A very pretty collection of books. It is great to finally have the copies of your printed Ph.D. thesis in your hands! The is a link to the pdf of my dissertation:
Judith ter Schure - ALL-IN meta-analysis.pdf
I've dedicated the work to Glenn Shafer, Stephen Senn, Peter Grünwald, and Daniel Lakens. Three of them are obvious, given their guidance during my Ph.D. research, one is less obvious, as I've also dedicated my Ph.D. dissertation to Stephen Senn for professional guidance! Even though we've hardly ever met, his books and blogs were essential in my growing appreciation for clinical trial statistics. And it thought me another valuable lesson: our profession needs storytellers that can express its beauty, especially in The Netherlands, where hardly any new statisticians prepare for a career in clinical trials.
Below I post the chapter that contains my Curriculum Vitae, so it has the necessary dates and distinctions, but also how I came to love statistics, writing, and clinical trials.
Judith was born in 1992 in the hospital of Meppel. During her high school Gymnasium education at RSG Tromp Meesters in Steenwijk (SEPT 2004 - MAY 2010), choosing between the sciences and humanities proved too difficult. So she cobbled up a collection of subjects including mathematics, chemistry, physics, and biology, with economics and history – although she cheated the languages by choosing Latin for its dictionary and similarity to math and philosophy instead of brute force remembering words. She now has to look up how to spell "brute force". A spell checker and calculator stay must-haves to prepare any writing or calculating in public. Still, good preparation is good practice, as she learned to be a committee and board member of mountain sports associations (USAC, NSAC, NKBV) during her university studies. It is this experience that she still builds on in governing the professional society VVSOR and her homeowner's society of 32 owners and tenants; building towards ideals of professionalism, sustainable involvement of members/owners, and energy efficiency.
Judith has always been fascinated by systems of argument and decision making and wrote a high school dissertation on rhetoric ("profielwerkstuk", cum laude) and a Bachelor's thesis on logical fallacies. She had to admit that there are limits to the logic of language, however. Most decisions have to resort to data and do not have a written truth (to paraphrase the "In God we trust, all others bring data" W. Edwards Deming quote). This moment arrived during a Bachelor's program in Artificial Intelligence at Utrecht University (SEPT 2010 - MARCH 2014, cum laude) that combined formal logic, philosophy, linguistics, and psychology with mathematics and computer science. On a Leiden University website, she read that according to Hal R. Varian (Google’s Chief Economist, 2009) "the sexy job in the next ten years will be statisticians (sic)". She thereupon entered the Masters program in Statistical Science at Leiden University (SEPT 2014 - MARCH 2017, cum laude) and completed this Ph.D. dissertation with the Leiden Alma Mater as well (MAY 2017 - JAN 2022)*.
It proved very fortunate to find her Ph.D. position with professor Peter Grünwald in his Machine Learning group at an institute that encouraged uncompromising ideals: CWI, the Netherlands institute for mathematics and computer science. Peter shared her interest in statistics' influence on society, helped her believe in the sweeping possibilities of their research results, and above all, taught her the value of the mathematical precision to produce them (and was ruthless to help her learn from her mistakes). With Peter, she got a very thorough education into the fundamentals of statistics and how to ground their new approach. Applications presented themselves in the clinical sciences, motivated by research waste opponents and the early work of evidence-based medicine enthusiasts (often the same very inspiring people). She was drawn to clinical biostatistics and specifically the regulatory clinical trials she believed to be the realm of statistical decision-making, only to find out that "statisticians need to become more like data scientists". This was uttered by EMA head of data analytics Dr. Peter Arlett, at a regulatory statistics meeting in 2021. (EMA is the European Medicines Agency that she believed to be the center of statistical decision-making to regulate new medicines.) This interesting world of data had lost its appetite for statistical decision-making, while she had just gained that appetite!
The large amounts of data going into Covid-19 pandemic decisions had solidified her love for statistics nevertheless, especially the kind that designs well-controlled randomized trials. Many interesting statistical problems remain and solving them can immensely improve scientific and medical decision making, as she learned by reading blogs and books by Stephen Senn. These lured her back to language; to the many great stories that still need to be told. Some of the best decisions during the Covid-19 pandemic were made by those designing the Covid-19 vaccine trials, and effective treatment platform trials like RECOVERY. Honestly, if statisticians need to be like anything other than statisticians, it is like the doctors and funders that pushed for well-designed, large, and simple clinical trials. A bit more ambition and initiative wouldn't hurt the profession. It can be a very sexy job indeed.
*Four full-time equivalent years: between May 1st, 2017 and February 1st, 2022 she spent 44 months working 80% of her working week (approximately 35 weeks full-time equivalent) on this Ph.D. research and 13 months working 100%, so 48 full-time months in total.
Fri, 18 March