Jean Opsomer is the Program Chair for the Joint Statistical Meetings in Boston, which will take place August 2–7, 2014. He writes:
JSM 2014 (http://www.amstat.org/meetings/jsm/2014/index.cfm) will take place at the Boston Convention and Exhibition Center in Boston, Massachusetts, August 2–7, 2014. As I am writing this, the members of the JSM Program Committee are hard at work finalizing next year’s invited program. Together with the plenary lectures, these invited sessions will form the backbone of the 2014 JSM activities. But if you will forgive a bad analogy, what is the use of a backbone without the rest of the body? What makes the JSM a truly rewarding conference is not just these carefully selected sessions, but also the breadth of activities in which attendees can directly participate, ranging from continuing education short courses and roundtable luncheons to poster and contributed sessions.
Unlike conferences in many other disciplines, JSM is based on the principle that every participant should have the opportunity to present his/her work. That said, given the sustained growth JSM has experienced in recent years, we are getting close to the point where the size of the program becomes overwhelming for participants. I will return to this topic below.
The theme of 2014 JSM, laid out by ASA President Nat Schenker, is “Statistics: Global Impact—Past, Present, and Future.” This theme seeks to emphasize, celebrate, and share information about the contributions our profession has made, currently makes, and will continue to make to important problems in the world. The website for JSM 2014 already contains much information about the program, other JSM activities, and the host city—and more content is added frequently.
The next stage of preparation for JSM involves building the contributed program. This part of the JSM program includes poster sessions, SPEED sessions, topic-contributed sessions, and contributed sessions. I will describe each, but want to first draw your attention to the following important deadline. To present your work in any format, you must submit an abstract and register for the conference by February 3. The website above will provide instructions and links for the submission process.
Presentation of a poster is a terrific way to participate in the meeting. A poster affords you and your audience the chance to have a conversation about the work, making for a much more personal experience compared to presenting in a contributed session. While poster sessions are still a somewhat new format at JSM, they have become more popular and well attended, so I encourage you to consider this format.
This new type of session is a “hybrid” between a poster and presentation session that was introduced last year on a trial basis in collaboration with five ASA sections (i.e., Biometrics, Statistics in Epidemiology, Statistical Learning and Data Mining, Biopharmaceutical Statistics, and Survey Methodology). A speed session consists of 20 oral presentations of five minutes each, with 10 minutes of break after the first set of 10 talks. The short oral presentations are then followed by a poster session on the same day. All indications are that the trial was extremely successful, so SPEED sessions will again be featured next year.
A topic-contributed session is organized around a common theme and includes five participants. The session can entail five papers, or four papers with discussant, or three papers with two discussants. Topic-contributed sessions have several advantages over contributed sessions:
• They provide a more cohesive set of papers with a common theme
• They are limited to five participants, rather than seven as in a contributed session, allowing up to 20 minutes for each speaker to present
• They are labeled as topic-contributed in the program, providing more visibility and attracting larger audiences
Topic-contributed sessions require more work up front, because the organizer has to select a theme, invite speakers, and submit a session proposal prior to February 3. Starting with JSM 2014, the number of topic-contributed sessions is going to be capped, with each participating society and the ASA section given an allocation based on past numbers. This is being done to help contain the growth of the JSM program, which has been particularly pronounced in this category. Because of this, it is expected that a selection of proposals will be done by the JSM Program Committee. Proposals not accepted can be converted to contributed sessions.
The final category in the contributed program is the contributed sessions. These do not involve the up-front planning that characterizes topic-contributed sessions. To present a contributed paper, you simply have to submit a title and abstract, along with an indication of which ASA section is most closely associated with the topic of your paper. These sessions consist of seven papers, each presented in 15 minutes, and are put together by the JSM Program Committee.
In closing, let me return to the subject of the size of the JSM program. In Montréal, there were almost 700 sessions overall, with 47 going on at the same time during most of the conference. While this is a clear testimony to the vitality of the statistical discipline, it is also reaching the point where the program becomes unwieldy for both conference attendees and organizers. The Committee on Meetings, which is comprised of representatives of all the sponsoring societies, and the JSM Program Committee (see http://www.amstat.org/meetings/jsm/2014/program.cfm) have been considering ways to manage the growth of the JSM program while maintaining the overall quality and high level of access embodied in the contributed portions of the program. The SPEED sessions, in addition to being an innovative and fun format, are meant to reduce the number of contributed sessions. Similarly, the newly introduced cap on the number of topic-contributed sessions is intended to slow the growth of the program. The latter decision was not taken lightly and is likely to be somewhat controversial, especially among participants who might see their session proposal not accepted. As we continue grappling with the issues surrounding future planning of JSM, I encourage you to contact us with your feedback and suggestions.
This article originally appeared in the December 2013 issue of Amstat News.