Darren Dahly

Darren Dahly

@statsepi.bsky.social

1064

Posts

Epidemiologist + Statistician | Clinical Research Facility - UCC | UCC School of Public Health | #ClinicalTrials #Epidemiology #Statistics #RStats
Views mine -> https://statsepi.substack.com/

Posts Replies Media

I will continue to make fun of linkedin, because it's hilarious, but I really do see genuinely useful links/papers/studies etc whenever I log in.

We have a problem in medical/health research whereby methodological expertise is downgraded as "technical", and respect instead flows to those that can effortlessly regurgitate what other authors feed to them via paper abstracts and conclusions.

Dan Ariely just posts on linked with impunity lol

I spent my early research career studying human growth and implications for public health and my jaw dropped when I first started hearing about these height data coming out of the UK.

14 years of Tory rule gives us this utterly horrifying graph:

The satisfaction of finally working out how to best organise the thing you are writing.

Reposted by

Lots of comments about parents. I meant to write a follow-up article about the research on parents, but never got around to it. Your intuitions are correct: the kids are alright, but their parents are not. To learn more, start with this short review paper; free version here: osf.io/preprints/ps...

There is a lot of research on this. Learning about racism doesn't make white kids feel guilty. Here is what it does. Summary of some of the studies here: fivethirtyeight.com/features/wha...

Reposted by

Congratulations to Susan Holmes @sherlockpholmes.bsky.social for being elected as EMBO Member! people.embo.org/profile/susa...

Reposted by

Announcing the winners of the 2024 SIPS Commendations! 👇✨ For information about our awards and to nominate projects for mission awards and commendations for 2025, see here: improvingpsych.org/mission/awar...

Awardsimprovingpsych.org The SIPS Awards Committee gives out two different types of awards to recognize projects that further the SIPS mission: SIPS Mission Awards and SIPS Commendations. SIPS Mission Award SIPS is solici…

Super proud to have played a small role in this project led by my colleague Dr Karen Matvienko-Sikar. x.com/KarenMSikar/... Thank you @improvingpsych.bsky.social @priyasilverstein.com for the very welcome acknowledgement of the work. ☺️

Principles and Practices of Open Research (PaPOR Trail) is an open educational resource for undergraduate & postgraduate students. open.ucc.ie/browse/all/c...

PaPOR TraIL: Principles and Practices of Open Research - 003CPD - Centre for CPDopen.ucc.ie

Reposted by

Currently attending this UNESCO event "Announcing the Global Diamond Open Access Alliance" www.unesco.org/en/articles/... Just heard a well-deserved shout out to PCI (Peer Community In), an innovative preprint-review-curate publishing model. Nice!

Announcing the Global Diamond Open Access Alliancewww.unesco.org The event will introduce and officially announce the Global Diamond Open Access Alliance, highlighting its vision, mission, and objectives, and to engage stakeholders in a collaborative effort to prom...

Reposted by

If you've come here from X we're not having the same issue on this platform as we have there. The link to this article is still being miscategorized as spam despite our reports to X. The link to the investigation below should work on Bluesky as it appears to be an X only issue.

Latest: An analysis of open source evidence, as well as multiple missile experts, have pointed to a Russian launched Kh-101 cruise missile being the weapon that struck a children’s hospital in Kyiv www.bellingcat.com/news/2024/07...

Russian Missile Identified in Kyiv Children’s Hospital Attack - bellingcatwww.bellingcat.com An analysis of open source evidence, as well as missile experts, have pointed to a Russian launched Kh-101 cruise missile being the weapon that struck a children’s hospital in Kyiv.

Reposted by

A great reminder of what "sensitivity analysis" is (or should be), before just diving in. Nomenclature in stat can be very risible (WOTD) but we have to try and stay on top of it! bmcmedresmethodol.biomedcentral.com/articles/10.... @timpmorris.bsky.social #StatsSky #Stats #RStats

Choosing sensitivity analyses for randomised trials: principles - BMC Medical Research Methodologybmcmedresmethodol.biomedcentral.com Background Sensitivity analyses are an important tool for understanding the extent to which the results of randomised trials depend upon the assumptions of the analysis. There is currently no guidance...

Me, crying tears of joy.

Reposted by

Biomedical science folks: Curious to learn more about peer-review and the scientific publishing landscape as a whole? Want hands-on, group reviewing experience with an editor-mentor? Apply now for our next run of Peerspectives, fully online, starting this fall: forms.gle/assQnYNgpaav... (due Aug 1)

Peerspectives Course - Winter Semester 2024/25 - Application Formforms.gle We are excited to announce the upcoming offering of our semester-long "Peerspectives" scientific peer review training course for a new cohort of enthusiastic participants. The next run of the course i...

Reposted by

Quick fix for one part of psychology’s underspecification crisis: Add a ‘target population’ subsection before the sample description

A nice paper you may have missed: pubmed.ncbi.nlm.nih.gov/30299451/ cc @epidbydesign.bsky.social @lizstuart.bsky.social @edwardsjk.bsky.social

Target Validity and the Hierarchy of Study Designs - PubMedpubmed.ncbi.nlm.nih.gov In recent years, increasing attention has been paid to problems of external validity, specifically to methodological approaches for both quantitative generalizability and transportability of study res...

Pure scam. Congrats to all the clueless dipshits in academia that helped make this happen.

Sam Altman and Arianna Huffington announce Thrive AI Health, a new startup backed by OpenAI and Thrive Global to build a "hyper-personalized AI health coach" (TIME) Main Link | Techmeme Permalink

Why the hell would it I take leave in July just when everyone else has f**ked off!? 😜

Reposted by

"We were right about how the data were altered, Gino’s prevailing explanation for the alterations does not make sense, and yet we are the defendants in this case." datacolada.org/118

[118] Harvard’s Gino Report Reveals How A Dataset Was Altered - Data Coladadatacolada.org As you may know, Harvard professor Francesca Gino is suing us for defamation after (1) we alerted Harvard to evidence of fraud in four studies that she co-authored, (2) Harvard investigated and placed...

Just when I thought all was lost, a glimmer of hope appears.

Reposted by

I am not commenting on the contents of the paper (www.nature.com/articles/s41...) but the article in the Guardian discussing it is pretty bad (www.theguardian.com/science/arti...) for a metagenomic test with a sensitivity of 65.30% and a specificity of 72.4%.

Autism could be diagnosed with stool sample, scientists saywww.theguardian.com Researchers found differences in the gut microbes of autistic people, raising hopes for faster diagnosis

It's a real challenge to maintain this level of humility when I am constantly proven correct about so many important things.

Reposted by

Made it to Salzburg for #useR2024 ! 🇦🇹 🏴‍☠️ Couldn't be more excited for my Keynote talk on Tues: 💫"Keep R Weird" 🤪 Come for the hexagon pantsuit, stay for the rant about non-standard eval. sched.co/1c9Ix (🦭and watch @rdatatable.bsky.social for how to grab dins and some data.table swag after! )

useR! 2024: Keynote: Keep R Weird - Dr. Kelly Bodwin...sched.co View more about this event at useR! 2024

Learn. To. Use. Calendar. Invites. Or. Else.

Reposted by

Well, well, well. WELL. That whole “protests made cops feel bad so they all quit” narrative? Looks like it was mostly myth. Excited to read this paper by Ben Grunwald, who compiled a huge dataset on post-2020 police employment. Agg decline was 1%… and bigger local declines not tied to protests.

Reposted by

And here's the no-longer-forthcoming blog post

I've been using Positron as much as possible over the past week and it's been delightful! Blog post (probably) forthcoming about how I have it all set up. #rstats github.com/posit-dev/po...

Fun with Positron | Andrew Heisswww.andrewheiss.com Combine the best of RStudio and Visual Studio Code in Posit’s new Positron IDE

Very happy I am registered to vote in NC.

In a normal campaign, this would have dominated the weekend’s news. I wrote about @gregsargent.bsky.social’s scoop about the Christian nationalist GOP nominee for NC Gov calling for murder of his political enemies: ‘Some folks need killing’: www.msnbc.com/opinion/msnb...

Opinion | ‘Some folks need killing’: North Carolina Lt. Gov takes Christian nationalism to dangerous lowwww.msnbc.com Mark Robinson’s newly explicit calls for killing perceived enemies show how seamlessly Christian nationalists can glide into promoting real violence.

Reposted by

Let's play "Spot the obvious errors in this table". I will pre-register what I can see from the most cursory visual inspection in the next post, encrypted so as not to spoil.

Reposted by

Abuse of power, herd mentality, and outright fraud have protected the scientists pursuing the amyloid hypothesis for decades, at the expense of other researchers… and millions of people with Alzheimer’s who deserve better

“But very few diseases of aging have just one cause. It’s just not logical,” www.nytimes.com/2024/07/07/o...

Opinion | All the Alzheimer’s Research We Didn’t Dowww.nytimes.com The hyper-focus on one theory for treatment has slowed progress.

Reposted by

I wrote about the latest attempt to sell genetic IQ prediction and why it is snake oil. A short 🧵: theinfinitesimal.substack.com/p/genomic-pr...

Genomic prediction of IQ is modern snake oiltheinfinitesimal.substack.com Predicting IQ from genetics is inaccurate, confounded, and sows confusion about genetic research.