Heather Turner, the useR! 2014 Interview

Heather Turner is a biostatistician and Associate Fellow at the University of Warwick, as well as a consultant. She is a former editor of the R Journal, was a local organizer for the useR! 2011 conference, and holds a wealth of information about the field of data science, the state of the technology, and the wider community.

As an academic with roots outside of the traditional Computer Science pathway, Heather holds fascinating insights on what makes the R programming language interesting and different. She became familiar with R during her PhD studies as an user, and a few years later she developed the Generalized Non-Linear Models package which is still in use today. Our conversation, in the video above, begins with this transition from R user to R developer.

There is certainly more in this video than I could mention here – don’t miss her new work using Shiny in statistical education and how to best encourage younger generations to enter the field. This is genuinely a must-watch video, full of great information combined with level-headed advice. I hope you find it as enlightening as I have!

VIDEO BREAKDOWN

Please click here to see the video directly on YouTube. Or if you prefer, subscribe to our podcast to get the audio!

The Biostatistics Workflow

Heather’s workflow would seem non-standard to R developers in biostatistics, for a variety of reasons. Her background in developing the CRAN Generalized Non-Linear Models package and her partnership with stricter, software-oriented collaborators set her upon on a slightly different track, for example. Her description of how she accomplishes her work is fascinating – including the fact that she prefers Emacs over RStudio!

Women, Minorities and R

A pleasant surprise at my first R conference (useR! 2014) was the seemingly proportional representation of women and minorities who use R for their work. Computer programming has historically struggled to truly include people of color and women, making it refreshing to see that R and the larger data science community indeed come from a variety of backgrounds. In this part of the video, I ask Heather how we can keep this trend moving towards the positive, along with her thoughts on the Women in R panel hosted at useR! 2014.

The Process of Package Development

One of the great features of R is CRAN, a repository of hundreds of packages providing pre-tested, pre-built, ready to run solutions for issues ranging from model building to data cleansing, interfacing to databases, and everything in between. Because Heather mentioned package development a number of times during the interview, I decided to dive a little deeper to better elucidate the process of package development in R.

Consulting

On top of her academic work, Heather is also quite active as a consultant. For those interested in what the consulting opportunities available to an R superstar can look like, she provides high-level insight into her current work and where these kinds of opportunities can be found.

Share This Post

1 Comment

  1. Thomas Speidel - September 15, 2014

    I’m truly enjoying the podcasts so far. I think Eduardo is doing an awesome job!

    I can’t help but notice an underlying element of surprise on the part of Eduardo – not just in this podcast, but in several others as well – at the realization that so many of the R titans and contributions come from the field of biostatistics, and that they have been using R (and S-PLUS before it) for several years if not decades. I feel that many coming from a programming background fail to realize R’s heritage or dismiss its history as one relegated to academic settings and perhaps not worthy of a production environment. But biostatistics is a very applied flavour of statistics, used to solve real and practical health related problems.

    See p.46 of Frank Harrell’s graphical course, where you can see what can be considered a precursor to Shiny done in S-PLUS (I suspect in the late 90’s – early 2000’s):
    https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CB0QFjAA&url=http%3A%2F%2Fbiostat.mc.vanderbilt.edu%2Fwiki%2Fpub%2FMain%2FStatGraphCourse%2Fgraphscourse.pdf&ei=hR4XVPqsF46wogSLyIH4Bg&usg=AFQjCNFHXqEg2OFJAK2qcrR5a-JFsFP3lA&bvm=bv.75097201,d.cGU

    David Smith sort of alluded to that, but didn’t go into details; so did Max Kuhn. So, I’m very happy that Eduardo is giving due credits to the serious contributions this and other fields has developed over the years, even long before R existed.

Leave a reply