Thomas Lumley
@tslumley.bsky.social
Biostatistician. Baritone. He/Him.
Product of more than one country.
May contain nuts.
Product of more than one country.
May contain nuts.
Auckland Museum in the distance behind engineering and trees
November 16, 2025 at 11:28 PM
Auckland Museum in the distance behind engineering and trees
Reposted by Thomas Lumley
Moral: Be careful what you ask for.
November 15, 2025 at 10:14 PM
Moral: Be careful what you ask for.
Reposted by Thomas Lumley
Reposted by Thomas Lumley
Interesting review paper on nonprob online samples by @jeremyfreese.bsky.social and Olivia Jin
www.annualreviews.org/content/jour...
www.annualreviews.org/content/jour...
Online Nonprobability Samples | Annual Reviews
Online nonprobability samples provide social scientists with opportunities to conduct surveys and experiments on large, diverse samples at modest prices. Researchers may find bewildering the options o...
www.annualreviews.org
April 11, 2025 at 9:51 AM
Interesting review paper on nonprob online samples by @jeremyfreese.bsky.social and Olivia Jin
www.annualreviews.org/content/jour...
www.annualreviews.org/content/jour...
Given that everyone has Epstein-Barr and very few people have lupus, it might be a necessary cause but it's surely not a sufficient cause
Every single case of systemic lupus erythematosus may be caused by the Epstein-Barr virus, according to a new study in Science Translational Medicine. It's estimated that 95% of people worldwide have been exposed to Epstein-Barr www.science.org/doi/10.1126/...
Epstein-Barr virus reprograms autoreactive B cells as antigen-presenting cells in systemic lupus erythematosus
Epstein-Barr virus reprograms autoreactive B cells as antigen-presenting cells to promote pathogenic antinuclear T and B cell responses in lupus.
www.science.org
November 15, 2025 at 12:46 AM
Given that everyone has Epstein-Barr and very few people have lupus, it might be a necessary cause but it's surely not a sufficient cause
Fence with load-bearing jasmine*
*ecologically-mandated fake jasmine substitute
*ecologically-mandated fake jasmine substitute
November 15, 2025 at 12:39 AM
Fence with load-bearing jasmine*
*ecologically-mandated fake jasmine substitute
*ecologically-mandated fake jasmine substitute
Double machine learning fits an adjustment model from a confounder set using correlation and so breaks down when offered variables that have good correlation just *because* they aren't suitable adjustment variables
#IWouldSimplyNotIncludeEndogenousVariables
#IWouldSimplyNotIncludeEndogenousVariables
Maybe you've missed the crushing, DAG-wielding critique. Of course, not by a star scholar... www.degruyterbrill.com/document/doi...
Double machine learning and automated confounder selection: A cautionary tale
Double machine learning (DML) has become an increasingly popular tool for automated variable selection in high-dimensional settings. Even though the ability to deal with a large number of potential co...
www.degruyterbrill.com
November 14, 2025 at 9:39 PM
Double machine learning fits an adjustment model from a confounder set using correlation and so breaks down when offered variables that have good correlation just *because* they aren't suitable adjustment variables
#IWouldSimplyNotIncludeEndogenousVariables
#IWouldSimplyNotIncludeEndogenousVariables
Reposted by Thomas Lumley
Spinoff asked local tyre shops where the bed places to eat are. And it's clear tyre sellers are the experts on ranking restaurants.
thespinoff.co.nz/kai/13-11-20...
thespinoff.co.nz/kai/13-11-20...
Michelin who? The finest places to eat in Aotearoa, according to local tyre shops
Tread experts across the country share their best local culinary tips for free.
thespinoff.co.nz
November 14, 2025 at 6:30 PM
Spinoff asked local tyre shops where the bed places to eat are. And it's clear tyre sellers are the experts on ranking restaurants.
thespinoff.co.nz/kai/13-11-20...
thespinoff.co.nz/kai/13-11-20...
"Negotiations between the Council of Australasian University Librarians and academic publishing giant Elsevier have broken down"
www.rnz.co.nz/news/nationa...
www.rnz.co.nz/news/nationa...
Universities in 'battle of the century' with journal publisher Elsevier
One New Zealand university told its staff all universities in New Zealand and Australia would "lose some degree of access" to the publisher's 1600 titles from the start of next year.
www.rnz.co.nz
November 14, 2025 at 5:59 AM
"Negotiations between the Council of Australasian University Librarians and academic publishing giant Elsevier have broken down"
www.rnz.co.nz/news/nationa...
www.rnz.co.nz/news/nationa...
Reposted by Thomas Lumley
From Matt Tyler at Rice on measuring support for beliefs in conspiracies: One idea is to look at whether respondents report consistent beliefs over multiple questions in the same survey, with different wording.
(Also: bonus Olympics content!)
(Also: bonus Olympics content!)
November 13, 2025 at 5:39 PM
From Matt Tyler at Rice on measuring support for beliefs in conspiracies: One idea is to look at whether respondents report consistent beliefs over multiple questions in the same survey, with different wording.
(Also: bonus Olympics content!)
(Also: bonus Olympics content!)
The world's most honest conference ad image
The @royalstatsoc.bsky.social annual international conference is now on bluesky and looking forward to our conference in Bournemouth next September: rss.org.uk/training-eve... #RSS2026conf
November 13, 2025 at 5:49 PM
The world's most honest conference ad image
Reposted by Thomas Lumley
@bsky.app ISSUE 4 IS OUT! with 3 survey methods and 2 statistics articles. We've got a backlog of 20 accepted articles -- some in preprint now -- but we're seeking more manuscripts for 2026, especially stat papers. So please consider submitting your next manuscript to JSSAM. #AAPOR
November 12, 2025 at 11:17 PM
So, when a scientific journals says "expression of concern", they don't mean 🤔 or 🙄.
They mean more like 🤯🤮🤬
They mean more like 🤯🤮🤬
November 12, 2025 at 10:59 PM
So, when a scientific journals says "expression of concern", they don't mean 🤔 or 🙄.
They mean more like 🤯🤮🤬
They mean more like 🤯🤮🤬
Christmas in the science atrium:
Star ✅
Red floofy balls ✅
Flashing lights ✅
Hearts with NZ birds ✅
Hearts with pohutukawa ✅
Jandals ✅?
Australian flowers (Grevillea)❔
Bombs or maybe giant glitter avocados❓
Star ✅
Red floofy balls ✅
Flashing lights ✅
Hearts with NZ birds ✅
Hearts with pohutukawa ✅
Jandals ✅?
Australian flowers (Grevillea)❔
Bombs or maybe giant glitter avocados❓
November 12, 2025 at 12:25 AM
Christmas in the science atrium:
Star ✅
Red floofy balls ✅
Flashing lights ✅
Hearts with NZ birds ✅
Hearts with pohutukawa ✅
Jandals ✅?
Australian flowers (Grevillea)❔
Bombs or maybe giant glitter avocados❓
Star ✅
Red floofy balls ✅
Flashing lights ✅
Hearts with NZ birds ✅
Hearts with pohutukawa ✅
Jandals ✅?
Australian flowers (Grevillea)❔
Bombs or maybe giant glitter avocados❓
Reposted by Thomas Lumley
If you're wondering why @bmj.com haven't already retracted this debacle: since 1989 Retraction Watch shows they have retracted a grand total of 7 research articles, so maybe they don't know how to.
This is one of the most remarkable academic debacles I've ever seen.
A large RCT got published in BMJ. There are currently 44 Pubpeer comments, mostly about the data, including...well. Read for yourself.
pubpeer.com/publications...
A large RCT got published in BMJ. There are currently 44 Pubpeer comments, mostly about the data, including...well. Read for yourself.
pubpeer.com/publications...
PubPeer - Prevention of acute myocardial infarction induced heart fail...
There are comments on PubPeer for publication: Prevention of acute myocardial infarction induced heart failure by intracoronary infusion of mesenchymal stem cells: phase 3 randomised clinical trial (P...
pubpeer.com
November 11, 2025 at 4:19 PM
If you're wondering why @bmj.com haven't already retracted this debacle: since 1989 Retraction Watch shows they have retracted a grand total of 7 research articles, so maybe they don't know how to.
Reposted by Thomas Lumley
It is truly shite that after running NCEA level 2 for 20 years NZQA still haven’t figured out how to get images into past exam papers. (And also that exemplars are images of handwritten scripts. Accessibility anyone?)
November 10, 2025 at 7:33 PM
It is truly shite that after running NCEA level 2 for 20 years NZQA still haven’t figured out how to get images into past exam papers. (And also that exemplars are images of handwritten scripts. Accessibility anyone?)
Reposted by Thomas Lumley
The New York Times profiles Gandalf, the Northland green fairy facing charges of cultivating, possessing and selling cannabis and cannabis oil. We need to change our cannabis laws. (This is a gift link you can read for free) www.nytimes.com/2025/11/10/w...
‘Gandalf’ the Medical Marijuana Grower: Folk Hero or Criminal?
www.nytimes.com
November 10, 2025 at 6:54 PM
The New York Times profiles Gandalf, the Northland green fairy facing charges of cultivating, possessing and selling cannabis and cannabis oil. We need to change our cannabis laws. (This is a gift link you can read for free) www.nytimes.com/2025/11/10/w...
Reposted by Thomas Lumley
Once again, Jason shows that Loeb is full of crap.
I've summarized the truth about Loeb's 10 "anomalies" about 3I/ATLAS in one post.
Thanks to @deschscoveries.bsky.social @michael-w-busch.bsky.social @cometary.org and @marshall-eubanks.bsky.social for contributing their expertise!
Thanks to @deschscoveries.bsky.social @michael-w-busch.bsky.social @cometary.org and @marshall-eubanks.bsky.social for contributing their expertise!
Loeb’s 3I/ATLAS “Anomalies” Explained
Avi Loeb continues to claim that 3I/ATLAS has many anomalous behaviors that lead to the conclusion that it “might” be an alien spacecraft. He carefully hedges the probability that it is a spacecraft ...
sites.psu.edu
November 10, 2025 at 6:32 PM
Once again, Jason shows that Loeb is full of crap.
Chris Thile has a new Bach Sonatas & Partitas album out. It's much less traditional than the first one. I think I like it?
November 10, 2025 at 6:17 AM
Chris Thile has a new Bach Sonatas & Partitas album out. It's much less traditional than the first one. I think I like it?
Reposted by Thomas Lumley
Stats NZ is “seeking a forward-thinking AI Privacy Advisor to lead the development and enforcement of privacy frameworks across our AI systems and data-driven operations. This role is critical in ensuring ethical AI deployment,”
jobs.govt.nz/jobtools/jnc...
jobs.govt.nz/jobtools/jnc...
Senior Privacy Advisor | Kaitohutohu Matua Tūmataiti | Wellington | NZ Government Jobs
Senior Privacy Advisor | Kaitohutohu Matua Tūmataiti | Wellington | NZ Government Jobs
jobs.govt.nz
November 10, 2025 at 2:22 AM
Stats NZ is “seeking a forward-thinking AI Privacy Advisor to lead the development and enforcement of privacy frameworks across our AI systems and data-driven operations. This role is critical in ensuring ethical AI deployment,”
jobs.govt.nz/jobtools/jnc...
jobs.govt.nz/jobtools/jnc...
Reposted by Thomas Lumley
What happened when I dug into some of the key studies and ideas quoted in The Tipping Point and its sequel… kucharski.substack.com/p/the-real-r...
November 9, 2025 at 1:44 PM
What happened when I dug into some of the key studies and ideas quoted in The Tipping Point and its sequel… kucharski.substack.com/p/the-real-r...
Reposted by Thomas Lumley
This profile in ‘Significance’ on DuckDB co-founder Hannes Mühleisen is quite interesting, and has helpful insights about data quality and the changing meaning of “big data.” Also some good professional advice in here for statisticians.
academic.oup.com/jrssig/artic...
academic.oup.com/jrssig/artic...
Is big data dead?
Abstract. Data, ducks and statistics – Sandra Alba gathers dispatches from Amsterdam and Auckland
academic.oup.com
November 9, 2025 at 5:47 PM
This profile in ‘Significance’ on DuckDB co-founder Hannes Mühleisen is quite interesting, and has helpful insights about data quality and the changing meaning of “big data.” Also some good professional advice in here for statisticians.
academic.oup.com/jrssig/artic...
academic.oup.com/jrssig/artic...
Reposted by Thomas Lumley
Approiximately my Belz Lecture in Melbourne from a couple of weeks ago (there may be a recording later?)
notstatschat.rbind.io/2025/10/30/l...
notstatschat.rbind.io/2025/10/30/l...
Laws and Orders - Biased and Inefficient
notstatschat.rbind.io
November 9, 2025 at 6:15 AM
Approiximately my Belz Lecture in Melbourne from a couple of weeks ago (there may be a recording later?)
notstatschat.rbind.io/2025/10/30/l...
notstatschat.rbind.io/2025/10/30/l...
Approiximately my Belz Lecture in Melbourne from a couple of weeks ago (there may be a recording later?)
notstatschat.rbind.io/2025/10/30/l...
notstatschat.rbind.io/2025/10/30/l...
Laws and Orders - Biased and Inefficient
notstatschat.rbind.io
November 9, 2025 at 6:15 AM
Approiximately my Belz Lecture in Melbourne from a couple of weeks ago (there may be a recording later?)
notstatschat.rbind.io/2025/10/30/l...
notstatschat.rbind.io/2025/10/30/l...
"Oh come on! That was not six months" -- Persephone, seeing Christmas decorations
November 9, 2025 at 3:06 AM
"Oh come on! That was not six months" -- Persephone, seeing Christmas decorations