Home » ASA Leaders Reminisce, Featured

ASA Leaders Reminisce: Sallie Keller

1 October 2015 No Comment
Jim Cochran

    In the 10th installment of the Amstat News series of interviews with ASA presidents and executive directors, we feature a discussion with 2006 President Sallie Keller.

    Sallie Keller

    Sallie Keller

    Q Sallie, thank you for taking time to talk with me. At one time, conventional wisdom was to think carefully about where you took your first position after completing your degree because moving into and out of academia in mid-career was extremely difficult. But you have successfully moved from academia to a federal lab and then back to academia. Do you think moving into and out of academia in mid-career has become easier? What advice do you have for students who are completing their degrees and hope to move across academia, government, and/or industry during their careers?

    A Anyone who knows me would likely say I have never followed conventional wisdom! I would like to believe it has become more porous to move between academia, industry, and government. I believe the barriers are less about technical fit and more about differences in salary structures. There are endless opportunities today given the data science boom.

    My advice to students and colleagues is to follow your passions and find the best place to realize them. Allow yourself to realign what you want to learn and accomplish during your career. Look for new challenges when you find yourself getting comfortable and even complacent. Don’t be afraid of change and developing new relationships. Keep your career network strong.

    Sallie Keller is professor of statistics and director of the Social and Decision Analytics Laboratory within the Virginia Bioinformatics Institute at Virginia Tech. She holds a PhD in statistics from Iowa State University of Science and Technology. Formerly, she was professor of statistics and academic vice president and provost at the University of Waterloo, director of the IDA Science and Technology Policy Institute, and professor of statistics and William and Stephanie Sick Dean of Engineering at Rice University. Her other appointments include head of the statistical sciences group at Los Alamos National Laboratory, professor and director of graduate studies in the department of statistics at Kansas State University, and statistics program director at the National Science Foundation.

    Keller has served as a member of the National Academy of Sciences Board on Mathematical Sciences and Their Applications and chaired the Committee on Applied and Theoretical Statistics and Committee on National Statistics. Her areas of expertise are social and decision informatics, the statistical underpinnings of data science, uncertainty quantification, and data access and confidentiality. She is a national associate of the National Academy of Sciences, a fellow of the American Association for the Advancement of Science, an elected member of the International Statistics Institute, and a member of the JASON advisory group. She is also a fellow of the American Statistical Association and a recipient of the ASA’s Founders Award.

    Q Almost two decades ago, you and Richard Becker published the article, “Presentation Myths,” in The American Statistician (1996). Since the publication of this article, technology has dramatically changed the way we give presentations. For example, PowerPoint was rarely used in 1996, but it is now ubiquitous. How would you update this article to reflect the changes in presentation technology over the past 20 years?

    A It is interesting you should ask this question. I picked that paper back up earlier this summer and was stunned at how relevant it is today, so much so I circulated it to our students and post docs. And, since you asked, I think we should remind everyone of the myths Rick Becker and I described:

    Myth 1: Presentations Require a Magic Number of Visual Aids. The ease of making slide-decks today requires even more thought to this point. More presenters should exercise the discipline of skipping slides when it is apparent they will never get to slide number 83. Also, be prepared to give your presentation without the slides in the event of technical failures, which is not uncommon today.

    Myth 2: The Audience Cannot Read. Don’t read your slides to the audience!

    Myth 3: Slides Are Preferred to Overhead Transparencies. Fortunately for statistics, slides never gained traction.

    Myth 4: Handwritten Computer-Generated Transparencies Are Preferable. Legible visual aids are preferable.

    Myth 5: Tabulated Data Are Informative. I remain firm on the point that if there is an important message embedded in a table, a graphical display will communicate it more effectively.

    Myth 6: Graphical Displays Are Obvious. Ha!

    Myth 7: Everyone Knows My Problem Is Important. Ha, ha!

    Myth 8: The Session Chair Will Not Cut Me Off. Ha, ha, ha!

    Myth 9: The Paper-Hiding Part of the Transparency Will Stay on the Overhead Projector. Today I would caution about the overuse of automation.

    Myth 10: A Vibrating Display Gets Their Attention. Here, too, I think our point is excellent and relevant, given the abundance of green and red lasers. Not knowing what else to do with our hands while we speak, it is natural to begin to point fast and furiously at something.

    What I would add to this paper is encouragement that you gain some comfort with extemporaneous discourse, both in formal presentations and in informal settings such as research team meetings. This is a learned skill and takes both courage and practice. The practice comes from doing and gaining a thick skin for the reactions. Remember it is never personal unless you decide to make it so.

    Q What were your first thoughts when you were asked to be a candidate for ASA president? What was your reaction when you learned you won the election?

    A First, I called both Jon Kettenring and Brad Efron for advice. It was like getting advice in time-delayed stereo—they both said you have no choice, run! This was followed with advising that I decide how I wanted to make a difference for the profession and simply do that. So, I accepted the nomination.

    I resolved, win or lose, that I would put forward a provocative statement clearly stating where I felt ASA needed to step up and lead. This was gaining a presence in science policy, as distinct from public relations. It would require expanding the ASA staff at a time when the association was feeling resource constrained.

    As you know, I was elected. I kept a sharp focus on this goal and am proud to say that, after lengthy discussions and debates, the ASA Board of Directors made science policy a priority. A full-time ASA position in science policy was created. It is currently filled by Steve Pierson and an additional position is to be added this year.

    However, it is not simply about adding staff to the ASA. Today, statistical sciences through the ASA leadership has become part of the voice of science policy. The ASA Board of Directors routinely issues science policy statements. The ASA participates with other leading science organizations to promote science with major stakeholders and has become an influential voice on Capitol Hill and in the president’s administration, advocating for the budgets and independence of the federal statistical agencies. Further, statisticians are more visible at the National Institutes of Health, the National Science Foundation, and the White House Office of Science and Technology Policy, making the case that statisticians improve scientific research. As one example of this increased visibility, the new U.S. chief data scientist, DJ Patil, recorded a message for attendees of the 2015 Joint Statistical Meetings extolling the work of statisticians. (Watch the video on YouTube.)

    Q In your presidential address at the 2006 Joint Statistical Meetings, you stressed the importance of statisticians becoming leaders in the science integration process. Has our discipline made substantial progress in this area since you gave this address? What can the ASA and its member statisticians do to become the stewards of the science integration process? What advice would you share with future candidates for president of the ASA?

    A You have now made me go back and read two papers I wrote! It is worth highlighting here what I meant by the “scientific process.” It means going beyond implementation of and protecting the scientific method. We need to put our work (research) into a decision context for policymakers to inform the leaders who shape the world. This ranges from world leaders to city managers.

    Lead by example is the advice I would share with future candidates for ASA president.

    My passion has been and remains engagement in the science process. During my career, I have studied natural disasters, terrorism, weapons of mass destruction, dual-use industrial innovation, technology insertion in developing economies, and urban analytics. Today, I have taken on the biggest challenge of my career in building a new lab, the Social and Decision Analytics Lab that is part of the Virginia Biocomplexity Institute at Virginia Tech. My lab is an intentional collection of statisticians and social scientists jointly focused on developing quantitative methods at a scale able to leverage all data to study the social condition. We are keenly focused on supporting policy and strategic decision-making under uncertainty. This requires team science, where the teams range from researchers to fire fighters to city demographers, all collaborating toward common goals to improve the quality of life where we all live, learn, work, and play. What could be more important?

    Please return to this column next month, when we will feature an interview with 1995 ASA President Mitchell H. Gail.

    1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)

    Comments are closed.