In our series Interview With A Team Member, we aim to introduce the people behind the JASP-project. Today we are interviewing Johnny van Doorn, one of our analysts.
Johnny van Doorn is a PhD candidate at the Psychological Methods department of the University of Amsterdam. At JASP, he is responsible for Bayesian nonparametric analyses. To contact Johnny, you can send him an e-mail.
What is your professional background?
After some academic meandering in liberal arts, I decided to pursue a degree in psychology, so I went to the KU Leuven to do my bachelor and master. At first I wanted to venture more into social psychology, but when I realized the critical importance of proper statistics and how fun they can be, I started to focus more and more on psychological methods (special shout-out to Francis Tuerlinckx and Wolf Vanpaemel for aiding me in this adventure!). During my master I did an internship with EJ at the University of Amsterdam, which was very exciting and turned into a PhD project on rank-based Bayesian inference. I am currently in my third year and still enjoying it very much!
What is your favorite statistical test?
That has got to be my first love: Kendall’s tau, the rank correlation coefficient. My first publication is an article outlining Bayesian inference on this correlation, but besides that, the calculation and interpretation is very elegant and intuitive, which is something i greatly appreciate in statistical tests.
What is your relation to JASP?
My main PhD project consists of developing new Bayesian rank-based tests, such as rank correlations and the Wilcoxon tests. When such a test has been developed, I add it to JASP so people can very easily use these methods. So far the Bayesian test on Kendall’s tau has been implemented, and the Wilcoxon tests are soon to follow! Besides that, I also allocate some time towards general JASP bug-fixing on the R/statistics level.
What feature of JASP do you like best?
I really enjoy the JASP output: when the analyses have been conducted, the output can be saved as a single JASP file, which contains the raw data, all the (annotated) output, and the settings used to create the output. This makes the inferential process very transparent and shareable, especially when combined with the Open Science Framework functionality that is integrated in JASP.
What aspect of JASP would you like to see improved in a future version?
For some modeling, the current possibilities are a bit limited, so one feature I would love to see is a module that allows the user to specify their own statistical models through JAGS or WinBUGS syntax. I think this will greatly enhance the flexibility of what the user can use JASP for.
Are you a Bayesian, a frequentist, an agnostic, a pragmatist, or perhaps something else?
When faced with conventional tests where the Bayesian counterpart is easily available, such as a t-test or ANOVA, I would opt for the Bayesian version, as the interpretation of the test is just that much more straightforward and meaningful. At the same time I realize that there are cases where either the Bayesian counterpart does not exist, or suffers some limitations, so it’s not a very black and white issue.
What question would you like to answer?
What is your favorite dinosaur?
What is your favorite dinosaur?
The Anscombosaurus of course!
br>
Like this post?
Subscribe to our newsletter to receive regular updates about JASP including our latest blog posts, JASP articles, example analyses, new features, interviews with team members, and more! You can unsubscribe at any time.
br>