Q&A with Dianne Cook

Dianne Cook

Dianne Cook is the editor of the Journal of Computational and Graphical Statistics. We asked her to tell us a little about herself, the journal, and what we can expect to read in future issues.

Where did you grow up and go to school, and what or who inspired you to be a statistician?

I grew up in a small town called Wauchope on the north coast of NSW [New South Wales], Australia. We also had a weekender farm at Dongdinglaong, where we raised beef cattle, corn, and pumpkins. I had my own calves to tend to on a regular basis, as well as a few rescued kangaroo joeys. I played a lot of sport. Our little town high-school girls’ hockey team beat all the bigger schools in the state to be champions; I played the center half position. In my late teens, I was also featured in the town newspaper as “The Lady and the Willow,” the first woman to play on the local A grade men’s cricket team.

People who inspired me to become a statistician are many. We had one female math teacher in high school, who was absolutely gorgeous, with really hairy legs! She was part of the flee the big cities, live in a commune generation that popped up all over the area. I was rather in awe of her. At the University of New England, I was drawn to statistics as a major by a brilliant professor, Eve Bofinger. I had been accepted into Sydney University, but the city was very scary to me, so I chose the closest university near home to do undergraduate studies.

Meeting Deborah Swayne and Andreas Buja at Bellcore in New Jersey helped with the decision to do graduate studies focusing on data visualization. Also, a colleague at the time, Andrew McDougall, and his artist mother, Gretchen Albrecht, were very supportive.

Why did you become interested in being editor for JCGS?

I graduated just as JCGS was born, and my close colleagues were so excited to have a good outlet for graphics research. We resolved to flood the journal with graphics papers. It hasn’t really happened, and we still desperately need more graphics research in the field of statistics and an increase of graphics submissions to JCGS. The journal is very precious to me.

I have been asked several times to be editor over the years, and it wasn’t until my son went to college that I agreed to the job. It was a good decision. It is a very time-intensive responsibility. It would have been, for me, very difficult and stressful to have this responsibility while raising a family. I hope that, in the future, it is less difficult for women to juggle demands like this, because statistics journals need female researchers in decision-making roles.

Have you made any changes to the journal, or do you plan to make any in the next year?

There are numerous changes I have made to the journal operations:

  1. Reproducibility is a very important aspect of scientific research today. Articles need to submit code to reproduce results with the initial submission. Associate editors, or their assigned reviewers, are responsible for checking code runs and producing the results reported in the paper as a part of the review. This has been introduced gradually. JCGS’s original mission statement included a commitment to provide code and data associated with articles, so this is a natural progression. I would like to see more submissions that utilize reproducible document formats like Rmarkdown, where code, technical details, and writeup are all in one place. In the future, the journal should perhaps adopt a docker approach, which makes a virtual setup that enables code to be run in perpetuity.

  2. Data examples are a very important component of research in statistics to cement statistics firmly as a data science discipline. I have been strongly encouraging the use of contemporary data problems to illustrate new research. Papers relying on decades-old data sets are discouraged.

  3. I have actively pushed a number of submissions into a short technical note format. Papers around 10 pages can be reviewed quickly and published faster. The purpose of having short technical notes is also so small advances to prior work can be released to the community quickly. It hasn’t really taken off—there are not very many submissions, or converted submissions, in this format. I’d like to see more.

  4. The backlog of articles when I took the position was over a year, maybe closer to two years from acceptance to print. Clearing this backlog has been a priority.

  5. Because data visualization is my field of research, I have been working on improving the quality of the plots published in the papers.

  6. The R project has made a huge contribution to the practice of statistics globally. Many academic researchers have committed their lives to the service of making these computational tools readily available to the masses. I have been strongly encouraging additional citations of the software tools used to conduct the research reported in the papers. This acknowledgement, and the academic credit that citations provide, is important to build a strong foundation of new statistical computing researchers.

  7. For the annual meetings, I have pulled the full database of information about the articles published and created reports using the knitr package in R. Doing the analysis myself is very satisfying, with one reason being that I can decide on the appropriate plots and summaries to make. I have also been doing secondary analyses like gender matching the author’s names. This has enabled checking for bias in the review process. In the last year, we saw an increase in the number of submissions with a female author, which is very encouraging. We also have been able to determine that our reviewers are not sufficiently diverse, that there are very few female academics asked to review JCGS papers. It is an area that we hope to improve in the next year. I have also mapped the geographic distributions of authors and reviewers, and there is room for improvement in building a global community. This past annual meeting, we experimented to include participation of some AEs virtually. Our AEs are spread across the globe, and having the option of participating remotely—although not as good as face-to-face—is important for getting input and improving communication as a whole.

What do you find is the most enjoyable part of being a journal editor?

It is really exciting to see new research happening!

What do you find is the most challenging part of your job as editor?

Rejecting papers is the toughest thing. After a day of deciding which papers to send forward for review and which to reject immediately, I am quite emotionally drained. I also feel a little like Lucille Ball on the chocolate line. The papers keep on coming, and coming, and coming! Unlike the chocolates, I can’t just eat a few!

What do you enjoy doing in your spare time?

Playing and watching tennis. Reading.