November 2015 • Issue #461 AMSTATNEWS The Membership Magazine of the American Statistical Association • http://magazine.amstat.org The ASA

2015ellow UPDATE Award

F ALSO: Statisticians in Sierra Leone Technological Advance and Economic Growth: Stagnation or Revival?

The Identity of Statistics in Data Science

AMSTATNEWS NOVEMBER 2015 • ISSUE #461 Executive Director Ron Wasserstein: [email protected]

Associate Executive Director and Director of Operations Stephen Porzio: [email protected] features Director of Science Policy 3 President’s Corner Steve Pierson: [email protected] 5 ASA Board Approves Expansion of ASA’s Science Director of Education Policy Efforts [email protected] Rebecca Nichols: Seeking statistician to serve as science policy fellow Managing Editor Megan Murphy: [email protected] 6 The ASA Fellow Award: 2015 Update

Production Coordinators/Graphic Designers 10 ASA Leaders Reminisce: Mitchell H. Gail Sara Davidson: [email protected] Megan Ruyle: [email protected] 12 ESRA Award Winner Talks About ‘Bringing the Invisible Superhero of Science to Light’ Publications Coordinator Val Nirala: [email protected] 14 Teaching Statistics and Data Science at a Community College Advertising Manager Claudine Donovan: [email protected] 19 The American Statistician TAS Special Issue: Undergraduate Curriculum Contributing Staff Members Pam Craven • Amy Farris • Rebecca Nichols • Eric Sampson 19 ASA Seeks Editor Nominations, Applications

Amstat News welcomes news items and letters from readers on matters of interest to the association and the profession. Address correspondence to Managing Editor, Amstat News, American Statistical Association, 732 North Washington Street, Alexandria VA 22314-1943 USA, or email amstat@ amstat.org. Items must be received by the first day of the preceding month to ensure appearance in the next issue (for example, June 1 for the July issue). Material can be sent as a Microsoft Word document, PDF, or within an email. Articles will be edited for space. Accompanying artwork will be accepted in graphics file formats only (.jpg, etc.), minimum 300 dpi. No material in WordPerfect will be accepted. Amstat News (ISSN 0163-9617) is published monthly by the American Statistical Association, 732 North Washington Street, Alexandria VA 22314- 1943 USA. Periodicals postage paid at Alexandria, Virginia, and additional mailing offices. POSTMASTER: Send address changes to Amstat News, 732 North Washington Street, Alexandria VA 22314-1943 USA. Send Canadian Teaching Statistics address changes to APC, PO Box 503, RPO West Beaver Creek, Rich Hill, Page 14 ON L4B 4R6. Annual subscriptions are $50 per year for nonmembers. Amstat and Data Science News is the member publication of the ASA. For annual membership rates, see www.amstat.org/join or contact ASA Member Services at (888) 231-3473. at a Community American Statistical Association College 732 North Washington Street Alexandria, VA 22314–1943 USA (703) 684–1221 • FAX: (703) 684-2037 ASA GENERAL: [email protected] ADDRESS CHANGES: [email protected] AMSTAT EDITORIAL: [email protected] columns ADVERTISING: [email protected] WEBSITE: http://magazine.amstat.org Printed in USA © 2015 20 MASTER’S NOTEBOOK American Statistical Association Statisticians in Sierra Leone This column is written for statisticians with master’s degrees and highlights areas of employment that will benefit statisticians at the master’s level. Comments and sug- gestions should be sent to Megan Murphy, Amstat News managing editor, at megan@ amstat.org. Promoting the Practice and Profession of Statistics® The American Statistical Association is the world’s largest 23 SCIENCE POLICY community of statisticians. The ASA supports excellence in Technological Advance and Economic Growth: the development, application, and dissemination of statistical science through meetings, publications, membership services, Stagnation or Revival? education, accreditation, and advocacy. Our members serve in This column is written to inform ASA members about what the ASA is doing to industry, government, and academia in more than 90 countries, promote the inclusion of statistics in policymaking and the funding of statistics advancing research and promoting sound statistical practice to research. To suggest science policy topics for the ASA to address, contact ASA inform public policy and improve human welfare. Director of Science Policy Steve Pierson at [email protected]. departments 25 statistician's view Online Articles The Identity of Statistics in Data Science The following articles in this issue can be found 27 meetings online at http://magazine.amstat.org. Join Us for the 5th Annual Conference on Statistical Practice Following the recent decisions of the NSF 28 Division of Mathematical Sciences (DMS) to education Educational Ambassador from Cuba, Nigeria Sought sunset two of its mathematical sciences institutes, DMS leadership responded to an invitation from Biennial Electronic Conference on Teaching Statistics the Society of Applied and Industrial Mathematics Call for Proposals by providing an open letter to the DMS community on its funding policies for the institutes. Because it has been published in full in SIAM News, http:// bit.ly/1W29rhS—we published—a portion on the Amstat News website at http://magazine.amstat. org. The reaction of SIAM President Pam Cook can be viewed here http://blogs.siam.org/feedback- solicited-on-dms-mathematics-institutes- update.

On STATtr@k this month, three fellows from the Data Science for Social Good (DSSG) program offer advice and respond to questions about their experiences, views on data science, and Page Join Us for the 5th Annual future plans. Visit http://stattrak.amstat.org. 27 Conference on Statistical Practice

Visit the ASA Calendar of Events, an online database of statistical happenings across the globe. Announcements are accepted from educational member news and not-for-profit organizations. To view the 29 People News complete list of statistics meetings and workshops, visit www.amstat.org/dateline. 30 Awards and Deadlines 32 Section • Chapter • Committee News Corrections 34 Professional Opportunities In the people news section of the October issue, we spelled Chaitanya Baru’s name incorrectly. It has since been updated online. Visit http://magazine. Follow us on Twitter amstat.org/blog/2015/10/01/people-news-for- @AmstatNews october for details. Join the ASA Community http://community.amstat.org/home

In the article “Statistics Degrees Continue Strong Like us on Facebook Growth,” the colors on the figures were incorrectly www.facebook.com/AmstatNews labeled. The red line represents master’s degrees Follow us on Instagram and the blue line represents bachelor’s degrees. Visit www.instagram.com/AmstatNews http://bit.ly/1jJTr2J for the correct charts.

2 amstat news november 2015 president's corner

STATS 101: Making Statistics Interesting for Students in Their First Course

ost of us became statis- Dick and his team are gath- Tell us about the content. ticians because we ering exciting examples and What principles will you knew statistics is an describing the process of look- include, and will you present excitingM and stimulating profes- ing at and learning from the them in a unique way? sion—one in which curiosity and data. They’ll capture the stu- There’s really nothing new here. initiative intersect and we can dents’ interest with the kind of The Data and Story Library make a difference in almost any detective work many of us have (DASL, http://lib.stat.cmu.edu/ industry that sparks our passion. learned to do and that makes DASL) did something similar But how many times have we been our jobs so relevant and inter- nearly 20 years ago. What we’re at a social event where someone esting. Read on to learn more hoping to do is provide fewer, but David Morganstein asked us what we do for a living? about what Dick has in mind. deeper, stories and show how all When we replied “statistician,” we the techniques one learns in an almost certainly got “The Look.” What do you hope to accom- intro statistics course are actu- You have seen this look many plish with STATS 101? Who do ally used. Some of the studies will times! It’s the look that says, “There you think will be most inter- go further than the typical intro is no way I could ever do that for a stats techniques—using multiple ested in it? living.” And it’s the look that is regression or logistic regression, almost always followed by stories of As we know, a huge number of for example—which will give the an incredibly boring or outrageou- introductory statistics classes, instructors a sense of where the particularly in nonstatistics sly difficult high-school or college intro course leads, as well. Dick De Veaux statistics class. departments, are being taught To address this all-too-often not only by people who don’t Why did you take up this Join Us for the 5th Annual negative perception of our pro- have PhDs in statistics, but challenge? Who are your Conference on Statistical Practice fession and help ensure that often by people who have nei- partners, and what interests many students’ sole encounter ther taken a course nor worked do they bring? with statistics is a more pleas- in statistics. These instructors A wonderful independent ant experience, Dick De Veaux have had to learn statistics “from schoolteacher, Joe Cleary of the accepted a homework assign- the book”—sometimes just before Loomis-Chafee school, lives part ment: He and his committee are their students—and have never of the year in Williamstown, creating “Stats 101.” It’s dynam- had the experience of carrying where I teach. I didn’t know ic, relevant, real-life case studies out a statistical analysis on a real him, but he came up to me one for use as teaching examples, problem. So, we’re developing a day and said he was using my a kind of life raft for instruc- series of about 10 case studies that AP book and confessed he really tors that the ASA will offer to will show statistics in “action,” didn’t know much about statis- schools and universities. solving real problems and how all tics other than what he’d learned Too often, the person asked to those techniques taught in intro from the textbook. He asked if teach that introductory course in a stats work together. We’re hop- he could “shadow” me for a day nonstatistics program such as psy- ing these will give the instructors or two and see what I really do. I chology, medicine, engineering, or not only a sense of how statistics thought it was a really great idea. business is someone relatively early is practiced in the real world, but After spending the day watching in their career. They may have little also some extra examples to use in me do some data analysis he said, or no hands-on experience actually class. Our target audience is any- “You have to show this stuff to conducting a data analysis. They one teaching an introductory sta- other high-school teachers. Now may rely primarily on a textbook tistics course who has no training I see why you care about things and present statistics as a set of for- or experience in our field. that I didn’t think were important mulas and procedures. and don’t really care about some

november 2015 amstat news 3 Unfortunately, many people there. We’re just trying to pro- Unfortunately, many people vide some examples for instruc- who had a single course in statis- who had a single course in tors who haven’t had the ben- tics came away with the impres- efit of doing lots of statistical sion that it is primarily about statistics came away with the analysis to both give them some formulas and theory. We believe motivation and context and pos- statistics is a vibrant profession impression that it is primarily sibly provide some real-world because it is primarily about examples for their students. As practice and application. We’re one of my colleagues in opera- hoping to provide instructors about formulas and theory. tions research who had to teach with exciting, interesting exam- intro stats once said to me, “You ples that will make the subject make the course really interest- more real to them and help make things that I did.” So, we devel- ing, but it’s unfair because you their teaching come alive and be oped a workshop for high-school have all these great examples.” more attractive and stimulating teachers to give them some experi- We’re trying to level the playing to their students. ence with complex data sets using field just a bit by putting some both traditional and data mining on the web. Will you touch on Big Data techniques so they could really see in any way? MORE ONLINE what statisticians do today. Will there be any video In a way, yes. We hope these View the STATS101 My colleagues for these case included? examples provide a precursor resources page at studies are Dave Bock, retired We’re hoping to eventually have to statistical thinking and get http://community. from Ithaca High School; Nick some videos going through the instructors and their students amstat.org/stats101/ Horton of Amherst College; examples, as well, but that’s thinking about statistics as “prob- home. The committee welcomes your review Danny Kaplan of Macalester getting ahead of ourselves a lem-solving.” We want people to and comments. College; Julie Legler of St. Olaf bit. We first need to finish the see statistics as part of the solu- College; and Deb Nolan of case studies. tion when confronted with big Berkeley. They are all wonderful unstructured problems. All the teachers with many years of expe- What do you consider as data sets are “real,” and most rience who know the value of in- modern methods for teach- are “modern” in the sense that depth applications for motivating ing statistics, improvements they’re the kind of data now students in both introductory and over the traditional? routinely collected. AP Statistics. As we all know, statistics in The bike-use data are a great practice is done using comput- example. As people return bikes, Can you give us a preview of ers with some sort of statistical several things are sent to the some of the case study topics? package, whether that be R, database—the station the bike We tried for a mix of applica- SAS, Minitab, JMP, or some- was taken from, the destination tions. We have examples that thing else. I’ve never seen any- station, demographic informa- range from a look back at the one in industry using statistics tion on the user, and the length famous butterfly ballot of 2000 in their job computing sum- of time the bike was out. If you’ve to examining patterns of bike use mary statistics on a calculator or ever used a system like this, you in the Washington, DC, bike- drawing a histogram by hand. know one of the big problems is share system to modeling how Yet, we spend time in the intro- that too many stations are either likely your next flight is to be ductory course teaching some empty—so no bikes are avail- delayed. In each case, the choice of these skills. With students able—or full—so you can’t leave of example has been based on going through the Common the bike. I’ve been late to a meet- availability of data, pedagogical Core, most of the beginning ing because I couldn’t find a place impact, and general interest. part of the traditional intro to put the bike. How can we help stats course will soon be redun- the city do a better job at placing How is STATS 101 different dant. We need to get on to the bikes and figuring out how large from the “usual” introductory exciting parts of modern statis- the stations need to be? course? tics—building complex models; Let me be very clear: This is not simulation and resampling for Are you hoping students who a “course.” There are many inter- inference; and discussing causa- see these examples will “jump esting introductory courses out tion, reproducibility, and data ship” and take up statistics? collection in the Internet age. I’m always hoping students will

4 amstat news november 2015 ASA Board Approves Expansion of jump ship and study more statis- tics, and from what I see across ASA’s Science Policy Efforts the country, it’s happening even before we wind up posting Seeking statistician to serve as these examples. I’m just hoping these provide some help for the science policy fellow instructor who feels a bit over- At its August meeting, the ASA Board of Directors approved a new whelmed trying to understand a position to amplify its science policy efforts. The science policy fel- subject with which he or she has low—to serve a one- to two-year fellowship—will work with the ASA no real world experience. If we director of science policy to raise the profile of statisticians in policy- can help them understand what making and advocate on behalf of the statistics community. it is statisticians really do and Modeled after fellowship positions at sister societies (e.g., American how we think, I’ll be thrilled. Astronomical Society, American Chemical Society, Geological Society Dick and his committee are of America), the fellow will be a statistician based at the ASA headquar- storing their materials at http:// ters in Alexandria, Virginia, who will be expected to initiate and lead community.amstat.org/stats101/ new advocacy projects in addition to contributing to current efforts. home. They welcome your review The fellowship, or “rotator,” model for the new position was chosen and comment. If you have any because of its potential to increase the number of statisticians experi- reactions, suggestions, or just enced in science policy matters. Such fellows of sister science societies feel supportive, contact Dick at often go on to related positions in the Washington, DC, area. [email protected]. Please watch ASA communications for a call for applicants. Questions and interest can be directed to Director of Science Policy Steve Pierson: [email protected].

november 2015 amstat news 5 The ASA

he ASA’s 62 new fellows were honored August 11 during the Joint TStatistical Meetings in Seattle, ellow AwardWashington. The new fellows are from 24 U.S. states, the District 2015Stephanie Shipp, Chair, ASA CommitteeUPDATE on Fellows of Columbia, and six other coun- tries. Fellows are recognized for outstanding professional contri- butions to and leadership in sta- Table 1—Number and Distribution of ASA Fellow Awards by tistical science. See the Amstat Employment Sector by Year, 2005–2015 News October issue (http://bit. Employment Sector ly/1Lu47go) for a list of the new (% of Membership) ASA fellows. ASA fellows are recognized Academe* Business/Industry* Government* for their accomplishments. To (55%) (35%) (10%) Year Total be honored, nominees must have F2005 38 (67.9) 8 (14.3) 10 (17.9) 56 established a reputation in the 2006 50 (83.3) 5 ( 8.3) 5 (8.3) 60 profession and made outstanding contributions to statistical sci- 2007 37 (62.7) 11 (18.6) 11 (18.6) 59 ence. The Committee on Fellows 2008 32 (60.4) 13 (24.5) 8 (15.1) 53 evaluates candidates’ contribu- 2009 36 (63.2) 15 (26.3) 6 (10.5) 57 tions to the advancement of 2010 43 (81.1) 5 (9.4) 5 (9.4) 53 statistical science—requiring evidence of their innovation, 2011 45 (77.6) 8 (13.8) 5 (8.6) 58 leadership, and impact—and 2012 37 (77.1) 7 (14.6) 4 (8.3) 48 their contributions to the profes- 2013 41 (69.5) 11 (18.6) 7 (11.9) 59 sion and society at large. Under 2014 44 (68.8) 15 (23.8) 4 (6.3) 63 ASA bylaws, the Committee on Fellows can elect up to one-third 2015 50 (80.6) 10 (16.1) 2 (3.2) 62 of one percent of the member- *Distribution of membership in 2015; 26% of members do ship as fellows each year. not identify their employment sector on the ASA’s demo- The fellows committee is graphic information form. composed of nine members, with three each from academia, government, and industry. Committee members serve a three-year term. Starting in early March every year, each member of the committee begins to read the applications, assigns their individual scores, and submits the scores to the chair. The chair normalizes the individual scores and then ranks the applicants by the average score. This provides a starting point for the discussion when the committee meets in Figure 1: Percentage of ASA fellows awarded by employment sector, 2005–2015 April to select the fellows.

6 amstat news november 2015 2015 ASA Committee on Fellows Members Stephanie Shipp (chair), Social and Decision Michael Larson, The George Washington Analytics Lab, Virginia Bioinformatics Lab University at Virginia Tech, and National Institute of J. Jack Lee, MD Anderson Cancer Center Standards and Technology (retired) Susan Hinkins, NORC This article continues the tra- Jeri Mulrow, National Science Foundation dition of providing a descriptive Mingxui Hu, Takeda Pharmaceuticals Estelle Russek-Cohen, U.S. Food and Drug analysis of the ASA fellows and Keith Rust, Westat adds data for the 103 nominees Administration and 62 fellows in 2015. Data are Sudipto Banerjee, University of Minnesota presented by employment sector and gender. Employment Sector In 2015, the percentage of ASA members in each employment sector was 55% academe, 35% business/industry, and 10% gov- ernment. The number of ASA fellow awards given by employ- ment sector for the most recent 11 years is presented in Table 1 and Figure 1. In 2015, 81% of the new fellows are from academe, 16% are from business/industry, and 3% are from government. The percentages of fellows Figure 2: Percentage of fellows awarded by employment sector relative to the percentage awarded by employment sector of ASA membership in that sector by year relative to the percentages of ASA membership are shown in Figure Table 2—Number and Percentage of ASA Fellow Nominations by 2, which shows that academics are nominated at a higher rate Employment Sector by Year, 2005–2015 and business/industry members Employment Sector are nominated at a consistently (% of Membership) lower rate relative to their respec- Academe* Business/Industry* Government* Total tive proportions of member- Year (55%) (35%) (10%) ship. Government nomination 2005 51 (57.3) 22 (24.7) 16 (18.0) 89 rates have fluctuated the most in 2006 81 (73.0) 19 (17.1) 11 ( 9.9) 111 relative terms, in part due to the smaller proportion of members. 2007 79 (65.8) 22 (18.3) 19 (15.8) 120 The yearly number and per- 2008 60 (64.5) 18 (19.4) 15 (16.1) 93 centage of fellow nominations 2009 59 (62.1) 23 (24.2) 13 (13.7) 95 in each employment sector are 2010 71 (72.4) 13 (13.3) 14 (14.3) 98 shown in Table 2. The 103 nominations received in 2015 2011 76 (72.4) 18 (17.1) 11 (10.5) 105 are close to the average of 101 2012 62 (75.6) 14 (17.1) 6 ( 7.3) 82 nominations submitted between 2013 71 (71.7) 15 (15.2) 13 (13.1) 99 2005 and 2015. The number and 2014 86 (72.9) 24 (20.3) 8 ( 6.8) 118 percentage of nominations from academe remained high in 2015, 2015 78 (75.7) 17 (16.5) 8 ( 7.8) 103 while both the number and per- Average 70 (69.4) 19 (18.5) 12 (12.1) 101 centage of nominations from *Distribution of membership in 2015; 26% of members do not identify their employment government were low. sector on the ASA’s demographic information form.

november 2015 amstat news 7 One question of continuing Table 3—Percentage of Successful ASA Fellow interest is the rate of success for Nominations by Employment Sector by Year the nominations in each sector. Employment Sector As shown in Table 3 and Figure Academe Business/ Government 3, the success rate of nominations Year Industry submitted from the government 2005 74.5 36.4 62.5 sector was below average in 2015. The success rate for academe and 2006 61.7 26.3 45.5 business/industry was above aver- 2007 46.8 50.0 57.9 age in 2015. 2008 53.3 72.2 53.3 Gender 2009 61.0 65.2 46.2 Table 4 and Figure 4 show the cur- 2010 60.6 38.5 35.7 rent ASA membership by gender 2011 59.2 44.4 45.5 and year of joining the ASA (or, if 2012 59.7 50.0 66.7 membership lapsed, rejoining the 2013 57.7 73.3 53.8 ASA). The summary is based on available data; more than 15% of 2014 51.2 62.5 50.0 ASA members do not enter their 2015 64.1 58.8 25.0 gender on their membership form. Average 59.1 52.5 49.3 Among the current members who provide data, there is an increasing trend over time in the proportion of female members. The number and percentage of ASA Fellow nominations by gender in each year from 2005 through 2015 are presented in Table 5. The number and per- cent of female nominations have varied since 2005, with a slight increase in recent years. The number and percentage of ASA fellow awards from 2005 through 2015 are presented by gender in Table 6, and the per- Figure 3: Percentage of successful ASA nominations by employment sector by year, 2005–2015 centages are charted in Figure 5. There has been a slight decline in the percentage of males and slight overall increase in females, although the numbers and per- centages vary each year. Among nominees, the chance of success has generally been higher for females in the range of years summarized (Table 7 and Figure 6). Submitting a Nomination The ASA Fellow Award is a signifi- cant recognition of contributions to the statistics profession. The Figure 4: Percentage of current ASA members by gender by year of joining the ASA nominee may be worthy of being an ASA fellow, but the nomination

8 amstat news november 2015 Table 4—Percentage of ASA Membership by Gender and Year of Joining Year Joined Female Male ASA ≤ 1974 16 84 1975–1984 20 80 1985–1994 27 73 1995–2004 31 69 2005–2015 40 60 All Members 35 65

Table 5—Number and Figure 5: Percentage of ASA Fellow awards by gender by year, 2005–2015 Percentages of ASA Fellow Nominations by Gender by Year, 2005–2015 Year Female Male Total 2005 24 (27.0) 65 (73.0) 89 2006 25 (22.5) 86 (77.5) 111 2007 22 (18.3) 98 (81.7) 120 2008 16 (17.2) 77 (82.8) 93 2009 12 (12.6) 83 (87.4) 95 2010 24 (24.5) 74 (75.5) 98 2011 19 (18.1) 86 (81.9) 105 2012 19 (23.2) 63 (76.8) 82 Figure 6: Percentage of successful ASA Fellows nominations by gender by year, 2005–2015 2013 31 (31.3) 68 (68.7) 99 2014 32 (27.1) 86 (72.9) 118 2015 34 (33.0) 69 (67.0) 103 Table 6—Number and Percentages Table 7—Percentage of Successful of ASA Fellow Awards by Gender ASA Fellow Nominations by by Year, 2005–2015 Gender by Year, 2005–2015 package must convince the com- Year Female Male Total Year Female Male mittee of this. Both the nomina- 2005 15 (26.8) 41 (73.2) 56 2005 62.5 63.1 tion and letters of support should highlight accomplishments and 2006 11 (18.3) 49 (81.7) 60 2006 44.0 57.0 provide evidence of impact. A long 2007 14 (23.7) 45 (76.3) 59 2007 63.6 45.9 nomination package is not neces- 2008 11 (20.8) 42 (79.2) 53 2008 68.8 54.5 sarily better, but a well-prepared 2009 7 (12.3) 50 (87.7) 57 2009 58.3 60.2 one is critical to success. Members of the ASA 2010 17 (32.1) 36 (67.9) 53 2010 70.8 48.6 Committee on Fellows encourage 2011 12 (20.7) 46 (79.3) 58 2011 63.2 53.5 those planning to submit a nomi- 2012 13 (27.1) 35 (72.9) 48 2012 68.4 55.6 nation to review the guidelines 2013 20 (33.9) 39 (66.1) 59 2013 64.5 57.4 and examples on the ASA website (www.amstat.org/awards/fellows. 2014 22 (34.9) 41 (65.1) 63 2014 68.8 47.7 cfm) and to submit the pack- 2015 19 (30.6) 43 (69.4) 62 2015 55.9 62.3 age before the deadline. These Average 62.6 55.1 guidelines have evolved over time to give everyone a fair chance of being considered. n

november 2015 amstat news 9 ASA LEADERS REMINISCE Mitchell H. Gail Jim Cochran In the eleventh installment of the Amstat News series of interviews with ASA presidents and executive directors, we feature a discussion with 1995 president Mitchell H. Gail.

Mitch, thank you for taking time to talk with Qme. You have what I suspect is a unique dis- tinction among ASA presidents. Several years before you earned your PhD in statistics from The George Washington University, you earned an Dr. Mitchell H. Gail is a senior MD from Harvard Medical School. What moti- investigator at the biostatistics vated you to return to school to work on a PhD in statistics several years after you earned your MD? branch of the Division of Cancer Epidemiology and After working as a medical intern in at the Peter Genetics at the National ABent Brigham Hospital in Boston for one year, I Cancer Institute (NCI). He joined the U.S. Public Health Service (USPHS) as a researcher at the National Cancer Institute (NCI). earned an MD from Harvard My first project was studying the motility of cells Medical School and a PhD in tissue culture. They moved in a two-dimensional in statistics from The George random walk, and I needed statistical methods to Washington University. His work at NCI has included studies characterize their rates of diffusion. John J. Gart at on the motility of cells in tissue culture, clinical trials of lung NIH showed me how to perform certain regressions, cancer treatments and preventive interventions for gastric plotting mean squared cell displacement against cancer, assessment of cancer biomarkers, AIDS epidemiology, time. Three years later, David P. Byar invited me to and models to project the risk of breast cancer. Gail’s current join his clinical and diagnostic trials section at NCI research interests include statistical methods for the design and to work on lung cancer clinical trials and problems analysis of epidemiologic studies, including studies of genetic in the immune-diagnosis of cancer. These projects required statistical knowledge, and the USPHS paid factors, and models to predict the absolute risk of disease. He is for some night courses at The George Washington also working on methods of calibration and seasonal adjustment University in DC. At first, I enrolled as a master’s for multi-center molecular-epidemiologic studies. student, but before I knew it, I was working with Gail is a Fellow of the American Statistical Association and Joseph L. Gastwirth on my PhD thesis. I was lucky member of the Institute of Medicine of the National Academy to be working in an environment that supported of Sciences. He has received several prestigious awards, continuing education and where my increasing sta- tistical skills were useful and appreciated. including the Spiegelman Award from the American Public Health Association, the Snedecor Award from the American This spring, you received the 24th annual Statistical Association, the Marvin Zelen Leadership Award QAACR-American Cancer Society Award for Excellence in Cancer Epidemiology and in Statistical Science from the Harvard T.H. Chan School of Prevention from the American Association for Public Health, the Nathan Mantel Lifetime Achievement Award Cancer Research (AACR) and American Cancer from the American Statistical Association, and the Award Society. The AACR press release cited a statistical for Excellence in Cancer Epidemiology and Prevention from model you developed in the late 1980s for esti- the American Association for Cancer Research and American mating the absolute risk for a white woman of a Cancer Society. specific age with specific risk factors to develop breast cancer. What led you to develop this model, which is commonly known as the “Gail model,” and what has been the impact of this model? One day, I went to lunch with an NCI col- Alaborator, John J. Mulvihill, who was advising women in a clinic for women at high risk of breast cancer. Many such women thought they had unre-

10 amstat news november 2015 alistically high risks of cancer because of strong to provide realistic estimates of absolute risk. Our ini- family histories, and John wanted to be able to give tial attempt was to combine race/ethnicity-specific them realistic estimates. I was fortunate to be work- SEER rates with the relative and attributable risks ing with other collaborators (Louise A. Brinton from the original Gail model, which was developed and Catherine Schairer) who were familiar with from data on white women, to compute race/eth- data on breast cancer risk factors from the Breast nicity-specific absolute risks. More recently, we have Cancer Detection and Demonstration Project and been able to obtain data on relative and attributable with statisticians (Dave P. Byar, Donald K. Corle, risks that are also specific to various racial/ethnicity and Sylvan B. Green) who knew a lot about relative groups. Fully specific models are available for African- risk modeling and attributable risk estimation. My American and Asian-American women, and we are contribution was to bring all these tools together to obtaining such data for Hispanic women. compute absolute, not relative, risk. Absolute risk In 2012, you served as chair of the Section is the chance that a woman with specific risk fac- Qon Statistics of the American Association for tors will develop breast cancer over a defined time the Advancement of Science (AAAS). What issues interval. Absolute risk has many applications in did this AAAS section face during your term as individual counseling and public health planning. its chair? One of the first uses of the model was to help design a large randomized intervention trial to The Statistics Section (Section U) of AAAS determine whether tamoxifen could prevent breast Awas formed in 1962. Jerzy Neyman described cancer. Statisticians at the National Surgical Breast the role of the section and stressed the opportu- and Bowel Project (NSABP) combined national nity for interdisciplinary exchange (Science, 1962; breast cancer incidence data with the relative and 138:1080-1083). Several issues related to this tra- attributable risks from the original model to pro- dition come to mind. The first was developing an duce what is now known as the “Gail model,” or exciting set of symposia from Section U with broad SNEAK PEEK the National Cancer Institute’s Breast Cancer Risk appeal to the scientific community on topics rang- Please return to Assessment Tool. This model correctly predicted ing from “Understanding and Communicating this column next the number of breast cancers that would develop Uncertainty in Climate Change Science” to month, when “Benefits of Randomized Experiments for Science we will feature over five years during the Breast Cancer Prevention an interview Trial, which showed that tamoxifen cut breast can- and Society.” A second was nominating and recog- with 1994 ASA cer incidence by half. Unfortunately, tamoxifen also nizing AAAS fellows for their contributions in sta- President Ronald caused some serious side effects, such as strokes and tistics and its applications. A third was increasing L. Iman. endometrial cancer. visibility of Section U and the role of statistics. In The absolute risks associated with each of the coordination with the ASA (and with a big boost risks and benefits of tamoxifen are needed to help a from Marie Davidian, former ASA president, and woman decide whether she stands to benefit from Ron Wasserstein, ASA executive director), we geared such an intervention. Typically, young women with up to celebrate the 50th year of Section U and the high breast cancer risks benefit most. Today, there International Year of Statistics (2013). This coop- are several good risk models for breast cancer with eration with the ASA has continued, and Section various strengths and weaknesses, and they are U has grown during a period when overall AAAS widely used in counseling and for making public membership has not. health recommendations. NCI’s website for breast What challenge that you faced during your cancer risk assessment is visited more than 3 million QASA presidency stands out? times per year. That would be the immense diversity of interests Aof ASA members. It seemed to me that one of the You have developed many breast cancer risk challenges of the ASA was to serve such a profession- Qassessment tools, including several SAS mac- ally diverse and talented membership. In my presiden- ros for breast cancer risk assessment that have tial address, I reviewed the contributions of Bradford been designed for different demographic/ethnic Hill and Jerome Cornfield to the development of the groups. What led you and your colleagues to case-control method and clinical trials. These great develop different breast cancer risk assessment statisticians combined the development of innovative tools for different demographic groups? methods with deep involvement in collaborations and The National Cancer Institute’s Surveillance, substantive scientific research and set an example that AEpidemiology, and End Results Program (SEER) many of us could try to emulate, regardless of our par- gathers breast cancer incidence data for racial and eth- ticular interests or specializations. n nic subgroups, and there is considerable variation in the age-specific incidence rates. Therefore, specialized models are needed for various racial and ethnic groups november 2015 amstat news 11 ESRA Award Winner Talks About ‘Bringing the Invisible Superhero of Science to Light’ Morteza Marzjarani

he Committee on the Excellence in Statistical Reporting Award selected Julie Rehmeyer as the winner of its 2015 award. Rehmeyer is a professor at St. John’s College in Santa Fe, New Mexico, as well asT a contributing editor for Discover. Members of the ESRA committee were especially impressed with the number and breadth of Rehmeyer’s articles. Her work spans two books, recurring contributions to ScienceNews.com and Discover, and an array of articles in science- and general-audience publications such as The New York Times and The Washington Post that help connect the public with science. In particular, the committee pointed to two articles that stood out: “Equation: Calculating Ballot Bungles Is All About the P-Value,” pub- lished online by Wired “Florence Nightingale: The Passionate Statistician,” published on ScienceNews.com by the Society for Science & the Public

Julie Rehmeyer What follows is an interview with Rehmeyer.

When did you become interested in statistics? What do you particularly enjoy about writing I’ve actually never taken a statistics course, and it about statistics? How is it different from writing took me quite a while to discover the joys of sta- about science? tistics. I did my graduate work in mathematics I wrote the Math Trek column at Science News at MIT, but I was a dumb kid and only saw the for quite a few years, and I took particular plea- appeal of the purest of the pure mathematics, being sure in finding statistics stories for my column. a mystic at heart. I sneered at applied subjects like It felt like I was bringing the invisible superhero statistics and engineering. My attitude started to of science to light. change when I designed and built my own house On a deeper level, I find it especially meaningful and the engineering I learned to calculate beam to write about mathematics and statistics because so sizes and such enchanted me. Using mathematics to many people have had wounding experiences with do something was really cool! them in school. It gives me such enormous plea- I became a professor of mathematics and the sure when someone says, “I read your article, and classics at St. John’s College in Santa Fe, New you know, I just don’t have a math brain and I hate Mexico. St. John’s is a teaching college, and its facul- math, but I think I understood your story, and it was ty members have little time for research. I loved the pretty fun!” teaching, but eventually found that I missed having To me, when people come to believe they’re ter- my own intellectual and creative work. But at that rible at math, they’re suffering from a wound that’s point, for a variety of reasons, I didn’t see myself ultimately a spiritual one. They’ve been convinced doing research mathematics. On a lark, I picked up that their ability to find pattern and meaning in the a statistics book over the summer and had a ton of world is somehow fundamentally flawed. The impacts fun learning statistics. But I ultimately decided to of that kind of wound go far beyond anxiety when it head in the direction of science writing. comes to calculating a tip or balancing a checkbook.

12 amstat news november 2015 In a background way, often unnoticed, it saps people’s chronic illness, and since I wrote many of my own power, makes it harder for them to access their deepest statistics stories while dealing with a serious chronic selves and bring that into the world. illness, that was especially moving to me. So when someone is able to follow one of my stories and find a bit of joy in it, I hope that wound What have you done since you won the award? heals a bit. I hope, in some small way, that it bol- I’m writing a book at the moment, a memoir about sters people’s confidence in themselves, and it shows the science and politics of poorly understood illness- them that the joy and power of mathematics is es. I was terribly ill with chronic fatigue syndrome available for them, too. for many years, eventually so sick that I was para- lyzed much of the time, and I eventually discovered Are there special challenges in writing about statistics? that for me, the central problem was a hypersensi- One challenge is that people are often so frightened tivity to mold. Science hasn’t served chronic fatigue of it, so as a writer, you have to work especially hard syndrome patients well, and the experience has to enter a different mindset and offer handholds affected my relationship to the institutions of sci- along the way. ence quite significantly. On the surface, this is far Another challenge is that statistics is all about removed from statistics, but some of the questions MORE ONLINE uncertainty, and it’s easier to tell stories with clear, at its heart are quite similar: How do we make sense Do you know a member of the evidence our bodies give us? How do we assess of the media eligible single outcomes. for the Excellence in I think one of the biggest tools to deal with both uncertainty? How do we make good decisions in the Statistical Reporting those challenges is narrative. Our brains are wired for face of that uncertainty? Award? Nominations story, so readers are better able to absorb information I’m also a Scripps environmental journalism fellow are being accepted until that way. And if you can weave the uncertainty into at the University of Colorado, Boulder. I get to have March 1. Visit http://bit. the story, it can acquire a power of its own, rather than the run of the university and take any class I please, ly/1Pvosn8. merely diluting the power of a tidy result. but I have no responsibilities. Pretty sweet gig! What do you think about ESRA? What are your other passions? I’m really thrilled about receiving it. As writers, we Recovering my health enough that I’m able to exercise often don’t see the impacts of our stories. We hear again is an enormous thrill, so I’m loving biking and from very few readers, and the ones we hear from are hiking and running and swimming. I take huge plea- far from representative: They’re often those who are sure in my dog, Frances. I got married two years ago, outraged. We throw our stories out there and move and I still feel very much like a newlywed (and also on to the next, hoping that they’re somehow adding like I can hardly remember life before him). I built up to something. my own straw bale house on 12 acres of streamside When I learned I was receiving the award, I was land outside Santa Fe, New Mexico, and I feel very pleased, but I initially had a hard time really absorbing connected to my house and land. n it. But being at the award ceremony, hearing the cita- tion read, seeing the list of people on the committee I didn’t know who selected me, receiving my cup—it was really moving for me. It’s wonderful to know that people appreciate my work. Do you have a hero in the discipline of statistics? The ASA’s extensive awards program recognizes statisticians who Florence Nightingale. My story about her, “The have made outstanding contributions through research, teaching, Passionate Statistician,” was one especially cited Awards consulting, and service to the association and statistical profession. for my award. We tend to think of her as “the lady with the lamp,” saving the lives of wounded soldiers The deadlines to nominate a colleague for many of these awards through her nursing. But she saved many more lives occur from December to March. through her use of statistics, revealing the changes that needed to be made in nursing as a whole. She also did Visit www.amstat.org/awards/awardsscholarships.cfm a huge amount of that work while fighting a serious and nominate a colleague today!

november 2015 amstat news 13 Teaching Statistics and Data Science at a Community College Statistics and data science degrees are some of the fastest- growing undergraduate programs in the country. We wanted to know if this growth has affected teaching statistics at community colleges and if there are any relevant changes professors have seen or experienced.

WEST VALLEY COLLEGE What do you like most about mathematics, I understand the Northern California teaching statistics in the com- depth and breadth of the sub- munity college setting? What ject area and the connections challenges do you experience between the various courses in that might be unique to com- the sequence of study. Not hav- munity colleges? ing the same background in sta- Statistics is one of my favorite tistics makes it difficult to advise classes to teach. It is so easy to students who might consider bring students’ interests into the statistics as a major. In addition, classroom and find topics they community colleges offer only find engaging. For example, one one course in statistics, limiting of my students last semester has a the ability of students to explore YouTube channel. She was excit- statistics as a possible major. Rebecca Wong has taught ed to learn how to understand all Describe an emerging trend you the data analytics YouTube sends mathematics and statistics have noticed related to teach- her about her viewers. As part of at West Valley College for ing statistics in community a class project, another student colleges. more than 20 years. She interviewed an anthropologist enjoys designing activities and was surprised to learn how One trend at the community to help students actively much statistics played a part in colleges has been the develop- explore statistical concepts the work he did. Based on the ment of a “pre-statistics” course. Currently, community college and encouraging students interview, this student was able to expand her professional net- students (based on placement test to apply those concepts to scores) have to take several levels areas of personal interest. work and find an internship opportunity. Stories like this of algebra before taking statistics. Rather than repeating an alge- Describe the statistics course(s) aren’t uncommon in my statis- tics course. There is an excite- bra curriculum, pre-statistics is a your college offers and any one-semester course that focuses recent changes you’ve made. ment and interest in my statistics course that is not replicated in on those skills that students need At West Valley College, we offer my algebra or calculus courses to be successful in the elemen- an elementary statistics course quite as easily. tary statistics course. We have that fulfills mathematics and Teaching statistics at a com- been offering the pre-stats course quantitative reasoning transfer munity college does have its for one year, and it has been suc- requirements for our students. unique challenges. I, like the cessful in accelerating students Our department offers a course majority of my colleagues, have through their transfer-level math- aligned with the GAISE stan- degrees in mathematics—not ematics course. dards, emphasizing conceptual statistics. I have to depend on understanding, interpretation, conferences, networking, and working with real-world data, reading to expand my back- and the use of technology. ground in the subject area. In

14 amstat news november 2015 CAPE COD COMMUNITY COLLEGE What do you like most about West Barnstable, teaching statistics in the com- state (Massachusetts), funding of Massachusetts munity college setting? What the community colleges is now challenges do you experience directly linked to student success that might be unique to com- and graduation rates. Statistics has most often been the focus of these munity colleges? acceleration/student success proj- I love that when a student in our ects, with some advocates arguing statistics course asks when he or that little material in the tradition- she will ever use statistics, the al algebra prerequisite is actually honest answer is “every day.” needed in a statistics course. A challenge I face at the com- From the Common Core munity college is that many perspective, a significant num- students simply don’t have the ber of topics we include in resources to buy the materi- Mary Kehoe Moynihan introductory statistics will be als required for many statistics taught in K–12. If the Common chairs the ASA/American courses. I currently use the mate- Core is implemented successful- Mathematical Association rials from a free online course. ly, it will dramatically affect the of Two-Year Colleges Students can rent a calculator topics covered in the introduc- (AMATYC) Joint Committee from our library, and the statis- tory statistics course. and is an active member tical software I’m using costs less Some prominent members of of the New England than $20. the statistics education commu- Mathematical Association Describe an emerging trend you nity believe statistical inference MORE ONLINE of Two Year Colleges have noticed related to teaching should be introduced with boot- The Guidelines for and AMATYC. She has statistics in community colleges. strap confidence intervals and Assessment and Instruction in Statistics taught mathematics and The introductory statistics course randomization-based hypoth- esis tests, rather than the tradi- Education (GAISE) computer science at Cape is the most volatile of the courses College Report has been tional parametric techniques. Cod Community College taught by the mathematics depart- influential in shaping the Recent trends have increased the modern statistics course. since January 1979. ment right now. There are pres- sures, expectations, and changes emphasis on inferential statis- You can download the tics in the introductory statistics report here: www.amstat. coming from many directions. org/education/gaise. From the developmental course. Therefore, fundamen- Describe the statistics course(s) mathematics perspective, many tally changing the way inference your college offers and any schools are “bundling” develop- is taught would substantially recent changes you’ve made. mental mathematics and statistics change the course. Elementary Statistics at Cape into a pathway designed to help There is a temporary ASA GAISE Cod Community College is a non-STEM students succeed in a committee reviewing the College Report non-calculus–based introduc- college-level mathematics course. . This workgroup tion to statistics geared toward For example, the goal of both the is updating the report, with rec- non-STEM majors. It fulfills the Carnegie Foundation’s Statway ommendations expected as soon GAISE three-credit mathematics gradua- and the statistical reasoning path- as the end of 2015. The College Report tion requirement. way in the Dana Center’s New was originally Mathways project is to improve published in 2005 and has been Are you seeing increasing influential in shaping the modern enrollment in your statistics student success in a college-level statistics course. States have tired statistics course. classes, and, if so, what do you Finally, the projection of think is driving it? of high-school graduates spend- ing years in developmental math- unfilled jobs caused by the explo- Statistics is a “service course” for ematics courses required by many sion of Big Data analytics and us—our STEM majors don’t colleges as the prerequisite to col- data science has many wonder- traditionally take it. We’ve seen lege-level mathematics courses ing if our non-STEM intro- a significant increase in our such as statistics. ductory statistics courses might enrollment largely due to other This epidemic was the focus provide fertile ground for future departments such as psychol- of the first National Summit on statistics/data science majors. ogy and nursing recommending Developmental Mathematics held Traditionally, most majors begin or requiring their majors to take just before the 2013 AMATYC their program with calculus, elementary statistics. conference in Anaheim. In my and most introductory statistics november 2015 amstat news 15 courses are non-calculus based. of America, ASA, and AMATYC. Describe the statistics course(s) Still, as we look for hundreds of There are free webinars and your college offers and any thousands of professionals trained resource pages available on their recent changes you’ve made. in Big Data analytics, exposing websites. I just Googled “statistics We offer a general introductory introductory statistics students to webinars teaching resource” and statistics course and a business/ some of the basic concepts of data found valuable links, including economics statistics course. science might inspire students to the AMATYC statistics resources Both courses have been using explore the field. page the ASA/AMATYC Joint open resources lately to comple- Note that several of these Committee oversees. ment textbook resources, and trends are pulling the introduc- some sections are completely tory statistics course in different What advice do you have for com- using open resources on a trial directions. The pathways projects munity college students regarding basis. Many sections of the often minimize the mathematics statistics and or data science? introductory statistics course are prerequisite for the course and Do the homework. You need to working with simulation-based the traditional algebra topics in practice your skills and carefully inference this semester, and a the paired courses, while the goal read assigned material to master biostatistics course is being con- of the Common Core is to pre- common statistical procedures, sidered for development. pare college-ready students who gain a conceptual understanding have already achieved many of the of statistics, and hopefully devel- Are you seeing increasing learning outcomes in the current op a statistical way of thinking. enrollment in your statistics college-level introductory statis- classes, and, if so, what do you tics course. Meanwhile, an effort think is driving it? to attract statistics/data science Yes. Some of that is due to the majors would target students with increase in overall enrollment at a strong mathematics background. MONTGOMERY our college. However, many stu- Job analysts predict statisti- COLLEGE dents are choosing to take statis- cians will be in high demand, Germantown, Maryland tics (as opposed to other courses and statistics will be the “hot” in our mathematics/statistics job over the next decade. I department) to meet their math- believe the modern introduc- ematics requirements for gradu- tory statistics course will adapt to ation and quite a few programs these influences, providing many require or strongly recommend challenges and opportunities for statistics for their degree. all faculty teaching the course. What do you like most about Characterize the students who teaching statistics in the com- tend to take statistics at your munity college setting? What campus/college. challenges do you experience Almost every student is taking Brian Kotz is a that might be unique to com- elementary statistics to fulfill mathematics and statistics munity colleges? their mathematics general edu- professor at Montgomery I like that we are serving and reach- cation requirement: “I’m taking College in Germantown, ing a community of dedicated the course because it’s required Maryland. A statistics learners who, for various reasons for my degree.” educator for more than 20 (e.g., finances, job, family), either cannot or choose not to immedi- years, he was appointed What advice do you have for ately participate in a four-year col- to the ASA/AMATYC Joint other statistics instructors at lege environment. This population community colleges? Committee in 2012 and of learners is quite large in num- A modern statistics course is most has authored numerous ber, at least in our community. likely radically different from any print and video resources. statistics course you took as a Describe an emerging trend you student. Try to keep up with the have noticed related to teaching most recent recommendations statistics in community colleges. for introductory statistics from I think the GAISE standards really the professional societies, includ- helped codify expectations and ing the Mathematical Association practices and guide instruction in

16 amstat news november 2015 both two-year and four-year col- COLLEGE OF THE increased dramatically, and we leges. Ultimately, I want the part- CANYONS expect this trend to continue. For nership between the ASA and Santa Clarita, California example, College of the Canyons’ AMATYC to remain strong; to pre-statistics and statistics offer- continue to evolve; and to devel- ings have grown from 59 in 2012 op in meaningful ways regarding to 106 currently. California, curricula, professional develop- the largest college system in the ment, open technology resources, United States with 112 com- and so on. munity colleges, is experiencing a similar increase. In California, What advice do you have for statistics is taught within mathe- other statistics instructors at matics departments; however, this community colleges? adds another challenging dimen- The opportunities for profes- Monica Dabos is a statistics sion because a master’s degree or a sional development are many, PhD in statistics does not qualify and the ASA statistics education instructor at College of the Canyons. Recently, the an individual to teach statistics (a community is a particularly car- mathematics degree is required). National Science Foundation ing, engaged, and helpful group. Lack of support, combined Do not hesitate to contact them funded her initiative, TANGO with little formal training in sta- (e.g., authors, colleagues at a Stat Ed (Training A New tistics, are two possible reasons nearby four-year college, ASA Generation of Statistics why instructors are reluctant to section members). They want to Educators), to improve the teach statistics, despite the avail- work with you! teaching and learning of ability of sections. Technology What advice do you have for com- statistics at the two-year is another obstacle that faculty, munity college students regarding community college level. especially part-time instruc- statistics and/or data science? tors, encounter. Many commu- nity colleges lack the resources I tell students that whether Describe the statistics course(s) they plan to use what they have required to implement new your college offers and any teaching methodologies. Using learned to better serve the pub- recent changes you’ve made. lic, improve their business, or simulations or data sets that have College of the Canyons has one serve as the foundation for the many variables and more than 20 college-level statistics course, next statistics course they will data points are difficult to imple- Math 140 “Intro to Statistics.” take upon transferring to their ment without access to appropri- The college has also developed four-year college, they are better ate software and training. a new statistics preparation informed citizens of our commu- course for non-STEM majors What do you like most about nity having been introduced to called Math 075, “Intermediate teaching statistics in the com- the discipline of statistics. I also Algebra for Statistics.” After suc- tell them they have the kernel of munity college setting? cessfully completing Math 075, a very in-demand skill set and I love teaching statistics at the com- non-STEM students can enroll encourage them to stay in touch munity college level because prior- in Math 140 and complete all the and consider pursuing further ity is given to teaching and helping transfer math requirements. We coursework in statistics or data students succeed. I also have the are considering offering another science, regardless of whether it opportunity to shine a light on statistics course that addresses is required of them. statistics as a possible career choice more advanced topics, but there when many students are still defin- are some state regulations limit- ing their educational goals. ing what we can offer. What advice do you have for Are you seeing increasing enroll- other statistics instructors at ment in your statistics classes, community colleges? Any advice and, if so, what do you think for community college students is driving it? Also comment on regarding statistics (and/or any emerging trends you have data science)? noticed related to teaching sta- The best advice for instructors tistics in community colleges. I can offer is to get connected! Demand for statistics courses at In isolation, it is easy to feel the community college level has lost and/or discouraged when november 2015 amstat news 17 students do not understand sta- Describe the statistics course(s) Characterize the students who tend tistical concepts. A few programs your college offers and any to take statistics at your college. are emerging to address this and recent changes you’ve made. We have significant numbers they are worth mentioning here. We teach elementary statistics from health sciences and busi- CAUSEweb.org offers free webi- from the book Basic Practice of ness studies, but also students nars. The TANGO Stat Ed pro- Statistics, 7th edition. Just this from many other areas in which gram and ASA Member Initiative year, we added ANOVA as a the faculty think their students program both offer mentors and required chapter, so we now will benefit more from a statis- regional hubs for developing and cover chi-squared tests, inference tics course than college algebra. strengthening statistical knowl- on regression, and ANOVA. To We have a substantial number of edge of instructors teaching sta- returning students, who need sta- do this, we are de-emphasizing tistics at community colleges. tistics as a prerequisite for gradu- calculations with a hand-held cal- Whether you are an instructor ate work in various fields. who has experienced success in culator after one-sample t-tests— teaching statistics or an instruc- particularly de-emphasizing What advice do you have for tor who would like help develop- calculations of SE. We made sev- other statistics instructors at ing better teaching tools, getting eral other changes to increase the community colleges? connected is the answer. Get con- emphasis on concepts. Emphasize thinking about data nected, get connected, get con- We also have worked with and the conceptual ideas. Statistics nected. This is my mantra. both the Carnegie Foundation students (in courses offered in For students, I’d tell them to offering Statway and the Dana math departments) usually think dive into statistics to see how rel- Center offering New Math the main characteristic of a cor- evant it can be to their daily lives Pathways StatPath. rect answer is to match the num- and career opportunities. Most bers in the solution key. Work importantly, I’d promise them Are you seeing increasing on moving them toward describ- that, with the right instructor, enrollment in your statistics ing what information the data is statistics could actually be fun! classes, and, if so, what do you giving about a real-life question think is driving it? of interest. AUSTIN COMMUNITY Beginning around 2011–2012, If your text supports discus- COLLEGE we saw substantial increases in sion of data ethics, embrace that. Austin, Texas enrollment. More disciplines, It is some of the most engaging particularly health sciences dis- material to my students. I have it on a discussion board forum avail- ciplines, began to specifically able most of the semester and try require statistics. to stay out of the conversation as What do you like most about much as reasonable to encourage teaching statistics in the com- them to discuss the ethics of vari- munity college setting? What ous situations with each other. challenges do you experience What advice do you have for com- that might be unique to com- munity college students regarding munity colleges? statistics and/or data science? Mary Parker teaches The small classes (36 or fewer) Large data sets are increasingly mathematics and statistics are helpful for engaging students. available, and people are needed at Austin Community There are a variety of challenges, to extract useful information from College and mathematical which are mostly related to stu- them. In your statistics classes, pay statistics at The University dents being overburdened with careful attention to what tech- of Texas at Austin. She work and family obligations and niques are useful to answer what participated in writing sometimes-unrealistic expecta- sorts of questions and why. Learn curriculum and teaching in tions of the amount of work a to use a flexible professional statis- tical package. n the Carnegie Foundation’s college course takes. Statway project and the Dana Center’s New Math Pathways project.

18 amstat news november 2015 THE AMERICAN STATISTICIAN HIGHLIGHTS TAS Special Issue: Undergraduate Curriculum Nicole Lazar

he November 2015 issue of The American thus at an appropriate juncture to step back, take Statistician is a special issue on statistics and stock of what we are doing, and think about what the undergraduate curriculum. You will we could be doing differently or better. Tread about innovations in statistics education Guest editors Jo Hardin and Nicholas Horton beyond the first introductory course. The authors have put together a stimulating set of articles from of these papers tackle a wide range of questions that some of the leading researchers in statistics educa- are central to the way in which we think about our tion, who provide thoughtful answers to many of undergraduate programs. What should the second the questions I’ve raised here. I hope you will find course look like? How should we train our majors? ideas in this issue to improve and reinvigorate your Can we expose undergraduate students to more programs and inspire your teaching. n advanced concepts? To larger data sets? And how can we tell if any of this is working? Several broad themes addressed in the 2014 ASA Curriculum Guidelines for Undergraduate Programs in Statistical Science (www.amstat.org/ ASA Seeks education/curriculumguidelines.cfm) are included in Editor Nominations, this special issue. An overview of “where we are” by George Cobb is sure to generate discussion. Indeed, Applications you can find an online discussion of his paper, fea- turing comments from 19 leading thinkers and edu- The ASA is seeking editor nominations cators (along with his rejoinder), as a supplement to and applications for the following the print edition. Other papers focus on computa- publications: tion, visualization, simulation, data science, theoreti- cal statistics, and program assessment. Journal of the American Statistical Now is a time of great and rapid change for our Association, Theory and Methods discipline in general, and the effects are being felt at the undergraduate level, as well. Data science, CHANCE Magazine Big Data, and the surrounding hype have all led to increased demand on our courses and programs. Journal of the American Statistical Association/ The American Statistician, Reviews

3,000 Statistics in Biopharmaceutical Research

2,500 Nominations and applications for CHANCE 2,000 Bachelor's should be sent to [email protected]. Nominations Master's and applications for JASA Theory and Methods, 1,500 PhD's JASA/TAS Reviews, and SBR should be sent to 1,000 [email protected]. Nominations should include the candidate’s CV, 500 if possible, and a short statement as to why the person Sta$s$cs and Biosta$s$cs Degrees 0 would be an excellent editor of the publication. 1987 1992 1997 2002 2007 2012 Interested persons are encouraged to apply. Applicants should send a CV; the names of three Statistics and biostatistics degrees at the bachelor’s, master’s, and doctoral levels in the United States. Data references; and a letter of interest in the position, source: NCES IPEDS. including a brief statement of your vision for the publication, directions you would pursue, and con- The undergraduate major continues to grow, tributions you would make if selected as editor. both in the number of programs offering the bach- Nominations should be sent by December 11. elor’s degree and in the number of degrees awarded, Applications should be sent by February 5, 2016. which quadrupled between 1987 and 2012. We are

november 2015 amstat news 19 columns

MASTER’S NOTEBOOK Statisticians in Sierra Leone Dana Haberling and Jodi Vanden Eng

Getting to Zero Everyone from CDC working in Freetown lived Dana Haberling is a statistician in and worked (we took over a conference room for an the Prion and Public Health Office in emergency operations center) in the same hotel. The the Division of High-Consequence average day felt a bit like being a freshman again: Pathogens and Pathology at the Centers roll out of bed, load up on a buffet breakfast, walk for Disease Control and Prevention in down the hall, stick my head in a laptop, take quick Atlanta, Georgia. In January 2015, she breaks for ramen or coffee, maybe join happy hour spent a month working in Sierra Leone with colleagues (including laptops, work phones, as part of the ongoing Ebola emergen- and usually talk of Ebola), have a quick Skype with cy response. She earned her MSPH in biostatistics from loved ones, and then work in my room until it was Emory University School of Public Health. late/early enough to get some sleep in preparation for the next day. Repeat 28 times. The days were As statisticians, we don’t often get opportunities to long. But, because I was so into the work and on get out in the field. But hey, I work for the Centers the go, it was a week before I realized just how close for Disease Control and Prevention (CDC), too! I to the beach I was. love public heath, too! And I think it sounds “fun” I ended up as the country data manager and to go out, too! So as soon as I could, I put my name statistician. In many ways, you can imagine what in for a deployment to West Africa to be part of the it was like: getting new data, trying to understand Ebola response. and clean it both for regular and ad-hoc analysis, In real life, I am a “health statistician” and I tend to trying to figure out what had been done previously, act something like a consultant to my group at CDC. interpreting hand-me-down code, and wondering My division covers many diseases, and so I see many how I might be able to improve anything for the types of data and studies. We’re heavily slanted toward next person. epidemiology, but the job really requires a statistician’s In the beginning, it was a little intimidating thinking and a statistical programmer’s capabilities. I to step into the middle of a big production that knew there were a lot of data going around in West seemed to be going along just fine without me. Africa and figured I would have something to offer But much like being a freshman, I quickly found that others did not. That said, for an outbreak, you my place. I got to know other CDC people (some know going in that you have to be flexible. from my own building in Atlanta and others from The amount of time and effort it takes to get any international offices), the Sierra Leone Ministry of one person out the door is staggering. On top of Health staff, and others from different organizations my typical week and preparing to have my current or NGOs. projects under control, I needed the following: It eventually became normal to see the doorway Official permission from my boss guards with bleach water and a thermometer, sign- ing up for a driver if I needed to go out, feeling com- Training for personal and health safety fortable with road blocks and check points, getting A visa for Sierra Leone dinner or takeout before the city-wide curfew of 6 Plane tickets p.m. (thereby avoiding another hotel dinner), and navigating new buildings and meeting protocols. Budgeting paperwork Several days of my week were spent working at Clothing that washes well in the sink and can the Ministry of Health. My main task there was air dry pushing to streamline the data acquisition from Food, soap, and whatever else I might need various districts, each with their own data manager, and helping get a weekly national data set pub- Any number of arrangements for the life I was lished. We provided both analytical and technical leaving behind for 29 days support for their daily count reports; others from This is all done with the help of CDCers in our epidemiology team taught the staff how to use Atlanta volunteering to work Ebola administration mapping software to reproduce what the CDC had roles. Two days after Christmas, I boarded a plane been providing. for Freetown, Sierra Leone.

20 amstat news november 2015 columns

Statistical work on the ground in Sierra Leone was has a large international component. In addition like working with the cartoon angel and devil on my to statistical analysis, much of my fieldwork has shoulders: The angel wanted me to examine my data, involved developing tools and training ministry follow assumptions, and produce organized and of health staff on the use of mobile data collec- beautiful graphics; the devil nagged me to rush and tion, specifically using PDAs and smartphones to just get something out the door! Professionally, this improve data collection for national household sur- was the biggest challenge. And everything required veys, surveillance, and other research projects. patience—patience with myself for not working fast I have traveled to West Africa several times. In fact, enough, patience with my co-workers who needed I had been to Sierra Leone four times previously to data/graphs/tables/p-values/something shiny right work on malaria prevention activities before being now for a presentation in five minutes(!), patience deployed for the Ebola response. I thought I was well learning how partner organizations worked and prepared. I was used to the late night ferry crossing how to fit in with that, patience learning the needs when you arrive at the airport; I was prepared for and abilities of the Ministry of Health staff, patience the hot harmattan weather (a dry, dusty wind); I was waiting for data updates. Patience. My biggest les- familiar with the geography, bumpy roads, traffic bot- son learned was that when you aren’t afforded the tlenecks, and dilapidated government buildings; and I time to double-check everything, you still always do already knew some of the local customs, greetings, and your best work. However, you do your best by keep- not-too-distant history of civil war. Moreover, I had ing it simple and trusting your abilities and statisti- experience working in two other emergency deploy- cal intuition. ments—one in 2005 after the tsunami in Indonesia Finally, I must also acknowledge the constant and one in 2010 after the earthquake in Haiti. and real interaction with people who were suffering Time Magazine had just released its Person of the with and dying from Ebola every day. By the time I Year edition, so I bought it at the airport and read it on arrived, the epidemic had been going on for months the plane. I was heartened to read accounts from Ebola without a break, and no one knew when it might fighters, nurses, and doctors—the heroes on the front slow down. The entire region suffered horribly, and lines risking their lives to treat patients with a deadly many thousands lost their lives or family. But what and devastating virus. It was exciting to see a full pro- I consistently saw were people who were grateful for file picture and comment from CDC Director Tom our presence, patient with our questions, and always Frieden at the end of the article, and it was sobering welcoming and gracious. to have a sense of security because I thought the clos- Those who volunteered to work on the est exposure I would have to someone with Ebola was ground—in any capacity—may have had different analyzing his or her data. motivations for going to West Africa, but there was I thought I was prepared, but when I arrived in comfort in knowing we were all there for the same Sierra Leone on December 21, 2014, for a six-week reason: to get to zero. deployment, the epidemic was at its peak. About 500 new cases were being reported every week and it seemed no end was in sight. The differences from my previous visits were Seeing the immediately apparent. Instead of people holding hands and hugging in general social situations, there Big Picture were nods and faint smiles. Curfews at night prevent- Jodi Vanden Eng is a statistician in ed social gatherings. Roadblocks with hand-washing the Division of Parasitic Diseases and stations and fever checks had become routine. And Malaria at the Centers for Disease drivers yielded to ambulances and any vehicle with Control and Prevention in Atlanta, EBOLA RESPONDER on the windshield. Georgia. She spent six weeks working in Sierra Leone During an emergency, flexibility and resourceful- as part of the ongoing Ebola emergency response. She ness are important. I thought I would be supporting earned her MPH in epidemiology from Yale University data management and analysis, but a colleague and I School of Public Health and MS in biostatistics from the were given the task of assessing the surveillance process University of Michigan Ann Arbor School of Public Health. and data flow for the country’s Ebola response. This seemed insurmountable. How do you describe the As a mathematical statistician at CDC, I consider surveillance and information systems of a newly myself a statistical programmer and consultant for introduced disease for an entire country? How do epidemiologists. I’ve had the opportunity to work you account for the variation between districts, internationally many times while working in the partners, and nongovernmental organizations? Division of Parasitic Diseases and Malaria, which Despite the obstacles, it had to be done, so we developed a small standardized set of questions to november 2015 amstat news 21 columns

determine the data inputs, uses, and outputs during This can get chaotic. Nobody is thinking, “This may the stages of an Ebola response. be better in a standardized relational database on the We visited Ebola response centers in several districts cloud.” Nonetheless, real-time data during an outbreak and assessed the data flow from beginning to end from are invaluable for coordinating response efforts. Not the patient’s perspective, including alerts (like 911), only do data describe the epidemiology of the disease, ambulances, case management, labs, quarantine, hold- but they also identify areas with the greatest needs and ing and treatment centers, contact tracing, and burials. allow donors to direct aid and resources efficiently and At each of these stages, we attempted to identify varia- cost effectively. tions among districts and potential data gaps and their I saw public health workers tirelessly visiting every causes. We assessed information technology infrastruc- health center to train and retrain staff on standard ture and resources. Most importantly, we obtained this infection control procedures and survivors staying at information by meeting with key individuals without treatment centers to provide care for other patients. interrupting their primary response efforts. Special unplanned meetings were held to figure out During this assessment, I sometimes had mixed where to place newly orphaned children or how to care feelings about my work. Data collection and statisti- for a pregnant woman going into labor while in quar- cal analyses in emergency response seem like a catch- antine. A volunteer set up a village hotline to enable 22. Data collection is often an extremely low prior- families to get feedback about their ill loved ones. ity—and rightfully so. Health centers need supplies; Although these types of efforts cannot be measured patients need beds, care, and transport; lab specimens with data, they are just as significant and crucial in a need testing; potential contacts need tracing and mon- response. itoring to stop the chain of transmission; and families Working on this data assessment allowed me to see need closure for the loss of their loved ones. When the big picture of the outbreak. In doing so, I real- people are dying, data are secondary. ized how much the little things matter. From well-wish Data are collected in the way they are most use- letters sent by U.S. school children to encourage us ful. In treatment centers without electricity, this to care packages sent by CDC colleagues and family means in a logbook and on a whiteboard. In urban members during the holidays, from the warm greet- areas with more resources, it means smartphones ings and smiles of gratitude shared with the Sierra and geographic information systems. At the national Leoneans to the waving hands of small children, we level, it means reconciling and integrating all the were reminded often that we had so much to work data sources to make meaningful interpretations. for and protect. n

22 amstat news november 2015 SCIENCE POLICY Technological Advance and Economic Growth: Stagnation or Revival? Gordon Reikard In March 2011, Gordon Reikard published a widely read Amstat News piece, “Stimulating Economic Growth Through Technological Advance,” (http://bit.ly/1PGR0dq) exploring the con- nection between investment in technology and economic growth. Revisiting the topic four and half years later, Reikard discusses the prospects for economic growth through the end of the decade. —Steve Pierson, ASA Director of Science Policy ver the last 10 years, the most commonly used measure of technologi- calO advance—total factor pro- ductivity (TFP)—has slowed considerably, especially when compared with the growth rates achieved during the previous two decades. This has led to a pro- tracted debate about the future of economic growth. From one per- spective, the United States is in an ongoing period of secular stagna- tion. The countervailing view is that this is a temporary hiatus and Gordon Reikard growth rates will pick up. Figure 1: Multifactor Productivity, 1949–2014. Left scale: Annual percent is a statistician To get some perspective on change. Black: Actual. Blue: 7-year moving average. at U.S. Cellular. His research this, it is useful to look at the interests include long-term data. TFP, or multi- and Statistics in 1957. Using this there was a second rise that lasted the economics factor productivity as defined by framework, over the long run, roughly until 2004. Since this of growth and the Bureau of Labor Statistics, is technical advance accounts for time, however, TFP has clearly application defined as output less the con- roughly 38% of growth in gross slowed down. of statistical methods in the tribution of labor and capital. In domestic product (GDP). The early post-WWII period physical sciences. the short run, this measure can be Figure 1 shows the percent has often been described as a gold- volatile, but over longer periods— change in TFP in the private busi- en age for the American economy. a full business cycle—the trend in ness sector from 1949 to 2014 Growth rates were consistently TFP approximates the rate of tech- and a seven-year moving aver- high, while recessions were brief. nological advance. This insight age. Several trends are immedi- Unemployment declined, and owes to the seminal paper pub- ately visible. From the late 1940s inflation remained stable until the lished by Nobel Laureate Robert through the late 1960s, TFP was Vietnam War. One key factor was Solow, “Technical Change and the consistently strong. There was the high growth rate of techni- Aggregate Production Function,” an extended dip in the 1970s, cal advance. From 1949 to 1973, published in Review of Economics but starting in the early 1980s, TFP averaged 2.16% per year.

november 2015 amstat news 23 columns

By the late 1960s, there were The economy also ben- Microsoft Office suite raised the indications that the boom was efited from lower inflation productivity of office workers in slowing down. The slowdown and a milder business cycle: the 1990s, but much more pow- hit with a vengeance in the mid- Expansions were long, while erful statistical and analytical pro- 1970s. Some of this was due to recessions were shallower. From grams—ranging from SAS to the shocks. OPEC raised oil prices by 1983 to 2004, TFP grew by open software package R—have 400% from 1973 to 1974, and an 1.4% per year. The heyday was now been widely disseminated. additional 150% from 1979 to the period from 1996 onward, The emergence of large data- 1980. The inflation rate increased when TFP grew by better than bases and the software to mine and the economy underwent deep 1.7% per year. them (“Big Data”) has excited recessions in 1975, followed by a But then, starting around considerable interest. To date, double recession in 1980 and 1981 2005, gains slackened again. most of the applications have been to 1982. From 1970 to 1982, TFP From 2005 to 2014, TFP grew in marketing, with limited impact slowed to only 0.49%. TFP nor- by 0.57%, only marginally high- on growth. Instead, the real poten- mally declines during downturns, er than in the 1970s. The 2008– tial lies in the sciences, particularly and the successive recessions were 2009 recession played a role, fields like biochemistry. unusually deep. but the business cycle is only A third area is biotechnology, But the energy shocks and part of the story. By the middle which has applications in sev- business cycle volatility masked part of the last decade, many of eral areas. These include health a more important underlying the gains from computers had care, agriculture, non-food use of problem. The gains in standards already been achieved. crops, and other products such as of living from the major techno- This has serious implica- biodegradable plastics and biofu- logical advances of the 20th cen- tions for long-term growth. els. Research in genetics has the tury had already been achieved. As the baby-boom generation potential to achieve major gains To revive, the economy needed retires, the labor force will level in medical care. new technologies. off, or even decline. As the An even more intriguing This came in the form of contribution of labor inputs possibility is that technological computers and software. By the falls to zero, all future growth advances are already showing up early 1980s, personal computers will have to come from capital in the national income accounts. were becoming widely available, investment or new technology. In 2014, income ran ahead of and they were increasingly being Adding TFP to the effect of product, and in the second quarter equipped with powerful soft- capital, the growth rate would of 2015, GDP picked up. In the ware. This led to a series of spill- average about 1.5% annually. mid-1990s, income also grew fast- over effects into other industries, At the same time, there are er than product, leading to a major which generated both gains in embryonic trends toward a new revision that revealed growth had efficiency and entirely new prod- wave of technological advances. been higher than previously real- ucts and services. Electronic bank- Artificial intelligence has now ized. One of the key elements of ing made capital more mobile. advanced to the point where this revision was the explicit mod- Computerized inventory control much greater automation will eling of qualitative improvements enabled manufacturers to reduce be possible using intelligent in information technology using fluctuations in stocks. CAD-CAM machines. Manufacturing was hedonic price increases. made it possible for engineers to already heavily automated during While one should not read too design new products online. Word the 1990s, but with AI, it will be much into a few quarters worth processing revolutionized offices. possible to integrate production of data, the bulk of the evidence Local area networks made it pos- lines with new techniques such as argues that a revival in technical sible to equip all the workers in a 3D printing and nanotechnology. advance is, in fact, quite pos- company with advanced technolo- Productivity has been consis- sible. During each prior wave, it gy. The Internet led to a revolution tently strong in manufacturing has taken some time—often well in worldwide communications. industries, in part because these over a decade—to translate new For statisticians, regressions could technologies can be deployed designs into usable products. In be run instantaneously, and even more easily to produce goods. this sense, the productivity slow- very sophisticated models such But even in the service sector, the down of the last decade may be a as neural networks could be pro- capabilities of software are only transitory lull before the next set grammed with a few commands. beginning to be exploited. The of technologies comes online. n

24 amstat news november 2015 statistician's view

STATISTICIAN’S VIEW The Identity of Statistics in Data Science Thomas W. Jones

ata science has been generating consider- of data management. Survey the requirements of able interest inside and outside of the sta- most graduate statistics programs. There is a core of tistics community. Within the statistics courses in measure-theoretic probability, theoretical Dcommunity, there is a debate about whether data statistics, and linear models. I am not saying compu- science and statistics are distinct disciplines. This tation, database management, and application foci conversation about data science betrays an anxiety are absent. But the degree to which such courses are about our (statisticians’) identity. emphasized, or even offered at all, is highly variable. In a July 2013 article in Amstat News, “Aren’t What proportion of programs require a scientific We Data Science?” former ASA president Marie databases course or a high-performance computing Tommy Jones is the Davidian summarizes these concerns: “I’ve been course? We are well trained in quantifying uncertain- director of data science told of university administrators who have stated ty and deriving asymptotics. We are poorly trained at Impact Research, LLC. He holds an their perceptions that statistics is relevant only to in the tools of modern data management. MS in mathematics ‘small data’ and ‘traditional’ ‘tools’ for their analy- What has driven this structural break? Data have and statistics from sis, while data science is focused on Big Data, Big proliferated. This isn’t about the volume of data in Georgetown University Questions, and innovative new methods.” a “Big Data” sense, but rather that data are more and a BA in economics Similarly, Norman Matloff titled his November popular. More data sets exist. More people are ana- from the College of Amstat News William and Mary. He 2014 editorial in “Statistics Losing lyzing data. It is no longer the case (if it ever was) is a PhD student in Ground to Computer Science.” He raised many that only scientists, trained to deal with complexity, the George Mason good points, but his title cuts to the heart of our anx- are the consumers of data products. The need for University Department of iety. Does “data science” mean we’re being replaced? compelling visualizations and narratives to convey Computational and Data I believe this anxiety stems from an overly- complicated stories has increased. Sciences. He specializes broad definition of statistics and an unclear defini- As models have become more accurate, they have in statistical models of language and time series tion of data science. For my part, I’ve come to see also become more complex. modeling. data science as supply chain management for “data Ensembles of models are often better predic- products.” This supply chain starts with real-world tors than any single model. Ensembles are empiri- problems and ends with a report, business decision, cally accurate, but their asymptotic properties are or software. The middle contains lots of statistics, often unknown. And an additional question arises: databases, programming, communicating, etc. Data Asymptotic to what? One could take any or all of science is fundamentally multidisciplinary. “But,” the number of observations, predictors, models in you may ask, “isn’t that just statistics?” the ensemble, etc. to infinity and possibly arrive at Davidian’s article is titled “Aren’t We Data different solutions. In the age of Big Data, asymp- Science?” after all. Randy Bartlett answered “We Are totic properties matter. Data Science” in a subsequent Statistician’s View. This Finally, data are bigger in a Big Data sense. “everything data” definition of statistics is popular Storing, moving, and processing terabytes of data is among statisticians. Former ASA President Robert neither simple nor all “statistical” in nature. There Rodriguez championed this view in 2012, offer- has long been a working relationship between statis- ing ASA as a “big tent.” The popular blog Simply tics and computer science. But now software engi- Statistics states, “Whenever someone does something neering knowledge is required if any useful analysis with data, we should claim them as a statistician.” is to come from a Big Data project. There is historical precedent for this claim. Whither statistics? Statistics as a discipline originated in the 18th century. Least squares dates to the early 1800s with Gauss and The More Things Change, the More Legendre. We statisticians were the only data game in They Stay the Same town, even as statistics became tied with mathematical In an age of Big Data, I believe statistics’ focus on probability in the 19th and early 20th centuries. probability and asymptotic properties is more valu- Yet times have changed. Judging by current sta- able, not less. As we move toward more complex tistics curricula, statistics is more closely tied to the statistical and machine-learned models, there is still mathematics of probability than to fundamentals a need to understand the properties of and to get

november 2015 amstat news 25 statistician's view

... [T]he fundamentals of probability and Wager, Trevor Hastie, and Bradley Efron propose methods to get prediction errors of bootstrapped and bagged learners (http://arxiv.org/pdf/1311.4555. mathematical statistics taught today pdf). Abhijit Dasgupta et. al show how to estimate effect size using nonparametric, “black box” mod- add tremendous value and cement our els (www.ncbi.nlm.nih.gov/pubmed/24581306). Andrew Womack, Elias Moreno, and George identity as statisticians in data science. Casella have shown that a popular model for text mining is an inconsistent estimator (www.stat.ufl. edu/~ajwomack/WMC-LDA.pdf). But Sometimes, Things Just Change Further Reading While many of the fundamental problems fac- “Aren’t We Data Science?,” http://magazine.amstat.org/ ing statisticians are the same, the applications and environment are different. Statistics education, par- blog/2013/07/01/datascience ticularly at the graduate level, must adapt. As data “Statistics Losing Ground to Computer Science,” http:// get “bigger” and research and applications become magazine.amstat.org/blog/2014/11/01/statistics-losing- more multidisciplinary, the need for statisticians to ground-to-computer-science communicate and collaborate with a wide range of professionals and laypeople increases. “We Are Data Science,” http://magazine.amstat.org/ Statistics education should require minimum blog/2013/10/01/we-are-data-science competency in fundamentals of computer science. “ASA President Robert N. Rodriguez Calls for Creating the ASA’s recent statement, “The Role of Statistics in ‘Big Tent’ for All Statisticians,” www.amstat.org/news/pdfs/ Data Science” highlights three data science skillsets: RodriguezSpeech8_13_12.pdf database management, statistics and machine learn- ing, and distributed and parallel systems. Statisticians “Statistics/Statisticians Need Better Marketing,” http:// must work closely with software engineers to develop simplystatistics.org/2012/08/14/statistics-statisticians-need- solutions that scale. We must understand the code so better-marketing that scaled solutions still have desirable statistical prop- erties. I believe that statisticians should have minimum “A Very Short History of Data Science,” http://www.forbes.com/ foundational training in database management and sites/gilpress/2013/05/28/a-very-short-history-of-data-science/ high-performance computing. “The Role of Statistics in Data Science,” http://www.amstat. In addition, examples and applications in intro- org/misc/DataScienceStatement.pdf ductory statistics courses may need updates. For example, ensemble methods will be at least as impor- tant as linear regression in the coming years. We may inferences from these models. A (computational) consider teaching concepts like Zipf’s and Heap’s data scientist once told me “statisticians will be laws early on, as analyses of linguistic data are grow- the ones to help us figure this mess out.” These are ing more common. questions at the heart of theoretical statistics. It is an exciting time to be a statistician. Statistical And in a world that is streaming data, careful models and methods are applied in ways unimagi- research design and data collection are as important nable only a decade ago. Airplanes fly themselves; as ever. A biased sample is still biased if it has a mil- doctors use statistical models to aid diagnoses; scien- lion observations. This is especially important when tific research involves mining massive data sets. The the data are born of the Internet and people implic- importance of these tasks makes understanding our itly or explicitly opt in. These are challenges survey models an imperative. Yet, fundamental statistical statisticians face regularly. properties of these models remain little understood. Recent research by statisticians is tackling some I am not convinced that statistics is data science. of these issues. Gerard Biau, Luc Devroye, and But I am convinced that the fundamentals of prob- Gabor Lugosi have demonstrated the consistency ability and mathematical statistics taught today add of averaging classifiers (http://machinelearning. tremendous value and cement our identity as statis- wustl.edu/mlpapers/paper_files/biau08a.pdf). Stefan ticians in data science. n

26 amstat news november 2015 meetings

Join Us for the 5th Annual Conference on Statistical Practice Amy Farris, ASA Director of Membership Development and Marketing

he fifth annual ASA Conference on Mentoring Program Statistical Practice will be held in San Diego, Participate as either a mentor or mentee.San Register Diego, for Californi a California, February 18–20, 2016. Once the conference and the ASA will automaticallyFebrua sendry 18–20 Tagain, CSP will provide opportunities for attendees 2016 you information about how to participate in the to learn new statistical methodologies and best mentorship program. There is no cost to participate, practices in statistical analysis, design, consulting, but mentees are selected via an application process and statistical programming. and participation is limited. LEARN MORE The steering committee has organized an inno- For conference updates, vative and engaging program, including concurrent Career Service follow @AmstatNews on Twitter and use sessions, courses, tutorials, posters, and practical CSP also provides an online career service system to #CSP2016. computing demonstrations. The conference will connect employers and applicants so they can pur- also host exhibits. sue informal meetings and interviews. All registered The intimate and personal format of CSP pro- CSP attendees will be able to post their résumés at vides a unique opportunity to strengthen your no charge. Applicants can access current job openings relationships within this unique community while and contact employers in advance of the conference. enhancing your career. Early registration for the conference is open Keynote Address through January 5, 2016. For details and deadlines, The keynote address, “Communicating the Value of see www.amstat.org/meetings/csp/2016. Our goal is Statistics” will be delivered by ASA President-elect to maintain the intimate and personal feel of the Jessica Utts. conference, so register before the conference and courses are filled. n

november 2015 amstat news 27 education Educational Ambassador from Cuba, Nigeria Sought

he ASA is seeking qualified Cuban and The selected ambassador will be expected to com- Nigerian nationals to serve as the 2016 plete the following by December 2017: Educational Ambassador (www.amstat.org/ • Attend the 2016 Joint Statistical Meetings (www. Teducation/educambassador.cfm). The selected ambassa- amstat.org/meetings/jsm/2016) in Chicago, Illinois, dor will be subsidized to attend the Joint Statistical to take CE courses in an emerging area of statistics Meetings to take one or more Continuing Education research (CE courses offered in 2015 are available at (CE) courses in an emerging area of research and www.amstat.org/meetings/jsm/2015/onlineprogram). return to teach the subject matter in his or her home country and/or region. • Return to your home country, and within the next year, teach a one-semester class with no fewer than The nomination deadline is December 31. 10 students on the subject matter of the CE course at the master’s level and write lecture notes in the BIENNIAL ELECTRONIC CONFERENCE ON TEACHING STATISTICS language locally deemed most appropriate. You Call for Proposals will need to repeat the class the following semester. Proposals are being accepted through February 1, 2016, for the (The committee understands there can be practical Electronic Conference on Teaching Statistics (eCOTS), to take place and organizational considerations that can make it May 16–20, 2016. The online conference will focus on undergraduate difficult to meet this requirement. Should this arise, statistics education. the committee will be prepared to work with the ambassador to devise an alternative plan. The ASA The theme is “Changing with Technology.” Technology is inspir- will also investigate the possibility of providing fur- ing change in many aspects of statistics education: what we teach (cur- ther assistance through recorded webinars related riculum), how we teach (pedagogy), who we teach (audience), and why we to the CE course material.) teach (goals). Proposals can address technology-induced change in any of these areas. • Submit a report by December 2017 to the ASA Also included in the conference are virtual panel discussions, breakout through the Committee on International Relations sessions, workshops, posters, birds-of-a-feather discussions, and—new this in Statistics and the ambassador’s home institution year—regional face-to-face mini-conferences connecting to eCOTS. describing how the transfer of knowledge has been The types of proposals sought include the following: accomplished and how you invited statisticians Breakout sessions, which will be half-hour webinars about a particular from your country to work with you on the subject topic. Audience engagement during these sessions is encouraged. In addi- matter of your course. tion to leading the actual session, some breakout session leaders may be The ASA will arrange and pay for the selected organized into panels addressing a common topic. Each submission will ambassador’s airfare, hotel, meeting registration, and be reviewed. CE course registration. The ASA also will reimburse Virtual posters, which will record a brief (maximum 5 minutes) pre- the visa application fee and provide a $500 stipend. sentation to be posted on the website and displayed throughout eCOTS. Candidates for the ambassadorship must hold a Conference attendees will be encouraged to view posters throughout the PhD or equivalent degree in statistics or a related week and leave comments or questions. Presenters are encouraged to check field and be employed at the level of assistant or asso- their posters frequently and respond to comments. ciate professor at a recognized university or research Birds-of-a-feather discussions, which will take place during the pro- organization promoting the teaching and application gram gaps built in for regional conferences and provide an opportunity for of statistics. those not located near a regional conference to discuss a particular topic. Interested candidates should send the following to Discussion leaders should be prepared to organize and facilitate discussion Geert Molenberghs at [email protected]: (not give a presentation) and use GoToMeeting software. All proposals should be related to the theme and include a title, brief • A curriculum vitae abstract, description of how the topic relates to the theme, and—for break- • A proposal (at most two pages) describing how the out sessions—a description of how the session will engage participants and ambassadorship will help disseminate knowledge of what you hope participants will take away from your presentation. statistics in your country and how the ASA’s expec- For details, visit www.causeweb.org/ecots/ecots16/proposals. tation set forth above will be implemented upon For questions, contact the program chair, Kari Lock Morgan, at your return home [email protected]. • A letter from the head of your academic unit (e.g., department chair, director, or the dean) indicating institutional support for your dissemination plan n

28 amstat news november 2015 people news

five master’s and seven PhD methodological research. students. Her research contribu- She also serves as associate tions in the areas of statistical director of the Statistical and theory and the application of Applied Mathematical Sciences statistics to medicine, immunol- Institute. As NISS direc- ogy, engineering, social science, tor, she will be based in the and education include more Washington, DC, office. than 100 primary research For more information about publications and co-authorship NISS, visit www.niss.org, fol- of four books. low @NISSSAMSI on Twitter, At NISS, Sedransk has con- and join the NISS group on n Nell Sedransk tinued to mentor postdoctoral LinkedIn and Facebook. fellows and engage in statistical National Institute of Statistical Sciences (NISS) Chair Mary Batcher recently announced that the institute’s new director is Obituaries Nell Sedransk. “The search committee felt Charles Edgar “Chuck” Gates that Nell was the best person for Charles Edgar “Chuck” Gates, 89, of Bryan, Texas, died peacefully September 8, this position. She brings a wealth 2015, at his residence. of experience and expertise to An ASA Fellow and longtime member, Gates was born March 6 in Rapid City, the role of director and has done South Dakota. He was a veteran of World War II and the Korean War and worked a tremendous job restructuring as a statistician and professor at the University of Minnesota for 10 years, and then the organization this past year to for 28 years at Texas A&M University. His specialties were in experimental design, help ensure its future success,” wildlife abundance estimation, and agricultural application. During his career, he was noted Batcher. an editor of The American Statistician and travelled extensively, visiting 79 countries In her new role, Sedransk will over the course of his life. lead the institute into its second Following his retirement, Gates served as an income tax volunteer for the AARP quarter century with its current and was a member of Covenant Presbyterian Church. His other interests included foundation of statistical research playing bridge, collecting stamps, playing tennis, bicycling, and dancing. and service to the NISS affiliate In addition to his wife, he is survived by three daughters: Greta Watkins and hus- organizations from academia, band, Randy, of Bryan, Texas; Gail Gates of Stillwater, Oklahoma; and Glenda Honea government, and industry. “I am and husband, Robert, of Danville, Virginia. very pleased to have this oppor- In lieu of flowers, the family requests donations to Hospice of Brazos Valley, tunity to help NISS continue to Covenant Presbyterian Church, or the Alzheimer’s Association at alz.org. grow and thrive,” said Sedransk. “I look forward to the coming months as we begin to focus on Alvan “Al” Zarate increasing research activities, Alvan “Al” Zarate passed away August 27, 2015, at Washington Hospital Center expanding the postdoctoral fel- due to complications from treatment for pancreatitis. He was 78. lowship program, and engaging Al worked for the National Center for Health Statistics (NCHS) as a confiden- with new collaborators and cli- tiality officer who was dedicated to protecting the confidentiality of NCHS survey ents for NISS.” respondents. In 2008, he was elected fellow of the American Statistical Association. A Sedransk has served as acting year later, he received the NCHS Program Mission Support Award for his leadership director since September 2014, in the field of data privacy and confidentiality. having been associate director Al earned a PhD from Brown University and taught sociology/demography at The since 2005. She came to NISS University of Texas, Tulane, and Ohio State before joining NCHS in 1978. from the National Institute Al was also a talented musician who loved to play the guitar and sing—which he of Standards and Technology, often did at NCHS. In the last several years, he sang with the Bowie Senior Chorale where she was chief of the statis- and was a member its smaller traveling group, the Choraleers, who did special perfor- tical engineering division. mances throughout the Bowie community. He loved performing and the community. Since earning her PhD from Al was always happy when he had a guitar in his hand. Iowa State University, Sedransk To read more about Al’s life or make a memorial contribution to charities Al sup- has spent much of her career in ported, visit the memorial networks obituary page at http://bit.ly/1MIsCr8. academia, where she directed

november 2015 amstat news 29 awards & deadlines

Deadlines and Contact Information for ASA National Awards, Special Lectureships, and COPSS Awards Deadlines and Awards Nominations Questions December 15, 2015

COPSS Fisher Lectureship and Award www.copss.org Pam Craven Edsel A. Pena ASA Noether Senior and Young Scholar Awards [email protected] [email protected] Pam Craven John L. Czajka Monroe. G. Sirken Award in Interdisciplinary Survey Methods Research [email protected] [email protected] January 15, 2016

COPSS Presidents’ Award www.copss.org

Elizabeth L. Scott Award www.copss.org

February 1, 2016 Karl E. Peace Award for Outstanding Statistical Contributions for the Pam Craven Paul S. Albert Betterment of Society [email protected] [email protected] Pam Craven ASA W. J. Dixon Award for Excellence in Statistical Consulting [email protected] February 15, 2016

Causality in Statistics Education Award [email protected] Pam Craven Mary J. Kwasny Harry V. Roberts Statistical Advocate of the Year Award [email protected] [email protected] Pam Craven Lynne Billard ASA Samuel S. Wilks Memorial Medal [email protected] [email protected] Pam Craven Bradley A. Hartlaub ASA Waller Distinguished Teaching Career Award [email protected] [email protected] Pam Craven Bradley A. Hartlaub ASA Waller Education Award [email protected] [email protected] Pam Craven Blaza Toman ASA W. J. Youden Award inInterlaboratory Testing [email protected] [email protected] February 20, 2016 Philip J. Ramsey Philip J. Ramsey ASA Statistics in the Physical and Engineering Sciences Award [email protected] [email protected] February 23, 2016 Pam Craven Eloise E. Kaizar ASA Gertrude M. Cox Scholarship [email protected] [email protected] March 1, 2016 Pam Craven Pushpal Mukhopadhyay ASA Edward C. Bryant Scholarship [email protected] [email protected] Pam Craven Morteza Marzjarani ASA Excellence in Statistical Reporting Award [email protected] [email protected] J. Jack Lee ASA Fellows www.amstat.org [email protected] Pam Craven David R. Morganstein ASA Mentoring Award [email protected] [email protected] ASA Outstanding Statistical Pam Craven DuBois Bowman Application Award [email protected] [email protected] March 15, 2016 Pam Craven David R. Morganstein ASA Founders Award [email protected] [email protected]

30 amstat news november 2015 awards & deadlines

Sirken Award The Monroe G. Sirken Award in Nominations Wanted for 2016 Don Owen Award Interdisciplinary Survey Methods The San Antonio Chapter is accepting nominations from ASA Research honors the accomplish- chapters in North America for the 2016 Don Owen Award, which ments of its namesake, recog- is presented to a statistician who embodies the three-fold accom- nizes the accomplishments of the plishments of Donald B. Owen: excellence in research, statistical awardee, and enhances awareness consultation, and service to the statistical community. among the statistical community Nominees must be a member of the ASA, but are not required of the scope and importance of to be a member of the nominating chapter. In addition to a cover Sirken’s contributions. letter highlighting the accomplishments of the nominee, the nomi- Each year, an individual is nation packet must contain the following supporting information: selected to deliver the Sirken Lecture at the Joint Statistical • Name of the nominee Meetings and receive an hono- rarium. All lecturers are selected • Degrees (titles, dates, schools) from nominations submitted by ASA members. • Present position(s), followed by significant former Additional information about positions (with dates) the award, including informa- tion about submitting a nomina- • List of major publications having statistical content tion, can be found at www.amstat.org/awards/ • List of activities related to teaching and dissemination monroesirkenaward.cfm. n of statistical knowledge

Deming, Noether Awards • List of consulting activities related to statistical The ASA Deming Lecturer problems or editorial contributions Award and Gottfried E. Noether awards honor the accomplish- • List of activities supporting the mission of the ASA ments of Gottfried E. Noether and related professional organizations and W. Edwards Deming, recognize the accomplishments To nominate someone, send the nomination packet as a PDF of the awardee, and enhance via email to Joel Michalek at [email protected] with “Owen awareness among the statistical Award” in the subject field. Alternatively, six copies of the nomi- community of the scope and nation packet can be mailed to Joel Michalek, Department of importance of Noether and Epidemiology and Biostatistics, University of Texas Health Deming’s contributions. Science Center at San Antonio, 7703 Floyd Curl Drive, San Each year, those selected Antonio, TX 78229. deliver a lecture at the Joint The deadline for nominations is February 5, 2016. Statistical Meetings and receive Before his death in 1991, Owen was distinguished profes- an honorarium. All lecturers are sor of statistics at Southern Methodist University in Dallas. He selected from nominations sub- authored seven textbooks, seven monographs, and more than 75 mitted by ASA members. articles in refereed journals; trained 19 doctoral and master’s stu- Additional information about dents; served as an applied statistician for 10 years at the Sandia each lecture, including informa- Corporation; and operated a private consulting firm that special- tion about submitting a nomina- ized in quality control. Owen was editor of Communications in tion, can be found at Statistics for both Series A and B, associate editor of Technometrics www.amstat.org/awards/ and JASA, and editor of more than 50 textbooks. awardsscholarships.cfm. n

november 2015 amstat news 31 section • chapter • committee news

Selection of Winners Papers will be reviewed by an MDD committee. Criteria for selection include novelty in theory/meth- ods/applications, significance and potential impact of sectionnews the research, and clarity of writing and presentation. Decisions of the committee are final. Biometrics Applicants will be notified by January 30, 2016. Edited by Sheng Luo, Section Publications Officer The winners (including those receiving honorable The Biometrics Section is seeking applications for mention) must submit abstracts and register for the 2016 David P. Byar Young Investigator Award. JSM by February 1, 2016, through the official JSM This annual award is given to an early-stage inves- abstract submission system (http://bit.ly/1LSQlzP). tigator for best paper to be presented at JSM. The Winners must present a talk under the auspices of winner will receive $2,000. For details, visit http:// the MDD on the topic of the winning paper. Those bit.ly/1LeEO0o. who are not selected are encouraged to submit a con- The section also funded an award for a PhD-level tributed abstract to JSM 2016. biostatistician interested in conducting methodologi- The best paper winner and honorable mentions cal or collaborative research in radiology/imaging will be presented with a certificate at the MDD meet- clinical trials to attend the Radiological Society of ing at JSM. For more information about the award, North America (RSNA) Clinical Trials Methodology visit www.amstat.org/sections/studentpaperawards.cfm. n Workshop in La Jolla, California, from January 9–15, 2016. This award went to Yin Xi of The University of Texas Southwestern Medical Center to cover up to Quality and Productivity $3,000 in travel costs. Students in statistics programs can benefit from par- For more information about the Biometrics ticipating in statistical conferences and other profes- Section, access the section’s web page by clicking on sional activities. To aid these students in their involve- “Section” at the top www.amstat.org. n ment, the ASA’s Quality and Productivity (Q&P) Section’s student scholarship program offers support to attend the Joint Statistical Meetings. This year, five Medical Devices and Diagnostics $400 scholarships were awarded to the following stu- The ASA Medical Devices and Diagnostics (MDD) dents to attend JSM in Seattle, Washington: Section will sponsor a student paper competition for • Ahmed Mohamed Eshebli, Missouri JSM 2016. A manuscript, suitable for journal sub- University of Science and Technology mission, is required to enter the competition. The winner will receive a $500 award. • Richard Seymour, Air Force Institute of To be eligible for the award, applicants must be Technology a member of the section and have been a full- time • Michael Lee Walker, University of Alabama student (undergraduate, master’s, or PhD) on or after September 1, 2016, or be within two years of gradu- • Yandan Yang, Southern Methodist University ation if the submitted work was initiated while the • Rong Zheng, University of Alabama applicant was a student. The applicant must be first author of the paper, and the adviser (or other coau- Recipients were recognized at the joint Q&P/ thor) must write a letter stating that the applicant had SPES mixer at JSM. primary responsibility for the research and write-up. Looking forward to 2016, Q&P will offer up to To apply for the award, send the following in PDF three travel awards of $400 each for students enrolled format by December 15 to MDDpaperJSM2016@ in a graduate program with a concentration in applied gmail.com: statistics and/or quality management to attend JSM in Chicago, Illinois, from July 30 to August 4. Student • A curriculum vitae applicants must show a demonstrated interest in qual- • A letter from the adviser (or other coauthor) ity applications as evidenced by course work, research verifying the student status of the applicant topic, or prior work experience. Applicants either pre- and briefly describing the applicant’s role in the senting a paper or participating in a poster session will research and writing of the paper receive extra consideration. Applications will be accepted December 1 to • A cover letter describing the contributions of February 1, 2016. Complete information about the paper to MDD the award and how to apply is posted at http://bit. • The paper—including all tables, figures, and ly/1GKIP8U. n appendices as a single PDF file

32 amstat news november 2015 section • chapter • committee news

Physical and Engineering Sciences Michael Crotty, SPES Awards Chair lenges participants to analyze a government data set The Section on Physical and Engineering Sciences is using statistical and visualization tools and methods. pleased to announce the results of its annual compe- There will be two award categories: professional and tition for contributed papers presented at the 2015 student. One or more topic-contributed or poster Joint Statistical Meetings in Seattle, Washington. sessions at JSM will be established for contest sub- missions. (Presenters are responsible for their own Outstanding Presentation Award registration, travel, and other attendance costs.) • Christine Anderson-Cook, Los Alamos The data set for the GSS Data Challenge 2016 National Laboratory, for “Using Multiple will be the Department of Transportation’s General Objectives to Select an Optimal Split-Plot Estimates System (GES), which is collected by Design for a Mixture-Process Experiment” the National Highway Transportation Safety Administration. The GES contains a representative • Jonathan Stallings (runner-up), North sample of police-reported motor vehicle crashes Carolina State University, for “Carryover of all types. Contestants are free to use any or all Designs Including Washout Periods” of the data in their analysis. The data and other Honorable Mentions information (data dictionaries, manuals, etc.) can be found at ftp://ftp.nhtsa.dot.gov/GES. • Shan Ba, Procter & Gamble, for “Maximum Contestants must email Wendy Martinez at Projection Designs for Computer [email protected] expressing their interest in Experiments” participating by January 4, 2016. For details, visit • Gwendolyn Eadie, McMaster University, for the GSS website at www.amstat.org/sections/sgovt. “Incomplete Data and Measurement Error The section also sponsors the following awards, in the Galactic Mass Estimation Problem” which are typically announced at JSM: • Greg Piepel, Pacific Northwest National • Jeanne E. Griffith Mentoring Award, given to Laboratory, for “Mixture Experiment Design federal, state, or local government statisticians Varying One, Two, and Three Components for efforts supporting the work and developing at a Time” the careers of junior statistical staff • Camilla Sammut-Powell, University of • Wray Jackson Smith Scholarship, to reward Manchester, for “Experimental Designs promising young statisticians and encour- Optimally Attuned for Time Trends” age them to consider a future in government statistics The outstanding presentation awards encourage excellence in presentation and have helped raise the • Roger Herriot Award for Innovation in federal SPES contributed sessions to a higher level of qual- statistics ity. All awards are based on audience evaluations of each speaker. • Pat Doyle Award, given to a person who con- Winners receive a certificate recognizing their tributes to the section in a way that leaves a accomplishment and a cash award. The awards for lasting effect on both the section and ASA the 2015 JSM best presentations will be presented • Student Paper Competition for student and at the SPES mixer during the 2016 meetings in post-graduate papers Chicago, Illinois. Start preparing now for JSM 2016 in Chicago, Consult the GSS website for deadlines and Illinois, and you may see your name listed among other details. the SPES Outstanding Presentation Award winners Finally, it is time to start thinking about the JSM next year. n 2016 contributed program. We hope you will con- sider organizing a topic-contributed (TC) session for Government Statistics possible GSS sponsorship. Proposals for TC sessions must be submitted to the online system by January 14, Wendy Martinez, GSS Section Chair 2016. While not required, it is a good idea to contact The Government Statistics Section is sponsoring our GSS program chair, Morgan Earp (earp.morgan Data Challenge 2016 during the Joint Statistical @bls.gov), with your idea for a TC session before Meetings in Chicago. The contest is open to any- you submit your proposal. The online system (www. one, including college students and professionals amstat.org/meetings/jsm/2016/tcsessions/index.cfm) will from the private or public sector. This contest chal- be open from December 1 to February 1, 2016. n

november 2015 amstat news 33 professional opportunities

Arkansas Professional Opportunity listings may not exceed 65 words, plus equal n The math department at the University opportunity information. The deadline for their receipt is the 20th of the of Central Arkansas is seeking applica- month two months prior to when the ad is to be published (e.g., May 20 for the tions for tenure-track assistant professor July issue). Ads will be published in the next available issue following receipt. in applied statistics, beginning August Listings are shown alphabetically by state, followed by international list- 2016. A PhD or equivalent required. Regular teaching load is 12 hrs/semester. ings. Vacancy listings may include the institutional name and address or be Applicants should upload cover letter, identified by number, as desired. CV, transcripts, teaching & research Professional Opportunities vacancies also will be published on the ASA’s statements, and three reference letters to: website (www.amstat.org). Vacancy listings will appear on the website for the https://jobs.uca.edu (F0472015). Review entire calendar month. Ads may not be placed for publication in the magazine will begin December 7, 2015. More infor- only; all ads will be published both electronically and in print. mation: www.uca.edu/math. UCA is an Rates: $320 for nonprofit organizations (with proof of nonprofit status), EO/AA Employer. $475 for all others. Member discounts are not given. For display and online advertising rates, go to www.amstat.org/ads. California Listings will be invoiced following publication. All payments should be n Department of statistics & applied made to the American Statistical Association. All material should be sent probability, University of California, to Amstat News, 732 North Washington Street, Alexandria, VA 22314- invites applications for a visiting assistant 1943; fax (703) 684-2036; email [email protected]. professor position in statistics; starting Employers are expected to acknowledge all responses resulting from pub- 7/1/2016. Qualifications: research/teach- lication of their ads. Personnel advertising is accepted with the understanding ing excellence; PhD in statistics, biostatis- that the advertiser does not discriminate among applicants on the basis of tics or related fields. Candidates who can race, sex, religion, age, color, national origin, handicap, or sexual orientation. contribute to the diversity of excellence Also, look for job ads on the ASA website at www.amstat.org/jobweb. of the academic community through research, teaching and service are par- ticularly encouraged to apply. Additional information at: https://recruit.ap.ucsb.edu/ apply/JPF00561. An EO/AA employer.

n RAND Corporation is seeking PhD UNIVERSITY OF CINCINNATI statisticians for exciting opportunities to Applications are invited for a Tenure-Track Assistant Professorship in Statistics collaborate on multidisciplinary public The Department of Mathematical Sci- Completed applications consisting of a policy research projects. Openings exist ences has a strong graduate program cover letter, CV, description of research for recent graduates and experienced off ering both MS and PhD degrees in program (if applicable), description of statisticians. See our ad in the September statistics, and in pure and applied math- teaching experience (if applicable), and Amstat News for details or go to www. ematics. The department is dedicated to three letters of recommendation should rand.org/statistics. Applications received excellence in both research and teach- be submitted @ https://www.mathjobs. by December 15, 2015, will receive pri- ing. Statistics faculty in the department org/jobs?job-122-7689. Applicants should are involved in both methodological and also submit a CV and cover letter on UC’s ority. Applications must be submitted interdisciplinary research. Their research recruitment system at the link listed in online following the instructions at www. interests include Bayesian methods and the detailed ad on mathjobs.org. Review rand.org/statistics/jobs.html (search PhD statistical methods for massive data, of applicants will begin on December 1, Statistician). Send questions to Lou_ in general, and computer experiments, 2015. The appointment will begin on Au- [email protected]. EO/AA Employer. high-dimensional data analysis, statis- gust 15, 2016. The position requires a PhD tical machine learning, and graphical degree in the area of statistics or equiva- n The San José State University models, in particular. lent awarded by August 14, 2016. Inquiries Department of Mathematics and Statistics concerning the position should be directed has one tenure-track assistant professor The normal teaching load for research-ac- to, Dr. Siva Sivaganesan, the Search Com- tive faculty is 6 credit hours per semester mittee Chair, or Dr. Shuang Zhang, Head, opening in statistics and computational with a 25% reduction for new faculty in Department of Mathematical Sciences. statistics starting August 2016. Subject each of the fi rst 2 years and a reduced to Budgetary Approval. Job Opening ID service role. The job duties include main- The University of Cincinnati is an equal (JOID): 23386. See the complete posi- taining an active research program In Sta- opportunity/affi rmative action employer tion description at www.sjsu.edu/math/ tistics to constantly seek external fund- with a strong commitment to diversity. We employment. PhD required by time of ing, to mentor graduate students, and to actively seek a broad spectrum of candi- teach both graduate and undergraduate dates including women, people of color, appointment. For full consideration, courses in Statistics and related areas. people with disabilities and veterans. submit all application materials at www. mathjobs.org by December 1, 2015. SJSU is an AA/EOE.

34 amstat news November2015 Delaware n Senior Research Biostatistician. The Nemours Center for Healthcare The University of Vermont Delivery Science seeks a senior research Tenure–Track Assistant Professor biostatistician to join its team in Wilmington, DE. The biostatistician will facilitate planning, conduct and analysis of independent and collabora- The Department of Mathematics and Statistics at the University of tive research. A doctoral degree in bio- Vermont invites applicants for a tenure-track faculty position at the statistics or closely related field and five rank of assistant professor in Biostatistics. Expertise in one or more years of related post-doctoral experience of the following areas is highly desirable: high-dimensional are required. Send CV and cover letter modeling, Bayesian analysis, modeling of failure time data, to Dana Berger: dana.berger@nemours. statistical learning, and statistical genetics. Interest and experience org. Apply online: http://careers.nemours. in cross-disciplinary collaborations is also highly desirable. The org/jobs/57067 EOE. successful candidate will be expected to develop an independent program of methodological research while cultivating collaborations Florida with other investigators, as well as contribute to the educational n University of Miami, School of mission of the university at the undergraduate and graduate levels. Business Administration, Coral Gables, Review of applications begins 12/1/2015. Florida. The department of manage- The University of Vermont is an Equal Opportunity/Affirmative ment science at the University of

Action Employer and actively encourages applications from women, Miami invites applications for two veterans and people with diverse racial, ethnic, and cultural tenure-track positions starting August backgrounds. 2016 subject to budgetary approval. Salaries are extremely competitive and For more information and to apply online, visit www.uvmjobs.com generous summer research support is (search position no. 005888). anticipated. A PhD degree in statistics, operations research or a related field is required. Applications should be sub- mitted by email to MASrecruiting@bus. miami.edu. The University of Miami is an AA/EOE.

n The department of management sci- ence at the University of Miami invites applications for a clinical-track faculty position starting August 2016 subject to budgetary approval. This is a non- tenure track position that focuses on business applications. A PhD degree is statistics, operations Research or a related field is required. Applications should be submitted by email to [email protected]. The University of Miami offers competitive salaries and a comprehensive benefits package including medical and dental benefits, tuition remission, vacation, paid holidays and much more. The University of Miami is an AA/EOE. Kentucky n Tenure-track position, assistant professor, starting August 2016. PhD in statistics, biostatistics, or a related area is required by the start of employ- ment. For details, see www.math.eku.edu. Submit letter of application, curriculum

November2015 amstat news 35 samsi NSF•Duke•NCSU• UNC• NISS

2016-17 Postdoctoral Fellowships at SAMSI

Six postdoctoral fellowships are available at the Statistical and Applied Mathematical Sciences Institute for two Research Programs in 2016-17: Optimization; and Statistical, Mathematical and Computational Methods for Astronomy. Ap- pointments will begin in August 2016 and typically last two years. Appointments are made jointly between SAMSI and one of its partner universities, where teaching is a possibility. Competitive salaries, travel stipends, and health insurance will be offered.

The Optimization Program aims to leverage the synergy amongst mathematics,statistics and operations research to produce groundbreaking advances in three categories: 1) Optimization for large-scale statistical analysis, 2) Statistical approaches for the numerical solution of large-scale optimization problems, 3) Applications of optimization.

For additional information on this program, please see: http://www.samsi.info/OPT

The Statistical, Mathematical and Computational Methods for Astronomy Program focuses on the vast range of statis- tical and mathematical problems arising in modern astronomical and space sciences research, particularly due to the flood of data produced by both ground-based and space-based astronomical surveys at many wave-bands. To cope with the current and future needs of astronomy missions requires concerted efforts by cross-disciplinary collaborations involving astrono- mers, computer scientists, mathematicians and statisticians. The main research areas would include: 1) Astronomical simulations and big data issues, 2) Exoplanets, 3) Gravitational wave astrophysics, 4) High-performance computing, 5) Lightcurve analysis in astronomy.

For additional information on this program, please see: http://www.samsi.info/ASTRO

The deadline for full consideration is December 15, 2015, although later applications will be considered as resources permit.

SAMSI is an AA/equal opportunity employer. All qualified applicants are encouraged to apply, especially women and mem- bers of minority groups.

To apply, go to mathjobs.org, SAMSIPD2016 Job #7448

36 amstat news november 2015 vita, statements of teaching and research phi- Williams College losophy, unofficial graduate transcripts, and three letters of reference online at https:// Assistant Professor jobs.eku.edu. Review of applications will The Williams College Department of Mathematics and Statistics invites applications for two begin 12/15/15. Inquiries should be sent tenure-track positions in statistics, beginning fall 2016, at the rank of assistant professor (in to Daniel Mundfrom, daniel.mundfrom@ an exceptional case, a more advanced appointment may be considered). We are seeking eku.edu. Eastern Kentucky University in highly qualifi ed candidates who have demonstrated excellence in teaching and research Richmond, Kentucky, is an EOE. and who are committed to working with an increasingly diverse student body. The candi- dates will become the fourth and fi fth tenure-track statisticians in the department, joining a vibrant and active statistics group with a newly established statistics major. The teaching Maryland load is two courses per 12-week semester and a winter term course every other January. In n Applications are invited for a tenure- addition to excellence in teaching, an active and successful research program is expected. track assistant professor of statistics for The candidate should have a Ph.D. by the time of appointment. We welcome applications fall 2016. Loyola seeks candidates with from members of groups traditionally underrepresented in the fi eld. a PhD in statistics who can demonstrate Candidates may apply via interfolio.com by uploading their vita and having three letters of the potential for significant scholarly recommendation on teaching and research uploaded to http://apply.interfolio.com/30206. activity and excellence in teaching. To Teaching and research statements are also welcome. Evaluations of applications will be- apply, please visit https://careers.loyola. gin on or after November 15 and will continue until the position is fi lled. All off ers of em- edu to submit a cover letter, CV, an essay ployment are contingent upon completion of a background check. Further information that includes a statement on research and is available upon request. For more information on the Department of Mathematics and on teaching philosophy, sample teaching Statistics, visit http://math.williams.edu/. evaluations, and list of references. EOE. Williams College is a coeducational liberal arts institution located in the Berkshire Hills of western Massachusetts. The college has built its reputation on outstanding teaching and Massachusetts scholarship and on the academic excellence of its approximately 2,000 students. Please n Tenure-track assistant/associate pro- visit the Williams College website (http://www.williams.edu). Beyond meeting fully its legal fessor, mathematical sciences. Bentley obligations for non-discrimination, Williams College is committed to building a diverse and University, a private business university inclusive community where members from all backgrounds can live, learn, and thrive. outside Boston, invites applications for

FACULTY POSITIONS IN BIOSTATISTICS

The Department of Biostatistics at The University of North Carolina at Chapel Hill is seeking applications for up to two tenure-track or tenured positions. The appointments will start in the summer or fall of 2016. Appointments may be at the Assistant, Associate or Full Professor level, depending on qualifi cations. At least one position will be at the Assistant Professor rank. A doctoral degree in Biostatistics, Statistics or equivalent is required. The focus of the search is on areas of biostatistics related to big data and complex data, including but not limited to electronic health record data, precision medicine, data linkage and integration, machine learning, medical imaging, among other areas. Applicants should have broad research and teaching interests, the potential to direct PhD-level research, and the ability to engage in collaborative research with other faculty members at the University. Review of applications will start in December, 2015. The positions will remain open until fi lled.

To apply, use the electronic submission website at http://unc.peopleadmin.com/postings/84247 and upload PDF versions of your CV, cover letter, and research and teaching statements. Candidates must also arrange for four letters of recommendation to arrive via email at [email protected] addressed to:

Faculty Search Committee c/o Vera Bennett Department of Biostatistics CB #7420, McGavran-Greenberg Hall University of North Carolina at Chapel Hill Chapel Hill, NC 27599-7420

The University of North Carolina at Chapel Hill is an equal opportunity and affi rmative action employer. All qualifi ed applicants will receive consideration for employment without regard to age, color, disability, gender, gender expression, gender identity, genetic information, race, national origin, religion, sex, sexual orientation, or status as a protected veteran.

november 2015 amstat news 37 a full-time position in applied statistics or related field for fall 2016. Bentley offers degrees in mathematical sciences, actuarial science, MS in business analyt- ics and business. PhD required in applied statistics or a related field by fall 2016. Interested applicants visit: www.bentley. edu. Bentley University is an Equal Opportunity Employer, building strength through diversity.

n BU mathematics statistics invites appli- cations for a tenured statistics professor. PhD required. Begins 7/1/16, pending final budgetary approval. Strong commit- ment to teaching and research essential. The ideal candidate would be interested in serving as director, program in statistics. Submit cover letter describing research and teaching interests, CV, contact infor- mation for three references: mathjobs.org. Direct questions to Tasso Kaper, tasso@ bu.edu. Review begins 11/16/15, con- tinuing until filled. Boston University is an Equal Opportunity Employer and a VEVRAA Federal Contractor.

n MIT Mathematics Department, Cambridge, MA, seeking to fill posi- tions in statistics at levels of assistant

TENURE-TRACK AND TENURED FACULTY POSITIONS DEPARTMENT OF BIOSTATISTICS AND BIOINFORMATICS ROLLINS SCHOOL OF PUBLIC HEALTH, EMORY UNIVERSITY

The Department of Biostatistics and Bioinformatics in the Rollins School serves as the Data Coordinating Center for several clinical trials, operates of Public Health at Emory University is recruiting outstanding faculty can- the Biostatistics Consulting Center; and houses the Center for Biomedical didates for at least two positions: a tenure-track Assistant Professor and a Imaging Statistics (CBIS). The department offers a PhD program in bio- tenured Associate/Full Professor. Successful candidates will contribute to statistics from the Laney Graduate School, and masters degrees in public active methodologic and collaborative research programs, expand graduate health and public health informatics. A concentration in bioinformatics, education at the masters and doctoral levels, and serve locally, nationally, imaging, and genetics is available at the doctoral level. and internationally. Applicants with strong potential and experience in all Successful candidates for both positions must have a doctoral degree and research areas are encouraged, with particular interest areas including big relevant experience in biostatistics, statistics or a related fi eld prior to being data, environmental and spatial statistics, infectious disease modeling, and appointed to this position. Successful candidates for the senior position must neuroimaging statistics. have documented success in research, teaching, and service commensurate Ongoing and expanding collaborative opportunities exist throughout the for tenure as evaluated by the Appointment, Promotion, and Tenure Commit- Woodruff Health Sciences Center including the Rollins School of Public tee of the RSPH. To apply, a cover letter stating which position the candidate Health, the School of Medicine, Nell Hodgson Woodruff School of Nursing, is applying for (tenure-track or tenured), a statement of research interests, a Winship Cancer Institute, Yerkes Primate Center, and the Vaccine Center. complete curriculum vitae, and three reference letters should be sent to: Atlanta has been called the `Public Health Capital of the World’ and is Biostatistics Search Committee (attn to either: home to the U.S. Centers for Disease Control and Prevention, American Tenure-track or Tenured), c/o Mary Abosi ([email protected]) Cancer Society, CARE, and the Carter Center. Emory University, Department of Biostatistics and Bioinformatics, The department has 26 doctoral faculty and 7 masters level associate fac- 1518 Clifton Rd., NE, Atlanta, GA 30322, ulty members with primary appointments. Research interests of the faculty http://sph.emory.edu/departments/bios/index.html include methods for high-dimensional data such as imaging, genomics and Applicants must also fi ll out an application on the Emory Careers Website at: electronic health records; survival analysis; quantile regression; categorical http://www.hr.emory.edu/eu/careers/ under faculty search for Assistant Pro- data analysis; missing and mismeasured data; causal inference; Bayesian fessor: Posting # 56384BR or Associate or Full Professor: Posting # 56383BR. methods; estimating functions; and spatial statistics. The department collaborates with the Atlanta Clinical and Translational Science Institute Consideration of applications will begin immediately, and applications (NCATS-supported Clinical and Translational Science Award), the Center will be considered until positions are fi lled. Emory University is an Equal for AIDS Research, and the Health and Exposome Research Center. It also Opportunity/Affi rmative Action/Disability/Veteran Employer.

38 amstat news november 2015 THE FLORIDA STATE UNIVERSITY professor or higher beginning July 2016 (for 2016–2017 year). Appointments DEPARTMENT OF STATISTICS based on exceptional research qualifi- cations; appointees required to fulfill TWO ASSISTANT PROFESSOR POSITIONS teaching duties. PhD in mathematics IN STATISTICS AND BIOSTATISTICS or related field required by employment (TENURE-TRACK) start date. See full classified for online submission for full consideration by December 1, 2015, at www.mathjobs.org. The Department of Statistics at Florida State University invites Massachusetts Institute of Technology applications for two tenure-track positions in statistics and biostatistics is an Equal Opportunity Affirmative starting August 2016. The Statistics Department has a growing teaching Action Employer. and research program with over 150 graduate students and 19 faculty n members. We offer both M.S. and Ph.D. degrees in Statistics and Statistical Programmer (Cambridge, Biostatistics. Candidates with interests in all areas of statistics and MA). Develops SAS programs regulatory report submissions & government inqui- biostatistics are invited to apply. A PhD in Statistics, Biostatistics, or a ries re: toxicity & efficacy in drug discov- related field is required. Candidates should have strong commitments to ery & development. Min reqs: MSc in excellence in teaching and research. stats/biostats/eng. or rel. fld. & 3 yrs exp. w/SAS pgrm. in pharma. industry. Reqs: Applications are being accepted online. Please visit our web site at knowledge (1) FDA EMA submission http://stat.fsu.edu to apply and for additional information about the guidelines; CDISC data standards. Mail Department of Statistics. Review of candidates will begin November 15, resume: Matthew Metivier, HR. Infinity 2015 and continue until the positions are filled. Pharmaceuticals, Inc., 780 Memorial Dr., Cambridge, MA 02139. ref. Code: An Equal Opportunity/Access/Affirmative Action/Pro Disabled & IPIS15. EOE. Veteran Employer. FSU's Equal Opportunity Statement can be viewed at: n MIT Mathematics Department, http://www.hr.fsu.edu/PDF/Publications/diversity/EEO_Statement.pdf Cambridge, MA, seeking to fill posi- tions in statistics at level of instructor

FACULTY POSITIONS IN STATISTICAL GENETICS/GENOMICS The Departments of Biostatistics and Genetics at The University of North Carolina at Chapel Hill are seeking applications for up to three tenure-track or tenured positions in statistical genetics/genomics, starting in the fall of 2016. Primary appointments will be in the Department of Biostatistics in the Gillings School of Global Public Health, with one of the appointments being joint in the Department of Genetics in the School of Medicine. Appointments may be at the Assistant, Associate or Full Professor level, depending on qualifi cations. At least one position will be at the Assistant Professor rank. A doctoral degree in Bio- statistics, Statistics or equivalent is required. Applicants should have broad research and teaching interests, the potential to direct PhD-level research, and the ability to engage in collaborative research with other faculty members at the University. The successful candidate will join a vibrant group of statistical geneticists in Biostatistics, Genetics and elsewhere at UNC. The University of North Carolina is among the nation’s top public research universities, with dynamic programs in biostatistics, genetics, genomics, bioinformatics, biology, epidemiology, medicine, pharmacogenomics, statistics, and toxicology. In addition, UNC has numerous externally funded projects that provide an excellent environ- ment for interdisciplinary research. Review of applications will start in December. The positions will remain open until fi lled. To apply, use the electronic submission website at http://unc.peopleadmin.com/postings/84285 and upload PDF versions of your CV, cover letter, and research and teaching statements. Candidates must also arrange for four letters of recommendation to arrive via email at [email protected] addressed to: Faculty Search Committee c/o Vera Bennett Department of Biostatistics CB #7420, McGavran-Greenberg Hall University of North Carolina at Chapel Hill Chapel Hill, NC 27599-7420 The University of North Carolina at Chapel Hill is an equal opportunity and affi rmative action employer. All qualifi ed applicants will receive consideration for employment without regard to age, color, disability, gender, gender expression, gender identity, genetic information, race, national origin, religion, sex, sexual orientation, or status as a protected veteran.

november 2015 amstat news 39 beginning July 2016 (for 2016–2017 professor. PhD required. Begins July year). Appointments based on excep- 2016, pending final budgetary approval. tional research qualifications; appointees Commitment to research and teaching ACADEMIA SINICA expected to fulfill teaching duties. PhD in both undergraduate and graduate level INSTITUTE OF STATISTICAL SCIENCE mathematics or related field required by essential. Submit cover letter, CV, research Regular Research Positions employment start date. See full classified statement, teaching statement, and four recommendation letters, (one teaching): The Institute of Statistical Science, Academia Sinica, for online submission for full consid- is seeking candidates for tenure-track or tenured eration by December 1, 2015, at www. mathjobs.org. Alternatively, submit materi- research positions at the level of assistant, associate mathjobs.org. Massachusetts Institute als: Statistics T-T Search, Mathematics/ or full research fellow available in 2016. Candi- Statistics, Boston University, 111 dates in all areas of Statistics will be considered. of Technology is an Equal Opportunity Candidates should have a PhD degree in statistics Affirmative Action Employer. Cummington Mall, Boston, MA 02215. or areas related to data science. Application mate- Deadline 12/1/15. Boston University is rials must include (1) a curriculum vitae, (2) three n Assistant Professor, Department of an Equal Opportunity Employer and a letters of recommendation, and (3) representative Biostatistics, Boston University School VEVRAA Federal Contractor. publications and/or technical reports. Additional supporting materials such as transcripts for new of Public Health (http://sph.bu.edu/bio). PhD degree recipients may also be included. Except Participate in methodological and col- n Boston University Mathematics and for the letters of recommendation, electronic laborative research in exciting areas such Statistics invites applications for a ten- submissions are encouraged. Applications should be as statistical genetics, clinical trials, and ured statistics associate professor. PhD submitted to: observational studies. PhD in biostatistics required. Begins 7/1/2016, pending Dr. Hsin-Cheng Huang or equivalent required. Email CV, let- final budgetary approval. Strong com- Chair of the Search Committee mitment and record of teaching and Institute of Statistical Science, Academia Sinica ter of intent including career objectives, 128 Sec. 2 Academia Road, Taipei 11529, Taiwan, research and teaching experience, inter- research essential. Submit cover letter, CV, R.O.C. ests, and goals, and three letters of rec- research statement, teaching statement, Fax: +886-2-27831523 ommendation to [email protected]. Boston and four recommendation letters (one E-mail: [email protected] University is an EOE. teaching) to mathjobs.org. Alternatively, Applications should be received by December 30, submit materials to: Statistics ASCP 2015 for consideration. n Boston University Mathematics and Search, Mathematics/Statistics, Boston Statistics Department invites applica- University, 111 Cummington Mall, tions for tenure-track statistics assistant Boston, MA 02215. Deadline 12/1/2015.

PRANAB K. SEN DISTINGUISHED VISITING PROFESSORSHIP

The Department of Biostatistics at the University of North Carolina at Chapel Hill (UNC-CH) is seeking to recruit an aspiring international scholar in statistical science for the Pranab K. Sen Distinguished Visiting Professorship in Biostatistics. The position is for the fall semester beginning August 2016 for a period lasting up to fi ve months. Preference will be given to applicants from developing countries as defi ned by the Departments of Biostatistics and Statistics and Operations Research and in consort with the dean of the Gillings School of Global Public Health. The visiting faculty, who will come from an area of statistical science, broadly defi ned, including bioinformatics, will come to UNC-CH to conduct research, possibly teach, and interact with faculty, students, and the university community. The major benefi t will be an intensive interaction of the visitor’s strong methodology background with applications to interdisciplinary research underway at UNC-CH. The visiting professorship, at the assistant, associate or full professor level, off ers salary support commensurate with the rank. Financial support for other expenses such as travel is negotiable. Ideally, the visiting scholar will return to his or her home country for at least one year upon completing the appointment to further expand the reach of interdisciplinary work begun in Chapel Hill. Applicants should hold a PhD in statistics or biostatistics or have commensurate educational background.

To apply, upload your CV, cover letter, and research statement in the application at http://unc.peopleadmin.com/ postings/79598 Candidates must also arrange for three letters of recommendation to arrive via email to Betsy Sea- groves ([email protected].) The deadline for applications is March 1, 2016.

40 amstat news november 2015 Boston University is an Equal Opportunity Employer and a VEVRAA Federal Contractor. Michigan n Grand Valley State University invites applications for assistant professor of statistics. Qualifications include a PhD in statistics or related field, evidence of com- mitment to teaching excellence and active scholarship. For information regarding ASSISTANT/ASSOCIATE BIOSTATISTICIAN FACULTY POSITIONS application materials, see our complete Department of Epidemiology and Biostatistics position description at www.gvsu.edu/ The University of Texas Health Science Center at San Antonio, stat. Applications only accepted online San Antonio, TX at www.gvsujobs.org; review of candidate materials begins on November 20, 2015. Th e Department of Epidemiology and care, and to inform health policy decisions. GVSU is an affirmative action, equal Biostatistics (DEB) at Th e University of Texas Depending on the research interests of the opportunity institution. Health Science Center at San Antonio (UTH- selected candidate, there are opportunities to SCSA) seeks two experienced, doctorate-level collaborate with colleagues within the medical, faculty biostatistician for two tenure-track nursing, and dental schools as well as other Missouri faculty positions at the rank of Assistant or health sciences departments. For candidates n Assistant/associate professor, statistics Associate Professor. Candidates are expected interested in cancer research, there are research to have expertise in longitudinal data, survival opportunities with the Cancer Th erapy Re- department, University of Missouri, ten- analysis, methods for large healthcare databas- search Center (CTRC), one of four NCI-des- ure-track assistant/associate position; all es, statistical genomics applied to clinical out- ignated Cancer Centers in Texas and the only areas of statistics are encouraged to apply. comes, or clinical trial design. Th ese positions NCI designated center in South Texas. Th e PHD in statistics required by August 15, provide an opportunity to collaborate with a Department also has strong research and edu- 2016. Apply online at http://hrs.missouri. multidisciplinary research team as well as to cational collaborations with the San Antonio lead independent methodological research in a Regional Campus of Th e University of Texas edu/find-a-job/academic. Upload cover setting with access to a large, ethnically diverse School of Public Health. letter, CV, research & teaching state- population. Th e UTHSCSA is situated in the Th ese positions are tenure-track and a ments. Deadline: December 1, 2015. dynamic city of San Antonio; a wonderful 12-month appointment. Appointments are Have three letters of recommendation place to live and work that is rich with both open rank (Assistant or Associate level). sent to hrs.missouri.edu/find-a-job/ culture and history. Preference will be given to someone with Th e DEB at UTHSCSA has 20 full-time fac- a demonstrated ability to both collaborate academic. The University of Missouri ulty and 46 staff with a number of additional on multidisciplinary research teams and to is an Equal Opportunity/Affirmative affi liate faculty. Research interests of the facul- successfully lead their own research program. Action/ADA Employer. ty include translational science, clinical- and Applicants must have a doctoral degree (PhD population-based investigation, the application or equivalent) in Biostatistics or Statistics, of epidemiological and biostatistical principles excellent communication skills, and a strong New Hampshire to clinical problem solving, formulation of interest in interdisciplinary collaboration. n The mathematics department at health policy, and the development of new Review of applications will begin immediate- Southern New Hampshire University epidemiological and statistical methods for ly and continue until the positions are fi lled. (SNHU) seeks candidates for two posi- clinical trial design, analysis of “Big Data”, Salary will be commensurate with qualifi ca- clinical decision support systems, and statisti- tions. Information about the DEB is available tions as assistant professor of mathemat- cal genomics. here: http://deb.uthscsa.edu/. For full consid- ics beginning September 1, 2016. The Th e mission of the DEB is to develop and eration applicants should send electronically successful candidates will help build enhance population-based, clinical and transla- or via mail: 1) a cover letter stating career goals upon the recent launch of new majors tional research in clinical and community and the relevance of the candidate’s training and minors in math and math educa- settings, to develop epidemiologic, biostatisti- and experience to the position; 2) a current cal and medical informatics resources, and to CV; 3) a personal statement describing the tion. Candidates should be passionate promote the educational mission of the School candidate’s research plan and teaching experi- about teaching undergraduates. A PhD of Medicine by teaching epidemiology, bio- ence; and 4) names and contact information of in mathematics/statistics with at least statistics, and critical appraisal of the medical three references to [email protected] or 18 graduate hours. Apply Here: www. literature to students, house-staff , and faculty. by mail to: Jonathan Gelfond, M.D., Ph.D., Th e DEB is also the home of the Biostatistics Associate Professor and Biostatistics Search Click2Apply.net/9g9t5h6xqc. EOE. and Informatics Core which includes faculty Committee Chair, UT Health Science Center and Masters trained researchers who serve the at San Antonio, Department of Epidemiology New Jersey UTHSCSA, along with public, community, and Biostatistics, 7703 Floyd Curl Dr.--MSC and private health entities by providing data 7933, San Antonio, TX 78229-3900. n The department of statistics and bio- management, biostatistical and epidemiolog- statistics and the Center for Integrative ical support for projects and programs that All faculty appointments are designated as Proteomics Research at Rutgers involve clinical or health data. security sensitive positions. Th e University of University seeks to hire a tenure-track Th e DEB plays an integral role in UTH- Texas Health Science Center at San Antonio is SCSA's research and educational missions by assistant or associate professor with an Equal Employment Opportunity/Affi rmative strong research and teaching interests enhancing the programs to prevent disease, Action Employer including protected veterans and to promote health, to deliver quality health persons with disabilities. in statistical analyses of biological and biomedical information. Applicants must

november 2015 amstat news 41 have a PhD in statistics or related field by Sept. 1, 2016. Apply online through DEPARTMENT OF STATISTICS the Rutgers Interfolio website link (http:// apply.interfolio.com/31548). AA/EOE. Columbia University New Mexico Faculty Positions Starting Fall 2016 n Assistant professor of applied statistics. The department of economics, applied he Department of Statistics Additionally, applicants must submit statistics, and international business at invites applications for two New Mexico State University invites T materials through Head Hunter. To applications for a tenure-track assistant faculty positions in applied/interdis- apply for the senior position go to professor of applied statistics, 9-month ciplinary statistics to begin July 1, this link https://editorialexpress.com/ position. Doctorate required. Specialty 2016. Th e positions may be fi lled at apply/columbia_statistics_senior_re- in data analytics preferred. Other fields any rank from tenure-track assistant cruiting_2016_a89 . considered. Position requires graduate and professor through full professor undergraduate teaching, research, and ser- with tenure. A Ph.D. in statistics or To apply for the junior position go vice. For full position announcement, see a related fi eld and commitment to to https://editorialexpress.com/apply/ jobs.nmsu.edu/postings/23210. EOE. high quality research and teaching columbia_statistics_junior_recruiting New York in statistics and/or probability are _2016_a90 n required. Candidates will be expect- NYU Stern School of Business Statistics Th e Department of Statistics posi- Group, tenure-track assistant professor ed to sustain an active research and appointment in statistics. Candidates publication agenda and to teach in tions will be visible in Head Hunter should have evidence of boundary- the departmental undergraduate by clicking on “Positions” after spanning interests across fields that reflect and graduate programs. Candidates logging in to the Candidate significant interfaces of statistics with areas interested in an affi liation with the Application Interface. of relevance in a business school. Expected Data Science Institute are strongly that candidate will be productive researcher encouraged to apply. In Head Hunter, applicants for this and effective teacher at both undergradu- position at the assistant professor ate and graduate levels. See apply.interfolio. com/30831 for full details, including infor- Th e Department currently consists or non-tenured associate professor mation on application procedure. EOE. of 30 faculty members, 45 PhD rank should submit a cover letter, students, and over 200 MA students. Curriculum Vitae, a brief statement Ohio Th e Department has been expanding of their research plans, one writing n The Ohio State University Department rapidly and, like the University itself, sample, and arrange for three letters of Statistics invites applications for an is an extraordinarily vibrant academ- of reference to be sent on their open-rank faculty/program specialist ic community. For further informa- behalf. Applicants at the tenured position beginning autumn 2016. This tion about the Department and our associate professor and full professor nontenure-track faculty position provides activities, centers, research areas, and rank should submit a cover letter teaching and program support in the uni- curricular programs, please go to our versity’s new undergraduate data analytics and Curriculum Vitae. webpage at: http://www.stat.colum- major, with focus on enhancement of student learning in statistics and analyt- bia.edu Applications will only be considered ics. PhD in statistics and excellence in for the position once the process teaching are required. Apply by November For information about the Data is completed both in RAPS and in 29, 2015, at academicjobsonline.org/ajo/ Science Institute, please see web page HEAD HUNTER. jobs/6046. To build a diverse workforce, at: http://datascience.columbia.edu Ohio State encourages applications from Inquiries may be made to dk@stat. minorities, veterans, women, and individu- Applicants at all ranks are required columbia.edu als with disabilities. Flexible work options to create an applicant profi le and available. EEO/AA Employer. The Ohio upload a CV through Columbia’s Review of applications begins on State is an NSF Advance Institution. online Recruitment of Academic December 2, 2015, and will Personnel System (RAPS). To begin continue until the position is fi lled. Pennsylvania the application process, please go to: n The Wharton Department of Statistics, https:// academicjobs.columbia.edu/ Columbia University is an Equal University of Pennsylvania, is seeking applicants/Central?quickFind=61490 Opportunity/Affi rmative Action employer. full-time, tenure-track faculty at any

42 amstat news november 2015 level: assistant, associate, or full profes- Texas n sor, appointment beginning July 2016. Tenure-track assistant professor in sta- Applicants should show outstand- tistics. The department of mathematics at ing capacity in research and teaching. University of North Texas invites applica- Applicants must have a PhD (expected tions for a tenure-track assistant professor completion by June 30, 2017, is accept- in statistics beginning fall 2016. For more able) from an accredited institution. Please information on the position, qualifica- visit our website to apply: https:// tions, and application instructions, see statistics.wharton.upenn.edu/recruiting/ www.math.unt.edu. Applicants must facultypositions. Questions should be sent apply online at http://facultyjobs.unt.edu. to [email protected]. The Review of applications begins November University of Pennsylvania is an EOE. 15, 2015; if not filled, applications will be reviewed weekly thereafter until the search n Assistant Professor (tenure-track), is closed. The University of North Texas Mathematics and Computer Science is an EOE/ADA/AA institution commit- Department, Duquesne University, ted to diversity in its employment and Pittsburgh PA, for fall 2016. PhD in educational programs, thereby creating a statistics or closely related discipline welcoming environment for everyone. expected by August 2016. Maximum n expected teaching load is three courses Tenure-track position, starting fall per semester, with pre-tenure sabbatical 2016, Department of Statistics, Rice and other potential reductions available University. Priority hiring is at assistant to support research. For complete job professor level; however, highly qualified, description and to apply online, visit experienced candidates may be considered www.mathjobs.org/jobs/jobs/7475. Full at associate professor level. Doctorate consideration to applications received in statistics or a related field required. by December 1, 2015. Duquesne Apply at http://facultysearch.statistics.rice. University is an affirmative action, edu. Inquiries to [email protected]. equal opportunity employer. Minorities Review of applications begins December and women are encouraged to apply. 16, 2015. Rice University is an Equal Opportunity/Affirmative Action employer.

DEPARTMENT OF BIOSTATISTICS The Department of Biostatistics at the Johns Hopkins Bloomberg School of Public Health is seeking an outstanding applicant to join our tenure track faculty. Rank of appointment will be commensurate with experience. New PhDs and recent postdoctoral fellows are encouraged to apply. Candidates should have a PhD or equivalent in statistics, biostatistics, or a comparable fi eld. Women and under-represented minority candidates are particularly encouraged to apply.

-Track Faculty Positions The Hopkins Department of Biostatistics, founded in 1918, was the fi rst degree-granting department of statistical science in the US and has ranked among the best throughout its history. The Johns Hopkins Health Institutions (Schools of Public Health, Medicine, and Nursing, and the Johns Hopkins Hospital) are among the top worldwide and provide a research environment in which energetic faculty can achieve scientifi c excellence. Today, the Department comprises 21 tenure track faculty members, 17 research track faculty members, 13 postdoctoral fellows, 50 PhD students, 14 full-time master degree students, and 12 students pursuing joint master degrees together with doctorates in other departments. Emphases of the department address a diverse variety of public health and medical specialty areas and stages of the lifespan: they include statistical inference and methods, massive and real-time measurement, data science, educational innovation, and subject-area content for genomics, population health, environmental health, and behavioral health. For detailed information, please visit www.biostat.jhsph.edu. TO APPLY Email cover letter, CV, letters from three references, a statement of research interests and goals, a teaching statement, and two manu- scripts or articles representing your most important work to: Faculty Search Committee at [email protected]. Interviews commence in mid-December and continue into early winter. The Johns Hopkins University is an affi rmative action/equal opportunity employer.

november 2015 amstat news 43 n Tenure-track assistant professor position in statistics or biostatis- tics beginning fall 2016 (www.math. ttu.edu/FacultyStaff/stat-ad.shtml). A PhD is required. Apply for requisi- tion ID5154BR at www.texastech.edu/ careers. Have three letters of reference sent to Misty Rangel, Department of Mathematics and Statistics, Texas Tech University, Lubbock, TX 79409-1042. As an Equal Employment Opportunity/ Affirmative Action employer, Texas Tech University is dedicated to the goal of building a culturally diverse faculty committed to teaching and working in a multicultural environment. We actively encourage applications from all those who can contribute, through their research, teaching, and/or service, to the diversity and excellence of the academic com- munity at Texas Tech University. The university welcomes applications from minorities, women, veterans, persons with disabilities, and dual-career couples. Utah n The department of mathematics at the University of Utah invites applications for the following faculty positions: Full-time tenure-track or tenured appointments at the level of assistant, associate, or full pro- fessor in all areas of statistics. Applications must be completed through the website www.mathjobs.org/jobs/Utah, and will be accepted until the position(s) have been filled. For more information, see www. math.utah.edu/positions. The University of Utah is an Equal Opportunity/Affirmative Action employer and educator. Minorities, women, veterans, and those with disabilities are strongly encouraged to apply. Veterans’ preference is extended to qualified veterans. Reasonable disability accommodations will be provided with adequate notice. For additional informa- tion about the university’s commitment to equal opportunity and access, see www. utah.edu/nondiscrimination. Virginia n Manager, Research Initiatives (Alexandria, VA). This position is new to the Club Managers Association of America and expected to produce high-quality research products to drive greater member value. Currently, the association conducts one annual survey process that produces 2–3 reports. This position is responsible for growing our report frequency to

44 amstat news november 2015 10–12 annual projects. jobs.amstat.org/ DEPARTMENT OF STATISTICS jobs/7482119/manager-research-initiatives. CMAA provides equal employment Columbia University opportunities (EEO) In addition to fed- eral law requirements, CMAA complies Limited-term Faculty Position Starting Fall 2016 with applicable state and local laws gov- erning nondiscrimination in employment The Department of Statistics invites applica- All applications must be submitted through in every location in which the company tions for a four-year term position at the rank Columbia’s online Recruitment of Academic has facilities. of assistant professor to begin July 1, 2016. A Personnel System (RAPS) and must include PhD in statistics or a related fi eld is required, the following materials: cover letter, curricu- n Department of statistics, University as is a commitment to high quality research lum vitae, statement of teaching philosophy, of Virginia, invites applications for one and teaching in statistics and/or probability. research statement, evidence of teaching effec- tenure/tenure-track position and one lec- Candidates will be expected to sustain an ac- tiveness, one writing sample or publication, tive research and publication agenda and to and the names of 3 references into the system. turer, fall 2016. Applicants must present teach in the departmental undergraduate and Applicants also should arrange for three letters evidence of outstanding accomplishments graduate programs. of recommendation to be uploaded on their in research/teaching. Candidates are behalf. For more information and to apply, expected to teach graduate/undergradu- The department currently consists of 30 fac- please go to: https://academicjobs.columbia. ate courses, provide service. We seek ulty members, 45 PhD students, and over edu/applicants/Central?quickFind=61471 candidates dedicated to UVA’s mission 200 MA students. The department has been of excellence in research and teaching. expanding rapidly and, like the University Inquiries may be made to [email protected] We expect substantial growth due to itself, is an extraordinarily vibrant academic Review of applications begins on January 1, UVA President’s Big Data Initiative. community. For further information about 2016, and will continue until the position Application instructions: statistics. the department and our activities, centers, is fi lled. as.virginia.edu EO/AA Employer. n research areas, and curricular programs, please go to our web page at: http://www.stat. Columbia University is an Equal Opportunity/ columbia.edu Affi rmative Action employer.

Non-Tenure Track Faculty Position - Teaching Assistant Professor

The Department of Statistics at North Carolina State Tenure-Track Faculty Positions University invites applications for a non-tenure track position at the Assistant Professor level to begin in The Department of Statistics at North Carolina State August 2016. University invites applications for three tenure track positions to begin in August 2016: one Assistant Applicants must have a Ph.D. in Statistics or Professor and two open-rank positions. For Biostatistics. Experience and excellence in teaching is outstanding candidates, the open-rank positions may required. The initial appointment is expected to be for be combined with named professorships. two years, but the position is eligible for subsequent appointments, and promotions in rank are possible. Applicants must have a Ph.D. in Statistics or Biostatistics. Responsibilities include teaching, To apply, go to https://jobs.ncsu.edu/postings/53972. research, and doctoral student research supervision. Questions may be directed to the Search Chair, Dr. Candidates for Associate or Full Professor must have Roger Woodard ([email protected]). an established record of funded research, collaboration, and exemplary teaching. Inclusiveness and diversity are academic imperatives and thus are university goals. To apply for the Assistant Professor position, go to Please visit http://www.stat.ncsu.edu for https://jobs.ncsu.edu/postings/55131, and for the information about the Department. Processing of open-rank positions, go to applications will begin November 20, 2015 and https://jobs.ncsu.edu/postings/55266. continue until the positions are filled.

NCSU is an equal opportunity and affirmative action employer. In Questions may be directed to the Search Chair, Dr. addition, NC State University welcomes all persons without regard Ana-Maria Staicu ([email protected]). to sexual orientation or genetic information.

november 2015 amstat news 45 University of Rochester Department of Biostatistics and Computational Biology Tenure-Track Faculty Position (Open Rank) The Department of Biostatistics and Computational Biology an established track record of peer-reviewed publications, (DBCB) at the University of Rochester invites applications for have demonstrated success in attracting extramural research an open rank tenure-track position. The successful candi- funding, and present evidence of teaching excellence at the date will be expected to have developed an active program in graduate level. methodological research, contribute to our Ph.D. program To apply: Candidates should send a cover letter, CV, and through teaching and advising graduate students, and have a 3 letters of reference to the address below. The cover letter major role in collaborative research with the Department of should specifi cally address the candidate’s expertise in areas Neurology and the Center for Human Experimental Thera- peutics (CHET). A key component of this position will involve directly relevant to this position. Review of candidate fi les working with investigators throughout North America in the will commence December 1, 2015. Please email all materials design and analysis of high profi le clinical trials in neurolog- in a single PDF fi le to [email protected]. ical disorders; we are especially interested in applicants that edu to ensure timely consideration of your application. Candi- have a track record of relevant methodological and collabora- dates preferring to send hard copies should mail these to the tive clinical research. following address: DBCB, Neurology and CHET have a long-standing collabora- Open Rank Faculty Position (Neuro) tive relationship with each other; faculty play leadership roles c/o Malora Zavaglia in national and international groups conducting multi-center University of Rochester Medical Center clinical trials and epidemiological studies in Parkinson’s disease, Department of Biostatistics & Computational Biology Huntington’s disease, and other neurologic disorders. 601 Elmwood Avenue, Box 630 Rochester, NY 14642 Position Qualifi cations: Doctoral degree in biostatistics, statistics or strongly related discipline. Candidates must have The University of Rochester is an Affi rmative Action, Equal excellent oral and written communication skills. Candidates Opportunity institution. Women and minority candidates are for Associate and Full Professor positions should also have strongly encouraged to apply.

46 amstat news november 2015 AMSTATNEWS ADVERTISING DIRECTORY

Listed below are our display advertisements only. If you are looking for job-placement ads, please see the professional opportunities sec- TEXAS TECH UNIVERSITY tion. For more job listings or more information about advertising, please visit www.amstat.org. Department of Mathematics and Statistics

Chair and Professor misc. products and services The Department of Mathematics and research university. Its strategic plan American Mathematical Society...... centerfold Statistics at Texas Tech University charts a course for becoming a great Joint Program in Survey invites applications for the position of public research university by 2020. Methodology...... centerfold Department Chair and Professor be- The university has the characteristics SIAM...... centerfold ginning fall 2016. The salary will be of a Carnegie-classifi ed doctoral competitive and commensurate with granting university with very high qualifi cations and experience. research productivity. professional opportunities Academia Sinica...... p. 40 Candidates must have demonstrated Applicants should apply at http:// outstanding vision, leadership, and www.texastech.edu/careers/ using Columbia University...... p. 42, 45 scholarship, possess strong commit- Requisition 4369BR. Applicants Emory University...... p. 38 ments to interdisciplinary research should submit a detailed letter of ap- Florida State University...... p. 39 and educational activities, and be a plication along with a current resumé Johns Hopkins Bloomberg School of collegial motivator and advocate for including externally-funded research, Public Health...... p. 43 faculty. A Ph.D. in mathematics or teaching, administrative experience, statistics is required. The successful publications. and four letters of profes- North Carolina State University...... p. 45 candidate is expected to work with sional reference. Questions about Penn State University...... p. 46 the faculty to develop the department the position and/or the application SAMSI...... p. 36 under a strategic vision, foster excel- process can be directed to Texas Tech University...... p. 47 lence in research and teaching, and [email protected] provide appropriate service to the U.S. Census Bureau...... p. 44 Review of applications will begin college, university, and profession. immediately. Applications will be ac- University of Cincinnati...... p. 34 The department currently has 45 cepted until the position is fi lled, with The University of North Carolina at full-time faculty members with active those received prior to November 15, Chapel Hill...... p. 37, 39, 40 research groups in pure and applied 2015, assured full consideration. University of Rochester Medical Center....p. 46 mathematics and statistics. Six degrees As an Equal Employment Opportu- The University of Texas Health are offered: B.A., B.S., M.A., M.S., nity/Affi rmative Action employer, Science Center...... p. 41 and Ph.D. in Mathematics and M.S. Texas Tech University is dedicated The University of Vermont...... p. 35 in Statistics, as well as several interdis- to the goal of building a culturally ciplinary and combined undergradu- Virginia Tech...... p. 35 diverse faculty committed to teach- ate and graduate degrees. For further Westat...... p. 38 ing and working in a multicultural information regarding the Department environment. We actively encourage Williams College...... p. 37 of Mathematics and Statistics, please applications from all those who can refer to the department’s website: contribute, through their research, http://www.math.ttu.edu. software teaching, and/or service, to the di- Texas Tech University, located in versity and excellence of the academ- Cytel...... p. 5 west Texas in the city of Lubbock, ic community at Texas Tech University. JMP software from SAS...... cover 4 was founded in 1923. The student The university welcomes applications Minitab Inc...... cover 2 population is 35,000 and is antic- from minorities, women, veterans, ipated to grow to 40,000. Texas persons with disabilities, and NCSS...... p. 22 Tech is a state-designated national dual-career couples. SAS...... cover 3

november 2015 amstat news 47 SOCIAL CHATTER What our followers are saying online

FOLLOW US

community. amstat.org

www.facebook. com/AmstatNews

@AmstatNews

www.instagram .com/AmstatNews

ASA The American Statistical Association Heard at the office: “A statistician is a mathematician broken down by age and sex.” What is your favorite #statjoke?

AmStat Philadelphia @amstatphilly Shayantani Pal I married a guy who begins Q: What do you call 1000 statisticians at a with R tea party? Rose Saint Fleur-Calixte We are mean lovers A: A Z-Party Milo Rodriguez Statistics may be dull Apothicaire Amoureux @PotardDechaine bit it has its moments “How is your wife?” Statistician: “Compared to what?” Robbie Emmet Two proportions had an Next month, argument; they were at odds. we’ll ask our followers to fill in the blank: Matt Baron This one is variable (insert your As a statistician, I am own number that follows): 42.75 of all thankful for______. statistics are made up on the spot. Follow us to read the responses or send us one of your own. Don’t forget to cc @AmstatNews.

48 amstat news november 2015