2021 COPSS AWARD WINNERS - AMSTATNEWSMay 2021 Issue #527 - American Statistical Association
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
May 2021 • Issue #527 AMSTATNEWS The Membership Magazine of the American Statistical Association • http://magazine.amstat.org 2021 COPSS AWARD WINNERS ALSO: ASA, International Community Continue to Decry Georgiou Persecution Birth of an ASA Outreach Group: The Origins of JEDI
AMSTAT NEWS MAY 2021 • ISSUE #527 Executive Director Ron Wasserstein: ron@amstat.org Associate Executive Director and Director of Operations Stephen Porzio: steve@amstat.org features Senior Advisor for Statistics Communication and Media Innovation 3 President’s Corner Regina Nuzzo: regina@amstat.org 5 ASA, International Community Continue to Decry Director of Science Policy Steve Pierson: pierson@amstat.org Georgiou Persecution Director of Strategic Initiatives and Outreach 8 What a Year! Practical Significance Celebrates Resilient Donna LaLonde: donnal@amstat.org Class of 2021 Director of Education 9 Significance Launches Data Economy Series with April Issue Rebecca Nichols: rebecca@amstat.org 10 CHANCE Highlights: Spring Issue Features Economic Managing Editor Impact of COVID-19, Kullback’s Career, Sharing Data Megan Murphy: megan@amstat.org 11 Forget March Madness! Students Test Probability Skills Editor and Content Strategist with March Randomness Val Nirala: val@amstat.org 12 My ASA Story: James Cochran Advertising Manager Joyce Narine: joyce@amstat.org 14 StatFest Back for 21st Year in 2021 Production Coordinators/Graphic Designers 15 Birth of an ASA Outreach Group: The Origins of JEDI Olivia Brown: olivia@amstat.org Megan Ruyle: meg@amstat.org 18 2021 COPSS Award Winners Contributing Staff Members 23 A Story of COVID-19, Mentoring, and West Virginia Kim Gilliam columns Amstat News welcomes news items and letters from readers on matters of interest to the association and the profession. Address correspondence to Managing Editor, Amstat News, American Statistical Association, 732 North Washington Street, Alexandria VA 22314-1943 USA, or email amstat@ amstat.org. Items must be received by the first day of the preceding month to ensure appearance in the next issue (for example, June 1 for the July issue). Material can be sent as a Microsoft Word document, PDF, or within an email. Articles will be edited for space. Accompanying artwork will be accepted 25 STATtr@k in graphics file formats only (.jpg, etc.), minimum 300 dpi. No material in WordPerfect will be accepted. On the Data Frontline: Biostatisticians Amstat News (ISSN 0163-9617) is published monthly by the American in the Hospital Research Setting Statistical Association, 732 North Washington Street, Alexandria VA 22314- 1943 USA. Periodicals postage paid at Alexandria, Virginia, and additional STATtr@k is a column in Amstat News and a website geared toward people mailing offices. POSTMASTER: Send address changes to Amstat News, 732 who are in a statistics program, recently graduated from a statistics program, North Washington Street, Alexandria VA 22314-1943 USA. Send Canadian or recently entered the job world. To read more articles like this one, visit the address changes to APC, PO Box 503, RPO West Beaver Creek, Rich Hill, website at http://stattrak.amstat.org. If you have suggestions for future arti- ON L4B 4R6. Annual subscriptions are $50 per year for nonmembers. Amstat cles, or would like to submit an article, please email Megan Murphy, Amstat News is the member publication of the ASA. For annual membership rates, News managing editor, at megan@amstat.org. 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 28 STATS4GOOD (703) 684–1221 Education Advocacy: Bending the ‘Moral Arc ASA GENERAL: asainfo@amstat.org of the Universe’ with Data for Good ADDRESS CHANGES: addresschange@amstat.org AMSTAT EDITORIAL: amstat@amstat.org This column is written for those interested in learning about ADVERTISING: advertise@amstat.org the world of Data for Good, where statistical analysis is WEBSITE: http://magazine.amstat.org dedicated to good causes that benefit our lives, our Printed in USA © 2021 communities, and our world. If you would like to American Statistical Association know more or have ideas for articles, contact David Corliss at davidjcorliss@peace-work.org. ® The American Statistical Association is the world’s largest community of statisticians. The ASA supports excellence in the development, application, and dissemination of statistical science through meetings, publications, membership services, education, accreditation, and advocacy. Our members serve in industry, government, and academia in more than 90 countries, advancing research and promoting sound statistical practice to inform public policy and improve human welfare.
w Ruixiao Lu Vice President, and Data Scienc Head of Statistics, e, Quantum Leap ASA Treasurer (2020 Clinical Data Mana Healthcare gement Celebrating ASIAN PACIF –2022) IC What inspi I am interested world, make good red you to in how data can become a stati help us better stician? AMERICAN decisions, and understand the help each other. HERITAGE MO What is the Own your career best career advice you’v NTH . Speak up. Don’t e received? give up easily. What are the leader to have most important Integrity. Empat ? qualities for a Claire Bowen hy. Accountabili Lead Data Scient ty. Courage. ist, Privacy and When trave Urban Institute Data Security, ling is possible you like to again, wher ASA Chair, Comm visit? e woul d ittee on Career Many places, includi Development ng JSM and other ences, plus the confer- What is the ASA headquarters! best career advice you’v departments the in-person I’ve missed Pursue what you interactions with my fellow love. This phrase e received? statisticians, data us justify makin sounds cliché, scientists, and g decisions that but many of leagues. I also physician col- we ultimately would like to visit do because it don’t want to China, where is what societ some of my family our own succes y says is succes members are s. sful. We define living. What is your What are the favorite most important way to relax leader to have qualities for ? ? a Reading with Empathy and the Pumpkin (my willingness to on my lap. older cat) “serve” others When trave ling is possible education again, wher e would you like to visit? Ireland What is your 30 Jing Cao way to relax Running with ? favorite my dogs or knittin Registration Open for USCOTS 2021 g Southern Metho dist University ASA Chair, Comm ittee on Professional Ethics What inspi red you to become The perspective a statisticia of using logical -statistical skills n? people from divers and working with e research fields. It is never boring . What is the best career Cultivate curios advice you’v ity; be a life-lon e received? g learner What are the leader to have Being able to most important ? motivate and and getting things qualities for a bring people togeth Ji Zhu done on time er; being organi Professor, Unive zed rsity of Michigan ASA Council of When trave Sections ling is possible where woul again, d you like to What inspi Machu Picchu visit? red you to become The opportunity a statisticia to work with real data and n? meaningful conclu What is your sions about real-lif make favorite way e problems to relax? What is the I have been teachin best career g myself piano received? advice you’v years—very enjoya for three e ble and relaxin Don’t be intimidated g. by people who you. Anybody are smarter than can do good resear contributions! ch and make unique What are the most important In celebration for a leader qualities of Asian/Pacific to have? American Heri Integrity and good we are paying tage Month, communication skills tribute to the our Asian- and contributions When trave of some of ling is possible Pacific Islander–A you like to again, wher These individual merican mem visit? e woul d s have been chos bers. I’d like to go to Beijing to visit meaningful leade en because of years old and my parents. They are almost 80 rship and com their I haven’t seen them in more mitment to the than a year. ASA. What is your favorite way Playing badminton to relax ? with my 10-yea r-old daughter APPposter.indd THE AMER 1 ICAN S TAT I S T I C AL ASSOCI AT ION 4/15/21 11:28 AM Look inside this issue for a poster celebrating Asian/ Pacific American Heritage Month! Biostatistics Professor Wins WSDS 2021 UPDATE Nathan Mantel The Women in Statistics and Data Award—Page 31 Science Conference is going virtual. This conference, now taking place October 6–8, 2021, celebrates and member news recognizes the contributions of women in statistics and data science. It will bring together leaders from academia, government, and industry; professional statisticians and data scientists; and 31 People News graduate students and postdocs. Join us from anywhere. Early registration opens May 27. ww2. 34 Awards and Deadlines amstat.org/meetings/wsds/2021/registration.cfm 36 Section • Chapter • Committee News 37 Professional Opportunities SIGN UP FOR ISI SHORT COURSES PROGRAMME Registration for the virtual 2021 ISI Short Courses Programme, organized in cooperation with the ISI Follow us on Twitter Associations, is now open. The courses will take www.twitter.com/AmstatNews place from May 6 to July 1. For information, visit Join the ASA Community https://bit.ly/3tekEio. http://community.amstat.org Like us on Facebook www.facebook.com/AmstatNews Follow us on Instagram www.instagram.com/AmstatNews Subscribe to our YouTube channel www.youtube.com/user/AmstatVideos 2 amstat news may 2021
president's corner Using Our Superpowers to Contribute to the Public Good I like to think of we statisticians as Jedi mas- tweaking. Think about the COVID vaccine rollout ters of uncertainty. We find ways to garner as a program if you want a real-world example. “The Force”—mathematical statistics—in Note that a formative evaluation (almost subduing the dark side of uncertainty in data to always involving qualitative research) illustrates achieve the common good: to gain insight and the rigorous version of this tweaking activity. In knowledge. But we are not omnipotent. Our an ideal world, one would want the new program superpowers are only as good as their underlying to mature and be running on all cylinders at the assumptions, assumptions that are all too often time of a quantitative evaluation that involves embraced with aplomb, yet cannot be proven. an RCT or quasi-experimental design. But that Photo by Errich Petersen Photography Some election polling in the past two presi- does not always happen—not by a long shot. Rob Santos dential elections has succumbed to the inad- It raises the question of what actually is being equacies of underlying assumptions (specifically, tested. Formally testing a novice program may likely voter models), in my humble opinion. not represent a fair assessment of its true impact Some COVID models of infections and deaths or potency and lead to a missed opportunity to performed well early on, only to falter as a result do good. of real-time changes in underlying assump tions tied to population behaviors (e.g., impact of lock downs, travel bans, mask wearing). When we are in our comfort zone and (implicit and explicit) assumptions hold, statis- Our superpowers are only as good as their ticians can appear invincible in their promises of cogent statistical inference. And while the sage words of our statistical Yoda, George Box, holds underlying assumptions, assumptions that true today that “all models are wrong, but some are useful,” we statisticians can face a real chal- are all too often embraced with aplomb, lenge knowing when some models are useful and when they are not. Let’s discuss with an example. yet cannot be proven. Consider a class of statistical investigations in the policy arena—randomized controlled trials. The rage over the past couple of decades has been the adoption of randomized controlled trials (RCTs) to demonstrate effectiveness of a given program in Then, there is the RCT itself. In the social addressing a particular social problem. Pick any and health program evaluation world, RCTs are program in housing, food insecurity, employment, pretty standard and randomly assign enrollees to education, recidivism, policing, etc. The usual spon- the new treatment program, or to either nothing sor-preferred approach is to conduct a demonstra- or “usual care” if some nominal forms of services tion and then follow that with an RCT. Simple and are being offered. We, as statisticians, are typical- rigorous, right? ly charged with developing power calculations to Well …, I always advocate to respect the life detect some (often unspecified) level of impact course of a program. Demonstration projects are that would be tantamount to a “successful” out- necessarily formative in nature. Especially for a come. And then we are sent on our way until first-time, nascent program, necessary adjustments it’s time to analyze the results and presumably are made on the fly because you cannot possibly declare victory of some sort. get everything “right” until you actually imple- The problem is that an RCT is defined by ment and see what’s working or not and what needs assigning folks to a treatment or control group, may 2021 amstat news 3
where the control is supposed to be the counter- measured impact of a program with respect to factual. Alas, there never really exists a true coun- alternative, unknown other treatments compli- terfactual—the absence of treatment and noth- cate the job of a statistician. You do need to care ing else. The reason is that something else almost enough to scratch below the surface of develop- always exists. If I am unemployed and fail to get ing power calculations or analyzing results and into a program to help me get employed, you can exploring the underlying influences on impacts. bet I will go find some other resources to help My illustration used RCTs for program evalu- me secure a job. If my family experiences food ation to discuss underlying assumptions, but insecurity and does not get into a program to help the lessons learned apply to most social science– provide food, I will not stop my efforts until I related endeavors. This is also true for data sci- find some way to get food on the table, and that ence projects and big data/AI projects. How do may well be through some alternative program. we know we are measuring the right thing? And The point is that, for social program evalua- is it being measured accurately? What assump- tions, there is almost never a true counterfactual; tions implicitly undergird the validity of results instead, RCTs typically measure the impact of a and associated inferences? new program against some other unknown single We, as statisticians, are in an awesome posi- treatment or combination of alternative treat- tion to help researchers think through these issues ments. Specifically, what is actually being mea- and understand the limitations and strengths of sured is a treatment program’s marginal impact the statistical inferences that flow from rigor- against unspecified other treatments. ous research studies. I often use such opportu- Such program evaluations typically do not nities to apply an equity lens to assess the cul- measure efficacy against a true counterfac- tural relevance and appropriateness of all aspects tual. That can wreak havoc on power calcula- of the design, from the underlying logic model tions because notions of how much of a mea- to data-collection modes, measures of efficacy, surable impact really defines “success” need to and intended analyses (and interpretations). The be rethought and often lowered in magnitude, most stimulating discussions stem from vision- which inevitably means a much larger trial and ing questions at the design stage when I ask, “If more time and expense. the program works as intended, what would be And then there is the reality of the environ- happening with the program staff, with the pro- ment. RCTs can suggest efficacy via statistical gram participants, and in the community?” The inference from a rigorous design, implementa- bottom line is that we can and should invoke our tion, and analysis. But many social programs own critical thinking into our statistical work, to help people in need are implemented at a whether we are a team member on a project or local level using community-based organiza- the project’s “episodic” consultant. We best serve tions (CBOs)—which could benefit much from our profession and our communities when we us volunteering our services, by the way. Local are thoughtful and humble. CBOs are not always as robust as we wish they Yes, I do believe in the power of The Force. were. It is not uncommon for professional And I do believe statisticians can be superheroes. staff—including senior staff such as program But, like everyone else, we are not infallible. directors and CEOs/presidents—to have limited Let’s use our superpowers with grace, honor, and tenures, leaving after a couple of years. The loss integrity to achieve the public good. of a “program champion” due to normal staff turnover can be devastating, even to the most effective program. I have personally seen this time and again throughout my career. Thus, we see basic assumptions such as the stability of staff infra- structure supporting a program, the nature of the counterfactual in an RCT, and the magnitude of 4 amstat news may 2021
ASA, International Community Continue to Decry Georgiou Persecution Steve Pierson, ASA Director of Science Policy, and Lynn Wilkinson, Friends of Greece F ollowing this winter’s Greek appeals court … It is also about the incentive structures for Greek ruling finding former Greek chief statisti- official statisticians now and in the long term. What cian Andreas Georgiou liable for slander, happens in this case, when it reaches the Greek ASA leaders denounced the latest legal setback Supreme Court, will have repercussions in Greece against Georgiou. “Persecuting a scientific govern- and in the wider EU, both for the rule of law and ment official for doing his job with rigor and human rights, and for the soundness of policies integrity to produce official statistics is deeply based on honest and reliable official statistics.” concerning,” said ASA President Robert Santos. The slander charge pertains to a public state- “Prosecuting Georgiou for stating what is wide- ment Georgiou made in his official role while ly validated and fulfilling his official responsibili- fulfilling his responsibility to defend the fully vali- ties is a travesty,” added ASA Executive Director dated official deficit and debt statistics for Greece Andreas Georgiou Ron Wasserstein. produced under his leadership. Echoing ASA Board statements, both ASA lead- ers urged Greece to end its now 10-year persecution Background of Georgiou and signal Greece’s commitment to “The game is over; we need serious statistical data.” accurate and ethical government statistics by fully ~ Jean-Claude Juncker, President, Eurogroup, exonerating him. 2005–2013 The international community also spoke out, including through social media (see sidebar). In a “I am seriously concerned about significant statistical March 29 opinion piece for the Greek newspaper discrepancies… [which] will require an open and deep Kathimerini, prominent Greek economist Miranda investigation of what has happened.” Xafa wrote, “Georgiou’s prosecution has done enor- Joaquin Almunia, Commissioner, EU Monetary Affairs, mous damage to Greece’s image abroad and has 2004–2009 called into question the country’s ability to under- stand that the huge fiscal imbalance was real, not a One of the conditions the European Union conspiracy to impose austerity.” (EU) imposed upon Greece in the beginning of John Bailer and John Pullinger, presidents of the 2010 during the early days of the Greek debt cri- International Statistical Institute and International sis was significant reform of the statistical office. Association for Official Statistics, respectively, In this context, further illustrated by the quotes echoed the calls for Greece to end the persecu- above from 2009, Georgiou was recruited to head tion of Georgiou and exonerate him, reiterating the recast national statistical office—the Hellenic in a joint statement their “grave concern that these Statistical Authority (ELSTAT)—with a mission continued prosecutions have damaged the scien- to modernize the production of official statistics, tific integrity of highly regarded work addressing apply fully the EU rules for the production of these Greece’s problematic fiscal statistical reporting from statistics, and ensure ELSTAT’s independence and the 2000s.” They added that Georgiou’s “case has overall implementation of statistical ethics. implications for the international statistical system, Georgiou’s priority when he took the helm of and the rights of government statisticians to defend ELSTAT in August 2010 was to address concerns their statistics,” saying that “[d]efending official repeatedly voiced by European and international statistics … should not lead to legal prosecution.” partners about the credibility of previous Greek eco- German data scientist Katharina Schueller in a nomic statistical figures. Therefore, one of Georgiou’s March 2021 LinkedIn article wrote, “In this case, first responsibilities was to oversee a recalculation not only the fate of a single statistician is at stake. of the public finance statistics for 2006–2009 that Eurostat, the EU’s statistical office, could not validate may 2021 amstat news 5
and which had been adjusted several times during Georgiou was found guilty of simple slander the lead-up to the debt crisis. It was the fully correct- in criminal court. His first appeal of this decision ed public finance figures produced under Georgiou’s failed, but the Greek Supreme Court annulled it leadership that allowed Greece to continue to qualify for extraordinary legal errors. Simple slander, under for the EU and International Monetary Fund (IMF) Greek law, means that even though the statements loans given to help its struggling economy through a Georgiou made were true, he should not have made tough fiscal adjustment process. them because they damaged the honor and reputa- However, because of the austerity measures that tion of the plaintiff. came with the loans, Georgiou quickly became a scapegoat, blamed by leaders across the Greek politi- 2021 Civil Appeals Court Ruling cal spectrum for the debt crisis, the full size of which on Slander his work had uncovered. As the effects of the austeri- The latest ruling concerned the civil suit for simple ty measures started to be felt, criminal investigations slander by the same plaintiff. In 2017, Georgiou was were initiated against Georgiou in September 2011. ordered to pay damages of €10,000 to the plaintiff Greece proceeded to prosecute Georgiou for doing and publish large parts of the convicting decision as his job to produce accurate, complete, and current a public apology. After numerous delays, the appeals public finance statistics, starting with charges of fal- court held its hearing on this case in September HOW TO HELP The ASA joins sifying the magnitude of the deficit and violation of 2020. In its 2021 decision, the appeals court reject- international partners duty (for not allowing a vote on the statistics). ed Georgiou’s appeal and upheld the decision of the in requesting your At the time the slander charges originated (2014), lower court, including damages of €10,000 (plus support of Georgiou Georgiou had already been investigated twice on interest) and the public apology. by posting to Georgiou said he plans to appeal the ruling to the LinkedIn and Twitter charges of falsifying the deficit and violating his duty. Both investigations concluded there was no Greek Supreme Court, adding, “The stakes in this urging Greece to end its persecution evidence to support the charges and, both times, the case are not only personal for me. They are about fol- of Georgiou and conclusion was ignored and another investigation lowing statistical principles in the various processes exonerate him was initiated. Georgiou was about to face a third of producing and reporting official statistics, includ- using the hashtags investigation and, at this time, the deficit figures ing defending validated official statistics in the face #AndreasGeorgiou and of destructive criticisms. They are also about incen- #JusticeForGeorgiou. produced under Georgiou’s watch had been accept- ed by Eurostat without reservations eight times in a tive structures for official statisticians in Greece now row over a span of four years (2010–2014). and for the long term. Moreover, the outcome of this Reacting to these continued, relentless attacks case can potentially indirectly affect the conditions on the statistics produced under his leadership, for official statisticians anywhere in the world.” Georgiou issued a strenuous public defense of the The Greek government has also funded a signifi- integrity of these statistics. He also noted the lack of cant portion of the plaintiff ’s legal costs to pursue recognized validity for the originally provided data the slander accusation against Georgiou. The law for 2006–2009 by Eurostat and others and pointed under which these funds were granted was meant out that no investigation of these statistics had ever to support official statisticians in their legal defense, been conducted. This fact remains true to this day. and it is a perversion of the law that these provisions Public comments about government statistics have been used for a prosecution. are a standard practice around the world and in Public funding for the head of ELSTAT is part of a accordance with the European Statistics Code of 2017 agreement between the European Commission, Practice, as well as the UN Fundamental Principles Greece, and The Bank of Greece for that year’s dis- of Official Statistics. Public comments are also bursement of €7.7 billion. The agreement held that in accordance with one’s democratic right to free Greece must “legislate the State indemnification of speech, including in Greece. the ELSTAT President—and other ELSTAT officials Nevertheless, both criminal and civil charges acting upon his/her authority—against legal and were instigated against Georgiou by the former other costs … incurred as a result of legal challenges/ (2006–2010) director of the national accounts divi- actions/proceedings taken or threatened against them sion of the statistics office, who was responsible in relation to decisions made and actions taken … for the production of the deficit and debt statistics pursuant to carrying out their official functions in Eurostat could not validate. compliance with applicable provisions and rules, 6 amstat news may 2021
Influencers Speak Out Following the latest legal setback for Georgiou, the international community questioned Greece’s continued hounding of Georgiou on social media using the hashtags #AndreasGeorgiou and #JusticeForGeorgiou. Among the scores of people who tweeted were the following significant influencers: Olivier Blanchard, Robert Solow Professor of Nicholas Christakis, Sterling Professor of Social and Economics Emeritus, MIT, 77,000 followers: Natural Science, Yale University, 181,000 followers: Search #JusticeForGeorgiou to find tweets from other influencers. statutory or otherwise.” The agreement referenced a narrative and persecution of an honest and highly 2012 Greek government commitment “to support capable official statistician as an injustice, a viola- ELSTAT in upholding confidence in Greek statistics tion of Georgiou’s human rights, and a detriment to and to defend them against any efforts to undermine Greece’s reputation and economy.” their credibility.” His American lawyer, Robert Kyle, working on a pro-bono basis, said in an email, “The persecution of Perpetuation of False Narrative Mr. Georgiou is an affront to truth and government In their statements, Santos and Wasserstein equated integrity that ultimately undermines the credibility Greece’s continued prosecution of Georgiou with of Greek statistics. The Greek Government should the perpetuation of a false narrative. As noted above, do everything in its power to right this wrong.” the official statistics produced under Georgiou have This false narrative can still be found in Greek been widely and repeatedly validated. Further, the media today (e.g., https://bit.ly/3uD2g34). ELSTAT methods Georgiou put in place for pro- ducing Greek official statistics are still in use. Human Rights Violations While understanding the hardships of the Greek Besides the injustice of the prosecutions, the harm people because of the austerity measures, Santos stat- to Greece’s reputation, and the undermining of ed Georgiou “was unjustly cast as the scapegoat for official statistics, Greece’s treatment of Georgiou is the austerity measures put in place with the EU and also a violation of Georgiou’s human rights. Indeed, International Monetary Fund loans from the early both the 2019 and 2020 US State Department 2010s to help Greece’s then-struggling economy.” Country Reports on Human Rights Practices for Referring to the broader impacts of the scape- Greece include several sentences about Georgiou. goating of Georgiou, Santos said, the ASA is “deep- The language for both is similar, with the 2019 lan- ly concerned that a false narrative has damaged the guage excerpted here: scientific integrity of highly regarded work address- ing Greece’s problematic fiscal statistical reporting Observers reported the judiciary was at times from the 2000s.” inefficient and sometimes subject to influence Wasserstein reiterated the point, saying, “The and corruption. … On February 28, the Council ASA sees the continued perpetration of a false of Appeals cleared, for the third time, the former may 2021 amstat news 7
head of the Hellenic Statistical Authority, Andreas Georgiou, of charges that he falsified 2009 budget What a Year! Practical data to justify Greece’s first international bailout. The Supreme Court prosecutor had twice revoked his Significance Celebrates acquittal by the Council of Appeals. … EU officials repeatedly denounced Georgiou’s prosecution, reaf- Resilient Class of 2021 firming confidence in the reliability and accuracy of Kim Gilliam, ASA Marketing and Communications data produced by the country’s statistical authority Coordinator A under his leadership. s the COVID-19 pandemic stormed on, the Rebecca Everly, director of the Committee Class of 2021 and its teachers had a year no on Human Rights of the National Academies of one could have imag- Sciences, Engineering, and Medicine, said, “We are ined. Adaptable, determined, deeply concerned that Dr. Andreas Georgiou has and resilient are just a few faced and continues to face legal proceedings solely adjectives often attributed to MORE ONLINE students and educators this To learn more, for carrying out his scientific work and responsi- bilities during his five-year tenure as president of year. For our May podcast, visit The Friends of Greece website: Greece’s national statistics office. Dr. Georgiou is a hosts Donna LaLonde and Ron https://bit.ly/2QIod22. deeply respected scientist whose service to Greece Wasserstein welcome to the should be supported, not targeted.” show three students earning Yuyu (Ruby) Chen degrees this month and two Praise for Georgiou [very busy!] faculty members. While also speaking against the latest ruling, The following degree can- Wasserstein thanked and praised Georgiou. “I’m didates reflect on this unprece- sorry for all Georgiou has had to endure, but I dented year, share their favorite also commend, and am grateful for, his defense classes and experiences, offer of official statistics. With his perseverance and advice for academic success, strength, he is an exemplary embodiment of the and tell us what’s next for them: commitment of an official statistician to produce • Yuyu (Ruby) Chen, the most reliable, impartial data possible.” Emma Godfrey Master of Science, The praise echoed a 2018 commendation Biostatistics, New York from six statistical societies for Georgiou’s “com- University petency and strength in the face of adversity, his commitment to the production of quality and • Emma Godfrey, Bachelor trustworthiness of official statistics and his advo- of Arts, Mathematics, cacy for the improvement, integrity and inde- Pomona College pendence of official statistics.” • Sallie M. Yassin, Master The ELSTAT data produced under Georgiou of Science, Biostatistics, and their specific methodology has been checked New York University Sallie M. Yassin and accepted 22 times since 2010 by Eurostat Additionally, Miles Ott, associate professor of in its semiannual quality assurance procedures statistical and data sciences at Smith College, and provided for in European law. In addition, the Amy Hogan, who teaches mathematics and statistics statistical processes and ethics characterizing the at Brooklyn Technical High School, (both self-pro- production and dissemination of these official claimed “Smithies”) commiserate over the challenges statistics under Georgiou’s watch are considered of teaching during the pandemic—what’s worked, by the international statistical community to strategies they’ll use going forward, and advice they be fully consistent with international statistical traditionally share with their graduates. Good stuff! principles and ethics. Listen to Episode 5 of Practical Significance as these For more information about Georgiou’s bat- ASA members relay how they ‘Zoomed’ through 2021 tle for justice, contact ASA Director of Science and join us in celebrating this milestone moment for Policy Steve Pierson at pierson@amstat.org. n Ruby, Sallie, and Emma. Follow Practical Significance on Twitter @TheASAPodcast. n 8 amstat news may 2021
Significance Launches Data Economy Series with April Issue Brian Tarran, Significance Editor W e live in a “data-driven world,” an “age of are timely, relevant, and reliable … We conclude analytics” in which “data is the new oil.” that official statistics systems should be broadened. These are phrases we’ve all become They should generate quality public statistics that familiar with—and perhaps grown tired of hear- can be used to help get answers ing—because of their repeat- to the many urgent questions ed use. Less frequently heard, about society and how we can though, is the story of how we sustainably improve our lives got to where we are now— and livelihoods.” how the “data economy” came Several other articles tackle to be. The April issue of COVID-related issues. In the Significance begins a new four- Notebook section, for exam- part series that tells this story, ple, Harrison Schramm, starting with the birth of cus- Isaac Rubin, and Norah tomer insight. Schramm investigate the This “history of the data econ- impact distance learning omy” is an idea the Significance may have had on the grades staff have been developing for a of students at a California while—to tell the story of how high school. Meanwhile, the business of buying, selling, and in the StatsComm section, profiting from data has developed Kevin McConway and over time and consider where it David Spiegelhalter offer might go next. It’s a story that spans tips to statisticians communicat- 200 years and features an eclectic ing through the media—a valuable and timely set cast of characters, from advertisers and social scien- of advice. tists to market researchers and statisticians. Of course, Significance is a publication in which Elsewhere in the April issue, Robert Matthews statisticians and data scientists can hone their com- reflects on the American Statistical Association’s munication skills. The aim of the magazine is to 2016 p-value statement, asking what—if any- demonstrate in an entertaining, thought-provoking, thing—has it achieved? In the five years since the and nontechnical way the practical use of statistics statement’s release, there have been debates and dis- in all walks of life and to show informatively and agreements, editorials and symposia, and a plethora authoritatively how statistics benefit society. Almost of ideas about how science could be changed for the all articles are written by statisticians eager to share better. But, argues Matthews, “The reality is that, their passion for statistics and explain their work— in terms of changing research practice, the ASA or the work of others—in a way that is accessible statement has achieved little. Yet the need for such and relatable to readers of different backgrounds. change has never been greater.” If you have an article idea you would like to dis- In the Perspectives section, Paul Allin and David cuss, send an email to significance@rss.org.uk. You J. Hand make the case for change of a different sort, can also review notes for contributors at significance arguing that as societies look to “build back better” magazine.com/contribute. in the wake of coronavirus, official statistics should The April issue of Significance is available in take center change. “The COVID-19 pandemic is a print and online. ASA members can read the maga- testing time for statistics,” they write. “It is showing zine for free online by following the instructions at that single sources struggle to provide statistics that significancemagazine.com/654. n may 2021 amstat news 9
CHANCE HIGHLIGHTS Spring Issue Features Economic Impact of COVID-19, Kullback’s Career, Sharing Data Amanda Peterson-Plunkett, CHANCE Executive Editor I t has now been more than a year since SARS- CoV-2 began spreading across the globe. JSM 2021 Session Researchers understand the virus better, and CHANCE will host a session at the Joint vaccines are being distributed—but what do we know about the effects the pandemic has had on Statistical Meetings in August, titled our economy and society? Authors Jon T., Nicholas “The Stories of CHANCE: Frontiersman, B., and Thomas Middleton study this question as Exoneree, Player, Spy.” it pertains to the United Kingdom in the article, “Modeling the Economic and Societal Impact of Non-Pharmaceutical Interventions During the to minimize the effect of these extraneous factors in COVID-19 Pandemic.” “Toward Reducing the Possibility of False Positive Keeping with the topic of COVID-19, we Results in Epidemiologic Studies of Traffic Crashes.” look at an application of Benford’s law, a tool that Listed as one of the 25 most-dangerous jobs has successfully been used in applications such in the US (based on 2019 data from the Bureau as detecting financial fraud and altering digital of Labor Statistics Census of Fatal Occupational images. In “Benford’s Law and COVID-19 Data,” Injuries), police work is not for the faint of heart. authors Chase Marchand and Dalton Maahs use it In “Police Officers Killed in the Line of Duty: A to analyze reported COVID-19 cases. Correspondence Analysis of Circumstances and In “The Secret Career of Solomon Kullback,” his- Time of Day,” Terry Allen investigates whether cer- torian Brenda McIntire details the US intelligence tain types of police activities are more dangerous at career of one of the developers of the Kullback- specific times of day. MORE ONLINE Leibler divergence. If you are intrigued by this arti- Moved by the 2018 attack on the Pittsburgh Tree Follow CHANCE cle, you may also enjoy the recent PBS special The of Life synagogue, Howard Wainer and Richard on Twitter Codebreaker featuring one of Kullback and Leibler’s Feinberg analyze data associated with hate crimes. @ChanceStatsMag. Submit articles at contemporaries, Elizabeth Friedman. In the Visual Revelations column article, “Looking https://bit.ly/39TsMxd. Societal benefits abound from sharing data and at Reported Hate Crimes,” they zero in on statistics machine learning models built on those data, but reported in New Jersey due to its uniqueness in hav- data have the potential to be biased, resulting in ing complete data going as far back as 1990. biased models. What are the possible repercussions? In The Big Picture column article, “The Shape Who should be held accountable? What methods of Things: Topological Data Analysis,” Nicole can be employed to avoid data bias? Charna Parkey Lazar and Hyunnam Ryu demonstrate the use of tackles these questions in “Who Is Accountable for topological techniques for exploratory data analy- Data Bias?” sis. If you work with complex data, this may be a In the February issue of CHANCE, we included technique to add to your toolbox. an article about the connection between road famil- Escape rooms aren’t just a social activity to iarity and traffic accidents. In this issue, authors enjoy with your friends. They can also be used in Adam Palayew, Sam Harper, and James Hanley the classroom to reinforce new concepts. In the consider a different angle on the topic of traffic Taking a Chance in the Classroom column article, accidents. They note the study of factors that influ- “The Data Science Instructional Escape Room: A ence accidents may be complicated by factors such Successful Experiment,” authors Valerie Nelson as the season, day of the week, and time of day of and Jason Crea explain their implementation of an the accident. The authors evaluate various methods escape room for a data science course. n 10 amstat news may 2021
Forget March Madness! Students Test Probability Skills with March Randomness T his year, the ASA’s outreach campaign ThisIsStatistics launched a new contest March Randomness Winners for its annual spring challenge: March Throughout the month, 214 student Randomness, a month-long competition that teams competed with peers around encouraged middle- and high-school students the world. After 16 rounds of 24-hour and college undergraduates to test their proba- bility intuition skills. exercises over four weeks, the winning Inspired by the Borel board game, the inau- teams are as follows: gural March Randomness challenge invited stu- BEST PRE-COLLEGE TEAMS dents to predict outcomes of daily probability experi- FIRST PLACE: A Fraction Ahead, with ments. Every Monday Alaina Smith and Delfina Szigethy of through Thursday Mt. Lebanon High School, Pittsburgh, throughout March, Pennsylvania ThisIsStatistics SECOND PLACE: Elidia R, with Elidia Reyes posed a new experi- ment and asked stu- of Wayne High School, Fort Wayne, Indiana dents to predict the THIRD PLACE: Mathster’s Stupefying 1 outcome. For exam- Stars, with Molly, Miles, Stella, Trevor, ple, “Flip a coin four Roman, Talon, Sam, Patrick, Myles, times; will you get a head Beckley, Timothy, Will, Sarah, Reece, exactly once?” Laurel, Macy, and Sarah of Mariemont To keep things interesting, ThisIsStatistics Junior High School, Fairfax, Ohio also asked the students to back up their predic- tion with a wager from their stash of “StatCoins.” BEST UNDERGRADUATE TEAMS After the betting window closed the next day, FIRST PLACE: a celebrity statistician conducted the challenge experiment and revealed the official outcome on SiegeChauffantDansLeMercobenz, with video. Teams watched their fortunes shrink and Scott McGuigan, Alexis Morel, Clément grow on the leaderboard as they competed with Verdier, and Chloé Le Chevalier of peers in their school or division—or even just Université Grenoble Alpes, against themselves. Grenoble, France Celebrity statisticians included Christoforos SECOND PLACE: Rah Anagnostopoulos from Borel, Lucy D’Agostino McGowan, Ellie Murray, Chris Franklin, Rob Prof. Aliyev!, with Yang Santos, Wendy Martinez, John Bailer, Brittany Hsia, Nicolas Shier, and Greene, Glen Colopy, Kathy Ensor, Roger Zachary Branner of Peng, Jenny Green, Emily Butler, Eunice Virginia Military Institute, Kim, Eric Laber, Claire Bowen, and Ron Lexington City, Virginia Wasserstein. Watch them bring their personal THIRD PLACE: Rambler, style to the probability experiments at https://bit. ly/3wNzYEP. with Charles Hwang of Loyola To learn more about March Randomness, University, Chicago, Illinois visit https://thisisstatistics.org/march-randomness. n may 2021 amstat news 11
MY ASA STORY James Cochran ASSOCIATE DEAN AND PROFESSOR I am the Rogers-Spivey Fellow and Professor of Statistics at the University of Alabama. I am also the associate dean for research for my college. My research spans theory, methodol- ogy, and application. I joined the American Statistical Association soon after joining the faculty at Wright State James Cochran is associate dean for research, Rogers- University (WSU). My primary field of study to Spivey fellow, and professor of statistics at the University that point had been economics, but I was hired to of Alabama. He is also the editor of the Encyclopedia of teach introductory statistics, operations research, Operations Research and Management Science. and calculus to undergraduate business majors and MBA students. society. I also saw how the ASA could serve as a tool Although my education had prepared me well in pursuing my career goals (such as working with for using statistics, I did not feel prepared to teach colleagues in developing nations to resolve societal the subject and I joined the ASA in the hope of issues). I became involved with the ASA in several developing a better understanding of statistical ways: I have served on and chaired various ASA com- concepts and learning approaches to teaching mittees; served as an officer of various ASA sections; them effectively. and been involved with ASA public outreach, scien- tific freedom, human rights, and advocacy efforts. My ASA experiences and the friends I have made through these experiences have been high- lights of my academic career, and selecting a single … the opportunity to help establish Statistics Without experience from these experiences is extremely difficult. However, if pressed, I would select the Borders [is] my most memorable and meaningful opportunity to help establish Statistics Without Borders (SWB) as my most memorable and mean- experience as an ASA member. ingful experience as an ASA member. After organizing workshops on making intro- ductory statistics education more engaging and encouraging the application of statistics to soci- etal issues in developing nations for several years, I I attended several local and regional ASA events began thinking about how to create a mechanism and quickly learned about all the ASA had to offer. for ASA members to get involved in the application After I left WSU to become the director of analytic of statistics to societal issues in developing nations. services for a large marketing research firm and I eventually discussed my ideas with ASA later entered the University of Cincinnati to pur- Executive Director Ron Wasserstein, who put me sue my PhD (in statistics and operations research), in touch with a group of three colleagues—Fritz I maintained my ASA membership. Scheuren, Gary Shapiro, and Steve Pierson (ASA After earning my PhD, I became active in the ASA Director of Science Policy)—who had begun because I saw how the ASA helped its members and working on a similar concept. These colleagues 12 amstat news may 2021
Photo courtesy of James Cochran James Cochran goes on a lion walk in South Africa. quickly invited me to join their effort as a co-chair with Gary. In shortly over a decade, SWB has become inter- nationally recognized and grown to several hun- dred capable and earnest volunteers who take on dozens of projects each year—and has never accept- ed payment or remuneration of any kind (100% of all donations are used for SWB projects and are never used for payments or reimbursements). You can learn more about SWB at the organi- zation’s website: https://swb.wildapricot.org. There, you can find out about past SWB projects, suggest potential clients and projects, and volunteer (you do not have to be an ASA member to join SWB). The work of this organization’s volunteers is uplifting and inspiring; whenever I feel anxious about the myriad political and societal problems we are facing, I visit the SWB website for affirma- tion of the basic goodness of people. In each of the past 36 years, I have reflect- ed on the value of my ASA membership when I receive my notice to renew. And every year, I quickly conclude that my ASA membership is a tremendous bargain. Without any deliberation, I Photo courtesy of James Cochran promptly renew. n James Cochran ‘volunteers’ to participate in a python Sukuma dance. may 2021 amstat news 13
Photo courtesy of The University of Texas Health Science Center at Houston Hadley Wickham, from RStudio, speaks at StatFest 2019. StatFest Back for 21st Year in 2021 F inding people like yourself along the career science. Additionally, StatFest will feature opportu- path you desire can be intimidating during nities for attendees to network with other attendees the first years in higher education, especially and connect with institutions from academia, indus- when the household names in those places do not try, and the nonprofit sector. seem to have paths that match yours. It is common StatFest is an ongoing initiative of the American to be unsure where your vague major fits into grad- Statistical Association through its Committee on uate programs or the job market. However, answers Minorities in Statistics (CoMiS). The committee might be closer than you think. seeks to foster participation in statistics and data StatFest 2021 is a free conference over two half- science by historically under-represented minorities days aimed at encouraging BIPOC (Black, indig- and focuses much of its effort on two key programs: enous, people of color) undergraduates who have StatFest, a pathway program, and the Diversity quantitative interests to pursue careers or graduate Mentoring Program, an early-career success program. studies in the statistical and data sciences. StatFest is StatFest 2020 attendees appreciated the welcom- currently in its 21st year and will be held virtually for ing and informative environment; opportunity to the second year due to the COVID-19 pandemic. hear people’s stories and experiences; and ability to This year’s event will take place September 18–19. connect with graduate program directors, professors, During the conference, students will be inspired and graduate students from many campuses, based by a keynote address highlighting the speaker’s on feedback from the post-event survey. career trajectory and informed by engaging presen- While the conference is free, registration is tations and discussions that explore opportunities in required. Visit the StatFest website at https://bit. statistics and data science across government, aca- ly/2RkcEON for more information and to register. demia, industry, and the nonprofit sector. Students StatFest is made possible through the financial will also have an opportunity to engage in a candid support of the ASA and several academic and indus- Q&A with a graduate student panel that explores trial sponsors. If you have questions about StatFest, the life of a graduate student, navigating the appli- contact StatFest 2021 co-chairs Brittney Bailey cation process, and the steps to thriving as an aspir- at bebailey@amherst.edu or Therri Usher at therri. ing researcher. At the same time, professionals will usher@fda.hhs.gov. If your organization is interest- come together to address issues related to promoting ed in supporting the event, contact Adrian Coles, diversity, equity, and inclusion in statistics and data CoMiS chair, at adrian.coles@lilly.com. n 14 amstat news may 2021
Birth of an ASA Outreach Group: The Origins of JEDI Jana Asher, Slippery Rock University, and Cathy Furlong, Statistics Without Borders R eaders of Amstat News are undoubtedly aware of the myriad communities operating Working Group Members under the umbrella of the American Statistical Association: Brittany Terese Fasy, Leslie McClure, Mark Daniel Ward, Gabriel Huerta, Renee Moore, • Committees – work on initiatives important to the strategic goal of the ASA or the profes- Monica Jackson, David Hunter, Jo Hardin, sional visibility of the ASA Gretchen Martinet, Gretchen Falk, Julia Sharp, Rebecca Nichols (ASA staff ), Donna • Sections – focus on a specific methodology or LaLonde (ASA staff ) application of statistical science • Interest Groups – serve a similar purpose as sections but are less formal In January of 2019, the working group began with • Chapters – serve members of specific regions a broad charge: to identify mechanisms to support diversity initiatives led by ASA members. Two main • Student Chapters – operate within a college initiatives arose out of their early discussion: 1) to or university develop a consortium with other professional societies • Outreach Groups – support a common inter- that would focus on increasing diversity and 2) to cre- est of ASA members that is not well addressed ate a single repository of opportunities—a “one-stop by the formal structure for a section or a shop” for aspiring statisticians from all backgrounds. chapter The consortium idea was a recognition that many existing organizations in mathematics and However, many readers probably do not know statistics and across the scientific fields already had the work involved to establish one of these groups. initiatives related to diversity, equity, and inclusion. The story of the Justice, Equality, Diversity, and They focused original outreach on three primary Inclusion Outreach Group (JEDI) begins with organizations: Math Alliance (mathalliance.org), Karen Kafadar during her year as ASA president- National Association of Mathematicians (NAM, elect in 2018. By tradition, each ASA president www.nam-math.org), and Society for Advancement establishes several presidential initiatives, and of Chicanos/Hispanics and Native Americans Kafadar wished to build her initiatives in the spirit in Science (SACNAS, sacnas.org). At the time, of Barry Nussbaum’s successful presidential initia- SACNAS was going through a leadership change, tive to engage Asian statisticians through Asian but representatives from the Math Alliance and statistical societies. Kafadar knew she wanted to NAM met with the task force in November of 2019 increase diversity within the ASA and asked Julia to begin discussing how to make such a consortium Sharp to lead the initiative in late 2018. a reality. Sharp, who was finishing her term on the ASA The working group was simultaneously explor- Board of Directors in 2019, had been outspoken ing the idea of a repository of information on exist- about her belief that diversity, equity, and inclu- ing initiatives related to diversity, equity, and inclu- sion were important issues within the statistics sion and information of interest to marginalized field. Together, Sharp, ASA Director of Strategic members of the statistics community. The original Initiatives and Outreach Donna LaLonde, ASA idea was that grants, research opportunities, fund- Executive Director Ron Wasserstein, and Kafadar ing for travel, and internships would all be posted in assembled a team that became the Diversity the repository. The members of the working group Working Group. began developing a Google sheet to list links to may 2021 amstat news 15
these resources and exploring how to create a more willing to run in an election for either chair or chair- robust platform to house this “one-stop shop.” elect later that year. At first, she was confused by the All this was progress toward creating a more just request. Sellers was active in several organizations environment in the statistics profession, but the work- pursing diversity-related initiatives and was aware ing group was not designed to be a long-term body of the ASA Anti-Racism Task Force. She requested a within the ASA. In October of 2019, at the Women Zoom meeting to discuss what JEDI was and what in Statistics and Data Science Conference, individuals she could contribute to the effort. interested in diversity, equity, and inclusion met for Sellers, Sharp, and LaLonde met two days a lunch meeting, during which the idea of an ASA later. Sellers requested the chair position to be two Special Interest Group (SIG) specifically focused on years long to allow enough time to create a strong diversity, equity, and inclusion arose. Participants infrastructure and for elections to be postponed at that meeting included Suzanne Thornton, Jack a year. She pointed out that the JEDI Executive Miller, Tami Massie, Saki Kinney, and LaLonde. Committee was trying to organize elections by the By early 2020, the group—which included both end of the calendar year and there wasn’t enough task force members and other interested parties— time to build a good slate of candidates or get the began drafting a charter for the proposed SIG. In word out to the general ASA membership. late spring of 2020, as COVID began shutting down Sharp and LaLonde brought Sellers’s ideas back to travel, they reached out to ASA members through the executive committee and they arranged to meet ASA Connect with a petition to start the SIG and with Sellers in November. After further discussion, discussed the idea with the ASA Council of Sections. Sellers and the executive committee members decid- The ASA Council of Sections suggested they con- ed the executive committee would serve as the inter- sider forming an ASA Outreach Group instead. im leadership for JEDI for 2021, allowing enough At this time, Sharp’s role as leader of the work- time for an election to occur in 2022. Sellers would ing group led to her becoming the interim leader serve as chair and Sharp would serve as past chair for of the new outreach group, which decided it would 2021 and 2022. Members of the executive commit- be called the JEDI Outreach Group (JEDI OG), tee would serve as chairs for five organizing com- to represent justice, equity, diversity, and inclusion. mittees: Programming, Students, Communications, With a revised charter and a petition with about 80 Liaison, and Professional Development. signatures, the group petitioned the ASA Board of In early 2021, JEDI went “live” by asking every- Directors for official status and was approved dur- one who had expressed interest or been nominated ing the July 2020 meeting. for a leadership role to become part of the leader- The work of setting up the outreach group was ship team—that is, become members of these five just beginning. Those who had worked to make the committees that would develop the structured activ- JEDI OG a reality knew they would need much ities of the outreach group. As a result, about 40 more assistance to make the group an effective agent individuals are serving either as part of the interim for change. They became the executive committee executive committee or on one of the five organiz- of JEDI and scheduled several meetings to discuss ing committees. In the past several months, each how the leadership of the group would be struc- of the organizing committees has met several times tured and refine the charter and goals of the group. and is on the way to accomplishing its goals. Following the 2020 Women in Statistics and The program committee, chaired by Gretchen Data Science Conference, the group’s first public Martinet, determines programming such as webi- meeting occurred, during which they announced nars that will involve and support JEDI OG mem- they were soliciting nominations for leadership bers and the broader community. JEDI is already positions. They then sent out a Google form to listed among the sponsoring organizations for con- everyone who expressed interest in JEDI through tributed abstracts at the Joint Statistical Meetings signing the petition or attending the meeting fol- in 2021. JEDI will also host a panel discussion at lowing WSDS to allow nominations to be made. JSM titled, “Justice, Equity, Diversity, and Inclusion At this point in the story, Kimberly Sellers, a (JEDI) in the ASA.” The program committee is professor at Georgetown University, entered the starting to plan an invited session for JSM 2022. picture. She was unknowingly nominated for chair On a smaller scale, the program committee is dis- of JEDI, so she was surprised when she received an cussing the organization of an ongoing contributed email on October 14, 2020, asking if she would be presentation series with annual local meetings. 16 amstat news may 2021
You can also read