Dr Matt Lloyd Jones (he/him)
@befriendabacterium.bsky.social
Researcher interested in evidence synthesis and other evidence-based things, most often to do with microbes (e.g. veterinary microbiology in livestock, pollution in bathing waters).
Associate Research Fellow at University of Exeter
Dw i'n dysgu Cymraeg 🏴
Associate Research Fellow at University of Exeter
Dw i'n dysgu Cymraeg 🏴
Pinned
Reposted by Dr Matt Lloyd Jones (he/him)
If you compute an effect size (say a Cohen's d type standardized mean difference) several different ways and then selectively report the largest one, what would you call that?
p-hacking
research misconduct
fraud
something else?
p-hacking
research misconduct
fraud
something else?
November 20, 2025 at 2:48 PM
If you compute an effect size (say a Cohen's d type standardized mean difference) several different ways and then selectively report the largest one, what would you call that?
p-hacking
research misconduct
fraud
something else?
p-hacking
research misconduct
fraud
something else?
Not sure if this is an advert or a threat
November 20, 2025 at 11:27 PM
Not sure if this is an advert or a threat
I like to think that Chris Van Tulleken’s war on ultra-processed foods began when he sat opposite a guy on the train going goblin mode on the worst train picnic
November 20, 2025 at 11:01 PM
I like to think that Chris Van Tulleken’s war on ultra-processed foods began when he sat opposite a guy on the train going goblin mode on the worst train picnic
sir you do know trying to bribe an ebay customer with haribo starmix is an offence, right?
November 19, 2025 at 3:29 PM
sir you do know trying to bribe an ebay customer with haribo starmix is an offence, right?
receive email from the uni that doesn't mention AI challenge
November 19, 2025 at 10:21 AM
receive email from the uni that doesn't mention AI challenge
Reposted by Dr Matt Lloyd Jones (he/him)
My #rstats cheat code for today is the binom.confint function in the binom package that will spit out *12* different ways of calculating a CI for a proportion.
Also, this is why you use R for statistics...
(and of course the correct CI method is bayes 😎)
Also, this is why you use R for statistics...
(and of course the correct CI method is bayes 😎)
November 16, 2025 at 2:52 AM
My #rstats cheat code for today is the binom.confint function in the binom package that will spit out *12* different ways of calculating a CI for a proportion.
Also, this is why you use R for statistics...
(and of course the correct CI method is bayes 😎)
Also, this is why you use R for statistics...
(and of course the correct CI method is bayes 😎)
All this chin stroking from the BMJ on how to resolve the crisis in peer review, and not once do they appear to consider paying people for this highly skilled labour as an option www.bmj.com/content/391/...
November 14, 2025 at 7:02 PM
All this chin stroking from the BMJ on how to resolve the crisis in peer review, and not once do they appear to consider paying people for this highly skilled labour as an option www.bmj.com/content/391/...
Reposted by Dr Matt Lloyd Jones (he/him)
Fed up with the John Lewis's website talking to me as if I'm a naughty dog.
November 14, 2025 at 3:46 PM
Fed up with the John Lewis's website talking to me as if I'm a naughty dog.
Reposted by Dr Matt Lloyd Jones (he/him)
Confusing stats language:
The "linear" is linear regression means "linear in the parameters".
It does not mean "can only fit straight lines", and things like polynomials and splines can be included in linear models.
online.stat.psu.edu/stat501/less...
1/2
The "linear" is linear regression means "linear in the parameters".
It does not mean "can only fit straight lines", and things like polynomials and splines can be included in linear models.
online.stat.psu.edu/stat501/less...
1/2
November 14, 2025 at 9:08 AM
Confusing stats language:
The "linear" is linear regression means "linear in the parameters".
It does not mean "can only fit straight lines", and things like polynomials and splines can be included in linear models.
online.stat.psu.edu/stat501/less...
1/2
The "linear" is linear regression means "linear in the parameters".
It does not mean "can only fit straight lines", and things like polynomials and splines can be included in linear models.
online.stat.psu.edu/stat501/less...
1/2
If my myneddiad-level Welsh serves me well, BBC Radio Cymru appear to have two presenters doing 24 hour danceathon for Children In Need, like some bizarre Welsh version of They Shoot Horses, Don't They? ("Maen nhw'n saethu ceffylau, onid ydyn nhw?"?).
November 14, 2025 at 10:59 AM
If my myneddiad-level Welsh serves me well, BBC Radio Cymru appear to have two presenters doing 24 hour danceathon for Children In Need, like some bizarre Welsh version of They Shoot Horses, Don't They? ("Maen nhw'n saethu ceffylau, onid ydyn nhw?"?).
Reposted by Dr Matt Lloyd Jones (he/him)
E.O. Fucking Wilson
November 13, 2025 at 5:56 AM
E.O. Fucking Wilson
science has ruined me
November 13, 2025 at 10:19 AM
science has ruined me
From “Will the Prime Minister grow a tash for Movember?” straight to “Will the Prime Minister reintroduce the Death Penalty?”. #PMQs really is a wild ride for a lunch break
November 12, 2025 at 12:31 PM
From “Will the Prime Minister grow a tash for Movember?” straight to “Will the Prime Minister reintroduce the Death Penalty?”. #PMQs really is a wild ride for a lunch break
oh no i just felt a bit of christmas spirit
November 12, 2025 at 10:29 AM
oh no i just felt a bit of christmas spirit
Love these Buzzfeed investigative journalism pieces
November 11, 2025 at 10:15 AM
Love these Buzzfeed investigative journalism pieces
Whoever you are, Dr Dan Chalk 🫡
PenARC is proud to be part of #RPySoc2025! 🎉
🗓️ Thu 13 Nov: Dr Dan Chalk – “Total Eclipse of the Excel Chart”
🗓️ Fri 14 Nov: Sammi Rosser – “Reusable by Default”
⚡️Plus lightning talks from our brilliant #HSMA programme graduates!
🔗 bit.ly/43ZGpXq
#DataScience
🗓️ Thu 13 Nov: Dr Dan Chalk – “Total Eclipse of the Excel Chart”
🗓️ Fri 14 Nov: Sammi Rosser – “Reusable by Default”
⚡️Plus lightning talks from our brilliant #HSMA programme graduates!
🔗 bit.ly/43ZGpXq
#DataScience
RPySOC 2025: NHS-R/NHS.pycom Open Source Conference
bit.ly
November 10, 2025 at 9:29 PM
Whoever you are, Dr Dan Chalk 🫡
English Language Stack makes programming stack look like a gentlemen’s tea party english.stackexchange.com/questions/12...
What prefix means "during"?
The prefixes "pre-" and "post-" refer to events before and after. For instance, "pre-season" and "post-season" or "pre-study" and "post-study". Is there a prefix fitting this pattern which means "d...
english.stackexchange.com
November 10, 2025 at 8:13 PM
English Language Stack makes programming stack look like a gentlemen’s tea party english.stackexchange.com/questions/12...
Reposted by Dr Matt Lloyd Jones (he/him)
A week before his defence @rug.nl, the publication of @euantheyoung.bsky.social's chapter on the lifespan cost of reproduction in @science.org Advances - with @hannahdugdale.bsky.social, @lummaalab.bsky.social and myself - couldn't have been better timed! 👍 🥳 🥂https://doi.org/10.1126/sciadv.adz6422
November 10, 2025 at 12:39 PM
A week before his defence @rug.nl, the publication of @euantheyoung.bsky.social's chapter on the lifespan cost of reproduction in @science.org Advances - with @hannahdugdale.bsky.social, @lummaalab.bsky.social and myself - couldn't have been better timed! 👍 🥳 🥂https://doi.org/10.1126/sciadv.adz6422
peer review is labour and it should be paid for peer review is labour and it should be paid for peer review is labour and it should be paid for peer review is labour and it should be paid for peer review is labour and it should be paid for peer review is labour and it should be paid for
Post-publication peer review: remedy or risk for publishing?
Scholars split on merits of post hoc reviewing as answer to increasingly “unsustainable” system
www.researchprofessionalnews.com/rr-news-worl...
Scholars split on merits of post hoc reviewing as answer to increasingly “unsustainable” system
www.researchprofessionalnews.com/rr-news-worl...
Post-publication peer review: remedy or risk for publishing? - Research Professional News
Scholars split on merits of post hoc reviewing as answer to increasingly “unsustainable” system
www.researchprofessionalnews.com
November 7, 2025 at 9:59 AM
peer review is labour and it should be paid for peer review is labour and it should be paid for peer review is labour and it should be paid for peer review is labour and it should be paid for peer review is labour and it should be paid for peer review is labour and it should be paid for
Today I tried to argue for the value of a pre-registered study even with deviations, analogising that it’s easier to trust a political party that deviates from its manifesto than one that never had one. Then I realised this isn’t true and that’s why open research practices need to be mandatory.
November 5, 2025 at 7:32 PM
Today I tried to argue for the value of a pre-registered study even with deviations, analogising that it’s easier to trust a political party that deviates from its manifesto than one that never had one. Then I realised this isn’t true and that’s why open research practices need to be mandatory.
One of my favourite things to do to entertain myself on days when doing research seems like an absolute drag is to ironically drink from this mug I found in a charity shop.
November 3, 2025 at 3:26 PM
One of my favourite things to do to entertain myself on days when doing research seems like an absolute drag is to ironically drink from this mug I found in a charity shop.
Reposted by Dr Matt Lloyd Jones (he/him)
'In February, Sheffield Hallam University, home to the Helena Kennedy Centre for International Justice (HKC), a leading research institution focused on human rights, ordered one of its best-known professors, Laura Murphy, to cease research on supply chains and forced labour in China.' 1/3
UK university halted human rights research after pressure from China
Exclusive: Leading professor at Sheffield Hallam was told to cease research on supply chains and forced labour in China after demands from authorities
www.theguardian.com
November 3, 2025 at 7:44 AM
'In February, Sheffield Hallam University, home to the Helena Kennedy Centre for International Justice (HKC), a leading research institution focused on human rights, ordered one of its best-known professors, Laura Murphy, to cease research on supply chains and forced labour in China.' 1/3
stuff of nightmares
November 3, 2025 at 8:41 AM
stuff of nightmares
Reposted by Dr Matt Lloyd Jones (he/him)
reminder... activistchecklist.org/links/
Why we need to remove everything after the
Plain language steps for digital security, because protecting yourself helps keep your whole community safer. Built by activists, for activists with field-tested, community-verified guides.
activistchecklist.org
October 22, 2025 at 6:36 PM
reminder... activistchecklist.org/links/
Reposted by Dr Matt Lloyd Jones (he/him)
i said to someone that talking about AI is a little like you've hired a shitty magician to perform at your kid's birthday party and also he has a gun. like many parts of his act are a fucking farce but also he might kill you and that's not because he's a fucking ✨amazing magician✨
October 6, 2025 at 3:02 PM
i said to someone that talking about AI is a little like you've hired a shitty magician to perform at your kid's birthday party and also he has a gun. like many parts of his act are a fucking farce but also he might kill you and that's not because he's a fucking ✨amazing magician✨