewmatix.bsky.social · 15/09/2026Ah, I meant the acceptance rate is too *high* to believe (given that median time to first decision). Even much less prestigious journals have much lower acceptance rates these days (lots of slop in editors' queues!) 110
Matt Williams @matthewmatix.bsky.social · 15/09/2026The IF, possibly. They publish few articles and the ones they do seem to get cited lots, with a couple of huge outliers. But median time to first decision of 2 days along with acceptance rate of 31% can't be right. Might be calculated with different subsets of submissions (e.g. solicited vs total) 210
Matt Williams @matthewmatix.bsky.social · 15/09/2026I think this bit plays in somehow but the numbers still seem hard to believe. I call shenanigans 110
Reposted by Matt WilliamsMatthew B. Jané @matthewbjane.bsky.social · 13/09/2026I’ve been quiet on social media lately, but today I’m launching Audit Clinical Evidence (ACE), an initiative I’ve been developing over the past year. Take a look: AuditClinicalEvidence.orgauditclinicalevidence.orgACEIndependent audits of published clinical research for statistical errors, data inconsistencies, and research integrity violations. 89232
Reposted by Matt WilliamsThomas Lumley @tslumley.bsky.social · 10/09/2026A regression model without a colinearity diagnostic is like a fish without a bicycle 2317
Reposted by Matt WilliamsLeonid Schneider @forbetterscience.bsky.social · 10/09/2026Congratulations to Elsevier on this insanity my colleague found. Is anything in this clinical trial from Spain and UK real? www.cell.com/the-innovati...cell.comQuantum evidence of nonlocal consciousness during clinical deathIf consciousness can operate under quantum principles, then the boundaries between life, death, and cognition are far more permeable than current science allows. In the first large-scale, randomized, ... 135423
Reposted by Matt WilliamsShane Littrell, PhD @metacognishane.bsky.social · 05/09/2026New article is posted! In this one, I explain what LinkedIn, "cataphoric teasers," and Dolly Parton can teach us about people's growing distaste of AI-generated text. bullshitology.substack.com/p/behind-the...bullshitology.substack.comBehind the Bullshit Machine: How LinkedIn, “Cataphoric Teasers,” and Dolly Parton explain the AI Slop-ocalypseThat’s not X, it’s Why 032
Matt Williams @matthewmatix.bsky.social · 04/09/2026 I definitely emphasize with your experiences and those of the grad students in that sample. But just a gentle reminder that we can't get valid causal inferences by asking people how they think they were affected by something. Better studies needed! 001
Reposted by Matt WilliamsJoshua Eaton @joshuaeaton.bsky.social · 02/09/2026“There’s a new Mr. Beast book.” You bolt awake in your workshop in Mainz. It’s 1450. You’re Johannes Gutenberg, and you’ve changed your mind. The future cannot come to pass. Your printing press must burn. 6979531450
Reposted by Matt WilliamsKieran Healy @kjhealy.co · 01/09/2026Yes it was.newyorker.comThey Studied Dishonesty. Was Their Work a Lie?Dan Ariely and Francesca Gino became famous for their research into why we bend the truth. Now they’ve both been accused of fabricating data. 37414
Reposted by Matt WilliamsData Colada @datacolada.bsky.social · 31/08/2026An influential paper by Ariely & Wertenbroch (2002) reported two studies with tampered data. Post 1 of 2. datacolada.org/138datacolada.org[138] Artificial Deadlines (Part 1): Evidence of Fraud in an Influential Study About ProcrastinationA new paper in Psychological Science (.htm) reports a failure to replicate Study 2 of Ariely and Wertenbroch’s influential article entitled, “Procrastination, Deadlines, and Performance: Self-Contr… 723799
Reposted by Matt WilliamsJulia M. Rohrer @dingdingpeng.the100.ci · 31/08/2026I am once again begging everyone who supervises student theses to stop making students conduct mediation analyses because “a regression isn’t complicated enough for a thesis” 😭 1519823
Reposted by Matt WilliamsMarcus Crede @marcuscrede.bsky.social · 25/08/2026Angela Duckworth's clams about "Grit" - based on misunderstandings of factor analysis, odds ratios, measurement principles, and effect sizes. The level of incompetence still leaves my flabbers gasted - particularly since it sparked the whole "grit" movement in education. So many $$$ wasted. 1162
Matt Williams @matthewmatix.bsky.social · 25/08/2026PSA: Asking participants how something affected them is not a causal identification strategy Don't make me point any fingers now 030
Matt Williams @matthewmatix.bsky.social · 24/08/2026Shout out to University of Waikato for this wholly sensible guidance on AI use (which seems to be backed up by plenty of in-person assessment). libraryguides.waikato.ac.nz/c.php?g=9805... 094
Reposted by Matt WilliamsBjörn Siepe @bsiepe.bsky.social · 21/08/2026Looking for a fun & useful science sidequest? Consider contributing to FLoRA. We did some coworking sessions in our lab this week, using their fun gamified app Learn more: forrt.org/replication-... Bonus: Reading interesting articles (I annotated one on the link b/w sleep & cruise tour enjoyment) 0166
Reposted by Matt WilliamsMel Andrews @bayesianboy.bsky.social · 21/08/2026You cannot produce knowledge while outsourcing your thinking to software. thecybernetician.substack.com/p/being-in-t...thecybernetician.substack.comBeing in the business of thinkingSoftware cannot think for you. 28117
Reposted by Matt WilliamsAndrea Howard @drandreahoward.bsky.social · 21/08/2026Enjoyed talking to @psychscience.bsky.social about my group’s new(ish) paper for the Observer. I’ve talked about this on here at an earlier stage. In 4 studies, we found no evidence that asking young adults to cut back on their social media improves well-being. Small thread 🧵 follows:psychologicalscience.orgFor Young Adults, Limiting Social Media Use Doesn’t Improve Well-BeingWhat happens when young adults are asked to slash their social media use to 30 minutes a day? In a new set of studies, most did not follow through, and those who did cut back … 46625
Reposted by Matt Williamslastpositivist.bsky.social @lastpositivist.bsky.social · 17/08/2026The Recurring Speed Camera Discourse is a bit exasperating really because, look, we all know what's actually happening is people just want to be able to commit the pertinent crimes and don't like being punished for it, but people come up with obvious post-hoc pseudo-woke rationalisations for this. 391221177
Reposted by Matt WilliamsShane Littrell, PhD @metacognishane.bsky.social · 07/08/2026What can Bill Shanter's hair piece teach us about bullshit detection? For reasons that may seem obvious, I thought it'd be a good time to revisit this article from last year (slightly revised). Enjoy! bullshitology.substack.com/p/oh-look-an...bullshitology.substack.comHow BSing Is Like A Bad Toupée (Redux)The Toupee Fallacy helps explain why some BSers get away with it and others don't 013
Matt Williams @matthewmatix.bsky.social · 07/08/2026The last major landmass to be settled by humans! 010
Reposted by Matt WilliamsJenna Norton @jenna-m-norton.bsky.social · 07/08/2026*Criticism* of ideas is not the same thing as *censorship* of ideas. It's literally our job as scientists to criticize each others' ideas. That's how we make science stronger. We poke holes in hypotheses and point out potential harms so we don't waste resources or harm people with fallible ideas. 1568
Matt Williams @matthewmatix.bsky.social · 30/07/2026When the link in your data availability statement is just to the publisher's guidance on writing data availability statements... 030
Matt Williams @matthewmatix.bsky.social · 30/07/2026www.aimos2026.org AIMOS 2026 will be in Wellington this year on 30 Nov to 2 Dec. Abstracts due 15 Aug. This is a very fun and very interactive metascience conference. Come along and hang out!aimos2026.orgAIMOS_2026 | Metascience Conference – Register TodayJoin the AIMOS 2026 Conference n Wellington, New Zealand, for three days of interdisciplinary meta-research, keynote speakers, workshops, and networking to improve research practices. 0108
Matt Williams @matthewmatix.bsky.social · 27/07/2026www.massivemagazine.org.nz/articles/cal... This isn't the point of this article, but it's really good to see our colleagues in the College of Sciences pushing to do the right thing, even when it's the hard thing. 010
Reposted by Matt WilliamsJay Van Bavel, PhD @jayvanbavel.bsky.social · 18/07/2026We need to change norms around sexual harassment Yesterday Tessa West spoke up about a sexual harassment complaint she filed against John Jost. Not only is he a senior colleague in the dept of psychology, but he is one of the most prominant figures in the field of social and political psychology. 322484
Matt Williams @matthewmatix.bsky.social · 13/07/2026To me the best way forward for asynchronous online courses is to include in-person invigilated exams. That's difficult & expensive & some students will hate it... but can be done. Massey used to have a network of exam centres all round the world. Wiped out in COVID, but gradually being clawed back 130
Matt Williams @matthewmatix.bsky.social · 13/07/2026This. Even the "pretend" option may not pan out for much longer... How long will governments be willing to pay universities for handing out degrees to students who haven't learned anything? 120
Matt Williams @matthewmatix.bsky.social · 29/06/2026I am keen at some point to do an updated meta-analysis on EMDR where we take out all the studies that have signs of untrustworthy (e.g. fake) data, using this tool... pmc.ncbi.nlm.nih.gov/articles/PMC... If you have any free time outside the clinic it would be fun to have you on board!pmc.ncbi.nlm.nih.govINSPECT-SR: a tool for assessing trustworthiness of randomised controlled trialsThe integrity of evidence synthesis is threatened by problematic randomised controlled trials (RCTs). These are RCTs where there are serious concerns about the trustworthiness of the data or findings.... 020
Matt Williams @matthewmatix.bsky.social · 26/06/2026"Impact" and "affected" are claims about causality, FYI 200
Reposted by Matt WilliamsAustralasian Society for Experimental Psychology @expsyanz.bsky.social · 23/06/2026Join Professor Gina Grimshaw on 2 July, 6:00–7:30 PM, for a public lecture discussing how emotions shape attention, memory, decision-making, and human connection. Register now on: visualneuroscience.auckland.ac.nz/epc-apcv-2026/ 084