Background

Judith ter Schure's interests lie in foundations of statistics as well as in applied work. She divides her time between research and consultancy (before at CWI and significanthelp.nl, now at Amsterdam UMC). Her general motivation is the effect of statistics on society, which also inspires occasional writing for a wide audience – previously published by De Correspondent – and participation in popular science events like Nacht van de Wetenschap, ScienceBattle and Pint of Science.

Research

Judith completed her Ph.D. dissertation on ALL-IN meta-analysis at CWI, which she defended successfully at Leiden University on April 7th 2022.  The dissertation can be found in the Amsterdam UMC repository, with an update published in January 2023 (details here).

An interview with a really great summary of her work on Accumulation Bias and ALL-IN meta-analysis was written by Nicos Starreveld for the 'In de verdediging' section of Nieuw Archief voor de Wiskunde (in English). A short news article about her Ph.D can be found on the Leiden University website (in Dutch).

ALL-IN meta-analysis helps scientists to prioritize research, interpret findings and design new studies – so efficiently accumulate scientific knowledge. The ALL-IN approach is bottom-up, especially suited to combine studies that have no common (top-down) stopping rule.

"Scientists need to make sure that all relevant trials are in when they hedge their bets on future research." [ALL-IN meta-analysis thesis]

ALL-IN meta-analysis is a statistical approach that is Anytime, Live, Leading and possible on INterim data. ALL-IN can be applied retrospectively as well as prospectively, to evaluate the evidence once or sequentially. Because the intention of the analysis does not change the validity of the results, the results of the analysis can change the intentions. Any ALL-IN meta-analysis can be turned into a living one, or even become ‘live’ or ‘real-time’ by including interim data from trials that are still ongoing.

The research is motivated by the 'replicability crisis' in empirical science and the movements to reduce Research Waste (http://rewardalliance.net/) and make Research itself more Evidence-Based (http://ebrnetwork.org/).

ALL-IN meta-analysis is part of the field of anytime-valid statistics based on e-values: E-values - Wikipedia

Read more about Judith's research in this interview with Het Parool/New Scientist (in Dutch), this longer interview with De Psycholoog (in Dutch) or this short Academic Story about her experience working at CWI (in English).

Open Science

All Judith ter Schure's scientific work is openly available. She publishes in preprints on ArXiv and MedRxiv, embraces the F1000 open access/open review publishing model and made all supporting code and scripts available on Easy Dans, OSF and ResearchEquals. Past articles can be found (Gold or Green open access) in the CWI repository and her articles for the Dutch Journal of Medicine (NtvG) are available in English translation in the VVSOR blog.

Contact

j.a.terschure@amsterdamumc.nl
@JudithterSchure@mastodon.social

statistics

meta-analysis, sequential analysis, adaptive clinical trial design, research waste