James Cochran

As reported in the October 2014 issue of Amstat News, the ASA’s quartoseptcentennial celebration featured the roundtable breakfast “Past Presidents and Executive Directors Reminisce” that provided past and current presidents/presidents-elect and executive directors an opportunity reflect on their terms in office. Because ASA members have shown a great deal of interest in this event and in these reflections, Amstat News established a monthly column that will feature extended interviews of ASA past presidents and executive directors. In this initial column, we feature an interview with 2013 ASA President Marie Davidian.

Marie Davidian is William Neal Reynolds Professor of Statistics at North Carolina State University (NCSU) and adjunct professor of biostatistics and bioinformatics at Duke University. She earned her PhD in statistics in 1987 from The University of North Carolina at Chapel Hill (UNC-CH). She is an elected member of the International Statistical Institute and has been named a Fellow of the American Statistical Association, Institute of Mathematical Statistics, and American Association for the Advancement of Science. Marie has served as chair of grant review panels for the National Institutes of Health (NIH), coordinating and executive editor of Biometrics, a member of U.S. Food and Drug Administration advisory committees, 2004 president of the Eastern North American Region of the International Biometric Society, and 2013 president of the ASA.


Marie’s interests include analysis of longitudinal data, methods for design and analysis clinical trials and observational studies, methods for making statistical inference in the presence of missing or mismeasured data, and causal inference and dynamic treatment regimes. She is co-author of the 1995 book Nonlinear Models for Repeated Measurement Data and is co-editor of the 2009 book Longitudinal Data Analysis. She received the 2007 Janet L. Norwood Award for Outstanding Achievement by a Woman in the Statistical Sciences; the 2009 George W. Snedecor and 2011 F. N. David awards presented by the Committee of Presidents of Statistical Societies; and the 2010 NCSU Alexander Quarles Holladay Medal for Excellence, the highest honor the university bestows upon a faculty member. She also has delivered several distinguished lectures, including a 2010 IMS Medallion Lecture.


Since 2004, Marie has served as director of the annual joint NCSU-Duke Clinical Research Institute Summer Institute for Training in Biostatistics, a six-week program to inspire U.S. undergraduates to pursue graduate training in biostatistics, funded by the National Heart, Lung, and Blood Institute. She is a principal investigator for several research grants from the National Cancer Institute, including a multi-institutional (with Duke and UNC-CH) P01 Program Project devoted to statistical methodology for cancer clinical trials.

A: As an undergraduate mechanical engineering major at the University of Virginia (UVA), I was bored, questioning if this field was for me, and looking for an interesting course to take. By luck, I took a statistics course taught by Dave Harrington, who was in his first position in the applied mathematics department in the engineering school. The course followed the book Statistics for Experimenters by Box, Hunter, and Hunter. Immediately, I was hooked. Talking with Dave about the relevance of statistics in the health sciences and the realization that, in this field, I could use my quantitative skills to make a difference iced it for me. I changed my major to applied mathematics—there was no statistics department at UVA back then—so I could focus on statistics.

A: I am fortunate to have had worked as a methodologist and a collaborator, and, in both realms, the most gratifying thing is to be able to solve problems to advance science. I spent the first six years of my career engaged in extensive statistical collaborations with scientists in a range of subject matter areas all over our campus, and nothing gave me more satisfaction than working with them to design experiments and analyze the results. Even if things didn’t turn out as my collaborators hoped, I was gratified knowing their use of sound and appropriate design and analysis methods meant the results were credible and they could move forward to learn from them with confidence. Likewise, developing a new method to address a data-analytic challenge and seeing it used in practice to gain new insights has been the most satisfying part of my statistical research career.

A: I joined the North Carolina Chapter of the ASA as a graduate student at The University of North Carolina at Chapel Hill in 1982 and I recall attending chapter meetings with fellow students in Research Triangle Park. When I took a position at North Carolina State University in 1987, I found that several of my new colleagues were active in the chapter. I continued to attend meetings regularly and became acquainted with members of the local statistical community whom I would not have met otherwise. This led to my first ASA volunteer roles as at-large representative, treasurer, and president of the chapter during 1989–1991.

My chapter involvement inspired me to participate in the ASA more generally. My first role was as program chair for the ASA General Methodology Section for the 1994 JSM. I remember receiving a huge stack of contributed paper abstracts (in the mail, on paper …) and sitting on the floor of my office sorting them into stacks, hoping to create contributed paper sessions. The following year, I was appointed as an associate editor for JASA Applications and Case Studies. Having gravitated toward biostatistics in the early 1990s, I also began to get involved in ENAR, and I was appointed to my second stint on the JSM Program Committee in 1998 as ENAR program chair. That same year, I was elected to the ENAR Regional Committee, through which I became acquainted with many colleagues who, because of their interests, were also active in the ASA Biometrics Section, of which I was a member. This led to my first major ASA elected position as Biometrics Section chair in 2005.

A few years earlier, I had the opportunity to broaden my ASA participation through my appointment in 2003 to the Wilks Medal Committee, for which I served as chair in 2007. I was appointed as chair of the ASA Committee on Nominations in 2006 and had the delight of inviting two wonderful members of our profession to run for president; little did I know that just a few years later I would be invited to do the same.

A: This is almost an impossible question to answer! There were so many once-in-a-lifetime experiences that it is hard to identify the high point. I was uniquely fortunate to be president during the International Year of Statistics, which led to many invitations to speak about the past, present, and future of our discipline. Researching the past and contemplating the future in preparation, and considering issues I had not previously given serious thought, led me to a whole new level of appreciation for our field. Having ASA members thank me for enlightening or inspiring them was one of the most rewarding aspects.

If I had to point to one specific event, it was being able to get someone of Nate Silver’s stature as the president’s invited speaker at the 2013 JSM. When I walked into the ballroom of the Palais des congrès in Montréal with Nate and saw the vast sea of 4,000 statisticians filing in and the throngs of mostly young people at the front waiting to take his picture and have him autograph their copies of his book, I was awestruck. I had hoped that having Nate as the speaker would be a highlight for JSM attendees, but the enthusiasm of the audience and the attention this event brought to our meeting in our profession and in the media were beyond anything I could have imagined. And I’d like to reiterate, Nate did not charge a speaker fee.

The success the ASA has had in engaging with the American Association for the Advancement of Science (AAAS) and seeing more and more ASA members joining AAAS were also major highlights for me. One of my presidential initiatives was to raise the visibility of statistics within the AAAS, the world’s largest general scientific society and publisher of the journal Science. Through the tireless efforts of Steve Pierson, Jeff Myers, and Ron Wasserstein, we established what I hope is an enduring connection between the ASA and the leadership of AAAS, the editorial staff of Science, and especially Science Editor-in-Chief Marcia McNutt. The most satisfying outcome has been the Science reproducibility initiative announced by McNutt in January 2014, a major part of which is the creation of the Science Statistics Board of Reviewing Editors in July 2014, which is responsible for evaluating the statistical integrity of promising Science submissions.

Q: David Morganstein has just entered his term as the 2015 ASA president. As a past ASA president, what advice would you offer him so he can make the most of his term?

A: The most important advice I can give to David is to focus on a few key things he wants to accomplish and devote his energy to them, because this year will go by very quickly. In January, it seems like you will have endless time to see your initiatives come to fruition, and then all of a sudden you are writing your last Amstat News column! And be prepared for surprises—you never know what events or opportunities will arise suddenly, for example, the chance to do a media interview or for the ASA to get involved in a new arena. Being ASA president is a tremendous amount of work, but is an unparalleled opportunity that very few of us are fortunate enough to have to learn so much about our discipline and to have an impact on its future. So most of all, have fun!

Please return to this column next month, when we will feature an interview with 2008 ASA President Peter Lachenbruch.