A fabricated dataset hosted on Kaggle was recently removed for breaching copyright, months after it was exposed in research led by #AusHSI PhD student @alexdgibson.bsky.social with @aidybarnett.bsky.social, @nicolewhite.bsky.social and Gary Collins. Find out more 👇

Retraction Watch@retractionwatch.com · last wk.

The online data repository Kaggle has removed a problematic dataset for violating intellectual property rights months after it was flagged by sleuths for containing celebrity photos and lacking information on data provenance or confirmed medical diagnoses.

Years ago we (@zoedoubleday.bsky.social) told scientists to stop creating new acronyms because people find them hard to read elifesciences.org/articles/60080. Now there's another reason, the machines find them hard to read too. If you want your work to be in LLM searches, then quit the TLAs.

Meta-Research: The growth of acronyms in the scientific literature

A study of 24 million articles has revealed that scientists have created more than 1 million acronyms since 1950, most of which have been used fewer than 10 times.

elifesciences.org

Papers with publicly available peer-review reports are less likely to be retracted than are those without public reports, a new study has found. My latest for @nature.com: www.nature.com/articles/d41... @aidybarnett.bsky.social, @jeroenson.bsky.social, @jessicapolka.bsky.social

Papers with open peer-review reports are less likely to be retracted

Researchers say that transparent publishing practices could be associated with research quality, but more evidence is needed.

nature.com

Need more time to get work done? Have you thought of FAMT? www.ft.com/content/0291...

😘❤️‍🔥😍

New blog on trying to chase down a potentially dodgy statistic from my local council medianwatch.netlify.app/post/homeless/

Sniffing out a bad statistic

A blog about statistics, meta-research, metascience and academia

medianwatch.netlify.app

A new blog on two innovations from funding peer review that could be transplanted to journal peer review. 1. 🔄 Distributed peer review. Quid pro quo. 2. 🎲 Lotteries. Acknowledging the limitations of human decision making with a bonus randomised trial. medianwatch.netlify.app/post/experim...

Experimenting with peer review

A blog about statistics, meta-research, metascience and academia

medianwatch.netlify.app

🚨 Protocol -> preregistration -> preprint -> publication 🚨 125 clinical prediction model articles developed on unreliable data on stroke and diabetes. Since preprinting our results 7 articles have been retracted. doi.org/10.1186/s129...

Dear Belgium, please win bigly www.abc.net.au/news/2026-07...

FIFA throws out US striker's ban after Trump intervention

US striker Folarin Balogun is free to play Belgium despite earning a red card in his last match, in a shock decision that delights Donald Trump and "astonishes" the Belgians.

abc.net.au

Until the ABS joins bluesky I'm just going to steal their memes from Facebook and post them here

Medical students are using a popular research tool to pump out misleading studies | Science | AAAS www.science.org/content/arti...

Medical students are using a popular research tool to pump out misleading studies

Critics say TriNetX’s easy analyses of electronic medical records fuel quick-and-dirty publications from inexperienced authors

science.org

The #WorldRefugeeDay #Telethon powered by the ASRC has begun! Let’s show that love not hate, community not division, welcome not fear is who we are as Australians. We are stronger together! Help refugee families today by calling 1300 692 772 to donate or www.asrc.org.au/telethon

Fox sports are clueless. #socceroos

Some record of ethical approvals should be publicly available. Research integrity is under enormous strain from instant AI made slop. We need to differentiate the real from the fake. medianwatch.netlify.app/post/transpa....

Ethics transparency

A blog about statistics, meta-research, metascience and academia

medianwatch.netlify.app

📢 Only one month until our statistical leadership symposium in Australia! With international keynotes, invited local speakers, and contributed presentations, it's shaping up to be a great event! agbarnett.github.io/stats_leader... Register here by 30th June:

Essential skills for statistics leadership in consultancy and research

A two-day Statistical Leadership Symposium for early and mid-career researchers across Australia

events.humanitix.com

Only two of the ten most populous nations are going to the #worldcup2026

📢 Statisticians in Australia from all sectors: Join us in July to share your experiences and learn from others in developing professional skills and becoming leaders. Please spread the word! Hope to see you there! @mischhub.bsky.social @nicolewhite.bsky.social @aidybarnett.bsky.social

Karen Lamb@kelamb.bsky.social · 2mo ago

There's still time to register for our new statistical leadership event in July 2026. We have in-person hubs in both Melbourne and Brisbane, with low registration costs thanks to the @science.org.au Theo Murphy Initiative. Join us by registering by 30th June: events.humanitix.com/essential-sk...

The submission portal for #AIMOS2026 is now open. Submit your lightning talks, discussion groups hackathons, and workshops now. www.aimos2026.org/submit

A blog on the evidence free call to "Bring back panels!" medianwatch.netlify.app/post/panels/. The call appeared in Science www.science.org/doi/10.1126/....

Beating the panel drum

A blog about statistics, meta-research, metascience and academia

medianwatch.netlify.app

New preprint showing that suspected paper mill papers get more citations (50 to 100% more) than non-suspect papers, but have fewer readers. This odd pattern is explainable by the mills citing their past papers in their new papers. www.biorxiv.org/content/10.6...

My season R package is getting a big overhaul thanks to @njtierney.aus.social.ap.brid.gy. My daughter has made a hex badge that I'm really happy with.

Flock of cockatoos

Amongst all the rubbish that I have seen in peer reviewed papers, this data is the worst. There are sketches, images of children (with no mention of ethics) and multiple celebrities, whilst the paper states: "A smart camera located in patient rooms has been used to collect images".

Retraction Watch@retractionwatch.com · 3mo ago

Scrolling through an online image dataset, Adrian Barnett pointed out a few familiar faces: Sylvester Stallone, George Clooney, Angelina Jolie and Daniel Craig. “You can see,” Barnett said, “this is just a comically bad dataset.” retractionwatch.com/2026/05/18/k...

Very interesting. In one fabricated citation the authors were "J Doe and R Smith". Some fraudsters have no idea that the LLMs are dropping these massive stink bombs throughout their "work".

Retraction Watch@retractionwatch.com · 3mo ago

Fabricated citations in the biomedical literature have increased 12-fold in two years, according to an audit of nearly 2.5 million papers published as a letter to The Lancet today.

If you've tried using INSPECT-SR to assess non-randomised studies and would be up for having a quick chat about your experience using it, let me know!

How much does it cost to buy your way onto a scientific paper? To answer this question (and better understand the global market for academic reputation manipulation), we compiled BuyTheBy, a massive dataset of thousands of advertisements from paper mills in seven different countries. Link below!

Outstanding work by @alexdgibson.bsky.social to uncover the use of highly dubious data sets from @kaggle.com being used in hundreds of research papers and potentially even informing clinical practice. If you're re-using data, take the time to confirm that it's real. www.nature.com/articles/d41....

Dozens of AI disease-prediction models were trained on dubious data

The models are designed to predict someone’s risk of diabetes or stroke. A few might already have been used on patients.

nature.com

More than one year after I raised serious concerns about a trial published in a @springernature.com journal, the journal contacts me to say they are taking it seriously. The trial appears in the Australian clinical guidelines for stroke and needs to be urgently examined. pubpeer.com/publications...

New preprint with Matt Spick. We examined differences between authors who chose open vs closed peer review. Articles with open review had ⬆️ citations and ⬇️ retractions. Authors that chose open review may be revealing a trust signal in the quality of their research. #allofplos osf.io/preprints/me...

Our local kids park is bee-themed, which is brilliant, but the new bees are going to cause some tears and nightmares.