WORLD STATISTICS DAY: “Connecting the world with data we can trust” (also, Open Science)

It’s World Statistics Day, with the tagline “Connecting the world with data we can trust“. Yes, Tuesday 20 October is the day. (Also the birthday of Lindy, my wife. Maybe there’s a message there somewhere.)

Science is under attack around the world, and vital data are being ignored–or totally rejected. Time for a good news statistics story. My bedside reading is a recent issue of Significance (unfortunate title!) magazine, which goes to members of both the Royal Statistical Society (U.K) and the American Statistical Association.

It’s mostly behind a paywall for 12 months, but, happily, this article is a free download: Science after Covid-19, Faster, better, stronger? Dare we hope?!

Simon Schwab and Leonhard Held, of the Centre for Reproducible Science, University of Zurich, describe how this year:

• 30 publishers agreed to make Covid-19 research papers and data freely available–no paywalls
• Uploading of covid-related preprints exploded, as the figure above tells
• Quick action is encouraging rapid and open reviews of preprints, e.g. via Outbreak Science

Schwab and Held also discuss:

• The value of peer review before studies are conducted. Some journals offer registered reports, and aim to review study plans within 7 days.
• Ways that fast and high quality peer reviewing can be supported.
• The need for rigour and best-practice methods, as well as speed, and prompt systematic reviews. Then presentation of evidence-based advice for public policy and practice.

They conclude “courses in good research practice should be widely adopted to address highly relevant topics such as study design, open science, statistics and reproducibility … and preparation must also include the training of teams for rapid synthesis of relevant evidence. We cannot be prepared enough for the next global health crisis.”

In other words, Open Science! Bring it on–on World Statistics Day, and every day.

Geoff