May 2012 • Issue #419 AMSTATNEWS The Membership Magazine of the American Statistical Association • http://magazine.amstat.org

JSM BOUND ALSO: When Statistics Education and State Policy Meet Statistical Consulting in a University Research Setting

AMSTATNEWS MAY 2012 • ISSUE #419

Executive Director features Ron Wasserstein: [email protected] 3 President’s Corner Associate Executive Director and Director of Operations Stephen Porzio: [email protected] 5 Highlights of the March 2012 ASA Board of Directors Meeting Director of Science Policy 6 Workshop on New Developments in Econometrics and Steve Pierson: [email protected] Time Series Director of Education Rebecca Nichols [email protected] 7 Call for Proposals for SAMSI Summer Programs and Workshops

Managing Editor 8 Chapter, Section Microsites a New ASA Benefit Megan Murphy: [email protected] 8 Leading Economic Forecaster Chosen as Economic Production Coordinators/Graphic Designers Luncheon Speaker Melissa Muko: [email protected] Kathryn Wright: [email protected] 9 Petitions Requested for New Section

Publications Coordinator 10 Technometrics Highlights Val Nirala: [email protected] Service Accessibility Analyzed with Novel Methods in May Issue Advertising Manager 11 Meet Economic Research Service Administrator Mary Bohman Claudine Donovan: [email protected] 13 The World Is Calling; Should We Answer? Contributing Staff Members Donna Arrington • Kalil Deschamps • Amy Farris 16 Nominations Sought for Deming Lecture, Noether Awards Rebecca Nichols • Rick Peterson

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 departments 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. 18 meetings 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. JSM Bound Amstat News (ISSN 0163-9617) is published monthly by the American Keynote Speakers Chosen for Knowledge, Dedication Statistical Association, 732 North Washington Street, Alexandria VA 22314- 1943 USA. Periodicals postage paid at Alexandria, Virginia, and additional Writing Workshop for Junior Researchers to Take Place at JSM mailing offices. POSTMASTER: Send address changes to Amstat News, 732 North Washington Street, Alexandria VA 22314-1943 USA. Send Canadian How to Get the Most Out of Your First Joint Statistical Meetings address changes to APC, PO Box 503, RPO West Beaver Creek, Rich Hill, ON L4B 4R6. Annual subscriptions are $50 per year for nonmembers. Amstat News is the member publication of the ASA. For annual membership rates, San Diego Offers JSM Attendees Much to Do, See see www.amstat.org/join or contact ASA Member Services at (888) 231-3473.

American Statistical Association 732 North Washington Street Alexandria, VA 22314–1943 USA (703) 684–1221 • FAX: (703) 684-2037

ASA GENERAL: [email protected] columns ADDRESS CHANGES: [email protected] 27 SCIENCE POLICY AMSTAT EDITORIAL: [email protected] When Statistics Education and State Policy Meet ADVERTISING: [email protected] This column is written to inform ASA members about what the ASA is doing to WEBSITE: http://magazine.amstat.org promote the inclusion of statistics in policymaking and the funding of statistics Printed in USA © 2012 research. To suggest science policy topics for the ASA to address, contact ASA American Statistical Association Director of Science Policy Steve Pierson at [email protected].

Contributing Editor

Promoting the Practice and Profession of Statistics Christine Franklin is the Lothar Tresp Honoratus Honors Professor in Statistics at the University of Georgia. She was the The American Statistical Association is the world’s largest lead writer for the ASA Pre-K–12 GAISE Framework that served community of statisticians. The ASA supports excellence in as the basis for the statistics strand in the Common Core State the development, application, and dissemination of statistical Standards. She is also a past AP Statistics chief reader and was science through meetings, publications, membership services, honored in 2006 with the Mu Sigma Rho National Statistical Education Award. education, accreditation, and advocacy. Our members serve in Franklin industry, government, and academia in more than 90 countries, advancing research and promoting sound statistical practice to inform public policy and improve human welfare. columns 29 STATtr@k Collaboration with Statisticians STATtr@k is a column in Amstat News and a website geared toward people who are in a statistics program, recently graduated from a statistics program, or recently entered the job world. To read more articles like this one, visit the website at http://stattrak.amstat.org. If you have suggestions for future articles, or would like to submit an article, please email Online Articles Megan Murphy, Amstat News managing editor, at [email protected]. Contributing Editor The following articles in this issue can be found online Danny Modlin earned his bachelor’s degree in mathematics from at http://magazine.amstat.org. Elon University. After teaching middle- and high-school mathemat- ics for six years, he earned his master’s degree in mathematics from Roderick Little will deliver the COPSS Fisher Lecture, The University of North Carolina at Wilmington and his master’s degree in statistics from North Carolina State University. Modlin is “In Praise of Simplicity, Not Mathematistry! Ten Simple, employed as a statistical training specialist at SAS Institute Inc. and Powerful Ideas for the Statistical Scientist,” in San working toward his PhD in statistics. Modlin Diego, California, during the Joint Statistical Meetings. Little has been a leading researcher in the modeling, evaluation, and analysis of missing data, sample survey, and causal inference. At the University of Michigan, 31 MASTER'S NOTEBOOK he has played important roles in various collaborative Statistical Consulting in a University Research Setting projects across campus. He currently serves as director This column is written for statisticians with master's degrees and highlights areas of of the biostatistics cores for three major research employment that will benefit statisticians at the master's level. Comments and sug- centers. A special article on Little’s accomplishments, gestions should be sent to Megan Murphy, Amstat News managing editor, at megan@ prepared by John Kalbfleisch and Ross L. Prentice, amstat.org. can be found online at http://magazine.amstat.org/ Contributing Editor blog/2012/05/01/roderick-little. Gregory J. Stoddard, a master’s degree–level biostatistician, is co- director of the Study Design and Biostatistics Center at the University The National Center for Health Statistics invites you of Utah, which provides statistical consultations on applied research to the 2012 National Conference on Health problems throughout the university. Statistics, to be held August 6–8, 2012, at the Renaissance Washington, DC, Downtown Hotel. Stoddard Registration is open. For details, visit www.cdc.gov/ nchs/events/2012nchs. There are many places to see when you are in San 34 175 Diego for JSM, some of which are an hour or more Reaching 20k away. Therefore, members of the JSM local area The ASA will celebrate its 175th anniversary in 2014. In preparation, column “175”— written by members of the ASA’s 175th Anniversary Steering Committee and other ASA committee compiled a list of longer trips within members—will chronicle the theme chosen for the celebration, status of preparations, San Diego County. To view, visit http://magazine. activities to take place, and, best yet, how you can get involved in propelling the ASA amstat.org/blog/2012/05/01/longer-trips. toward its bicentennial. Contributing Editor Monica Johnston is chair of the ASA Committee on Membership Retention and Visit the ASA Calendar of Events, an online Recruitment and cofounder of the Section for Statistical Programmers and Analysts. She database of statistical happenings across the globe. is an independent consultant and serves as director at Mostly Math, an education services Announcements are accepted from educational and center in Walnut Creek, California. not-for-profit organizations. To view the complete list . of statistics meetings and workshops, visit www.amstat. org/dateline.

Many of the sections and committees sponsor member news events and host workshops and meetings. For details 35 People News about these events and other news, make sure you 39 Section • Chapter • Committee News visit our section, chapter, and committee pages online at http://magazine.amstat.org. 41 Calendar of Events 42 Professional Opportunities

2 amstat news may 2012 president's corner

Photo courtesy of the NCSU Department of Statistics From left: Faculty members Alison Motsinger-Reif, Montserrat Fuentes, and Peter Bloomfield congratulate PhD recipient Shenek Alston at the 2011 commencement ceremony of the department of statistics at North Carolina State University. Graduation Time: Is Your Commencement Speech Ready?

ay is graduation month, a time to cele- Their answers were insightful. One student brate the accomplishments of new grad- wrote, “My education in statistics gives me the uates in the field of statistics and and tools to approach and solve complex problems.” encourageM them as they commence their careers in Another stated, “My education in statistics gives me our profession. While few of us will have the oppor- the power to see the truth hidden behind numbers.” tunity to speak at a commencement ceremony, When the students asked how I would complete most of us either know students who are graduat- the sentence, my answer was the word “choices.” A ing or will soon meet a recent graduate. Each of us statistical education provides career choices, but it should be prepared with a personal commence- also requires that graduates make choices if they are ment speech—and after reading this column, you to develop and contribute as professional statisticians. will be ready! And so “choices” became the theme for my speech. What Can You Say in 10 Minutes? Career Choices Last spring, the department of statistics at North Today, our graduates have an unprecedented range Carolina State University invited me to give the of career choices in research, government, and busi- commencement address at its annual graduation ness. A McKinsey Global Institute report released Robert Rodriguez ceremony. I was asked to keep it brief, but the in 2011 projects that by 2018, the United States theme was up to me. As I thought about what to will need 140,000 to 190,000 more people with say, I realized I should know more about the stu- “deep analytical” skills, typically experts in statisti- dents who would be in my audience, so I arranged cal methods and data analysis technologies. More to meet with some of them in advance. To launch recently, 46% of firms surveyed by Bloomberg the discussion, I asked each student to complete Businessweek reported that they plan to increase the following sentence: “My education in statistics their analytical talent. gives me [blank].”

may 2012 amstat news 3 The demand has never been greater for the 2,200 graduates to contribute time and energy as active students graduating with advanced degrees in sta- members of our chapters and sections (see Page 34 tistics. And the opportunities are just as strong for for Monica Johnston’s 175th column). By starting the increasing numbers of graduates with bachelor’s now, they will have the opportunity to join our cel- degrees in statistics. ebration of the ASA’s 175th anniversary in 2014. Our graduates are in high demand because our And when we celebrate our 200th anniversary in society is drenched in raw data. Much of the data are 2039, today’s graduates will look back on a quarter being collected in massive databases by government, century in which they have learned much, enriched health care, marketing, and manufacturing organiza- our profession, and benefited society in ways we can- tions and consist of both unstructured data—from not imagine. sources such as email and social media—and struc- For new graduates, the choice to give also means tured data—from observational and planned stud- contributing to their institutions of higher learning ies. All those data must be understood, analyzed, and as a means of paying forward to the next generation. translated into the bases for scientific discovery, gov- Graduates should realize that only a fraction of the ernment policymaking, and business decisionmaking. cost of their education was covered by the tuition Managers and executives who hire statisticians tell their families paid or the work they did as teaching me they are impressed with the core statistical skills assistants. The majority of that cost was covered by of our new graduates. But they always add that their other sources, including the institution, government main problem is finding statisticians who can not appropriations, research grants, and alumni giving. only analyze data, but also explain to others how the Graduates can begin to make this choice by results serve the goals of their organization. thanking their professors for their commitment to education. As they progress in their careers, gradu- The Choice to Communicate ates should keep in touch with the faculty because Our students discover in their consulting courses that teachers are always gratified to know about their the language of statistics does not easily translate into students’ accomplishments. And once they have the the languages of other disciplines. And wherever their means to do so, graduates should contribute finan- careers take them, they will find that statistics is an cially to their institutions and serve as mentors for interdisciplinary activity. So if the future contribu- future students. tions of our graduates are to be valued and visible, they must choose to improve their abilities to write Consequence of Wise Choices effectively and to give presentations that are clear, con- The choices to communicate, belong, and give require cise, and relevant to their audiences. This can require effort and dedication, but bring great rewards: years of practice and coaching, but it is an investment with huge long-term dividends. Graduates who choose to improve their commu- nication skills will go beyond analyzing data and The Choice to Belong building statistical models. They will make their Graduates also must choose whether to become work relevant for the rest of the world. professional statisticians. That means far more than adding the word statistician to a LinkedIn profile. Graduates who choose to become professional stat- It means keeping up with new developments in isticians will do more than earn a good living. They the field. It means participating in the growth of will develop their abilities, enrich the field of statis- the field—through research, teaching, or statistical tics, and contribute to society. practice. It means leading and influencing others. It means adhering to ethical standards and principles Graduates who choose to give back will reap more of sound practice. And it means belonging to a com- than personal accomplishments. They will gain a munity of statisticians who have common interests, share in the future success of their institutions and values, and goals. our profession. The world’s largest community of statisticians is the American Statistical Association, and more than Choosing Your Own Words 4,500 of our members are students. Graduation is the When it comes time to give your own commence- perfect time for these students to make our association ment speech to a new graduate, I hope you will their permanent professional home. express the enjoyment and excitement you have expe- The Choice to Give rienced in your career as a statistician. Nothing could be more encouraging! New graduates also must choose whether to give back and “pay it forward.” We need to challenge

4 amstat news may 2012 Highlights of the March 2012 ASA Board of Directors Meeting SA President Bob Rodriguez conducted his first meeting as board chair, leading board members through an agenda guided by the 2012 ASA Board of Directors ASA’sA strategic plan. The board met March 30–31 at the ASA office in Alexandria, Virginia. Here are Bob Rodriguez (President) the highlights: Marie Davidian (President-elect) • The board approved editorial appointments for Nancy Geller (Past President) several of its journals (JASA, JBES, JCGS, JNPS, SADM, and Technometrics). New editors will Rod Little (3rd-Year Vice President) officially assume their roles in 2013, though the Mary Mulry (2nd-Year Vice President) actual transition begins later this year. David Morganstein (1st-Year Vice President) • Funding proposals received through the ASA’s Member Initiative Program were reviewed; several Jessica Utts (3rd-Year Council of Sections Representative) were selected to receive funds. John Bailer (2nd-Year Council of Sections Representative) • The board heard the report of the exploratory Janet Buckingham (1st-Year Council of Sections Representative) committee it appointed last fall to consider Tom Short (3rd-Year Council of Chapters Representative) whether to establish a prestigious international prize for statistics. The executive director and the Bonnie LaFleur (2nd-Year Council of Chapters Representative) Development Committee will now investigate Nick Horton (1st-Year Council of Chapters Representative) the feasibility of raising funds for such a prize. The ASA is collaborating with several interna- Ray Chambers (International Representative) tional societies in considering this effort. David Banks (Publications Representative) • As always, the board heard reports from the Keith Ord (Treasurer) Council of Sections Governing Board and Ron Wasserstein (Executive Director) Council of Chapters Governing Board. The COSGB reported on the allocation of invited sessions to sections for JSM 2013 and 2014. The COCGB is working on ways to interact with chapter representatives outside of JSM. • President-elect Marie Davidian introduced her strategic initiatives ideas for 2013. She is • The board had the honor of hosting Jim Crowley, planning efforts to encourage students to study executive director of the Society for Industrial and statistics (education component of the strategic Applied Mathematics (SIAM). Crowley informed plan), increase the visibility of statistics in the board about SIAM and its activities. The ASA the science community (visibility and impact works with SIAM on a book series and new jour- component), and address issues related to the nal and interacts with SIAM through the Joint number of concurrent sessions at JSM (meetings Policy Board for Mathematics. component). The board provided feedback to • The 2011 audit report was received by the board. shape and sharpen Davidian’s initiatives. Once again, the ASA received a clean audit. The • The board received reports from the 2012 strate- board thanked staff, especially Associate Executive gic workgroups: Director and Director of Operations Steve Porzio, for another successful audit. —Jeri Mulrow reported on the activities of the “In-Reach” Workgroup, which is devel- • Treasurer Keith Ord presented his regular report oping materials to improve communication on the status of the ASA’s finances and invest- with members about the ASA’s activities. ment portfolio. The organization has a healthy Preliminary materials were pilot tested using balance sheet, and investments have continued board members as presenters and audience. to recover value.

may 2012 amstat news 5 —Christy Chuang-Stein reported on the work • David Banks, ASA publications representative to of the 175th Anniversary Steering Committee. the board, reported on discussions of a subcom- She asked for feedback about three major ini- mittee of the Committee on Publications on the tiatives the committee is considering as part of future of publishing. At the suggestion of the sub- the 175th anniversary activities. The feedback committee, the board will form a panel to con- she received will guide future planning. sider these matters in detail. —Bob Starbuck informed the board about • Ron Wasserstein updated the board on the plans being developed by the Career Success activities under way for the International Year of Factors Workgroup. Four factors have been Statistics, a collective effort of the International identified: presentation skills, influence skills, Biometric Society, Institute of Mathematical personality training and team building, and Statistics, International Statistical Institute, career planning. Starbuck said the value prop- Royal Statistical Society, and ASA. He said osition for the ASA and its members includes hundreds of societies, universities, schools, gov- developing skills that increase impact and ernment agencies, and businesses have already career potential of statisticians through course- expressed their interest in participating. (See work specifically designed for them. “Train the http://statistics2013.org.) He noted that the trainer” sessions are planned for JSM. steering committee for Statistics 2013 is work- ing vigorously on activities to kick off and wrap —John Bailer updated the board on the activi- up the International Year of Statistics. ties of the Education Workgroup, which is developing a set of learning outcomes for ter- • The board continued its discussion of the ASA as minal master’s programs leading to employ- the “big tent for statistics,” regarding the develop- ment in business and government. ment of a better understanding of the diversity of statisticians in the ASA and how to meet their • The board received the annual reports of the needs. This discussion started at JSM and has con- Membership Council. David Morganstein updat- tinued since. During this meeting, the board: ed the board on the activities of the committees within the council. The annual reports of the —Approved a statement on the importance of councils provide a key mechanism for the board continuing professional development (CPD) and the many ASA committees to stay in touch and the role of the ASA in providing it. The with each other. statement will appear on the ASA website soon and in a future Amstat News issue. • In other committee-related activities, the board approved the creation of an ASA-AMATYC —Created, in conjunction with the state- joint committee (to replace the Committee on ment, an ad hoc group to more fully explore Statistics in Two-Year Colleges), the dissolution the ASA’s role and activities in CPD. of the Committee on Outreach Education, and the reorganization of committees related to the —Reviewed the purpose and function of the Conference on Statistical Practice. SPAIG Committee and discussed how to bet- ter formulate statistical partnerships among academe, industry, and government. Workshop on New Developments —Heard remarks about “big data” and the roles statisticians and our association should in Econometrics and Time Series play in the big data era. Sallie Keller, Stephanie The Collaborative Research Center “Statistical Modelling of Shipp, Sastry Pantula, and Bob Rodriguez Nonlinear Dynamic Processes” (SFB 823), supported by the offered various perspectives. Deutsche Forschungsgemeinschaft, is organizing an international —Steve Pierson, ASA director of science pol- workshop titled “New Developments in Econometrics and Time icy, updated the board on activities related to Series,” to be held in Rome September 10–11. The aim is to bring science and statistical agency budgets, climate together renowned experts in these areas with the researchers of SFB change, forensic science, statistics for congres- 823 to review recent developments and initiate scientific exchange. sional staff, and more. The workshop is in part a continuation of the successful series of Brussels-Waseda seminars on time series and financial statistics. Details and registration forms can be found at www.statistik. The board next meets on June 22 in Alexandria n tu-dortmund.de/1740.html. for its annual budget meeting.

6 amstat news may 2012 Call for Proposals for SAMSI Summer Programs and Workshops Richard Smith, SAMSI Director he Statistical and Applied Brief outline of how the pro- Mathematical Sciences gram or workshop will be Institute (SAMSI) structured Tannounces a new call for propos- als to organize a 2013 summer The program or workshop program or workshop focused on will take place at SAMSI, and a topic within the broad spec- the final number of partici- trum of new or recent research pants should be fewer than 50. themes covered by SAMSI. A The proposal should be sent to summer program is a 1–2-week [email protected] by program or workshop, including research program that is consis- July 27. budgeting, local arrangements, tent with the SAMSI core theme Proposals will be evaluated by and the coordination of invita- of bringing together statisticians, the SAMSI directorate, which tions. All organizational arrange- applied mathematicians, other may contact the applicants for ments will be made by SAMSI. mathematical scientists, and further information. A decision For more information about researchers in other disciplines, should be made by the end of SAMSI, visit www.samsi.info. though it is not necessary for September. Once a proposal is Reports from past programs every proposal to include all accepted, the SAMSI director- (including summer programs) these elements. A workshop is ate will work closely with the can be downloaded from www. typically 2–3 days and focused organizing committee members samsi.info/pgm-rpts. n on a specific topic. on further development of the A summer program or work- shop can form the seed from which a future year-long pro- gram develops. Therefore, a sum- Webinar Series Sponsored by mer program or workshop offers Quality & Progress and the Section on an opportunity to develop a new SAMSI research theme without Physical and Engineering Sciences the workload required for a year- long program. TITLE: Flexible and Powerful Approaches to Process Optimization using Bayesian Methods To apply, prepare a docu- PRESENTER: John Peterson (GlaxoSmithKline) ment of roughly 3–5 pages that DATE AND TIME: Wednesday, June 20, 1:00 p.m.- 3:00 p.m. Eastern time includes the following: DESCRIPTION Title and nature of proposed Bayesian methods provide straightforward solutions to important process optimization activity (e.g., whether it is a problems that are diffi cult to solve outside of the Bayesian approach. Th e Bayesian ap- two-week summer program, proach has provided published solutions to multiple-response process optimization with or one-week summer program, or without noise variables. Th e Bayesian approach has also provides model-robust solutions to process optimization by way of Bayesian model averaging. A recent paper by Edwards and workshop of 2–3 days) Fuerte (2011) provides a solution to the multiple-response ascent direction problem. Recent Names and affiliations of the research on hierarchical models has application to multiple-response split-plot and multi- batch experiments for process optimization. organizing committee members CONTACT INFORMATION Rationale and objectives of the Ananda Jayawardhana: [email protected] or Jo Martinez: [email protected] program or workshop Leading researchers whom the Medical University of South Carolina Healthcare Simulation South Carolina program is expected to involve Harborview Tower, 19 Hagood Avenue, Suite 706 (commitments from specific PO Box 250812 individuals are encouraged, but Charleston, South Carolina 29425 not required)

may 2012 amstat news 7 Chapter, Section Leading Economic Forecaster Chosen Microsites a NEW as Economic ASA Benefit Luncheon Speaker

The ASA’s Business and Economic Statistics Section recently recruited Edward Leamer to be this year’s Economic Outlook Luncheon speaker. Leamer is the Chauncey J. Medberry Professor of Management, professor of economics, and profes- new benefit of ASA mem- Microsites start with a standard sor of statistics, as well as former bership is the microsite template, but can be custom- chair of the economics department component of the ASA ized to a chapter or section’s at the University of California, Los Community,A which allows chap- Angeles. He is the author of more needs, look, and feel. If there is ters and sections to create a a specific image or logo associ- than 100 articles and several books, website for public use free of ated with a chapter or section, including Macroeconomic Patterns charge. Recently, the Kansas- it can be incorporated into the and Stories and, most recently, The Western Missouri Chapter site through a banner, and all Craft of Economics. (http://community.amstat.org/ the information on the previous As the director of the well- KWMChapter/Home) created a website can be migrated. After respected UCLA Anderson microsite after viewing the the microsite is set up, chapters Forecast, Leamer is uniquely quali- Chicago Chapter’s site (http:// and sections can add or update fied to speak on the economic community.amstat.org/Chicago_ content independently to suit outlook. For 50 years, the UCLA Chapter/Home). their needs. Anderson Forecast has provided “Our previous web page Microsites serve as a great tool forecasts for the economies of supported by the ASA was nei- for informing the public of news California and the United States ther administrator friendly nor and events. To take advantage of and is recognized as one of the most very attractive,” said Ananda this, all news items, announce- accurate and reputable independent Jayawardhana, Kansas-Western ments, and events posted in a sources of U.S. economic forecasts Missouri Chapter representa- chapter or section’s community available. Leamer is thus both a tive. “We started with a standard will be posted to its microsite leading economic forecaster and a template and customized it to our automatically. leading academic researcher in eco- chapter needs. … Maintenance of For more information about nomics and econometrics. the new web page is very easy, and this member benefit or to The luncheon, sponsored by the only a few emails were sent back get started now, contact Rick Business and Economic Statistics and forth to get all this done.” Peterson, Continuing Education Section, will be held in San Diego, The microsite feature is and chapters and sections associ- California, on July 31. The fee to becoming popular due to its ate, at [email protected]. n attend is $40 and includes a meal. functionality and easy set up. To register, visit www.amstat.org/ meetings/jsm/2012.

8 amstat news may 2012 Petitions Requested for New Section

n his book, Statistics on the Table: The History of Statistical Concepts and MethodsI , Stephen Stigler wrote, “Statistics and psychology have long enjoyed an unusually close relationship—indeed more than just close, for they are inextrica- bly bound together.” The same is true of statistics and education, in part because psychology and education are closely related. In an effort to embrace this shared tradition of statistics and mea- surement in psychology and edu- cation, a new ASA section, tenta- tively titled Statistics and Measurement in Psychology and One requirement for a new Education, is being considered. section is to have petitions from The section, if approved, will at least 100 ASA members say- allow ASA members to cooper- ing they would join the section ate with other ASA sections and and actively participate in its related professional organizations activities. If you are interested in to advance, discuss, understand, joining the section and partici- and publicize advances in sta- pating in its activities should it tistics and measurement as they be approved by the ASA, send a relate to psychology and educa- petition via email to Ken Kelley tion, as well as related disciplines at [email protected]. The email (e.g., sociology, management, should contain the following nursing, behavioral medicine). statements and information Some statistics and measure- for ASA to consider the email a ment issues are unique to study- signed petition: ing psychological and educa- tional phenomenon, and related I am currently a member of phenomena, which will be the the ASA and I support the principal focus of the section. formation of the Statistics and The scope will be broad because Measurement in Psychology both psychology and education and Education Section. are multidisciplinary domains of I intend to be actively involved study and many disciplines rely in the section and pay dues if on psychological and educational the section is approved and theory and data in a variety of becomes a full-fledged section ways. At the core of the Statistics of the ASA. and Measurement in Psychology Member Name and Education Section will be the development, improvement, Member Email Address and evaluation of statistical and ASA Member # measurement methods that have the potential to advance research Additionally, the name of the in psychology, education, and section is tentative. Any com- related disciplines. ments would be appreciated. n

may 2012 amstat news 9 TECHNOMETRICS HIGHLIGHTS Service Accessibility Analyzed with Novel Methods in May Issue ervice accessibility is the access of a community to Once estimated, a statistical reliability model can help the nearby site locations in a service network con- plan burn-in tests, in which every part is tested as a way sisting of multiple geographically distributed ser- of eliminating infant mortality. In “Degradation-Based viceS sites. Existing studies have had limited scope, both Burn-In Planning Under Competing Risks,” Zhi-Sheng geographically (e.g., towns) and temporally (e.g., a one- Ye, Min Xie, Loon-Ching Tang, and Yan Shen develop year period). In “Clustering Random Curves Under and illustrate such a planning framework for competing Spatial Interdependence with Application to Service risks, enabling both infant mortality and normal failure Accessibility,” Huijing Jiang and Nicoleta Serban modes. The failures, themselves, can be either degra- develop new statistical methodology to estimate and dation threshold failures or catastrophic failures. The classify service accessibility patterns varying over a large authors build three degradation-based burn-in models geographic area and a period of 16 years. The focus of and derive the optimal cut-off degradation levels. this study is on financial services, but it generally applies The issue finishes with three papers from different to any other service operation. areas of experimental design. The first concerns com- The paper introduces a model-based method for puter experiments. Computer simulators can be effec- clustering random time-varying functions that are spa- tive, but computationally intensive models of real sys- tially interdependent. The underlying clustering model tems. In such situations, experimental designs enable the is nonparametric with spatially correlated errors. The efficient exploration of the input-output relationship. In authors assume the clustering membership is a realiza- “Non-Collapsing Space-Filling Designs for Bounded tion from a Markov random field. These assumptions Non-Rectangular Regions,” Danel Draguljic, Thomas enable the model to borrow information across func- J. Santner, and Angela M. Dean invent a design algo- tions corresponding to nearby spatial locations, result- rithm that can fill unusually shaped regions and be non- ing in enhanced estimation accuracy of the cluster collapsing over input dimensions that may be irrelevant. effects and cluster membership. The paper demonstrates the technique in applications Several discussions and a rejoinder by the authors with constrained design regions, including a total elbow accompany the article, examining the relationship with replacement and tool-coating study. other approaches for functional and spatiotempo- Generally, experimental designs seek to maximize ral data and identifying possible directions for future information or minimize expected uncertainty. In “A research. The discussants are Ciprian M. Crainiceanu, Bifocal Measure of Expected Ambiguity in Bayesian Ana-Maria Staicu, C.B. Dean, Cindy X. Feng, Nonlinear Parameter Estimation,” Emanuel Winterfors Gareth M. James, Wenguang Sun, Xinghao Qiao, and Andrew Curtis take a new view of this problem, Bo Li, Xiao Wang, Jiaping Wang, Haipeng Shen, developing a “bifocal” measure of ambiguity based on and Hongtu Zhu. analyzing pairs of parameter estimates. Applicable to The remainder of the issue includes papers about both linear and nonlinear models, the new measure is regression, reliability, and experimental design. John equivalent to expected posterior variance in the linear H.J. Einmahl and Maria Gantner develop the “half- case and to a related measure in the nonlinear case. An half (HH) plot,” a new graphical method to investi- earth sciences application involving the location of wave gate qualitatively the shape of a regression curve. The sources in a medium of inhomogeneous velocity dem- empirical HH plot counts observations in the lower and onstrates the method. upper quarter of a strip that moves horizontally over the Experiments with factors at two levels are popular scatterplot. The plot displays jumps clearly and reveals for investigating main effects and two-factor interac- further features of the regression curve. tion effects. However, cost considerations may make Two papers in reliability follow. In “Testing for it infeasible to run resolution V designs, which would Monotone Trend in Recurrent Event Processes,” J.F. enable estimation of all main effects and two-factor Lawless, C. Cigsar, and R.J. Cook examine a general interactions. If some two-factor interactions are negli- concept of “trend,” showing the behavior of tests can gible, then resolution IV designs may allow estima- depend on the assumed definitions of “no trend” and tion of all effects of interest. In “Fractional Factorial “trend,” and on the observation periods for the process- Designs with Admissible Sets of Clear Two-Factor es. The paper also presents robust tests for trend across Interactions,” Huaiqing Wu, Robert Mee, and Boxin multiple processes, extends them to deal with interval- Tang improve on earlier approaches to the search censored event times, and compares them with other for such designs, using a graph representation of the well-known trend tests. design to reduce the set of designs under consideration to a more manageable size. n 10 amstat news may 2012 Meet Economic Research Service Administrator Mary Bohman

mstat News invited Mary Bohman of the Economic Research Service to respond to the following questions so readers could learn more about her and the agency she leads. Look for other statistical Aagency head interviews in past and forthcoming issues. What about this position rural economies. That means nontechnical, audience is a con- appealed to you? anticipating emerging issues and sistent challenge. This year, we providing substantive research unveil a redesigned website that The opportunity to lead a princi- and analysis on these issues to organizes and displays material pal federal statistical agency that inform policy decisions. On the for easier access to information informs critical public policy other side, it’s to maintain the and data. and societal issues was the major confidence of our customers and Our communications strategy draw. Today, perhaps more than stakeholders in the objectivity of includes investment of time and ever, the issues involving food ERS’s work. resources to improve data visual- and agriculture are central con- Another top priority is to ization and clarity and to enhance cerns of the national and interna- capitalize on technological devel- GIS-based tools such as our Food tional community. The challenge opments that enhance our ana- Desert Locator and the Atlas of of feeding a growing world popu- lytical capabilities. That includes Rural and Small-Town America. lation is an example. ERS analy- our ability to continue generat- Our new daily “Charts of Note” ses of agricultural productivity, ing high-quality data and models series demonstrates the broad the role of agricultural research in such as those used in our long- range of data ERS offers. Each productivity growth, the factors term baseline projections for the chart is accompanied by text to driving global food prices, and agriculture sector. provide context—often of a poli- Before becoming global and domestic food secu- We need to raise our visibility cy nature. The media have taken director of the Economic rity have already informed pol- and reach out to new custom- notice, with several news articles Research Service, Mary icy and provide the foundation Bohman served in several ers and partners, beyond those generated by individual charts. for further work to address this positions at the agency, who already value our work. challenge. Our research is poised most recently heading Partnerships with land grant and How can the statistical com- to provide objective analysis to ERS’s Resource and Rural other universities, other gov- Economics Division. She inform other national debates as munity help you? ernment agencies, and private served a detail position at well—around the next Farm Bill, ERS has long recognized the research institutions like the Ford the White House Office of childhood obesity, renewable importance of close links with Science and Technology Foundation have been mutu- energy, and adaptation to climate the statistical community. Policy and was on the ally enriching. Anticipating the change, to name a few. These mutually beneficial con- agricultural sciences needs and contributions of other tacts can lead to highly relevant faculty at the University of British Columbia. Describe the top 2–3 priori- customers and partners—being policy research. customer- and stakeholder-driv- ERS and other USDA agen- ties you have for the Economic en—will be a priority in the next Research Service. cies, for example, collabo- few years. rated with the Environmental A consistent priority is for the What do you see as the big- Protection Agency (EPA) when agency to remain relevant and EPA was developing regulations independent. My vision on one gest challenges for ERS? on animal waste management side of this coin is for ERS to con- Communicating our data and under the Clean Water Act. The tinue to addresses major policy findings effectively is as impor- USDA team provided analysis of issues under the umbrella of the tant as performing incisive alternative policy options. Data U.S. Department of Agriculture’s analysis on relevant topics. The from ERS on livestock operators’ (USDA) broad mission in the effort to communicate mate- production costs was a useful areas of food, agriculture, and rial clearly to an informed, but tool in evaluating cost-effective

may 2012 amstat news 11 Fast Facts Reports to Under Secretary for Research, for ERS customers. The Census and relevance, whether provid- Bureau, Bureau of Economic ing briefings to policymakers or Education, and Economics, U.S. Analysis, Bureau of Labor preparing in-depth reports. And Department of Agriculture Statistics, Centers for Disease the agency has been ahead of the Website: www.ers.usda.gov Control and Prevention, and curve in anticipating and inform- National Agricultural Statistics ing front-burner issues. FY12 budget: $77.7 million Service (NASS) are among the Our annual report on U.S. Staff size: 385 agencies whose data we use. household food security is used ERS’s Food Environment Atlas, by federal and state policymak- which compiles geospatial data ers in shaping laws and pro- on communities’ food options grams regarding food assistance, and health outcomes, is an exam- other public assistance, and pov- options for animal waste man- ple of one product incorporating erty. ERS, as part of the U.S. agement. ERS’s own research data from a number of sources— Food Security Measurement on market mechanisms of con- in this case, to explore the inter- Project—a collaboration among servation used data from EPA’s action of multiple factors in food federal agencies, university National Section 303(d) List of choices and diet quality. researchers, and private orga- Impaired Waters. nizations—played a key role in ERS draws on data from Prior to your tenure, what do developing the survey measure, USDA and other Federal agen- which has become a national and cies, as well as universities and you see as the biggest recent international standard. For more private-sector sources to use in accomplishments of the agency? than 10 years, ERS has overseen our analytical work, our forecasts, The Economic Research Service the survey—conducted by the and in data products we package has sustained its independence Census Bureau—and is respon- sible for analyzing and publish- ing the results and ensuring con- tinued reliability of the survey methodology. Each year’s report is widely cited by stakeholders and the media. I’m also particularly proud of ongoing technological enhance- ments in our Agricultural Resource Management Survey (ARMS). ARMS is a joint ven- ture with NASS and the primary source of information on the financial condition, production practices, resource use, and eco- nomic well-being of America’s farms and farm households. Information from ARMS is used extensively by policymakers and researchers. ERS and NASS have developed dynamic, easy to use web-based data delivery tools to expand access by customers. The two agencies are currently pro- viding remote access to autho- rized users to sensitive ARMS microdata in a confidential, pro- tected environment in collabora- tion with the National Opinion Research Center. n

12 amstat news may 2012 The World Is Calling; Should We Answer? Roger W. Hoerl, GE Global Research

watershed event in human history occurred Table 1—Copenhagen Consensus, 2004 in Copenhagen, Denmark, from May 24–28, 2004, during a meeting referred to The World’s Greatest Challenges asA the Copenhagen Consensus. Bjorn Lomborg, the Hunger and Malnutrition Swedish economist, brought together a “blue rib- Global Warming bon” panel, including three Nobel laureates, from Communicable Diseases Population and Migration around the globe to discuss the world’s greatest Poor Governance and challenges. Rather than prioritizing problems, the Conflicts and Arms panel was asked to prioritize potential solutions. Proliferation Corruption Specifically, the panel was asked to answer the fol- Global Financial Instability Sanitation and Access to lowing question: If one had $50 billion, how could Clean Water this money best be spent to improve the human Access to Education condition? Therefore, the panel had to consider not Subsidies and Trade Barriers only how severe the problems were, but also what impact could be made with this level of expendi- death for African-American women between 25–34 ture. For example, virtually everyone is appalled by years of age. Few people seem to have taken notice human trafficking, but what impact could be made of these trends, sometimes referred to as the “femi- to end it with this amount of investment? nization of AIDS.” The panel began with a list of the greatest chal- The cover story of the July 20, 2010, New York lenges facing the world, reproduced in Table 1. Times discussed a randomized trial conducted in Editor’s Note: This They then reviewed suggested solutions that had article was adapted Malawi. In this study, the treatment group of teen- from the Deming been proposed by globally recognized authorities age girls was given $1–$5 per month if they stayed Lecture, given at the in each field, which included some estimate of the in school, and their parents were given $4–$10 per 2011 Joint Statistical expected impact of their proposals. Finally, mem- month for household expenses. The control group Meetings in Miami bers of the panel met to prioritize the solutions, of girls received no money, nor did their parents. Beach, Florida. based on the impact they felt each was likely to Girls were randomly assigned to groups. At the have. In his summary of the gathering, Lomborg end of 18 months, the data showed that girls in the wrote the following: control group had an HIV infection rate 2.5 times that of the treatment group. This difference was Combating HIV/AIDS should be at the top of the statistically significant using any reasonable test. A world’s priority list. That is the recommendation jugular question that requires answering is whether from the Copenhagen Consensus 2004 expert this result is acceptable. Is it acceptable that a girl’s panel of world-leading economists.… About 28 chances of acquiring HIV are in large part depen- million cases could be prevented by 2010. The cost dent on access to $10 per month? would be $27 billion, with benefits almost 40 times My own experiences investigating the AIDS as high…. pandemic are discussed at length elsewhere. In short, I received a six-month sabbatical from GE What Has Happened in the Battle Global Research as part of my Coolidge Fellowship, Against HIV/AIDS Since 2004? which I used to study why AIDS seemed to defy Interestingly, Lomborg organized a second resolution, despite the expenditure of billions of Copenhagen Consensus meeting in 2008, at which dollars, and what might be done about it. After a proposed solution to global malnutrition was the considerable reading; interacting with AIDS activ- top priority. One might logically ask if perhaps HIV/ ists and researchers; and spending a month travel- AIDS is no longer a major issue, given existence of ing through Uganda, Zambia, and South Africa, antiretroviral drugs (ARVs). Sadly, the facts paint my co-author, Presha Neidermeyer, and I came to a different picture. For example, the World Health a number of conclusions, including the following: Organization announced in 2009 that AIDS- Money is necessary, but not sufficient to related illnesses are now the leading cause of death resolve this pandemic. It is sometimes diffi- globally for women of childbearing age (15–44). In cult for westerners to accept that they cannot the United States, AIDS is now the leading cause of “buy” a solution to a major problem.

may 2012 amstat news 13 Pastor Titus Sithole of Mamelodi, South Africa—a black township of about 1 million people outside of Pretoria—was publicly test- ed for HIV in front of his 2,000-member con- gregation, along with all his leaders, to make the point that AIDS is just a disease, and should not be stigmatized. This act of cour- age was and is virtually unheard of. Titus has subsequently opened an AIDS clinic within the community, as well as a children’s home for AIDS orphans and a primary school with approximately 450 students.

Addressing Large, Complex, Unstructured Problems Given the types of problems the world faces, such as Roger Hoerl speaks during JSM 2011 in Miami Beach, Florida. AIDS and the other challenges listed in Table 1, we might ask why statisticians are not working on these Similarly, “made in America” solutions often problems. The short answer, of course, is that we fail outside a U.S. context. Malawi is not are. I certainly want to acknowledge the outstand- Kansas, and solutions that work in Kansas ing work being done by statisticians around the often simply do not work in Malawi. globe who are focusing on finding solutions to can- This issue is not about charity, but rather cer, AIDS, climate change, providing safe drink- about justice. While western celebrities often ing water in resource-limited areas, and so on. run charitable events to raise money for They are doing precisely what I am suggesting. worthwhile causes and visit developing coun- Unfortunately, I don’t think there are enough of tries for photo shoots in impoverished areas, them. We need to do more, and I think we can too often such efforts seem more motivated by do more. what researchers call a “god complex,” rather The types of problems listed in Table 1 are than a desire for justice. A god complex is the not typical textbook problems in statistics. For subconscious belief that one is fundamen- example, in virtually every problem in statistical tally superior to those in abject poverty, and textbooks, there is a correct answer that can be because such people are typically wealthy and calculated precisely, without any understanding western, they often come as “saviors” rather or knowledge of any other discipline. In other than servants to the developing world. words, the problems are well defined and narrow. The same could be said of many other disciplines. Presha and I discuss a number of Africans we While the reasons for such an approach are under- have met personally who we feel are inspirational standable—certainly we would not ask a Stat 101 heroes, each of whom began significant and suc- student to solve the global AIDS pandemic—over cessful efforts to address AIDS and/or poverty in time it creates an expectation that problems should their communities with no western help. Of course, be well defined, of manageable size, and have a western aid can enable such people to help thou- “correct” answer—an optimal solution. Of course, sands, rather than hundreds, of people, so clearly the real world does not work this way. The unfor- money does have an important role to play. Two tunate result is that researchers often walk past the such people are: critical but messy problems begging for a solution, preferring to find much smaller, less important Jolly Nyeko, a woman from Kampala, problems that are well defined and solvable in a Uganda, who quit a paying job to counsel short timeframe. adolescent victims of sexual abuse on a volun- teer basis. She gradually expanded her efforts Science and Engineering and organization to include AIDS testing and treatment, legal referrals, education, and In January of 2010, Susan Hockfield, president of micro-loans. Her nongovernmental organiza- Massachusetts Institute of Technology (MIT) and tion (NGO), Action for Children, has now member of the GE Board of Directors, gave a talk affected the lives of roughly 30,000 Ugandans. at the GE Global Research Center in Niskayuna, New York. One of the topics she spoke about was

14 amstat news may 2012 Table 2—Attributes of Large, Complex, Unstructured Problems the need for better integration between science and engineering. Loosely quoting from my notes, • Big payoff, mission critical problems Hockfield stated: • Impact is broad—process performance, financial, customer, social, and environmental Around the dawn of the 20th century, physicists discovered the basic building blocks of the • Several departments, groups, and functions are involved universe, a “parts list,” if you will. Engineers said, • Problem has high degree of complexity involving both “We can build something from this parts list,” technical and non-technical challenges and produced the electronics revolution, and, subsequently, the computer revolution. More • Multiple sources of data and information are used recently, biologists have discovered and mapped • More than one statistical technique is required for solution the basic “parts list” of life—the human genome. Typically nonstatistical techniques are required Engineers have said, “We can build something from this list,” and are producing a revolution in • Creative use of information technology is needed for success personalized medicine. • Long-term successes require embedding solution into work Hockfield’s comments on science and engineer- processes, typically through: ing reminded me of a recent article by Xiao-Li Use of custom software Meng. In this article, Meng noted that Harvard Integration with other sciences and disciplines had recently added a course, Stat 399: Statistical Problem Solving, to its curriculum. This is a course that, in Meng’s words “… emphasizes deep, broad, and creative statistical thinking, instead of technical problems that correspond to a recogniz- and experimentation to discover, understand, and able textbook chapter.” This caught my attention, explain natural phenomena. Published definitions since the types of problems in Table 1 certainly do of engineering are typically some variation of the not correspond to a recognizable textbook chapter following: Engineering is the study of how to best in any discipline. use scientific and mathematical principles for the Meng’s article also resulted in significant dia- benefit of humankind. In other words, engineering logue with a professional colleague, Ron Snee, in does not advance scientific laws—science does— which we discussed how one should structure such but rather attempts to develop a theory of how these a course as Stat 399. There does not seem to be any laws can be better used for practical benefit—that consensus as to how one should attack large, com- is, how to build something of importance from the plex, unstructured problems. What approaches to scientific parts list. attacking large problems should be included in the course, and how would one know they are the best Statistical Engineering methods? By what theory would we answer these Based on these definitions, Snee and I defined the questions? It seemed to us that while significant term “statistical engineering” as the study of how theory exists about individual statistical methods, to best use statistical concepts, methods, and tools very little theory or research has been developed and integrate them with information technology on how these methods should best be linked and and other relevant sciences to generate improved integrated in order to attack large problems. This results. We think it is consistent with dictionary line of thinking led us to conclude that the science definitions of engineering, discussed above. We of statistics needed to be integrated with engineer- have further argued that the statistics profession has, ing approaches, per Hockfield’s comments. That is, in general, primarily focused on advancing statis- we think there needs to be a greater emphasis on tical science—the development and application of building something of interest to society from the new methods—while not recognizing that statistical statistical science “parts list” of tools. engineering is the “other side of the coin” that could Semantics are important in this discussion, so enable statistics to have broader societal impact. I will define my terms. A problem need not be as Relating this to Hockfield’s comments, it difficult as world hunger to be considered large, occurred to us that while there currently is a lot complex, and unstructured. Snee and I developed of emphasis within the profession on advancing some typical attributes of these problems, which the “parts list” of statistical methods, there is con- are listed in Table 2. Science can be defined many siderably less emphasis on “building something of ways, but typically refers to the use of observation importance to society” from this parts list. I refer,

may 2012 amstat news 15 for example, to building overall approaches to prob- approaches should be deployed. The natural result lemsolving or process improvement that involve is that statisticians, even experienced statisticians, multiple methods. Certainly, there are counter- have had to “reinvent the wheel” when faced with examples such as the Six Sigma methodology, but new problems having the attributes listed in Table there is little theory as to how this should best be 2. For example, I would argue that virtually no rig- done (i.e., what works, what doesn’t, and why). orous research or accepted theory exists on how to Many experienced statisticians have figured best integrate multiple statistical tools, or how to this out on their own, through trial and error, and integrate statistical and nonstatistical tools. applied statistical engineering without labeling it as Since publishing a definition of statistical engi- such. Unfortunately, this has typically been done neering and discussing it with statisticians in a on an ad-hoc basis, with little theory or research number of venues, we have been asked one question documented as to how such statistical engineering more than any other. This question has been asked in a few ways, such as: Is statistical engineering just another term for applied statistics? Isn’t this what statisticians have always done? Is this an attempt to rebrand applied statistics? These questions are cer- tainly reasonable and to be expected, as we are using the term statistical engineering in a new way and Nominations Sought there is a clear overlap with applied statistics. Using my own career as an example, I think I did a good for Deming Lecture, job of applying the statistical methods I learned in graduate school while an intern with the DuPont Applied Statistics Group in the summers of 1981 Noether Awards and 1982. However, I certainly did not build any- Nominations for the 2013 Deming Lecturer Award are due thing novel from the statistical science parts list of July 15. The awardee will give the Deming Lecture (an invited tools; in other words, I applied individual tools paper) at the Joint Statistical Meetings and receive a $1,000 to well-defined problems, but did not integrate honorarium, award plaque, and money for travel expenses. the tools in any innovative ways to achieve break- To be considered for the award, an individual must have through results. I did good applied statistics, but no made significant contributions in fields related to those in statistical engineering. which Deming devoted his career (e.g., survey sampling, sta- A number of commentators have pointed out tistics in the transportation industry, quality management, and that while most academic science departments are quality improvement), or more broadly, have made significant becoming narrower and more compartmentalized, contributions through effective promotion of statistics and sta- the truly groundbreaking research being conduct- tistical thinking in business or industry. ed today is almost exclusively cross-disciplinary in For details, visit www.amstat.org/awards/deminglectureraward. nature. Addressing the large, complex, unstruc- cfm or contact Marilyn M. Seastrom, selection committee chair, tured problems we currently face will require both at [email protected] or (202) 502-7303. breakthroughs in various sciences, including statis- Nominations can be sent by email to ASA Executive tics—such as the development of newer and better Secretary Pam Craven at [email protected] or by mail to statistical tools—and also the effective integration the ASA office, ATTN: Award Nominations, 732 North of diverse scientific methods into broader engi- Washington St., Alexandria, VA 22314–1943. neering approaches. The IBM computer Watson, which successfully competed on the U.S. game Noether Senior and Young Scholar show Jeopardy! and is now being applied to prob- lems in public health and consumer finance, is one Nominations are being accepted for the 2013 Noether Senior example. Watson integrates technology from a vari- and Noether Young Scholar awards. Visit www.amstat.org/ ety of disciplines, including computational science, awards/gottfriednoetherawards.cfm for more information and a natural language processing, statistics, and machine nomination form. If you have questions, contact the committee learning, to name a few. chair, Pranab K. Sen, at [email protected] or (919) 966-7274. In summary, it is clear that the world has numer- Nominations should be sent before December 1 to Pam Craven ous complex challenges and is in need of leadership in the ASA office at [email protected] or 732 N. Washington to address them. I believe statisticians can answer St., Alexandria, VA 22314, ATTN: Award Nominations. the call and have broader societal impact, if we choose to. A balanced emphasis on both statistical science and statistical engineering would help con- siderably in this effort. n

16 amstat news may 2012 Mathematics and Science Teachers (www.amstat.org/education/mwm) Sponsored by the American Statistical Association (ASA) 2011 Joint Statistical Meetings (JSM)* ” ” ” Based on the Common Core State Standards for Mathematics (corestandards.org) and Guidelines for Assessment and Instruction in Statistics Education (GAISE): A Pre-K–12 Curriculum Framework (www.amstat.org/education/gaise)

Dates: Tuesday, July 31, and Wednesday, August 1, 2012, 8:00 a.m. to 3:30 p.m. Place: San Diego Convention Center, located at 111 West Harbor Drive, San Diego, CA 92101 (workshop meeting room location to be announced) Audience: Middle- and high-school mathematics and science teachers. Multiple mathematics/science teachers from the same school are especially encouraged to attend. Note: Experienced AP Statistics teachers should register for the Beyond AP Statistics (BAPS) workshop. See www.amstat.org/education/baps for more information. Objectives: Enhance understanding and teaching of statistics within the mathematics/science curriculum through conceptual understanding, active learning, real-world data applications, and appropriate technology Content: Teachers will explore problems that require them to formulate questions; collect, organize, analyze, and draw conclusions from data; and apply basic concepts of probability. The MWM program will include examining what students can be expected to do at the most basic level of understanding and what can be expected of them as their skills develop and their experience broadens. Content is consistent with the Common Core State Standards, GAISE recommendations, and NCTM Principles and Standards for School Mathematics. Presenters: GAISE report authors and prominent statistics educators Format: Middle-school and high-school statistics sessions One-day pass to attend activities at JSM* (statistics education sessions, poster sessions, JSM exhibit hall) Activity-based sessions, including lesson plan development Provided: Refreshments One-day pass to attend the Joint Statistical Meetings Lodging reimbursement (up to a specified amount) for teachers from outside the San Diego area Handouts Certificate of participation from the ASA certifying professional development hours Optional graduate credit available Cost: The course fee for the two days is $50. Please note: Course attendees do not need to register for the Joint Statistical Meetings to participate in this workshop. Follow up: Follow-up activities and webinars (www.amstat.org/education/k12webinars) Network with statisticians and teachers to organize learning communities Registration: More information and online registration available at www.amstat.org/education/mwm. Space is limited. If interested in attending, please register as soon as possible. Contact: Rebecca Nichols, [email protected]; (703) 684-1221, Ext. 1877

*The Joint Statistical Meetings are the largest annual gathering of statisticians, where thousands from around the world meet to share advances in statistical knowledge. The JSM activities include statistics education sessions, posters sessions, and the exhibit hall. JSM BOUND Keynote Speakers Chosen for Knowledge, Dedication s the ASA membership gears up for JSM 2012, those planning to attend the annual meetings must decide which sessions to attendA and which activities to undertake. There are several to choose from, but the keynote addresses are special. Each speaker was chosen specifically for his or her vast knowledge of statis- tics and dedicated work in the field. Here, we introduce these speakers and hope you will con- sider attending their addresses.

18 amstat news may 2012 meetings

ASA Presidents Invited Address ASA Deming Lecture

Alan B. Krueger Jeff Wu Monday, July 30, 4:00 p.m. “Quality Evolution and Revolution: From Autos and Chips to Nano and Bio” Tuesday, July 31, 4:00 p.m.

Alan B. Krueger is chair of President Barack Obama’s Jeff Wu’s research contributions span the full range Council of Economic Advisers and a member of the of statistics, from theory to application, and touch Cabinet. He is on leave from Princeton University, many applied domains, from sample surveys to where he is the Bendheim Professor of Economics nanotechnology. They are notable for their com- and Public Affairs. He has held a joint appointment bination of novelty, technical strength, and far- in the economics department and the Woodrow reaching vision. He has made especially significant Wilson School at Princeton since 1987, and he contributions to experimental design. As one of his served as chief economist at the U.S. Department supporters wrote, “In view of professor Wu’s con- of Labor in 1994–1995. He earned his bachelor’s tribution to design of experiments, it is particularly degree, with honors, from Cornell University’s fitting for him to deliver a lecture that honors R. A. School of Industrial and Labor Relations and an AM Fisher, commonly regarded as father of the modern and PhD in economics from Harvard University. theory of experimental design.” Krueger was a member of the board of directors of the MacArthur Foundation and the Center for Economic Research and Graduate Education at Charles University in the Czech Republic, as well as a senior scientist for the Gallup Organization. He also is a fellow of the Econometric Society, the Society of Labor Economists, the American ASA Presidential Address Academy of Arts and Sciences, and the American Academy of Political and Social Science. In addition to receiving the Mahalanobis Memorial Medal by Robert N. Rodriguez the Indian Econometric Society and the Kershaw Prize by the Association for Public Policy and Tuesday, July 31, 8:00 p.m. Management, he has authored or coauthored sev- eral books about economics.

ASA President Robert Rodriguez is senior director of research and development at global software firm SAS. He earned his master’s and doctorate degrees in statistics from The University of North Carolina at Chapel Hill, where he also is an adjunct pro- fessor of statistics and operations research. Before joining SAS, he worked at General Motors Research Laboratories as a research scientist. He has chaired and served on a number of ASA committees, as well as committees of the North Carolina Chapter of the ASA. He is an ASA Fellow and, in 2009, received the prestigious ASA Founders Award.

may 2012 amstat news 19 meetings

IMS Medallion Lecture I IMS Medallion Lecture III

Yoav Benjamini Donald Geman “Selective Inference” “Order Statistics and Gene Monday, July 30, 10:30 a.m. Regulation” Wednesday, August 1, 10:30 a.m.

Yoav Benjamini is a professor of statistics in the Donald Geman earned his BA in literature from the Department of Statistics and Operations Research University of Illinois and PhD in mathematics from at Tel Aviv University. His areas of interest include Northwestern University. He was a distinguished pro- biostatistics, statistical genetics, bioinformatics, fessor at the University of Massachusetts until 2001, meteorology, zoology, data mining, and quality. He when he joined the department of applied mathemat- has published numerous papers and is a member of ics and statistics at The Johns Hopkins University. the Israel Statistical Association, American Statistical There, he is a member of the Center for Imaging Association, The Institute of Mathematical Science and Institute for Computational Medicine. Statistics, and the Eastern Mediterranean Region of His main areas of research are statistical learning, the International Biometric Society. computer vision, and computational biology. He is a fellow of the Institute of Mathematical Statistics and Society of Industrial and Applied Mathematics. IMS Medallion Lecture II

Emmanuel Candes COPSS Fisher Lecture “Sparse and Low-Rank Models in Statistics: Theory and Roderick Little Applications” “In Praise of Simplicity, Not Tuesday, July 31, 10:30 a.m. Mathematistry! Ten Simple, Emmanuel Candes is the Simons Chair in Powerful Ideas for the Mathematics and Statistics and professor of electri- Statistical Scientist” cal engineering (by courtesy) at Stanford University. Wednesday, August 1, 4:00 p.m. Prior to this, he was the Ronald and Maxine Linde Professor of Applied and Computational Roderick Little is the Richard D. Remington Mathematics at the California Institute of Collegiate Professor in the biostatistics department Technology. He earned his PhD in statistics from at the University of Michigan and associate director Stanford University in 1998. His research interests for research and methodology and chief scientist at are in computational harmonic analysis, mathemati- the U.S. Census Bureau. He earned his bachelor’s in cal optimization, statistical estimation and detection, mathematics at Cambridge University and his mas- with applications to the imaging sciences, signal ter’s and PhD in statistics and operational research processing, scientific computing, and inverse prob- at Imperial College, London. His research interests lems. Candes has received numerous awards, most include incomplete data, sample surveys, Bayesian notably the 2006 Alan T. Waterman Medal. Other statistics, and applied statistics. His primary research awards include the 2005 James H. Wilkinson Prize interest is the analysis of data sets with missing val- in Numerical Analysis and Scientific Computing, ues. An ISI highly cited researcher, he has refereed awarded by the Society of Industrial and Applied more than 170 publications, notably on methods for Mathematics (SIAM); 2008 Information Theory the analysis of data with missing values and model- Society Paper Award; 2010 George Polya Prize, based survey inference. In 2005, he received the ASA awarded by SIAM; and 2011 Collatz Prize, award- Wilks Memorial Award. To read more about Little, ed by the International Council for Industrial and visit http://magazine.amstat.org/blog/2012/05/01/ Applied Mathematics. He has given more than 50 roderick-little. plenary lectures at major international conferences.

20 amstat news may 2012 meetings Writing Workshop for Junior Researchers to Take Place at JSM

he National Institute of Statistical Science (NISS) and the American Statistical Association will hold a writing workshop Tfor junior researchers. The goal of the workshop is to provide instruction for writing journal articles and grant proposals. Participants will be required to provide a recent sample of their writing, which will be reviewed by a senior mentor. The sample could be a current draft of an article to be submitted for publication, or it could be an early version of a grant proposal. (Submission of the manuscript will be required as part of the registration process. Prior experience suggests that the best results come from submitting an early draft of something written sole- ly or primarily by the participant.) Mentors will be former journal editors and pro- gram officers, who will critique (a portion of) the submitted material. Individual feedback will be provided as part of the opening session, and par- ticipants will be expected to prepare a revision in response. The workshop will open with a one-day session of general instruction on effective writing techniques and close with discussion and debriefing at a follow-up lunch. The full-day session is scheduled for Sunday, July 29, in San Diego, California. At the close of the formal activities, mentors will meet individually with participants to go over their writing samples. Each participant will then prepare a revision of a critiqued portion of the paper and return it to the mentor by Tuesday evening, July 31. Mentors and participants will meet again in conjunction with a lunch on August 1 to discuss the success of the revi- sions. The lunch program also will include general This workshop is designed for researchers with feedback to participants, mentors, and organizers. a recent PhD in either statistics or biostatistics. Top Attendance is limited and will depend on the priority will go to those who have held the PhD number of mentors available. An online application for 0–3 years. The limited available funding will be form is available at www.amstat.org/meetings/wwjr/reg- used to support attendance by researchers at U.S. istration/index.cfm?fuseaction=ShowApp. Applications institutions. Current PhD students who are com- are due by June 1, and successful applicants will be pleting their degree before the end of the summer notified by June 30. Applications received after June and who will be at U.S. institutions in the fall also 1 will be considered if space is available. There is no will be considered. If space is available, research- fee to participate. Participants will receive lunch on ers at institutions outside the United States will be Sunday, July 29, and Wednesday, August 1. They admitted, but not provided with travel support. must agree to attend both the full Sunday session For more information, contact Keith Crank at and the Wednesday lunch. We anticipate funding for [email protected]. n partial travel support.

may 2012 amstat news 21 meetings How to Get the Most Out of Your First Joint Statistical Meetings Christopher Bilder, University of Nebraska-Lincoln

Before JSM Prepare before you leave. First, you should decide whether you want to give a presentation. For new attendees who choose to present, most give a con- tributed presentation, which is either an orally pre- sented paper or poster. The deadline to submit a corresponding abstract is usually February 1, and all are accepted. Additional proof of progress (e.g., drafts of a paper) for the presentation must be sub- mitted by mid-May. A preliminary program listing the presentation schedule is available online in April. Because there may be more than 40 concurrent presentations at any time, it is best to arrive at JSM with an idea of which to attend. This can be done by examining the session titles and performing keyword searches in the online program prior to JSM. Presentations are separated into invited, topic- contributed, and contributed sessions, each lasting 1 hour and 50 minutes. Invited and topic-contrib- uted sessions include groups of related presentations that were submitted together and selected by JSM Program Committee members. Oral presentations each last for 25 or more minutes for invited and 20 minutes for topic-contributed. Contributed paper ses- sions include groups of 15-minute oral presentations. Unlike invited and topic-contributed sessions, contributed presentations are submitted individu- ally and then grouped by JSM Program Committee members. Poster presentations (most are within he largest congregation of statisticians in the contributed sessions) involve speakers being avail- world happens every August during the able for questions next to their displayed poster dur- Joint Statistical Meetings (JSM). More than ing the entire session. T5,000 people attend these meetings, which are Online registration for JSM begins around May Christopher Bilder sponsored by seven statistical societies, including 1. For members of a sponsoring statistical society, is an associate the American Statistical Association. The meetings the cost is $380 in 2012 during the early registra- professor in the offer a variety of activities such as attending research department of tion period. The cost increases to $460 if you reg- presentations, interviewing for jobs, taking statistics at the ister at JSM. Registration for student members is Continuing Education courses, and browsing the University of only $85 in 2012, and this rate is available at any exhibition hall. With so many opportunities, new Nebraska-Lincoln. time. Also starting around May 1, you can reserve He earned his PhD in attendees can be easily overwhelmed by their first a hotel room through the JSM website. A number statistics from Kansas JSM experience. of hotels near the convention center are designat- State University. Based on my experience attending meetings for ed as official conference hotels, and they discount the last 12 years and the experiences of student their normal rates. However, even with a discount, groups I have led, I am going to tell you how to get you can expect to pay $200 or more per night for the most out of JSM. If you would like to share your a room. Most meetings also offer a less-expensive own recommendations, I encourage you to submit lodging option for students, usually housing at a a comment at http://stattrak.amstat.org/2012/04/10/ nearby university or hostel. mostjsm-2.

22 amstat news may 2012 meetings

Important Links Below are a number of web links corresponding to the Attending JSM can be expensive. Students have topics described in this article: several options to reduce the cost burden. First, ask your adviser or department for funding. Many JSM 2012: www.amstat.org/meetings/jsm/2012/index.cfm departments offer financial support for students who present their research at JSM. Students also Program: www.amstat.org/meetings/jsm/2012/ may qualify for funding from the student activities onlineprogram/index.cfm office on their campus. For example, when I was a student, my department’s statistics club received Job seekers: www.amstat.org/meetings/jsm/2012/index. funding this way, which paid for most of my first cfm?fuseaction=applicants JSM expenses. In addition to school-based resources, many ASA Continuing education courses and monitors: www.amstat. sections sponsor student paper competitions that pro- org/meetings/jsm/2012/ce.cfm vide travel support to award winners. For example, the Biometrics Section sponsors the David P. Byar Student paper competitions: www.amstat.org/sections/ Young Investigators Award, with $1,500 awarded to studentpaperawards.cfm a chosen student. Most competitions require a com- pleted paper to be submitted prior to JSM. At JSM JSM begins on a Sunday afternoon in late July or attend a course for free by volunteering prior to early August. Business casual clothing is the most JSM to be a monitor. Monitors perform duties such prevalent attire, but some attendees wear suits and as distributing and picking up materials during the others wear T-shirts and shorts. When you arrive at course. As an added benefit, monitors can attend JSM, go to the registration counter at the conven- one additional course for free without any duties. tion center to obtain your name tag and conference Keynote addresses at JSM are usually sched- program book. The program book will contain a uled for late afternoon on Monday through map of the convention center that can be useful for Wednesday. On Tuesday evening, the ASA presi- finding session rooms. dential address is given, along with a number of To welcome and orient new attendees, the awards and introductions of the new ASA Fellows. JSM first-time attendee orientation and reception The Fellows introduction is especially interesting is scheduled for early Sunday afternoon. Also, because approximately 50 ASA members (<0.33% the opening mixer on Sunday evening provides of all members) are recognized for their contribu- drinks, hors d’oeuvres, and the opportunity to tions to the statistics profession. meet other attendees. In addition to presentations, the JSM EXPO The main sessions start on Sunday at 2:00 p.m. features more than 50 companies and organizations Many of the research presentations are difficult to exhibiting their products and services. Many exhibi- understand completely. My goal for a session is to tors give away free items (e.g., candy, pens, etc.). have 1–2 presentations in which I learn something All of the major statistics textbook publishers and relevant to my teaching or research interests. This software companies are there. Textbook publishers may seem rather low, but these items add up after offer a sizable discount on their books during JSM, attending many sessions. and this discount is usually available for a limited For attendees who teach introductory cours- time after JSM. Software companies sometimes give es, the sessions sponsored by the ASA Section on away free trial editions of their programs. Statistical Education are often the easiest to under- The JSM Career Placement Service provides a stand. Many of these sessions share innovative ideas way for job seekers and employers to meet. This about how to teach particular topics. service offers an excellent way to interview with Introductory overview lectures are another many companies during a short time period. Pre- type of session that has easier-to-understand top- registration is required, and the fee is discounted ics. Recent lectures have included introductions to if you register before mid-July. The service works missing data, spatial analysis, and multiple testing. by providing an online message center for job seek- There are also many Continuing Education courses ers and employers to indicate their interest in each available for an additional fee. However, you can other. Once a common interest is established, an

may 2012 amstat news 23 meetings

Don’t let what happens at JSM stay at JSM, After JSM though. The first thing I do after the meetings JSM ends in the early afternoon on a Thursday. Don’t let what happens at JSM stay at JSM, though. The is prepare a short review of my activities. first thing I do after the meetings is prepare a short review of my activities. Using notes I took during ses- sions, I summarize items from presentations I want to examine further. I also summarize meetings I had interview can be arranged during the meetings. with individuals about research or other important Other activities at JSM include the following: topics. Much of this review process starts at the airport while waiting for my return flight. Shopping at the ASA Marketplace to purchase a If you give a presentation at JSM, you may statistics-themed T-shirt or mug submit a corresponding paper to be published in Attending an organized roundtable discussion the conference proceedings. Papers are not peer- during breakfast or lunch about a topic of interest reviewed in the same manner as for journals, but (pre-registration is required) authors are encouraged to have others examine their paper before submission. The proceedings are Using the free Internet access at the Cyber Center published online around November. Authors retain the right to publish their research later in a peer- Taking a little time off from JSM to go sightseeing reviewed journal. n

24 amstat news may 2012 meetings San Diego Offers JSM Attendees Much to Do, See

an Diego calls itself America’s finest city, and the variety of attractions and activities do make it a fine vacation destination. Following areS some highlights of local things to do, as endorsed by the ASA local area committee. For more, visit the convention and visitor’s bureau at www.sandiego.org. The visitor’s bureau site has information about all the activities listed below. If you set the calendar to the dates that interest you, you can find many of the activities scheduled for those days. Within walking distance of the convention center is the city’s waterfront, the Embarcadero. Here, you can tour an aircraft carrier at the USS Midway Museum, visit a collection of historic ships at the floating Maritime Museum, or catch a cruise of the bay. The historic Gaslamp Quarter, adjacent to the convention center, is best known for its nightlife, but features boutique shopping and great dining, too. Seaport Village, catering to tourists, is also just a few steps from the convention center. It is a nice place to catch some sun and souvenir shop between sessions. Balboa Park, a few minutes by car from down- town, was the site of the 1915 Pan-American Exposition and is America’s largest urban cultural park. While you may plan to visit the park to see the world-famous San Diego Zoo, consider taking in the rest of the park, too. It contains a multitude of smaller museums, the Old Globe Theater, an artist village, an outdoor organ pavilion, and numerous Playful orangutan at the San Diego Zoo gardens. The Prado restaurant is rated highly for outdoor dining. If you visit the zoo, set aside several hours, bring good walking shoes, and be prepared The peninsula of Point Loma is home to for lots of hills (unless the bus tour is more your Cabrillo National Monument. Here, you can speed). If you plan to visit the zoo and the rest of catch sweeping views of the city, see a historic light- Balboa Park, allow at least two full days; alterna- house, and take a short hike along the bayside trail. tively, plan for a full-day speed walk. Another great place for a walk along the ocean is Coronado Island sits in San Diego Bay and is Sunset Cliffs Natural Park. The beach community accessible by car or ferry from the Embarcadero. It of Ocean Beach boasts restaurants renowned for is home to one of the best beaches in America and their burgers and fish tacos. the Hotel Del Coronado, where visitors can check Old Town San Diego is convenient to the out the hotel and shops and enjoy seaside drinks or convention center via trolley. The Old Town dining at several restaurants. Coronado is a great State Historical Park, housing many buildings place to get an ice cream cone and stroll along the from the early days of the city, gives a glimpse of Pacific, or rent a bike to cover more of the island. 19th-century life in San Diego. Visitors can catch The public Coronado Golf Course is one of the tours and demonstrations and find more souvenir most affordable and beautiful in the area. shops here. Old Town has a high concentration

may 2012 amstat news 25 meetings

of Mexican restaurants and cantinas. Southern California’s history is deeply entwined with that of the missions that dot the coast. California’s first mission, Mission San Diego de Alcala (restored), is a few miles east of Old Town in Mission Valley. Mission Bay is the site of Sea World San Diego, a marine life adventure park where guests can find rides, shows, exhibits, animal encounters, and nightly fireworks. Mission Bay also offers miles of waterside trails and is a good place to try out kayaking or paddle boarding away from the crash- ing ocean waves. Head up to the cross on Mount Soledad for a beautiful city view; it’s a nice place to see the city lights at night. The city of La Jolla has a great deal to offer visi- tors. The village has high-end shopping and din- ing, while you can get a close-up view of the sea San Diego Cityscape. Convention center is in the foreground lions and harbor seals at the Children’s Pool in the cove. La Jolla Shores is another great family- friendly beach you may want to consider for a kayak tour or surfing lesson. Here, you may be able to snorkel with the leopard sharks that are often by shore in early August. The Scripps Institute of Oceanography’s Birch Aquarium is nearby. Also More … nearby is the University of California, San Diego, which has diverse art scattered around the open Museums: www. areas of the campus, a spectacular library building, sandiegomuseumcouncil. and sculpture gardens. org/museum_directory North of La Jolla is Torrey Pines, famous for its golf course and the location of a state park with a Queen Califia’s Magical Circle: Niki de St. Phalle sculpture gorgeous beach and hiking trails. Those looking for garden in Kit Carson Park, Escondido: www.queencalifia. a longer hike should head here. Adventurous souls org/MagicalCircle.pdf. Admission is free. Hours are 9 a.m. to may want to check out the glider port, where you sunset daily. It is most convenient if you visit this on a trip to can ride tandem with an experienced paraglider. the San Diego Safari Park or another site in Escondido. (760) The less adventurous may enjoy watching the col- 839-4691. orful sails floating and swooping along the coast. Hikers looking for more elevation may want to Niki de Saint Phalle’s statuary: Displayed outdoors in San visit Mission Trails Regional Park for a morning Diego. For locations: www.queencalifia.org/SanDiego.pdf. or evening hike up Cowles Mountain. The face on the convention center grounds, titled “Coming If you enjoy exercising your knowledge of prob- ability at the track, JSM overlaps with racing season at Together,” is by her. Del Mar Racetrack, “where the turf meets the surf.” Maritime Museum of San Diego: Open daily 9 a.m. to 8 p.m. The track, opened by a partnership between Bing Crosby and friends sits on Los Peñasquitos lagoon. Independent of the USS Midway. Being on the Pacific coast, San Diego offers USS Midway: Retired aircraft carrier with about 20 aircraft on many opportunities for boating. Short harbor Maritime Museum deck. Open daily 10 a.m. to 5 p.m. cruises originate at the , about a 20-minute walk north from the convention center. Visions Art Museum: The Art of the Quilt: www.quiltvisions. Longer cruises, whale-watching, and sport fishing org. Admission is $5; children under 12 are free. Open trips originate at Point Loma, about a five-minute cab ride from the convention center, just west of the Tuesday through Sunday from 10 a.m. to 4 p.m. San Diego Airport. n

26 amstat news may 2012 columnscolumns

SCIENCE POLICY When Statistics Education and State Policy Meet Christine Franklin, Lothar Tresp Honoratus Honors Professor in Statistics at the University of Georgia

Through the passionate efforts of Christine Franklin, Georgia now includes statistics throughout its K–12 performance standards. For this month’s science policy column, Franklin responds to the following questions, sharing her experiences, challenges, lessons, and observations. She also shares her advice to those seeking to have more statistics in their state’s education curriculum. ~ Steve Pierson, ASA Director of Science Policy

We understand you were instrumental Framework was the basis for the resulting statistics to the Georgia Department of Education standards that appeared in the new GPS for math- including statistics in their K–12 curricu- ematics. The statistics strand was considered one of the major strands of the GPS. Georgia became a lum standards. How did you get the ear of leading state for the inclusion of statistics and prob- the GA DoEd people and convince them? ability at K–12. The implementation for the new GPS began in the 2005–2006 school year. In the early 2000s, Georgia began the process of reviewing and rewriting the state mathematics stan- dards. When the GA Department of Education What were the main hurdles you faced in released the draft K–12 mathematics standards urging the state to include statistics? (called the Georgia Performance Standards [GPS]) The main hurdle was convincing the committee that for review, I read the standards and was surprised at the curriculum needed more statistical topics while what were being called the statistics and probability maintaining a balance of the necessary mathematical standards, especially at the high-school level. I was topics. Part of the justification for this balance was to seeing mathematics topics such as Venn diagrams, let data motivate the teaching of many mathematical sequences, and series placed under the statistics stan- topics. Statistics is a natural way to teach integrated dard. There was no indication of the statistical inves- mathematics supporting and complementing good tigative process and how to carry out this process. mathematical reasoning. The GA high-school cur- Fortunately, one of my mathematics education riculum was built around integrated mathematics. colleagues during that time at UGA, Brad Findell, A second hurdle was educating the writers that was a member of the writing team for 6–12. He statistics was NOT the mathematical probability understood the need for improvement in the statis- traditionally seen in K–12 mathematics textbooks. tics strand and advocated to the 6th–12th advisory committee that I be invited to speak. I was asked to Once statistics was included in the state travel to Atlanta and give a two-hour presentation. Also fortunate at that time, the ASA Pre-K–12 curriculum standards, what challenges did Guidelines for Assessment and Instruction in you encounter? Statistics Education (GAISE) Framework Strategic I discovered that most teachers had received training Initiative was approved by the ASA Board and made in only the very basics of statistics (or no statistics at public on the ASA website. I was able to use this doc- all). It became a personal campaign to help teachers ument as the basis for my presentation in explaining teach at a level that was higher than simply computa- what is meant by statistical thinking and the impor- tional skills. It was also important to help our teachers tance of all high-school graduates being able to use understand the statistical investigative process and the sound statistical reasoning. I advocated that the sta- learning progression of statistical concepts (such as the tistics standards be rewritten for 6th–12th. mean interpretation and variation from the mean at The advisory committee of dedicated teachers different grade levels). These are big ideas promoted and leaders in mathematics embraced this challenge by the GAISE Framework. and asked that I join a subgroup of writers to rewrite There was also the issue of teachers searching the statistics standards. The ASA Pre-K–12 GAISE for ready-made resources in teaching the statistics

may 2012 amstat news 27 columns

Science Policy Actions The ASA signs onto letters in support of FY13 budgets for the in-service teachers. I have learned much from the U.S. Census Bureau, BEA, BTS, EIA, ERS, and NASS. my colleagues being on this committee and working in a collaborative effort. The ASA executive director gives testimony to House Commerce, Justice, Science Appropriations Subcommittee What are you experiencing as a result of in support of budgets for NSF, Census Bureau, BEA, BJS, and statistics being part of the GA state cur- forensic science at NIST. riculum standards? The ASA signs letters expressing concern for making the American Community Survey voluntary. An exhilarating exhaustion! It has been rewarding to see where our committed teachers have progressed in the past few years, growing in confidence over standards. Helping teachers find these resources teaching the statistics curriculum. It is fulfilling to in an efficient way and to use these resources also hear the consistent comment that the students are became a personal campaign. most attentive when statistics is being taught, and Finally, there were simply not enough statisti- the teachers now understand the importance of the cians involved with assisting in the rollout of the students becoming statistically literate. The teach- new GPS and the statistics standards. ing of statistics is becoming part of the teachers’ belief system. How did the increased statistical content in What is the impact, if any, of Common Core the standards affect state assessments and State Standards (CCSS) including substan- professional development of Georgia teach- tial statistics content in grades 6–12? ers? How were you involved in these efforts? A big impact, since most states currently include The statistics frameworks (tasks) written for teach- very little in the way of statistics content. There ers to use in their classrooms have not always been is going to be a tremendous need for professional of the ideal quality with respect to statistical rea- development nationwide in statistics. For Georgia, soning. The state department was challenged with CCSS seems to have less statistics than the GPS. finding individuals familiar with the philosophy of But, hopefully, GA will use part of the 15% flexibil- the GAISE Framework to write these frameworks. ity to keep all of our current statistics standards— The tasks have tended to focus more on probability most especially what we have at K–5. The writers of than data analysis. The statistics items on the state CCSS elected not to include statistics at K–5. assessment tests have some of the typical skills-based questions, but an effort has been made to ask more What advice would you have for others conceptual statistical questions. Professional development has been huge. For interested in urging their state to include the past six years, I have traveled all over the state more statistics in their state education of Georgia and taught many professional develop- standards, improve the statistics questions ment workshops (some week-long in the summers and several one-day workshops during the school on the state assessments, or get involved year). My colleague, Gary Kader at Appalachian with professional development efforts? State University, assisted me with the summer work- (How does CCSS fit in?) shops. I have also provided support to the math spe- cialists at the Regional Educational School Agencies If you are in an academic setting, make connections (RESA) in Georgia. I have been fortunate to work with colleagues who have interaction with the state with fabulous math specialists who then carry on the department of education. Be willing to volunteer to professional development of statistics within their serve on test-writing committees, advisory commit- RESA’s and local schools. The University of Georgia tees, or as the writer of statistics resources for teach- also has a mathematics curriculum team consisting ers. Although this work on top of the regular job of two mathematicians, three mathematics educa- commitments can be a time challenge, my work tors, two statisticians, two master teachers, one high- with the teachers, RESAs, and state department in school math specialist, and one educational psychol- GA has been some of the most rewarding in my pro- ogy assessment expert who meet once a month to fessional career. It is the leaders and teachers at K–12 work toward the improved preparation of the pre- that will make the difference in spreading statistical service teachers and the professional development of literacy to our children. There is no better reward. n

28 amstat news may 2012 columns

STATtr@k Collaboration with Statisticians Danny Modlin, SAS Institute Inc.

esearchers and statisticians exist in a symbi- otic world of learning and discovery. To complete degree requirements, researchers Rare required to complete an introductory statistics course, as statisticians are required to take science courses. You, as a researcher, are an expert in your fields, as we, as statisticians, are in ours. So how do we get experts in diverse specialties to find common ground so our team can quickly achieve a mutual understanding of the problem at hand and effec- tively pursue a solution? Fortunately, as a statistics student at North Carolina State University (NCSU), I was exposed to an effective collaboration method thanks to NCSU’s statistical consulting course. Students and faculty across campus could connect to our departmen- tal web page and request statistical assistance with their projects or research. Students in the consulting class then met with the requesting individuals and assessed their problems. To assist me in writing this Communication Is Key article, I asked two NCSU faculty members—Jason During typical daily interactions, most would agree Osborne and William Hunt—for their input about that communication skills are something we all have collaborations. The following are a few viewpoints I the chance to practice. Whether it be telling a joke, gained from them and the consulting course. catching up on what you did over the weekend, or giving directions to some location, the goal is to be Don’t Wait Until the End to Ask for a able to convey your thoughts to someone else in Statistician’s Assistance such a way that they can understand. With all this There are many who think they are fully capable of practice, why is communication between research- performing the statistical analysis needed to solve ers and statisticians sometimes difficult? any of their problems with their brief introduction I posed the question, “What is the most impor- to statistics, and there are some statistical topics that tant thing that a researcher could do that would researchers can tackle without issue. However, both ease their collaboration with statisticians?” to Osborne and Hunt agree that a statistician should both NCSU faculty members. Osborne replied, be involved as soon as possible. “Communicate clearly. Explain things in basic “If an experiment is to be undertaken and the terms. Explain the big-picture first and then get into design is not simple, then the statistician should be details about sampling or experimental designs.” included as early as possible,” said Osborne. Hunt I would concur with this. I have had collabo- stated, “The statistician [should be] trained to be a rations in which my customer’s first question was team member from the very beginning.” a design question. I first asked for an overview Getting a statistician involved at the end of a of the project/experiment and the question(s) to project could result in, “I’m sorry, but the data you be answered. I would like to add that everyone collected will not be able to answer your question(s) involved needs to remain patient. You may have of interest.” The hassle of going back to square one to explain a concept a couple of times, but that with just weeks left before a deadline is an unwanted repetition will aid the statistician in their under- complication to a researcher. For this reason, statis- standing of the problem and better enable them to ticians should be a part of your projects from the solve your problem. A ‘good’ statistician, as I see it, beginning. We can assist in designing the experi- is one who repeats back to the researcher, in his/ ment and ensure the proper tests are performed her own words, their understanding of the prob- under validated assumptions. lem. A ‘good’ researcher ensures that this statement

may 2012 amstat news 29 columns

conveys the problem accurately. Your patience will For the researcher who may not be familiar with pay dividends later. the concept of repeated sampling, I simply restate the Are there specific concepts researchers need to need for a statistician to be involved as early in the know that would assist in collaboration? Hunt said, problem as possible. “It would be helpful if they had both introductory statistics and a course in the design of experiments.” Statisticians Are Researchers, Too I definitely agree with the design of experiments Those who have studied statistics do not spend their suggestion. As a student at NCSU, I took our design days consumed by the calculations of probability of experiments course and was pleasantly surprised density functions. We statisticians do have interests in to see that the majority of students were not from science, too. For example, I have always had an inter- the statistics department. When the semester began, est in hurricanes, so much so that my PhD research I was unsure of how this balance of students would involves hurricane modeling. My interactions with determine the focus of the course. It made the course meteorological researchers have been much smoother one of the best I had the opportunity to take at than if a different statistician had been involved with- NCSU. Each student benefited from other’s diverse out some interest and basic knowledge of weather. backgrounds. From the statistics perspective, we saw As a researcher, you may not have the good for- real-world applications brought in by nonstatistics tune of finding that ideal statistician who has an students for the experimental designs discussed. The interest in and prior knowledge of your field, but do nonstatistics students, beyond learning other experi- not let that stop you from finding a good communi- mental designs, were exposed to the statistical con- cator who is willing to listen and learn more about versations concerning the problems they presented your subject matter. to the instructor. This immersion started the lines of communication between our groups. Final Thoughts Osborne answered the question by saying, “The Statisticians and researchers are two groups of people concept of ‘repeated sampling’ can sometimes be elu- who need each other. Despite this need, communi- sive. I often ask how things would be different if the cation between us is not always fluid. With a little exact same experiment were repeated, controlling for work, and perhaps taking a few of the suggestions as much variation as possible.” mentioned above, a more stable bridge can be built. n columns MASTER'S NOTEBOOK Statistical Consulting in a University Research Setting Gregory J. Stoddard, University of Utah

y first job after graduation was as a will slow their disease progression. Can you help biostatistician for a medical device sub- me with this study?” sidiary of Becton Dickinson, Inc. The directorM of statistics from the corporate headquar- Situation Three: A researcher from the sociol- ters commented to me that he found working as a ogy department comes to you with the request, “I statistician in the medical device industry much have a grant that is due in one week. I would have more exciting than his experience in the pharma- come to you sooner, but I just found out about the ceutical industry. In the pharmaceutical industry, grant opportunity one month ago and I wanted to he explained, it takes several years to complete a have a good part of it done so you could see what clinical trial, while for medical devices, the clinical it is about.” trials are completed in less than a year. So, you get to see the results of your work much faster and then Common Theme One move on to a new problem. That seems like an The common theme of these situations is that you amusing contrast now. In the university setting, I are now on the spot. You have to display confidence frequently complete consulting projects in less than and take charge, or you will lose your credibility as a week, and the projects just keep coming. an expert. These researchers are depending on you Here are three hypothetical situations you might to make their projects successful. You are now in a find yourself in, so you can evaluate your skill level very applied situation, which requires you to think and interest. way outside of the statistical theory box. You have to be much more than a statistician—you have to be a Situation One: A researcher from the surgery researcher and think like a scientist. You have to not department comes to you with the consultation be intimidated by the research client’s field, which is request, “I have some data, and I now need some unfamiliar to you. Even with your own field, which help with the analysis.” The researcher gives a short is statistics, you cannot get away with specializing in description of the study. It involves a retrospec- just one or a few topic areas. Your consulting clients tive data collection from n=50 patient records who rely on you to be an expert in all areas of applied underwent Therapy A and n=50 patients from the statistics, providing the best solutions available and same clinic who underwent Therapy B. It was not a tailoring these statistical methods to their specific randomized study, but a series of patients who went research problems. through the clinic during the past year. The investi- That is pretty much how your job interview will gator wants to show that Therapy A is better. You ask go, as well. The interviewer, who is trying to pre- to see the study protocol. In the statistical section, it dict your success in an academic applied statistical states simply, “The data will be analyzed with chi- consulting, will provide you with a research prob- square tests and analysis of variance. The sample size lem and ask how you would solve it. Your statistical of n=50 per group provides adequate power10.” The theory classes are essential training, but questions reference number 10 is a surgery research article, not about theory are not likely to come up in the job an article on sample size determination. interview. To prepare, if you are still in school, you should take as many applied statistics and research Situation Two: A researcher from the art depart- design classes as you can from departments all across ment comes to you at the design stage of a study. campus. Also, getting work experience while earn- She says, “I have developed an art therapy program ing your degree is crucial. that I think will help children with muscular dys- Now let us test your skill at grasping what is trophy. As you probably know, muscular dystrophy involved with these three scenarios. is a disease in which muscles of the body get weaker and weaker and may slowly stop working. I think if Situation One Solution: From the sample size I can get the kids meaningfully engaged in express- determination of the protocol, you can tell a statis- ing themselves through drawing and painting, it tician was not involved in designing the study. The

may 2012 amstat news 31 columns

For more information about statistical consulting, visit the ASA Statistical Consulting Section online at www.amstat. “observational” study, confounding is going to org/sections/cnsl/index.cfm. be present. Since it is a case series, some form of selection bias is likely to be present, as well. The researcher apparently found an article in a surgical concepts of confounding and bias are taught in epi- journal, not necessarily involving the same thera- demiology courses—every statistician should take pies the researcher is studying, in which a signifi- an epidemiology class, or at least read one epide- cant difference was observed using a sample size miology textbook. The control for confounding is of n=50 per group. It is surprisingly frequent for going to require some version of a multivariable nonstatisticians to think this qualifies as a sample regression model. Addressing bias might require size determination. The statistical methods of the a sensitivity analysis. protocol appear to be some statistical tests thrown in to sound credible, so it will be up to you to sug- Situation Two Solution: Since a degree in art gest better statistical approaches. does not require research training, you will probably One thing you should bring up in your meeting have to take the lead with all research aspects of this with this investigator is whether it would be suf- project. The study is not a bad idea, as some benefit ficient to show that Therapy A is no worse than, or to the children with this disease could come out of noninferior to, Therapy B. If the two therapies are it. It is not likely that art therapy will successfully essentially identical in effectiveness and complica- slow the disease progression, or at least the effect tions, but Therapy A is less expensive, a noninferi- will be very small. It would require an impractically ority study approach is a better study design than large sample size to have adequate statistical power. the traditional superiority, or “find a difference,” So, you should make physical function a secondary approach. Most researchers do not think about this, variable, being content to show a “trend in the right so it is up to you to bring up the topic before look- direction” descriptively. Then, make something ing at the data. like “life satisfaction” or “self-esteem” your primary Something else you should recognize is that outcome, using an existing standardized scale, and since the study was not randomized, being an power the study for that outcome.

32 amstat news may 2012 columns

Abstracts Sought for What to use for a control group will take some American Statistical Association Conference on thought. What will help with that decision is deter- mining what the researcher would want to say in STATISTICAL PRACTICE her article when she publishes the result. It is possi- ble that the improvement in self-esteem is due large- A limited number of presentations will be accepted for the 2013 ly to the attention the researcher gives the children Conference on Statistical Practice, to be held in New Orleans, during the art therapy sessions, rather than the art Louisiana, from February 21–23. Abstract submission will take activity. If that is fine with the researcher, then you place from May 1 to June 1 at www.amstat.org/meetings/csp/2013/ would only need to compare against a control group abstracts.cfm. Abstracts can focus on any one of the following four in which no art therapy is given. If it is important broad tracks: to the researcher to conclude that the effect can be attributed to the art activity itself, then a control Communication, Impact, and Career Development group in which the researcher is giving the children • Communication skills the same amount of attention, but with a different • How to have an effect on your organization type of activity, would be required. This is a good • How to listen example of the statistician needing to think way • Ethics beyond statistical tests. • How to advance your career Situation Three Solution: Grant writing is a • How to adapt and learn very advanced statistical consulting skill. Frequently, you have to help the researcher revise his study aims Data Collection and Analysis as the first step, which requires that you be a better • Useful data analysis techniques researcher than the research client to even recognize • Econometrics and time series methods that the aims are not correctly conceived. In the • Nonparametric methods end, the grant reviewer for the funding agency, in • Modern regression methods deciding if the project should be funded, will look • Categorical data analysis to see if there is perfect alignment between aims, • Survival and reliability methods testable hypotheses, statistical method, and sample • Statistical engineering size determination. • Experimental design You cannot just cut and paste statistical methods • Computer experiments into the grant; they have to be perfectly tailored, • Survey design, distribution, automation, analysis and you have to convince the reviewer why those statistical methods are the best methods for testing Prediction and Analytics the hypothesis. It is almost always a rushed process, • How to process and use big data so you do not have time to learn as you go—you • Determining prediction reliability have to be an expert from the start. Since funds are • Data mining limited, the reviewer must find critical faults with • Is there such a thing as too much data? 90% of the grants, so these can be eliminated from • Discovery versus prediction the competition. A lot is riding on your perfor- • Analytics as a process or system mance, frequently more than $1 million in poten- • Determining valuable information from misleading or “junk” information tial funding. If you make a mistake, that money will be lost, since the statistical sections of a grant are an Software and Graphics easy target for the reviewer to find an excuse for not funding the study. • Software tools • Interactive visualization Common Theme Two • Visualization A second common theme of these scenarios is that there is a lot more to consulting than just analyzing Learn more about CSP 2013 at www.amstat.org/csp. data. The analysis part is there, to be sure, but you cannot get by with just crunching numbers. You American Statistical Association Conference on have to have a “big picture” perspective and exper- tise of the whole research process. If this sounds like a fun challenge and you do okay with deadline stress, STATISTICAL PRACTICE New Orleans, LA February 21–23 then statistical consulting is a great career path. n 2013

may 2012 amstat news 33 columns 175 Reaching 20k Monica Johnston, ASA Committee on Membership Retention and Recruitment Chair

Did you know the he American Statistical Association is a strong, graduation, especially if their term of service does vibrant organization with approximately ASA has a website not coincide with their graduation date. 18,000 members and the largest community We encourage chapters and sections with resourc- for individuals Tof statisticians in the world. In the April issue, the ASA es to consider offering grants to transitioning stu- who recently was described as a “big tent” organization with room dent-professionals for continuing education and graduated from a for growth. The goal of the Committee on conference travel not specifically hinged on submit- statistics program? Membership Retention and Recruitment (CMRR) is ting a paper to a conference. Grants help transition- Check it out at to help the ASA reach 20,000 members by its 175th ing student-professionals experience the benefit of anniversary in Boston in 2014. To meet this goal, we http://stattrak. quality continuing education, while leadership oppor- are working diligently to identify viable ways to tunities help them experience professional develop- amstat.org. increase the number of student members who renew ment; both can have a positive effect on retention. as regular members after they graduate and increase CMRR is working with the ASA marketing recruiting in areas where statisticians have a growing department staff to develop exit packages for gradu- professional presence. ating statistics students. Exit packages will contain The ASA has about 4,800 student members. About benefit information and incentives for keeping 59% of students renew their annual membership membership profiles updated and new professionals compared to 86% of nonstudents. Furthermore, we connected to the ASA. estimate that only 25% of students convert to regular A second focus of the CMRR is to increase mem- membership after graduation. Since the ASA offers bership from growing groups of statisticians or areas reduced membership fees for students and recent of application. The CMRR is exploring how the ASA graduates, and the full regular membership fee is less can attract distance learners of graduate programs in than 0.5% of a typical first-year salary, CMRR is satis- statistics. In the last 12 months, the department of sta- fied that fees—for most—are not a barrier for renewal. tistics at Texas A&M University received 1,133 appli- Instead, there are other factors affecting conversion. cants, of which 718 applied for a master’s degree and Members of CMRR believe students determine 415 applied for a four-course certificate. Currently, what a membership benefit is. ASA members who are the department has 17 traditional learners and 240 established professionals may itemize and exalt ben- distance learners. Other universities are developing efits of membership, but if a student doesn’t value a similar programs. Distance learners may represent particular benefit, then it isn’t genuinely considered a a good source for recruiting. Other sources include benefit. An important aspect of communicating mem- statisticians who are working in business analytics bership benefits is listening to prospective members and data mining. describe benefits that they value or expect. Therefore, CMRR is exploring whether professional groups in addition to showing a slide show on benefits of within social media networks are good sources of new membership during the JSM Student Mixer, CMRR members. For example, Meetup.com has a data min- is exploring how student focus groups can improve ing group that attracts nearly 1,000 members from the our understanding of factors that promote or hinder San Francisco Bay Area. Members of CMRR will be conversion from student to regular membership. working with the ASA’s marketing staff to increase the Additionally, members of CMRR think creat- ASA’s visibility within such groups. We also encour- ing opportunities for students to be active members age you to help grow membership through the ASA’s promotes conversion. For example, if you are an Member-Get-A-Member Drive (www.amstat.org/ officer or active member of a chapter or section, you membership/mgm). can help increase retention of transitioning student- As members of chapters, sections, and committees, professionals by encouraging greater student involve- it is important to remember the difference between ment in chapter and section activities. Officers describing a benefit to students and early-career pro- can use online membership reports (along with fessionals and having them experience that benefit help from the ASA) to identify current (or lapsed) through active membership in the ASA. Help us reach student members. When students serve as elect- our goal of 20,000 members with timely creation of ed or appointed officers, event organizers, ses- opportunities for transitioning student-professionals. sion chairs, etc., they may encourage other stu- If you have ideas for increasing membership, dents to get involved. Additionally, students who please convey them to your chapter, sections, hold office or have other responsibilities may be and committees, or contact Monica Johnston at more likely to renew as regular members after [email protected]. n

34 amstat news may 2012 people news

Technometrics and Computational Statistics and Data Analysis and has served as board or advisory member of many other associa- tions and committees. The San Antonio Chapter is proud to honor McCabe for his excellence in research, statistical consultation, and service to the statistical community. The Don Owen Award is presented annu- ally by the San Antonio Chapter and Taylor & Francis. n

The San Antonio Chapter vice president, Jonathan Gelfond (right), presents the Don Owen Award to George McCabe of Purdue University (left) on Jerry L. March 2 for his outstanding contributions to statistical research, applica- ASA member tions, and teaching. Oglesby was selected by the Texas A&M College of Science The 2012 Don Owen Award, McCabe has done an enor- for its highest alumni honor, given by the American Statistical mous amount of statistical induction into the college’s Read about Association’s San Antonio consulting work not only in Academy of Distinguished your colleagues Chapter, was presented to the university setting, but also Former Students. and friends in George P. McCabe on March involving more than 70 govern- Oglesby currently serves 2 at the Conference of Texas ment agencies, organizations, as senior director of Global the news. Go Statisticians by chapter vice presi- and companies. He is an elected Academic Programs and Global to www.amstat. dent, Jonathan Gelfond. Fellow of the American Statistical Certification within the educa- org and click on McCabe is a professor of Association and American tion division of SAS Institute Inc. “Statisticians in Association for the Advancement statistics and associate dean for He earned his doctorate in statis- the News.” academic affairs in the College of Science. He is a senior mem- tics from Texas A&M under the of Statistics at Purdue University. ber of the American Society for direction of Ronald R. Hocking. He earned his doctorate in math- Quality and president of the He served 13 years as both ematical statistics from Columbia National Statistics Honorary an assistant and associate University in 1970 and has Society, Mu Sigma Rho. He professor of statistics at the served as assistant, associate, and served as associate editor for University of West Florida and, full professor—as well as the director of the consulting center in the department of statistics— at Purdue. McCabe has published more 2012 Conference of Texas Statistician than 175 articles about statisti- Kumer Pial Das cal methodology and in a variety of application fields, including The Conference of Texas Statistician (COTS) was hosted at Lamar University from engineering, medicine, agricul- March 2–3. Kumer Pial Das and Mary E. Wilkinson from the math department, as ture, veterinary medicine, and well as Steve Doblin and Joseph Nordgren from the College of Arts and Sciences, pharmacy. He has taught numer- spoke at the welcome session. ous graduate and undergraduate In addition to George McCabe, who was presented with the Don Owen Award, courses and directed a number Dennis Cox of Rice University, Robert Serfling of The University of Texas at Dallas, of doctoral students. His book Cecil Hallum of Sam Houston State University, David Kahle of Baylor University, with David Moore, Introduction Mohammed A. Shayib of Prairie View A&M University, and Eddy Kwessi of Trinity to the Practice of Statistics (now University gave talks. in its 7th edition), “has been a Shaymal De of The University of Texas at Dallas received the first-place award of major influence in changing $100 in the student poster contest. The second-place award of $75 went to Gerald statistical education and how Ogola of The University of Texas at Dallas, and the third-place award of $50 went it is perceived…,” as stated to Anthony Daspit of Lamar University. Twenty-two posters were submitted at the by Rebecca W. Doerge in her COTS this year. nomination letter.

may 2012 amstat news 35 peoplecalendar news of events

in 1978, became founder and Karanatic University, Dharwad, study of statistics and has plans chief executive officer of SCI in January of 2012; and DLitt to establish a museum of statistics Data Systems Inc. Before join- from International Sanskrit for public education of statistics. ing SAS in 1996, he worked for University, Tirupati, in February He is the recipient of Monsanto Chemical Company of 2012. These are his 34th, numerous national and inter- as plant statistician. 35th, and 36th honorary degrees. national awards, including the Oglesby serves on several Rao was the director of the India Science Award, National advisory boards in support of Indian Statistical Institute, Medal of Science, Guy Medal statistics, computer science, Kolkata, where he worked in Gold of the Royal Statistical and data mining, including for 40 years before retiring Society, and the Wilks Medal. five years on the Center for the at the mandatory retirement Rao worked with P. C. Management of Information age of 60. He is the founder Mahalanobis to establish state Systems Board in the depart- of the new institute, CRRAO statistical bureaus to make ment of information and oper- AIMSCS at the University India one of the most advanced ations management at Texas of Hyderabad. At the age of countries in the collection of A&M Mays Business School. 92, he is still active trying to official statistics. He initiated He also serves as co-chair for develop the institute as one educational and PhD programs several of SAS Education’s data of national importance for in statistics. In the November mining, forecasting, and dis- advanced research in mathe- 2012 issue of Significance mag- covery conferences. n matics, statistics, and computer azine, editor Julian Champkin science. He is currently Eberly said, “C. R. Rao is a great Professor Emeritus at Penn name from the golden age of C. R. Rao recently received State and research professor at statistics. His work was done in three honorary doctorate the University at Buffalo. India; his intellect shaped sta- degrees: DSc from the University Rao started the Statistic tistics worldwide.” n of Colombo, Sri Lanka, in Olympiad program, the first December of 2011; DSc from in the world, to encourage the

36 amstat news may 2012 people news

Obituary College and completed his became head of the new depart- bachelor’s degree in education ment of statistics. He served as Franklin Arno Graybill in May of 1947, with math- department head until 1975. ematics and physics as his main Frank was president of Prepared by Duane Boes, Jay subjects. It was during this WNAR of the International Breidt, and Hari Iyer period that Frank met Jeanne Biometric Society in 1967. In Bunting, whom he married on 1968, he served as an associ- May 24, 1947. ate editor of the Annals of In September of 1947, Frank Mathematical Statistics; he was and Jeanne moved to Stillwater, editor of Biometrics from 1972– Oklahoma, where Frank earned 1975. He was president of the his master’s degree in math- American Statistical Association ematics at Oklahoma A&M. in 1976. In the fall of 1949, he went Frank taught graduate courses to Iowa State University to do in linear models, experimental his PhD in statistics, where he design, and matrix theory. He would work under the direction advised 33 MS and 26 PhD of Oscar Kempthorne. Frank students during his tenure at wrote his dissertation, titled CSU. Frank also published a “Quadratic Estimates of Variance number of books, both theoreti- Components,” and graduated cal and applied, including Theory with a PhD in 1952. He then and Application of the Linear went back to Oklahoma A&M Model, Matrices with Applications as an assistant professor of statis- in Statistics, Introduction to tics and stayed until 1960. the Theory of Statistics, An In 1959, Frank attended Introduction to Statistical Models the IMS meeting in Laramie, in Geology, and Confidence Franklin Arno Graybill, profes- Wyoming, where he met Elmer Intervals on Variance Components. sor emeritus and founding chair Remmenga from the depart- In 1981, Frank began work- of the department of statistics ment of mathematics at CSU. ing part time, but continued to at Colorado State University Remmenga informed him that teach and advise until he retired (CSU), died February 17, 2012, the dean of faculty at CSU, in 1997. at the age of 90. Andrew Clark, and the head of The success of the statistics Frank was born to Arno and the mathematics department, program at CSU is in a large Lula Graybill on September 23, Les Madison, were interested part due to Frank’s efforts, astute 1921, in Carson, Iowa, a farm- in creating a graduate program hiring during the 1960s, vision ing community of about 600 in statistics and looking for a for the department, creation of inhabitants. He attended Carson suitable candidate to lead this the statistical laboratory (which Consolidated School and then effort. Shortly thereafter, Frank now bears his name), and lead- attended two years of college at interviewed for the position ership and hard work over the Tabor Junior College. and received an offer to join years. To thank him for his In September of 1941, the department. In August of many contributions to the devel- he enrolled at William Penn 1960, he began his tenure at opment of statistics at CSU, the College and completed one CSU, where he stayed for the department annually presents semester of coursework before next 37 years. a graduate student with the joining the Army Air Corps. Soon after Frank’s arrival, Graybill Award for Excellence After the war, Frank attend- the department of mathematics in Linear Models and hosts the ed summer school at Colorado changed its name to department Graybill Conference Series. State Teachers College. of mathematics and statistics. Frank will be remembered by However, the college would During his initial years at CSU, his students and colleagues not not accept his course credits Frank created the statistical only for his professional achieve- from Tabor Junior College, laboratory and initiated the PhD ments, but also as a warm, kind which had gone out of busi- program in statistics. In 1971, man with great devotion to his ness during World War II, so the department of mathemat- family and a wonderful, down- he returned to William Penn ics and statistics split and Frank home sense of humor.

may 2012 amstat news 37 people news calendar of events Obituary widespread use of dietary supple- noted University of Washington Paul Levy ments in children from national professor Jack Goldberg, one of Paul S. Levy, a prominent bio- data. His textbook, Sampling Paul’s former students. “I remem- statistics educator and researcher, of Populations: Methods and ber how he would always say that died in a convalescent center in Applications, is one of the most if you went back to the basic tools Durham, North Carolina, from widely cited books in this field. of biostatistics and epidemiology, consequences of a spinal cord In recognition of his achieve- you could usually find an answer injury he suffered following a fall ments in the development, evalua- to your problem without having three years ago. He was 75. tion, and dissemination of sample to do anything too complicated.” Paul was an outstanding leader survey methodology and for the Paul earned a doctorate in bio- in public health and medical design and implementation of statistics from The Johns Hopkins research. During his career, he surveys of major public health University, a master’s degree in authored more than 200 pub- importance, Paul was selected as a mathematical statistics from lications and 30 book chapters Fellow of the American Statistical Columbia University, and a bach- and monographs in the fields of Association. In addition, he elor’s degree in mathematics from biostatistics and epidemiology. was named a Senior Fellow at Yale University. His expertise in sampling meth- Research Triangle Institute (RTI) He was a founding faculty ods allowed him to affect a broad International in 2006. This des- member of the school of public range of topics. ignation represents the highest health and the first director of Early in his career, he devel- scientific honor RTI bestows on the division of epidemiology and oped methods to measure the its own researchers. biostatistics at the University of effectiveness of the Illinois Trauma In addition to a thriving career, Illinois at Chicago (UIC). Prior to System in getting severely injured Paul demonstrated excellent his career at UIC, he was a faculty patients to trauma centers. As physical endurance and stamina. member at the Harvard Medical part of this work, he pushed the Between the ages of 60 and 70, School, as well as the University science in new directions, most he competed frequently in ultra- of Massachusetts School of Public notably in devising measures of marathon events that involved Health, an epidemic intelligence injury severity that would allow running for distances of 50 miles. officer at the Centers for Disease surgeons to compare outcomes When he was 66, he climbed 96 Control and Prevention, and across hospitals. floors in Chicago’s John Hancock a mathematical statistician at Paul was principal investiga- Building in 15 minutes. the National Center for Health tor for the Patient Registry for Paul was a beloved colleague, Statistics. Most recently, Paul was Primary Pulmonary Hypertension professor, and mentor through- employed by the RTI, where he (PRPPH). His work on sur- out his illustrious career. He focused on the development and vival of patients with primary approached his life and his profes- dissemination of statistical meth- pulmonary hypertension was sion with precision, excellence, odology, particularly in the areas published in highly prestigious and a great sense of humor. of sample survey methodology medical journals and has been “Paul gave me my first job. and epidemiology. cited by thousands of researchers. His advice was to work hard Paul served as president of the His co-PI, Stuart Rich, recently and fast, before people discov- ASA Chicago Chapter and won commented that the registry is ered I was around. I did,” said the award for Teaching Excellence considered a gold standard for Northwestern University profes- at the University of Illinois School information about PPH. Paul was sor Borko Jovanovic, a former of Public Health in 1978. instrumental in developing data colleague. “His second piece of In lieu of flowers, the family collection and analysis tools for advice was, ‘Focus on problems requests contributions be made to the Centers for Disease Control useful to other people.’ I did, the UIC School of Public Health, and Prevention’s Behavioral Risk and I have done well in my where a fund in Paul’s name has Factor Surveillance System, the career ever since. His third sug- been established to assist students largest ongoing telephone health gestion was, ‘If it is not fun, it is in the division of epidemiology survey in the United States. As probably not worth doing.’ He and biostatistics. Checks should part of his research, he proposed was the best boss I ever had.” be made out to the University a simple software tool designed “He had a unique way of get- of Illinois Foundation/Paul to measure the use of dietary ting to the heart of a problem. Levy Fund and sent to the UIC supplements in the United States Paul had a gift for finding simple School of Public Health, 1603 population. In parallel, he was solutions to knotty statistical W. Taylor, Room 1189, Chicago, one of the first to document the problems that stumped others,” IL 60612.

38 amstat news may 2012 section • chapter • committee news

Massachusetts-Amherst; and Cheng Zheng, University of Washington. The awards ceremony is scheduled to take place that evening during the section’s business sectionnews meeting and social mixer; there will be a door raffle at the event, and all members are welcome to attend. Biometrics The section’s scientific program includes many informative sessions about risk versus propensity The Biometrics Section will sponsor four short courses score–based matching methods, bias reduction, and during JSM 2012 in San Diego, California. The first, causal modeling. The section’s topic-contributed panel Statistical Analysis with Missing Data, will be taught on scoring systems for person-reported outcome mea- by Roderick Little and Trivellore Raghunathan and surements will feature experts from the field who will focus on methods for the statistical analysis of data sets lead an interactive discussion. Also included is a full with missing values. lineup of roundtable discussions: Smoothing Splines: Methods and Applications will be taught by Yuedong Wang and concentrate on Causal Direction and Model Evaluation in a particular class of modern nonparametric regression Structural Equation Models—Douglas David methods called spline smoothing for estimating func- Gunzler, Case Western Reserve University tions of one and several variables. What Do Statisticians Have To Do with Patients? The third course—Statistical Methods for Genome- —Laura A. Hatfield, Harvard Medical School Wide Association, Copy Number Variants, and Rare Issues in Adaptive Bayesian Dose-Finding Designs Variants Analysis—will be taught by Hongzhe Li and —Thomas M. Braun, University of Michigan Wei Pan and introduce statistical issues and methods related to analysis of genome-wide association data. Health Care Policy, Entitlements, and the The fourth course, Design and Analysis of Pharmaceutical Industry—T. Ceesay and Darcy Biomarker Studies for Risk Prediction, will be Hille, Merck & Co., Inc. taught by Tianxi Cai and Yingye Zheng, who will Do You Want To Visit Australasia?—Louise M. present recent statistical development for construct- Ryan, Commonwealth Scientific and Industrial ing and evaluating risk prediction model (markers) Research Organization with censored data. Preparing To Work in a Developing Country— In addition to the CE courses, the section is spon- Mark Griffin, Australian Development Agency for soring six invited sessions. To view the JSM 2012 Statistics and Information Systems online program or register for a course, visit www. amstat.org/meetings/jsm/2012. The Impact of Comparative Effectiveness Research The section also needs ideas for JSM 2013 invit- on the Clinical Research Enterprise—Sheldon ed sessions, to take place August 3–8 in Montréal, Greenfield, Health Policy Research Center Québec, Canada. Anyone interested in organizing an Methods for Analyzing Patient-Reported Outcomes invited session or who has ideas for one should con- —Kelly Zou, Pfizer Inc. tact Timothy D. Johnson at [email protected]. Submit For questions or comments about the HPSS pro- your ideas for short courses to Annie Qu at anniequ@ gram or activities in general, email section officers illinois.edu. Chris Schmid at [email protected], Yulei He Invited session ideas are also welcome for the at [email protected], or Frank Yoon at fyoon@ ENAR 2013 conference, which will take place March mathematica-mpr.com. 10–13, 2013, in Orlando, Florida. Send your ideas to For detailed information about the HPSS JSM Daniel Scharfstein at [email protected]. program, visit www.amstat.org/meetings/jsm/2012/ For detailed section news, visit http://magazine. onlineprogram. For information about the section, amstat.org/?cat=17. visit http://magazine.amstat.org/?cat=17.

Health Policy Statistics Statistical Programmers and Analysts The Section on Health Policy Statistics is set for The Section for Statistical Programmers and Analysts the 2012 Joint Statistical Meetings in San Diego, (SSPA) is looking forward to being in San Diego California. The student paper competition winners and participating in JSM with relevant Continuing will present their papers as part of the regular pro- Education courses; exciting roundtable discussions; gram on July 30. Winners are Ludi Fan, University of and various presentations on statistical programming Michigan; Hwanhee Hon, University of Minnesota; in medicine, health care, and finance. So Young Kim, The University of North Carolina The section also offers student travel awards to at Chapel Hill; Gregory Matthews, University of cover the registration fees for students and recent

may 2012 amstat news 39 section • chapter • committee news

graduates who want to attend an ASA conference. in statistics in epidemiology presented at the meet- Details and an application are available at http:// ings and are open to all current and recent graduate goo.gl/JMrQD. We encourage you to go to the JSM students in statistics, biostatistics, and epidemiology. Student Mixer on July 30, from 6–8 p.m. to meet Each award consists of $500 and a certificate. other students. A section officer will be there to To apply, you must have already submitted an answer questions about SSPA and the role of statisti- online abstract for JSM 2012. Papers to be presented cal programmers in early-career jobs. should be submitted by June 1, along with a cover Visit the SSPA booth in the JSM exhibit hall letter stating where you are a current student or your for details about upcoming webinars on technical year of graduation if you are a recent graduate, to Jaya and soft skills and travel grants. If you are interest- Satagopan at [email protected]. Questions about ed in volunteering at the booth, visit www.doodle. the award can be addressed to Melissa D. Begg, sec- com/4hargzvnz43mn84g. tion chair, at [email protected]. For detailed section news, visit http://magazine. The section is also sponsoring a short course at amstat.org/?cat=17. To view the JSM 2012 program this year’s JSM called Propensity Score Matching To view or register, visit www.amstat.org/jsmregistration. in R, which will be taught by Ben Hansen. This section news in course introduces propensity score matching and its entirety, visit Statistics and the Environment related ideas conceptually and methodologically, http://magazine. with complementary programming exercises using R amstat.org. Members of the Section on Statistics and the and its optmatch and RItools add-on packages. For Environment are preparing for an exciting program ASA members, the cost is $370; for nonmembers, at this year’s JSM in San Diego, California. They are $490; and for students, $255. To view a complete sponsoring the following roundtables: course description and register, visit www.amstat.org/ Spatial Thinking for Environmental Processes, meetings/jsm/2012/onlineprogram. with Alan Gelfand of Duke University For section details, visit http://magazine.amstat. org/?cat=17. Bayesian Methods in Genetic and Environmental Epidemiology, with Bhramar Mukherjee of the University of Michigan Survey Research Methods How Do We Adjust for Confounding Factors in Did you know the Survey Research Methods Section Environmental Health? with Francesca Dominici sponsors a webinar series? This month, Roderick of Harvard University Little of the U.S. Census Bureau and the University A business meeting and mixer also are planned. of Michigan will present “The Calibrated Bayes For details, visit the online program at www.amstat. Approach to Sample Survey Inference.” org/meetings/jsm/2012/onlineprogram. To register, visit to www.amstat.org/sections/SRMS/ In addition to this year’s JSM, the section is host- webinar.cfm. This website provides information ing a workshop on environmetrics, to take place about all upcoming webinars and the slides from October 11–13 at North Carolina State University. previous SRMS webinars. The theme for the workshop is “Spatial Modeling In addition to the webinars, the section has five and Inference for Environmental Science,” and the invited sessions planned for JSM 2012. Four are focus of the workshop is state-of-the-art statistical paper sessions and one is a panel: methods motivated by important research problems International Census Coverage, organized by in modern environmental science. Two half-day short Vincent T. Mule courses will be offered, introducing researchers to Poverty Mapping with Complex Survey Data, important topics in methods for environmental sci- organized by Sam Hawala ence. There also will be a poster session, providing an opportunity for graduate students and other young Multiple Faces of Multiple Imputation: 25+ researchers to participate. Years of Development, Innovation, and To view detailed section news, visit http:// Application, organized by Joerg Drechsler magazine.amstat.org/?cat=17. Dual-Frame RDD vs. Address-Based Sampling: Conducting Scientific Surveys in the 21st Century, organized by Meena Khare Statistics in Epidemiology Panel: Multi-Level Data Methods to Detect and The Section on Statistics in Epidemiology invites Adjust for Nonresponse Bias in Sample Surveys, applications for young investigator awards from organized by Tom W. Smith junior researchers who will be presenting papers at For detailed section news, visit http://magazine. the 2012 Joint Statistical Meetings (JSM) in San amstat.org?cat=17. Diego, California. The awards honor the best papers

40 amstat news may 2012 calendar of events 2012 25–29—SAMSI Summer The following events are the latest additions to June Program: Nonlocal Continuum the ASA’s online calendar of events. Announce- Models for Diffusion, ments are accepted from education and not-for- *19–20—Human Error Mechanics, and Other profit organizations only. To view the complete list Prevention, Garden Grove, Applications, Research Triangle of statistics meetings and workshops, visit www. California Park, North Carolina amstat.org/dateline. For more information, visit For details, visit www. * Indicates events sponsored by the ASA or one of www.samsi.info/workshop/ HighTechnologySeminars.com or its sections, chapters, or committees contact Ben Marguglio, POB 8, Cold nonlocal-continuum-models- Spring, NY 10516; (845) 265-0123; diffusion-mechanics-and-other- [email protected]. applications or contact Karem ›› Indicates events posted since the previous issue Jackson, 19 T.W. Alexander Drive, 20–22—Using Lean Six RTP, NC 27709; (919) 685-9324; Sigma to Prevent Avoidable [email protected]. Readmissions, Chicago, Illinois For details, visit www.wcbf.com/ *26–28—4th European DOE Mirpuri, 30 S. Wacker Drive, 22nd quality/5112 or contact Selina User Meeting, Vienna, Austria Floor, Chicago, IL 60606; (800) 959- Mirpuri, First Floor, Jubilee For details, visit www.statease. 6549; [email protected]. House, Merrion Avenue, London, com/4theuropeanmeeting.html *28–29—Statistical Challenges International HA7 4RY, UK; (800) or contact Heidi Hansel Wolfe, in eCommerce Research, 959-5346; [email protected]. 2021 E. Hennepin Ave., Suite 480, Minneapolis, MN 55413-2726; (612) Montréal, Québec 20–24—8th International 746-2033; [email protected]. For information, visit www.scecr.org Symposium on Statistics, West or contact Jui Ramaprasad, 1001 Lafayette, Indiana 27–29—Using Lean Six Sigma Sherbrooke West, Montréal, Québec For information, visit www.stat. and Process Improvement to H3A1G5, Canada; (514) 398-7426; purdue.edu or contact Diane Improve Patient Satisfaction: [email protected]. Martin, 250 N. University Ave., West Improving Quality, Safety, To view and Experience, Boston, Lafayette, IN 47907; (765) 494-3141; the entire list [email protected]. Massachusetts July For details, visit www.wcbf.com/ of statistics 23–25—International quality/5113 or contact Selina 1–4—IMS Asia Pacific Rim meetings and Conference on the Interface Mirpuri, First Floor, Jubilee House, Meetings, Tsukuba, Japan workshops, visit Between Statistics and Merrion Avenue, London, HA7 4RY, For information, visit http:// Engineering, Tainan, Taiwan www.amstat. UK; (+1)(800) 959-5346; selina. ims-aprm2012.org/index.html or For details, visit conf.ncku.edu.tw/ [email protected]. org/dateline. icise or contact Shuen-Lin Jeng, No. 1, contact Runze Li, Penn State University Road, Tainan, International 27–29—WCBF’s Using Lean, Six University, University Park, PA 701, Taiwan; 886-6-2757575-53600; Sigma, and Process to Improve 16802-2111; (814) 865-1555; [email protected]. Patient Satisfaction: Improving [email protected]. n Quality, Safety, and Experience, 24–27—32nd Annual Chicago, Illinois International Symposium For details, visit www.wcbf.com/ on Forecasting, Boston, quality/5113 or contact Selina Massachusetts For details, visit http://forecasters. org/isf/index.html or contact Pam Stroud, 53 Tesla Ave., Medford, MA 02155; (781) 234-4077; isf@ Science and Technology Policy Conference Planned forecasters.org. The seventh Gordon Research Conference on Science and Technology Policy 25–27—3rd Annual Pacific will take place August 5–10 with the theme “S&T Policy in Global Context.” Coast Statisticians and Sheila Jasanoff and Raphael Kaplinsky will make keynote presentations. Pharmacometricians Innovation All participants are encouraged to present their own work or their Conference (PaSiPhIC), San Luis organization’s in the poster sessions. Also, there will be a student-led, early Obispo, California career seminar August 3–4 at the same site on a related theme. For information, visit www.pasiphic. Buses are available from the Boston airport to the conference site at calpoly.edu or contact Brian Smith, One Amgen Center Drive, MS 38-3- Waterville Valley Resort in Waterville Valley, New Hampshire. Applications for B, Thousand Oaks, CA 91360; (805) this meeting must be submitted by July 8. 447-1378; [email protected]. For details, visit www.grc.org/programs.aspx?year=2012&program=scipolicy.

may 2012 amstat news 41 professional opportunities

Professional Opportunity listings may not exceed 65 words, plus equal oppor- tunity information. The deadline for their receipt is the 20th of the month two months prior to when the ad is to be published (e.g., May 20 for the July issue). Ads will be published in the next available issue following receipt. Listings are shown alphabetically by state, followed by international listings. Vacancy listings may include the institutional name and address or be identi- fied by number, as desired. and analysis of population-based regis- tries. Details for all positions, as well as Professional Opportunities vacancies also will be published on the ASA’s application instructions, are on our web- website (www.amstat.org). Vacancy listings will appear on the website for the site: www.lerner.ccf.org/qhs/jobs. Cleveland entire calendar month. Ads may not be placed for publication in the magazine Clinic is a AA/EOE. only; all ads will be published both electronically and in print. Rates: $320 for nonprofit organizations (with proof of nonprofit status), $475 for all others. Member discounts are not given. For display and online advertising rates, go to www.amstat.org/ads. Texas Listings will be invoiced following publication. All payments should be n The Department of Preventive made to the American Statistical Association. All material should be sent Medicine and Community Health at the to Amstat News, 732 North Washington Street, Alexandria, VA 22314- University of Texas Medical Branch is 1943; fax (703) 684-2036; email [email protected]. recruiting a tenure-track professor or asso- ciate who will also serve as the director of Employers are expected to acknowledge all responses resulting from pub- the office of biostatistics to play an active lication of their ads. Personnel advertising is accepted with the understanding leadership role in their teaching, research, that the advertiser does not discriminate among applicants on the basis of race, and service responsibilities. Applicants sex, religion, age, color, national origin, handicap, or sexual orientation. must have a PhD in biostatistics or Also, look for job ads on the ASA website at www.amstat.org/jobweb. applied statistics. Submit letter of inter- est and CV to [email protected]. UTMB Health is an equal opportunity, affirma- tive action institution that proudly values diversity. Candidates of all backgrounds are encouraged to apply. Minnesota n IDeaS is looking for highly motivated individual to develop high-quality statisti- International cal algorithms programming & designing n Senior instructor of rank comparable to new algorithms and carrying out statisti- associate professor position in the Dept. cal data analysis in support of research. of ISOM. Applications will be accepted Qualified candidates will have 3+ years until the position is filled. Excellence in experience in statistical analysis & data teaching and PhD required by employ- modeling & 3+ years advanced program- ment start date. The successful applicant ming skills. Master�s degree in mathemat- is expected to play an important role in ics, statistics, or equivalent (PhD and SAS teaching and developing business statis- experience preferred). Apply at www.sas. tics courses for undergraduate and MBA com/jobs/USjobs/search.html referring to programs of the business school. Submit job 12001421. EOE. CV and three references to stat11@ust. hk. The Hong Kong University of Science and Technology, School of Business and Management is an Equal Opportunity Ohio Employer. n n The Cleveland Clinic Department of Quantitative Health Sciences is recruiting for faculty, postdoctoral, and master’s-lev- el positions. Many areas are being sought, including biostatistics, data mining, health economics, health status measures,

42 amstat news may 2012

44 amstat news may 2012 may 2012 amstat news 45 Christus Schumpert Health System Endowed Chair Employment Opportunity – Biostatistician/Epidemiologist

The Department of Neurosurgery at Louisiana State University Health Science Center is seeking an biostatistician/epidemiologist to join as the Christus Schumpert Health Sys- tem Endowed Chair focusing on neuroscience research, stroke and outcomes data. This position will be at the Associate Professor/Professor Faculty level, tenure track, as commensurate with experience. Our research program includes a broad array of ap- plied epidemiologic, clinical services and scientifi c research.

The successful candidate will be able to take leadership in a program of applied re- search; mining our large clinical database to produce successful research activities culminating in the writing and publication of scientifi c papers and the acquisition of applicable grants. They will be responsible for developing research plans, conducting analysis of existing databases, assessing gaps in data collection and analysis, oversee- ing research intended to acquire new data, and will focus on collaborative, applied re- search with research team colleagues in neurosurgery and neurology which will require good communication among the multiple departments.

Qualifi cations include: PhD in Biostatistics/Epidemiology or a closely related fi eld with interest in neurosciences; advanced knowledge of research principles, theories, and concepts; applied statistical skills; applied knowledge of statistical software for analysis of complex data; and a proven record of achievement including peer reviewed publi- cations and national presentations.

Salary, startup package, and space allocation will be commensurate with faculty rank and extramural funding. Applicants should have a PhD and/or MD with relevant post- graduate experience. Applications will be reviewed as they are received until the posi- tion is fi lled. Shreveport is a progressive modern city with excellent schools, numerous fam- ily activities, and a very low cost of living. Interested individuals who wish to participate in a unique, diverse, academically productive medical center should send a C.V. with a letter describing research or clinical interests and with three letters of reference to:

Anil Nanda, MD, FACS Professor and Chairman, Department of Neurosurgery LSU Health Sciences Center, 1501 Kings Highway, Shreveport, Louisiana 71130-3932, FAX: 318-675-4457 [email protected]

LSU Health Sciences Center is an Affi rmative Action/Equal Opportunity Employer. Connect With Your Colleagues at JSM! REGISTRATION OPEN

Join the largest gathering of statisti- cians in the world and take advan- tage of hundreds of sessions about the state of the art of statistics this summer in San Diego, California.

You’ll enjoy: • Technical sessions

• Special receptions and gatherings

• An exhibit hall

• Continuing Education courses

• The Career Placement Service

Statistics: Growing to Serve a Data-Dependent Society

www.amstat.org/ meetings/jsm/2012

may 2012 amstat news 47 MAY 2012 • ISSUE #419

AMSTATNEWS ADVERTISING DIRECTORY

Listed below are our display advertisements only. If you are looking for job- placement ads, please see the professional opportunities section. For more job listings or more information about advertising, please visit www.amstat.org.

misc. products and services Northwestern University...... p. 32

professional opportunities Cedars-Sinai Medical Center...... p. 45 Capital One...... p. 45 The Georgia R School...... p. 47 Louisiana State University Health Sciences Center...... p. 46 NORC at the University of Chicago...... p. 44 SAS...... p. 43 St. Jude Children's Research Hospital...... p. 44 U.S. Census Bureau...... p. 44 Westat...... p. 47

software Cytel Inc...... p. 36 JMP, a business unit of SAS...... cover 3 Minitab Inc...... centerfold NCSS...... p. 12 Salford Systems...... p. 9 SAS...... cover 4 Stat-Ease, Inc...... p. 30 StatSoft...... cover 2 Statistical Solutions...... p. 24

48 amstat news may 2012