In many internal company meetings I would be the only person in the room without the right to use the title of Doctor - in which I include physicians, as well as lawyers, the latter having that right in parts of Europe. It was often the case that I would be talked down to. Though unjustified it hurt
I worked for 10 years in industry with an MSc before undertaking a part-time PhD. Technically, I felt no need of a PhD, my future supervisor assuring me that the statistical research I was undertaking in industry was at, or above, PhD level. My reason was more prosaic.
Not been a great fan of -isms since the emergence of Thatcherism.
New Year's Eve Fondue
Samuel Galton of Birmingham ?
I will be adding this to my extensive collection of non-significant excuses / significant near-misses :
" ...lagged values of Padding Time fall just shy of having a statistically significant impact on Minutes Late at Arrival with a p-value of 0.102."
From: doi.org/10.1080/0003...
This sounds familiar. Here's a referee's report from the German journal Archiv der Toxicologie in 1983:
Agreed. I introduced the term "Eierlegendewollmilchsau” (Egg-laying-woolly-milk giving pig) into adaptive design literature (doi.org/10.1080/1054...) to describe a clinical trial that can answer many questions but none of them well
"I would like to predict that in the next century statisticians will he one-third Bayesian, one-third data analyst, and one-third scientist, i.e. they will view statistical theory and practice and scientific background as a single entity."
Tom Leonard, 1983
"It has been stated by many that the first month of our presidency... is the most successful in the history of our nation"
Donald Trump, Joint address to Congress on March 4, 2025.
Sound familiar !
Both could be true.
and played the opening song from his set Tonight's the Night a new song about the death from heroin of a long-time roadie. The only saving grace was that the band opening for him was the Eagles on their 1st UK tour who were brilliant.
That reminds me of a 1973 Neil Young concert at the Rainbow in London. He had had a huge hit with Harvest the year before bur refused to play those songs. At one point when the audience was booing and demanding Harvest songs he said "I'll play you a song you've heard before" ...
does not represent the probability that the null hypothesis is true. Is it our fault as a profession for not explaining, is it their fault for not understanding, or is it the fault of the concepts themselves?"
A.P.Grieve (1992) Royal Statistical Society News and Notes, 18(7), 3-4.
From the same source as yesterday: "It is 43 years since the recognised beginning of the modern era in clinical statistics; 43 years since Bradford Hill was successful in introducing Fisherian ideas into medical research, yet 43 years in which clinicians have still not grasped that a p-value
Pain in the arse, right?
"If we are unable to educate clinicians then merely persuading them to use CI's rather than p-values is to replace
the unthinking use of one technique with that of another."
A.P.Grieve (1992) Royal Statistical Society News and Notes, 18(7), 3-4.
The discipline of statistics itself cannot influence, cannot persuade, cannot interact, cannot design, cannot analyse and cannot interpret. Statisticians can and do."
A.P.Grieve (2006). The professionalisation of the shoe clerk. JRSS Series A, 158, 639-656.
"Many of the addresses have tackled issues that are associated with widening the influence of statistics- essentially through outreach. One aspect of this laudable objective that is not always considered is the statisticians themselves....
At times I've used all three, but I'm a medievalist at heart.
Couldn't agree more. Grieve's 1st Law of Influential Statisticians: design, design, design = D to the power 3.
doi.org/10.1002/pst.5
I'm convinced the solution is to train more statisticians and include them as team members. Had this discussion with a UK MRC toxicologist 20 years ago who wanted more statistical support to prevent the copying from subject matter journals of incorrect, out-of-date methods.
From the New York Times.
Used to make these at University in the early 70s. Delicious.
Sad to hear of the passing of Danny Thompson. First heard him with Alexis Korner's Blues Incorporated in the 1960s, followed by Pentangle. It was only couple of weeks ago that I listened again to the BBC recordings of his trio with John McLaughlin from 1967.
Fieller's theorem by Edgar Fieller.
Should it be pronounced F - ee- ller's theorem or F - eye- ller's. I know how his son Nick, statistician at Sheffield University, pronounced it.
I remember seeing "Barefoot in the Park" at boarding school, and skipping out of school on a Satrurday night to go and see Butch Cassidy and the Sundance kid with these guys in 1968, it's me on the left, in white jumper.
Back in the Netherlands, for September. 1st beer of the trip.
I checked my own papers to make sure I haven't committed this sin. I haven't yet. I have used the phrase "seminal paper" three times in describing Armitage, McPherson and Rowe's paper on multiple sequential testing, Ross Prentice's on surrogate endpoints and Harold Wikensky's on professionization.
