By: Jonathan Porat, Regulatory Economist
At the beginning of August, some of the biggest rock stars in the world descended on Chicago for Lollapalooza– a multi-day massive world-renowned music festival. Arguably more excitingly, that same week, some of the biggest rock stars in the world of statistics and data visualization came to Chicago for the 2016 Joint Statistical Meetings (JSM)– the Lollapalooza of statistics and data visualization.
The JSM is a conference of all of the major US and international statistics organizations. It boasts having over 6,000 attendees and over 600 sessions on topics ranging from new developments in statistical software programs to ways to better communicate information through graphics. Advocacy was invited to participate. Research Economist Richard Schwinn presented his contribution to Advocacy’s cutting-edge transformation of its State Economic Profiles. Data Visualization Designer Sarah Coleman, Regulatory Economist Jonathan Porat, and Research Economist Miriam Segal attended the JSM to expand their analytical capabilities and help promote some of Advocacy’s great research and design work.
Advocacy presented as part of a larger session on Administrative Records and Data Disclosure that was sponsored by the Government Statistics Section of the Committee on Applied Statisticians. The session was chaired by Statistician and Mathematician Jioashen You of the US Department of Transportation and included presentations from other government agencies on the role of data presentations in government. Jessica Helfrand of the Bureau of Labor Statistics presented on the policy implications of methodological changes in its quarterly employment and wage data. Erin Tannenbaum from the National Opinions Research Center (NORC) at the University of Chicago described how data visualization of the Centers for Disease Control data can promote good policy by quickly and effectively determining geographic disparities in key indicators of good health.
Schwinn presenting at JSM.
Schwinn presented his contributions to the 2016 State Economic Profiles by focusing on the value of reproducibility in government data and how he was able to use statistical software to improve the reproducibility of Advocacy’s publications. His presentation was greatly received by the audience and catalyzed an interesting discussion between other government practitioners and academics in the audience. The audience was particularly interested in Schwinn’s presentation of a simple statistical software template that could be applied in future research to maintain high standards in transparency and clarity.
In addition to Schwinn’s fascinating presentation, Advocacy attended many sessions to improve its research capabilities. These included:
- A course in best practices in data visualization taught by Teresa Larsen of the Foundation for Scientific Literacy. The course demonstrated ways to better communicate statistical information through shape and color. Advocacy will apply these techniques to improve its graphics’ readability.
- A presentation from Carson Sievert, a PhD student from Iowa State University, explaining how to make interactive web graphics using statistical software packages. The presentation included a variety of basic graphing techniques and a detailed discussion of the positives and drawbacks of each approach.
- A case study on What to Do with Messy Data? by John Emerson from Yale University. The case study discussed statistical techniques, particularly those related to pattern matching and regular expressions, which have already saved Advocacy time and resources in producing data-driven publications.
- A meeting with Yihui Xie, the (statistical software program) R pioneer who brought literate programming to the R language thus making the 2016 Small Business Profiles redesign possible. His presentation on making interactive web-based visualizations inspired countless possibilities for potential future projects.
- A talk by Andrei Sheinkman from FiveThirtyEight focusing on the role that data visualization plays in the news. He noted that his graphically advanced organization uses proprietary software and open source web tools, and uses Adobe Illustrator to edit most charts.
- A presentation by Jeffrey Leek of Johns Hopkins University on two experiments that explored the influence of wording and visual clarity on how participants interpreted data. The researchers found a small change in presentation led to participants interpreting the same data differently across a variety of dimensions. In some cases even with training these presentational changes had a significant impact.
A few of the impressive data visualization examples that were at the conference.
Advocacy was privileged to participate in and contribute to the 2016 JSM. It will be exciting to see how all the valuable information and experience from the JSM will be crystalized into future publications. Advocacy looks forward to continuing to contribute its great work and expand its research capabilities at more events in the future.