Peter Dal­gaard

Originally trained in Biostatistics (mostly with applications in Medicine), I joined the R project Core Team in 1997.  

This project aimed to provide researchers in both Applied and Theoretical Statistics with tools that were free to use and easy to extend. The project has been a huge success and is by now the “lingua franca” of statistical computing, with several million users and more than 20000 packages developed for it.  

The advent of R removed several frictions in the world of statistical methodology. At the time, annual software fees for a single-user license for mainstream statistical software could exceed the price of the computer itself, and dealing with site licenses for teaching was a major hassle. New statistical software often came in the form of standalone programs written for one platform and requiring major effort to get be runnable elsewhere. These issues were largely solve by R and the accompanying package system and the CRAN software repository.

My work in R has focused on release management, where I have been staging about four releases per year, while ensuring a stable and safe, not overly rushed path forward. Since R has applications in e.g. drug development, results of large investments can depend on correct statistical analysis, so it is important that we do our utmost to avoid introducing errors and keep user code working. On the other hand, the computing landscape is constantly changing, so we need to adapt to new hardware, etc.

Apart from my work on the R Core Team, one major contribution to the R world has been the book “Introductory Statistics with R” which came out originally in 2002 with a 2nd edition in 2009. Together the two editions have sold over 40000 copies and the book has been translated to Japanese and Chinese. I am currently working on a 3rd edition, to be published in 2026.