Stephen Senn

Looking N from Piz Scalottas to Piz Danis (L) and Piz Raschil

Looking S from Piz Scalottas

I am in Lenzerheide in canton Graubunden #skiing #Switzerland

Fascinating stuff. However, IMO, the value of double blind trials is not just dealing with patient- centred biases but physician-biases and this is often overlooked. www.linkedin.com/pulse/blind-...?

However, you can console yourself with the thought that nearly all of those you meet will have no idea what a Chorley Cake is, let alone have tasted one.

I believe the ultimate authority on whether accents were perceived as mellifluous or not was professor John Honey. I read his book, Does Accent Matter?, more years ago than I care to remember. I think his advice to all was, if you want to be understood, develop a paralect to your dialect.

Bruce or Sheila?

My wife, son and daughter have all independently visited the island. S & D both did the NW trek solo. Daughter Helen spent her 19th birthday (in June) alone on the trek. I am the only wimp in the family, it seems

Lots of new followers recently, so it's good to remind you that I wrote a book all about data centres, AI and local communities and software developers fighting back. It's available all over and here's a handy link for digital and print books2read.com/humansoftware #booksky

Death of the Author: How Postmodernism Changed Literature and Media In 1967, French literary theorist Roland Barthes published his seminal essay The Death of the Author, arguing that a text’s meaning is not…

But not in postmodern publications.

medium.com/@philosophy....

You won’t feel a thing. It’s just a ….

Die, Dichotomy We have studied 21 435 unique randomized controlled trials (RCTs) from the Cochrane Database of Systematic Reviews (CDSR).Of these trials, 7224 (34%) have a continuous (numerical) outcome and 14 211 (...

Die, Dichotomy.
www.linkedin.com/pulse/die-di...
A brief post encouraging you to read the original paper with @erik-van-zwet.bsky.social and @f2harrell.bsky.social .
onlinelibrary.wiley.com/doi/epdf/10....

Possibly a confusion over the law of the excluded muddle.

“They who wish to take up statistics must choose whether to be frayesian or bequentist.” Sayings of Confuseus

Nothing Rhymed - Wikipedia

Nothing unrhymed? Whatever happened to Gilbert O’Sullivan?

en.wikipedia.org/wiki/Nothing...

How do you react to assonance?

Dining philosophers problem - Wikipedia

What would Dijkstra say?

en.wikipedia.org/wiki/Dining_...

A Random Trio Three kinds of random effects in designed experiments The analysis of variance in the paper, however, is very defective. I suspect that Fisher did not give his full attention to it.

I am not entirely happy with this because the consequence of declaring an effect as random can depend on whether it is of a block, a treatment or an interaction of the two.

www.linkedin.com/pulse/random...?

After all, there are a peck of data-idiots in the world and these days some of them will be “Bayesians”.

In fact I can only make sense of Frank’s statement if he is comparing a good Bayesian analysis to a default frequentist one. OTBE I’d prefer an analysis by Frank (which would be Bayesian) to some random frequentist alternative but the good old frequentist notion of confounding applies.

In fact the situation is so bad that I am going to have to write a blog about and probably a letter to the editor.

I have sat on data safety-monitoring boards for flexible trials (sigh) and can assure you that the good old frequentist trick of declaring a block for every allocation rule and conditioning on it is easier to understand than the Bayesian hand-waving being offered instead.

Dundee restaurant.
Waiter: Will ye hae a sweet?
Diner: Yes ell he a sweet.
W: will ye hae an aiple peh or a meringue?
D: nae yer not wrang, ell he the aiple peh.

Logic and methodology
Theory of statistics
Fortran programming
Demography
Economics of Innovation

Another Approval and More NextGen Covid Vax News (Update No 37) - Absolutely Maybe This month, Arcturus’ self-amplifying mRNA vaccine was approved for use in the UK. There were also new clinical trial results for another…

Lots of news in my latest next generation Covid vax update:

- Self-amplifying mRNA vax Kostaive approved in the UK

- New trial for mucosal vax starting in Finland

- Results from AstraZeneca phase 1 trial for mRNA vax ...

absolutelymaybe.plos.org/2026/01/31/a...

1/2

#Vaccines #Covid #Covid19

When it comes to dividing distributions into four equal parts may the fourth be with you and give no quarter to those who refer to them as quartiles.

A Random Trio Three kinds of random effects in designed experiments The analysis of variance in the paper, however, is very defective. I suspect that Fisher did not give his full attention to it.

Especially since there are (at least) three types of random effect.

www.linkedin.com/pulse/random...?

Don’t get me started on Senntiles

View of W Kipp

W Kipp (L) and E Kipp (R) from the snow speckled slopes of S Black Hill

Rocky path that leads SE from White Craig Heads

Looking back at Carnethy Hill a wreath of cloud near the summit.

Today’s hike was from 8 mile burn to Flotterstone via S Black Hill, Scald Law, Carnethy Hill & the martyr’s monument at Rullion Green.

Indeed, but it’s not just fancier it’s wrong. If you use “quintile” to mean “fifth” what do you call the four boundaries that divide the data into five equal parts?

Slide 35 builds to show that the dose for children is exactly the same as for adults. Pharma is still completely bedazzled by using omics to personalise medicine but ignores simple information like age and weight.

See slide 35 www.slideshare.net/slideshow/pe...

Interesting postdoc opportunity in biostatistics at the University of Zurich to work on patient recruitment in clinical trials data.snf.ch/grants/grant... with Malgorzata Roos
www.ebpi.uzh.ch/en/research/...
#biostatistics #postdoc #Switzerland