"Model failed to converge" #rstats #medsky #statsky

Can we actually train ourselves to be more open-minded and harder to fool? New research says yes. A short message warning us about closed-mindedness made people better at spotting misinformation, less likely to believe conspiracy theories, and more thoughtful about what they shared #MisinfoResearch

🌧️ Forest loss and flooding A new study shows that deforestation and forest fires in Australia reduce rainfall interception, increasing river flooding across vast regions. 🔗 www.nature.com/articles/s43... #SciComm #ClimateCrisis #Floods 🧪

Thanks to everyone that weighed in on this - there are some strong arguments for imputing first and then subsetting, especially when the subsetting variable set contains missingness. But I am wrestling with how to implement this across the entire analysis pipeline in practice.

Ok #skyhivemind and #statsky #rstats - I am analysing a subset from a clinical quality registry. Both the total registry dataset and the subset have missing data that needs to be imputed. Should I i) impute the full registry set and then subset or ii) subset from the get-go and discard the rest?

Having reviewed for @bonejointjournal.bsky.social and others and generally having a low rate of "recommend accept" to the editor, I think its time to talk about "why my paper got rejected from journal x". Join me on this journey (rant) of what can go wrong with clinical science. #clinicalresearch

At the risk of revealing my silly-goose status in all things #rstats a word to the wise for anyone using mice for multiply imputation - you really need to convert your character columns to factors otherwise mice doesn't recognise them and sets them to constant.

I hear your valid commentary on clustering @cameronpat.bsky.social and raise you - it remains alive and well! Don't worry, l've got my own publications in case anyone thought i had the moral high ground here!

Heard from #AHRQ that the entire grants staff has been let go! They are unable to issue funds for grants already funded, let alone fund new grants. This is outrageous and will make Americans poorer and sicker. Why? A brief thread about just some recent AHRQ-funded research: #MedSky #HealthPolicy

Is there somewhere in the scientific record between "this paper should never have seen the light of day" and "we shall cherish this forever, regardless of its flaws"? Methods evolve constantly, maybe we should consider the retirement of papers that were good in their time. #medsky #orthosky

I think I've been doing this long enough that someone somewhere might benefit from my thoughts on certain topics. If you've been looking for someone talking about things you might relate to - I've made a start. #medsky #orthosky #rstats

Hello brains trust - can anyone point me to resources on power analysis of case-mix adjustment models, preferably with mixed effects? I'm using simr in R - but I'm having trouble relating the examples to my application #rstats #clinicaloutcomes #stats

To my American friends - the phrase is "I couldn't care less" implying that you have maxed your laissez-faire credit card and you have no more cares to give about an issue. Butchering the phrase by saying "I could care less" leaves many English speakers thoroughly confused.

Did you see the Choice sunscreen testing results? Here's an SPF deep dive! Key points: Sunscreens are very effective, these results do not indicate you should lose faith in them SPF testing is imprecise - this is an inherent limitation of the test, lots of variation even if nothing's gone "wrong"

I wake up more often in a cold sweat with feelings of dread the more I think I'm implementing better analysis practices and continue to review journal submissions...thanks @statsepi.bsky.social now you've ruined it for everyone...I wish I had the same universal confidence in the literature

If you work in the area of research integrity, you should submit an abstract to and attend the Internationa Research Integrity Conference being held at NeuRA this year: neura.edu.au/news-media/e...

U.S. attorney demands scientific journal explain how it ensures 'viewpoint diversity' Letter raised concerns it was designed to suppress academic and scientific freedom. @evanbush.bsky.social reports @nbcnews.com www.nbcnews.com/science/scie...

Many data sets held by NOAA are being "retired" in may, including datasets relevant to ocean sciences and climate. If you have need for these data, or are an archivist, it would be good to download this now. PLEASE SHARE WIDELY. www.nesdis.noaa.gov/about/docume...

On behalf of all Australians, I want to apologise to the world for giving the neo-nazis ideas on offshore detentions... If that seems like a wild claim, please look up the Aus government relationship with Nauru and Christmas island

This should be a wakeup call for anyone who has Sex or Gender as a line item in Table 1. This attitude doesn't effect just those based in the US, it's the international collaborations, the 2nd-order effects that will be the most impactful for all of us, whether you work in "science" or not.

Hi everyone. If you see news articles about the science crisis that we should repost, please share with us. And pl share gift links if the source is paywalled like WSJ or STAT, or if you can share text we can excerpt/screenshot. Thanks! 🧪

Pubmed is currently still working for me but not sure how long this will hold. Here is @hildabast.bsky.social ‘s blog post on what to do if Pubmed becomes unreliable absolutelymaybe.plos.org/2025/02/14/w...

Thank you for this! I would urge all authors to submit their author's final version of all articles to their institutional repository, a subject repository, and or zenodo.org I've been an academic librarian in this space for a longtime, and am happy to help anyone--DM me!

Time to start making this an issue in the upcoming Aus election...what are the candidates offering in terms of protecting access to material?

Right, so I'm now a member of Statistical Society of Australia. Should I book with a talent agency now or should I handle all the media attention by myself? How does one connect further @visnut.bsky.social now that I'm in the club with the cool kids? No bsky for ssa? #rstats #noflashphotography

Journals sending me a notice to cancel a reviewer invitation I never asked for, like I'm a problematic Tinder date, is quite the display of hubris in this day and age.

The scientific community in Aus should be taking notice when our "candidates" start sprouting Trump-lite slogans on the campaign trail. Gutting health research by censoring terms, regardless of context affects every single person - not just "them" or "others"

Science is a noble but fragile endeavour that must be protected from political control and censorship. Scientific communities, around the world, should show solidarity with our colleagues in the US and denounce this chilling attack on science and scientists. Share this if you agree! #ScienceSky

Our suite of “2 week systematic review” tools (mentioned in the Nature News article) are free at sr-accelerator.com#/

Is there anyone working on #Orthopaedic #musculoskeletal outcomes|science|biomechanics| patient reported outcomes? and has anyone put together a starter pack? I have some findings to get put to the world and would be psyched to connect with the right people