Julia Watzek



Who are you, and what do you do?

Hallo! My name is Julia Watzek (Twitter: @watzoever, Github: jwatzek). I’m a cognitive & behavioral scientist wrapping up my PhD and venturing onto the job market. I study why humans and other animals make irrational decisions and how that might fit the social and physical environments we encounter. I like using data to tell stories and help people make better decisions.

How do you use R? What are your favorite packages?

R and I started out as coworkers when my professor in my very first undergrad statistics class made us use it – a sharp break from SPSS, which is usually taught in psychology. I don’t know how exactly but one day I realized that base R had moved in and I was in love. R has been my go-to for data analysis ever since, and the more I learn the more I feel that anything is possible with R. ggplot2 and the tidyverse in general coming around was mind-blowing and I recently started contributing to #TidyTuesday to practice data wrangling on different data sets than the ones I typically work with and to add more creativity to my data visualizations than academia usually allows.

Picking favorites is hard but I’ll give the honor to beepr, which lets you play a fun sound e.g. when your long simulation or analysis is done. It always makes me smile. My workhorse packages are the tidyverse collection with ggforce, ggtext, cowplot, and patchwork among others for data viz. My data usually call for multi-level regression models, for which lme4, emmeans, and rstan are my go-tos.

What hardware do you run R on? Do you use any other software tools?

I do most of my work on a 2012 [MacBook Pro])(https://en.wikipedia.org/wiki/MacBook_Pro) and some on an old Dell tower running Windows at the office, but am looking for an excuse to run more intensive work on my friend’s Threadripper build via remote access.

It’s hard to beat RStudio (& Notebooks) for R and Jupyter Lab (& Notebooks) for Python, but I use Sublime Text as an IDE for everything else (and often for R and Python as well).

Do you have any projects that you’re working on right now?

My dissertation, various side projects, and job applications! ;) I’ve recently done a lot more machine learning work, for which there are R packages, but I tend to use Python for this more. The reticulate package has been amazing in letting R and Python talk to each other in the same place. When I get a chance, I want to start putting some of that work into occasional short blog posts on my website.