Stephen Senn
@stephensenn.bsky.social
Swiss & British statistician.
This www.slideshare.net/slideshow/ha...
Is from some years ago
Is from some years ago
Has modelling killed randomisation inference frankfurt
Stephen Senn discusses the evolution of statistical analysis from a design-driven to a more pragmatic approach, highlighting the tension between randomization inference and model-based inference. This document emphasizes the importance of resolving the dichotomy between these approaches, particularly in medical trials, and critiques the limitations of both methodologies. Senn argues for a framework that accommodates the strengths of both randomization and modeling to improve statistical inference. - Download as a PPTX, PDF or view online for free
www.slideshare.net
November 11, 2025 at 10:17 AM
This www.slideshare.net/slideshow/ha...
Is from some years ago
Is from some years ago
Fraudulent research should be withdrawn.
November 8, 2025 at 6:59 PM
Fraudulent research should be withdrawn.
OK So here’s a list of errors in meta-analysis in (mainly) Cochrane Collaboration work. bmcmedresmethodol.biomedcentral.com/articles/10....
I don’t think any of them have been withdrawn. Which ones should have been?
I don’t think any of them have been withdrawn. Which ones should have been?
Overstating the evidence – double counting in meta-analysis and related problems - BMC Medical Research Methodology
Background The problem of missing studies in meta-analysis has received much attention. Less attention has been paid to the more serious problem of double counting of evidence. Methods Various problem...
bmcmedresmethodol.biomedcentral.com
November 8, 2025 at 1:36 PM
OK So here’s a list of errors in meta-analysis in (mainly) Cochrane Collaboration work. bmcmedresmethodol.biomedcentral.com/articles/10....
I don’t think any of them have been withdrawn. Which ones should have been?
I don’t think any of them have been withdrawn. Which ones should have been?
Especially as a younger author I had many letters to the editors published pointing out errors in statistics. The original papers were never withdrawn and I never expected that they should be. My letter would nearly always be accompanied by an author’s reply. That’s what I think should happen.
November 8, 2025 at 7:28 AM
Especially as a younger author I had many letters to the editors published pointing out errors in statistics. The original papers were never withdrawn and I never expected that they should be. My letter would nearly always be accompanied by an author’s reply. That’s what I think should happen.
Some mistake surely. Any Frenchman can tell you “la philosophie ovale” is rugby.
November 6, 2025 at 8:10 AM
Some mistake surely. Any Frenchman can tell you “la philosophie ovale” is rugby.
Having worked in both academia and pharma, I have the following point of view. “You should treat all RCT based claims with caution, especially those coming from academia.”
November 3, 2025 at 7:38 AM
Having worked in both academia and pharma, I have the following point of view. “You should treat all RCT based claims with caution, especially those coming from academia.”
I decline to answer that.
November 3, 2025 at 7:31 AM
I decline to answer that.
I don’t necessarily disagree but you can make the inclusion criteria as wide as you like but it is no guarantee that you will get a wide selection of patients. If you do want to use covariates to define applicability, it’s the values of patients you did include not those you might have that matter.
November 2, 2025 at 6:41 PM
I don’t necessarily disagree but you can make the inclusion criteria as wide as you like but it is no guarantee that you will get a wide selection of patients. If you do want to use covariates to define applicability, it’s the values of patients you did include not those you might have that matter.
I think you’ll find that crowns, pounds, guineas, pearls and rubies are more traditional. www.poetryfoundation.org/poems/52706/...
When I Was One-and-Twenty
When I was one-and-twenty I heard a wise man say, “Give crowns and pounds and guineas But not your heart away; Give pearls away and rubies But keep your fancy free.” But I was one...
www.poetryfoundation.org
November 1, 2025 at 6:50 PM
I think you’ll find that crowns, pounds, guineas, pearls and rubies are more traditional. www.poetryfoundation.org/poems/52706/...
Brave of you to go West. I remember when I used to live in Fife, people would speak in hushed voices trying to hide their dread when they had to go West.
November 1, 2025 at 6:47 PM
Brave of you to go West. I remember when I used to live in Fife, people would speak in hushed voices trying to hide their dread when they had to go West.
It reminds me of the two sorts of manager. Those who run a tight ship and the other sort.
November 1, 2025 at 9:46 AM
It reminds me of the two sorts of manager. Those who run a tight ship and the other sort.
Heaven forfend that they should do anything that went against the grain.
October 31, 2025 at 8:41 PM
Heaven forfend that they should do anything that went against the grain.
And of the agriculturalists who like a hoedown you say nothing.
October 31, 2025 at 8:36 PM
And of the agriculturalists who like a hoedown you say nothing.
Let’s hear it for flawed paradox.
October 31, 2025 at 6:41 PM
Let’s hear it for flawed paradox.
I remember being taught to always draw a flow diagram before I wrote a program and so of course I always do that or at least I would do if I hadn’t lost that plastic template in 1973.
October 31, 2025 at 5:40 PM
I remember being taught to always draw a flow diagram before I wrote a program and so of course I always do that or at least I would do if I hadn’t lost that plastic template in 1973.
I think it sounds normal now but would have sounded strange some years ago.
October 31, 2025 at 12:32 PM
I think it sounds normal now but would have sounded strange some years ago.