In the past two issues of Amstat News, I have focused on building the ASA Leadership Institute. This month, I want to highlight another 2018 presidential initiative—expert witness training. The idea for such a program came from our membership.
Early in 2017, Executive Director Ron Wasserstein heard from several members about whether the ASA could help prepare consulting statisticians for service as expert witnesses in a trial or deposition. Around the same time, I had occasion to talk with a former University of North Carolina biostatistics student, Naomi Brownstein, and she described her interest in being an expert witness. Now a statistics faculty member at Florida State University, Naomi had been approached about serving in this capacity. So, Ron and I put our heads together to assess the need for this kind of training and came up with a proposal.
There are areas of the law that involve quantitative expertise. Ensuring there are qualified statistical professionals in the courtroom or otherwise involved with the legal process in those areas would improve the quality of the legal process and increase recognition of the important contributions of statistics and statisticians.
There are many leaders in our field who regularly step up to serve the courts on a variety of important topics, and our sister fields—such as mathematics—are also stepping up to contribute. Gerrymandering of legislative districts, for instance, is one topic that has drawn attention of late. Gerrymandering struck a chord with me, living in two states (Maryland and North Carolina) that have been accused of extreme partisan gerrymandering—but in opposite political directions. Other topics include investigations of Medicaid fraud by medical providers and using “risk-limiting audits” to detect problems with elections. These topics often end up in court, and statisticians should be prepared to chip in.
One aspect of the FDA’s mission to protect and promote the public health that I did not fully appreciate until working there was the need to ensure product claims made in labeling and advertising were accurate and did not mislead the public. Evaluating the evidence to determine whether product statements are misleading is the subject of FDA guidance documents, but these determinations often end up in court, where a company’s first amendment rights in making claims about a product are weighed against the agency’s need to ensure such claims do not mislead the public.
A recent example is the Federal Trade Commission’s suit against Quincy Bioscience that questioned the use of secondary analysis to support a claim about a treatment for memory loss after the trial failed on its primary endpoints. The FTC lost the claim, and no statisticians were involved as expert witnesses due to the nature of the suit, but statistical theory about inference in general and multiplicity in particular formed the basis of the FTC’s argument.
Given the interest in training among ASA members, requests for ASA’s assistance as expert witnesses from other parties, and my own interest in this topic, Ron and I presented our proposal for an expert witness training program to the board, and thus was born a presidential initiative.
Our vision is to develop a program that provides the general skills and knowledge a statistician should have to be an expert witness, as well as prepare participants to speak as an expert on at least one subject matter area. To this end, we formed a working group made up of leading statisticians Joseph L. Gastwirth, Mary W. Gray, Nicholas P. Jewell, and Rochelle E. Tractenberg and asked Kathy Ensor of Rice University to be the chair. The assembled team includes a lawyer and statisticians with courtroom experience.
I asked Kathy, as chair, to report on the deliberations of the working group to date, and here is what she had to say:
Leaders in our field have often provided expertise to the courts and Congress, many learning by experience. The objective of this training program for statistical expert witnesses is to help our community, especially those new to our profession, expedite the learning curve on how to best serve the courts as an expert statistician.
There were exciting suggestions for what should and should not be included in a training workshop. Although our experiences varied, several common themes emerged, including the following:
- What it means to be an expert witness
- What it means to be an expert statistical witness
- Voicing a clear unbiased statistical opinion at all stages of the legal process
- Ethical considerations as practicing professional statisticians
- Common mistakes and pitfalls to avoid
The role our profession has played in the courts throughout history is laudable. The role of the expert statistician emerged strongly in the 1980s. As a young statistician, I recall reading with great enthusiasm the text “Statistics and the Law” by [Morris] DeGroot, [Stephen] Fienberg, and [Joseph] Kadane and then later Jay Kadane’s 2008 book, “Statistics in the Law: A Practitioner’s Guide, Cases, and Materials.” I guess this speaks to my love of statistics and its application, as I read the books with the deep immersion great novels require.
The vast array of areas in which statisticians interact with the courts simply boggles the mind—areas such as employment discrimination, DNA, medical practice, environmental issues, patent challenges, economic risk, and financial fraud.
The recent creation of the National Institute of Standards and Technology–supported Center for Statistics and Applications in Forensic Evidence (CSAFE) recognizes the important role statisticians play in forensic science and hence the legal system. The expert witness working group also noted emerging areas that include statistical and machine learning algorithms potentially guiding decisions of the courts and the criminal justice system. Equally as broad as the societal issues statisticians are asked to address are the areas of statistics in which statisticians serve as experts. And given the demand for our expertise, we as a committee were reminded that one important component to serving as an expert is knowing when to decline a request.
We are developing a training program with the following key goals in mind:
- Quality – Develop a program that provides excellent training and meets, or even exceeds, the needs of our members
- Impact – Develop a program that can reach a substantial number of people over time
- Sustainability – Develop a program that can be offered regularly and support itself financially
An RFP for training program development will be announced, pending approval by the ASA Board. The general goal is to begin the program either in the fall of 2018 or spring of 2019. Once developed, this program will become part of the ASA Leadership Institute, offered at a frequency deemed helpful to our community.
My thanks go to Kathy and her team for the work accomplished so far, and I look forward to seeing this program roll out in the coming months. I believe the program will be a valuable resource to ASA members, and we have certainly heard from several who are anxiously awaiting its inception. This is yet another example of an area in which strong statistical leadership can have an impact that extends far beyond our membership.
So, here’s to statisticians leading with justice for all!
Department of Mathematics and Statistics, Georgetown University
The George Washington University, DC: PhD, Statistics (2001)
University of Maryland, College Park: MA, Mathematics (1998)
University of Maryland, College Park: BS, Mathematics (1994)
Kimberly Sellers is a statistician and associate professor of mathematics and statistics specializing in statistics at Georgetown University in Washington, DC, and a principal researcher with the Center for Statistical Research and Methodology Division of the US Census Bureau. A DC-area native, she completed her BS and MA degrees in mathematics at the University of Maryland College Park. During her graduate studies in mathematics at Maryland, she became inspired to study statistics, thanks to instruction by Professor Piotr Mikulski. After completing her MA degree, she earned her PhD in mathematical statistics at The George Washington University, partly through support as a Gates Millennium Scholar (one of the inaugural cohort recipients).
Sellers held faculty positions at Carnegie Mellon University as a visiting assistant professor of statistics and the University of Pennsylvania School of Medicine as an assistant professor of biostatistics. She was also a senior scholar at the Center for Clinical Epidemiology and Biostatistics before her return to the DC area.
Sellers’ research areas of interest and expertise are in generalized statistical methods involving count data that contain data dispersion and image analysis techniques, particularly low-level analyses including preprocessing, normalization, feature detection, and alignment. Her primary research centers on the Conway-Maxwell-Poisson distribution. Sellers is the leading expert on this distribution, having developed various statistical methods associated with distribution theory, generalized regression models, control chart theory, multivariate distributions and analysis, and stochastic processes for count data expressing data dispersion. Her single-authored and collaborative works have been published in prominent journals, and she is a nationally and internationally invited speaker.
Sellers is likewise recognized both nationally and internationally for her service activities. She is an associate editor for the Journal of Computational and Graphical Statistics and The American Statistician. She is also an active contributor to efforts to diversify the fields of mathematical and statistical sciences, both with respect to gender and race/ethnicity. Last but not least, she is the 2017–2018 chair for the American Statistical Association’s Committee on Women in Statistics and an advisory board member for the Black Doctoral Network.
Who are you, and what is your statistics position?
My name is Claire Bowen, and I am a statistics PhD candidate in the applied and computational mathematics and statistics department at the University of Notre Dame.Tell us about what you like to do for fun when you are not being a statistician.
When I’m not being a statistician, I participate in endurance races such as marathons and triathlons.What drew you to this hobby, and what keeps you interested?
During high school, I was overweight and could barely run a mile. Originally, I joined the cross country team as the student manager and started running to keep up with the team better. Through the encouragement of the team and coach, I changed to a runner by the end of the season.
Since then, I discovered I enjoy long-distance races. I completed my first marathon when I was 19 and my first half-Ironman when I was 24. My goal is to complete an Ironman before I am 30.
I continue this hobby to help maintain a healthy lifestyle and for the overall positivity I’ve experienced from running and triathlon communities. Being overweight before, my training keeps my physical and mental health in check more easily. I love that I can do other physical activities like hiking and snowboarding without being exhausted afterward. My stress has gone down considerably, and training breaks up my day when I’m stuck on a research problem (or sometimes I figure out the problem during a run).
I joined the local biking and triathlon clubs, where I have the opportunity to meet people in different social groups. In these endurance race communities, everyone is encouraging and positive about the sport they’re in, because it doesn’t matter how you perform in the races. What matters is that you are swimming/biking/running and you are doing it for you, becoming the best you can be.
Vice President, Data Science, Talkspace
Columbia University (Office of Naval Research Graduate Fellow): PhD, Statistics (1991)
Baylor University: BSc, Mathematics (1985)
Bonnie Ray currently leads data science activities at Talkspace, a NYC-based startup that enables improved mental health for all by providing an affordable, accessible, and secure platform for messaging-based psychotherapy. Previously, she led the data science team at Arena, a Baltimore-based startup focused on analytics-driven hiring assessments. From 2001–2015, she held positions of increasing responsibility at IBM Research, where her role immediately prior to moving to the start-up world was director of cognitive algorithms and she led teams charged with developing machine learning methods for next-generation AI applications.
A graduate of Columbia University and Baylor University, Bonnie started her career in academia as a post-doctoral fellow at the Naval Postgraduate School working with the late Professor Peter Lewis on his study of sea surface temperatures. She went on to hold assistant and associate professor faculty positions in the applied mathematics department at the New Jersey Institute of Technology, during which time she received three National Science Foundation awards, including a CAREER grant awarded to promising young scientists that funded her educational and research activities related to environmental time series analysis.
Bonnie was born and raised in the Deep South, living first in Mississippi and then in northern Louisiana before attending college in the heart of Texas. She always had a love of mathematics, but was introduced to statistics as a junior in college and knew almost immediately that a continuing study of pure mathematics was not to be. Summer internships at Texas Instruments after her junior and senior years helped her appreciate the importance of mathematics to address business challenges and the power of computing to obtain efficient solutions, both of which continue to serve as career touchstones.
Bonnie has published more than 60 refereed papers, holds 12+ patents, and is an elected Fellow of the American Statistical Association. In her spare time, she loves to swim, watch independent films, and relax with a good book.
Director, Mathematical Statistics Research Center, Office of Survey Methods Research, Bureau of Labor Statistics
Joint Program in Survey Methodology Certificate in Survey Statistics (2015)
George Mason University: PhD, Computational Sciences and Informatics (Computational Statistics area) (1995)
The George Washington University: MS, NASA Langley Research Center, Aerospace Engineering (1991)
Cameron University: BS, Physics and Mathematics (1989)
Wendy Martinez was born and raised in Detroit, Michigan. After high school, she served as an active-duty member of the US Army Signal Corps, where she had the opportunity to be stationed in Germany for several years.
Martinez has been serving as the director of the Mathematical Statistics Research Center at the Bureau of Labor Statistics (BLS) for six years. Prior to this, she worked in several research positions throughout the Department of Defense. She held the position of science and technology program officer at the Office of Naval Research, where she established a research portfolio comprised of academia and industry performers developing data science products for the future Navy and Marine Corps.
Her areas of interest include computational statistics, exploratory data analysis, and text data mining. She is the lead author of three books on MATLAB and statistics. These books cover topics ranging from classical approaches in statistics to computationally intensive methods and exploratory data analysis. She became interested in data science when pursuing her PhD under Edward Wegman (GMU), who had founded a new curriculum in computational statistics that included most of the courses in what is now considered data science.
Martinez was elected a Fellow of the American Statistical Association in 2006 and is an elected member of the International Statistical Institute. She was recently honored by the American Statistical Association when she received the ASA Founders Award at the 2017 Joint Statistical Meetings.
Human Rights Data Analysis Group
Rice University: BA, Statistics and Mathematics (2006)
Duke University: PhD, Statistical Science (2010)
Kristian Lum is the lead statistician at the Human Rights Data Analysis Group (HRDAG). Previously, she worked as a data scientist at a small technology startup and was a research assistant professor in the Virginia Bioinformatics Institute at Virginia Tech.
Lum’s research primarily focuses on examining the uses of machine learning in the criminal justice system, developing new statistical methods that explicitly incorporate fairness considerations, and advancing HRDAG’s core statistical methodology—record-linkage and capture-recapture methods for estimating the number of undocumented conflict casualties. Although statistics and other quantitative disciplines are not the typical path to social impact, Lum is drawn to applications in which she can use her knowledge of statistics to address important societal problems and amplify the voices of marginalized groups and individuals.
In her work on statistical issues in criminal justice, Lum has studied uses of predictive policing—machine learning models to predict who will commit future crime or where it will occur. In her work, she has demonstrated that if the training data encodes historical patterns of racially disparate enforcement, predictions from software trained with this data will reinforce and—in some cases—amplify this bias. She also currently works on statistical issues related to criminal “risk assessment” models used to inform judicial decision-making. As part of this thread, she has developed statistical methods for removing sensitive information from training data, guaranteeing “fair” predictions with respect to sensitive variables such as race and gender. Lum is active in the fairness, accountability, and transparency (FAT) community and serves on the steering committee of FAT, a conference that brings together researchers and practitioners interested in fairness, accountability, and transparency in socio-technical systems.
Lum’s work on record linkage and capture-recapture focuses on solving problems that arise in HRDAG’s conflict casualty estimation work. In capture-recapture, she has proposed methods to obtain reliable casualty estimates, even when some victims are unlikely to be recorded. She is also the primary developer of the DGA package for R, which implements a popular Bayesian method for capture-recapture.
Lum is originally from Auburn, California—a small town in the foothills of the Sierra Nevada mountains. For as long as she can remember, she has enjoyed math and logic puzzles. Her interest in math took off in high school when she took calculus from an excellent professor at the local community college, Sierra College. For the first time, she felt she was learning why math worked, rather than how to manually implement an algorithm to solve an equation. In college, she took a few statistics courses as electives for her math degree and grew interested in discovering patterns in data that explain real-world phenomena. This led her to graduate school in statistics, where she first became involved with HRDAG after “cold emailing” the organization’s founder. She has been focused on applying statistics to important problems in human rights ever since.
The American Statistical Association and Council of the American Mathematical Society (AMS) issued a joint statement to inform discussions and planning around the drawing of voting districts as we approach the 2020 census. This marks the first time in recent history the two organizations have issued a joint statement of broad interest to the American public.
The statement is organized around the following three facts:
- Existing requirements for districts generally do not prevent partisan gerrymandering.
- It has become easier to design district plans that strongly favor a particular partisan outcome.
- Modern mathematical, statistical, and computing methods can be used to identify district plans that give one of the parties an unfair advantage in elections.
“While these points may be common knowledge in some circles, it’s important they be stated by objective and respected authorities like the AMS and the ASA and for them to be more widely known in the redistricting discussions around the 2020 Census,” noted 2018 ASA President Lisa LaVange.
AMS President Ken Ribet said, “Our community is poised to play a central role in ongoing discussions about methods for creating voting districts and the evaluation of existing and proposed district maps. It has been a pleasure for me to observe the recent explosion in interest in this topic among colleagues and students in mathematics and statistics. I anticipate that the new statement by the ASA and AMS Council will lead to increasing transparency in the evaluation of districting methods.”
“Statistical and mathematical standards and methods can be very helpful to inform decision-makers and the public about partisan gerrymandering,” remarked the statement’s main architect, Jerry Reiter, chair of the ASA Scientific and Public Affairs Advisory Committee. “The statement acknowledges the value of partisan asymmetry as a standard, and it highlights some methods for measuring partisan asymmetry. The statement does not endorse any one method, as ultimately this issue is determined by policymakers and the courts.”
In issuing the statement, the two societies also offer to connect decision-makers and policymakers with mathematical and statistical experts.
More than 350 statisticians, methodologists, and health policy experts gathered January 10–12, 2018, at the Marriott Hotel in Charleston, South Carolina, for the 12th International Conference on Health Policy Statistics (ICHPS). This was a record-breaking year in terms of number of conference abstracts and attendance.
ICHPS is held every two years, jointly sponsored by the ASA and Health Policy Statistics Section (HPSS). Conference co-chairs Laura Lee Johnson (US Food and Drug Administration) and Bonnie Ghosh-Dastidar (RAND Corporation) were supported by a 30-member organizing committee, including two student representatives and past conference chairs.
The theme—Health <-> Statistical Science <-> Care, Policy, Outcomes—reflected the interactive relationship between health services and outcomes research and innovative statistical methodology to facilitate informed discussions regarding health reform and other efforts to improve health care in the United States.
ICHPS 2018 was supported by grant number R13HS025884 from the Agency for Healthcare Research and Quality and by the Patient-Centered Outcomes Research Institute (PCORI) through a Eugene Washington Engagement Award. Project officers took an active role, challenging attendees to find effective mechanisms for disseminating results so they translate into actual policy and practice.
Presentations and posters covered a range of topics, including fraud detection methods, causal inference and treatment heterogeneity, real-world evidence, pragmatic clinical trials, and comparative effectiveness. The Alan Alda Center for Communicating Science conducted a workshop using improvisational theater techniques developed to help people speak more vividly and expressively. Wednesday included a career panel and networking lunch for students. Thursday afternoon included town halls and roundtable discussions to allow for idea sharing and informal networking. Town halls were on global real-world data, the VA, engaging with community partners, Medicaid payment reform, and health care delivery system transformation.
Networking dinners and meet-ups afforded opportunities to mingle in a relaxed environment while enjoying Charleston’s hospitality and delicious food during the local restaurant week.
Workshops included sequences on causal inference, complex survey analysis, and patient-reported outcomes, plus workshops on social network analysis and an introduction to several data sets from the US government, including the Medical Expenditure Panel Survey, National Health and Nutrition Examination Survey, and Medicare Beneficiary Survey.
Rousing addresses were delivered by Robert Califf, “Evidence Generation in the Era of Ubiquitous Information,” and Suchi Saria, “A Methodologist’s Quest to Improve Health Care.” Each offered advice on opportunities to redesign the health care delivery system in an evidence-driven way with incentives. Both emphasized high-risk, interdisciplinary problems of real interest requiring scalable interventions.
Conference proceedings will be published in Health Services and Outcomes Research Methodology.
HPSS presented its Long-Term Excellence Award to Sally C. Morton (Virginia Tech University) and Paul Rosenbaum (University of Pennsylvania). Anirban Basu (University of Washington) received the Mid-Career Award.
ICHPS also provided 21 student travel awards, supported by grants, the ASA’s Biopharmaceutical Section, and the ASA’s Mental Health Statistics Section. Conference activities were supported by grants, awards, and multiple industry and institutional partners, including AbbVie, Amplexor, Pfizer, and Research Triangle Institute (RTI).
Dean, Mellon College of Science, Carnegie Mellon Univeristy
Cornell University: Post-Doctoral Scholar (1993–1995)
North Carolina State University: PhD, Statistics (1993)
University of Utah: MS, Mathematics (1988)
University of Utah: BS, Mathematics (1986)
Rebecca Doerge is dean of the Mellon College of Science at Carnegie Mellon University. Prior to joining both the department of statistics and department of biology at Carnegie Mellon , she was the Trent and Judith Anderson Distinguished Professor of Statistics at Purdue University. Doerge joined Purdue University in 1995 and held a joint appointment between the colleges of agriculture (department of agronomy) and science (department of statistics).
Doerge’s research is focused on statistical bioinformatics, a component of bioinformatics that brings together many scientific disciplines to ask, answer, and disseminate biologically interesting information in the quest to understand the ultimate function of DNA and epigenomic associations.
Doerge is the recipient of the Teaching for Tomorrow Award, Purdue University, 1996; University Scholar Award, Purdue University, 2001-06; and Provost’s Award for Outstanding Graduate Faculty Mentor, Purdue University, 2010. She is an elected Fellow of the American Statistical Association (2007), an elected Fellow of the American Association for the Advancement of Science (2007), and a Fellow of the Committee on Institutional Cooperation (2009). She has published more than 120 scientific articles and two books, as well as graduate 23 PhD students.
Doerge was born and raised in upstate New York. As a first-generation student, she studied theoretical mathematics at the University of Utah. It was there she gained interest and experience in both computing and human genetics (Howard Hughes Medical Institute). She earned her PhD in statistics from North Carolina State University under the direction of Bruce Weir and was a postdoctoral fellow with Gary Churchill at Cornell University.
Doerge is a member of the Board of Trustees for both the National Institute of Statistical Sciences and the Mathematical Biosciences Institute. She is a member of the Engineering External Review Committee at Lawrence Livermore National Laboratory and the Global Open-Source Breeding Informatics Initiative (GOBII) Advisory Board.
I know it is a little late, but Happy New Year!
I’ve been a SPES member for at least 20 years, and I still didn’t realize how active SPES is until I took on the role of chair this year. I am taking this opportunity to remind you of the many activities SPES sponsors and invite you to participate in at least one this year.
Networking is vitally important in any career, and you can meet your colleagues in SPES at three conferences. First up is the 2018 Joint Research Conference on Statistics in Quality, Industry, and Technology. This is a joint meeting of the 25th Spring Research Conference on Statistics in Industry, which SPES jointly sponsors with the Institute of Mathematical Statistics, and the 35th Quality and Productivity Conference, which is sponsored by the ASA’s Section on Quality and Productivity (Q&P). The joint conference is being held June 11–14 in Santa Fe, New Mexico. This is a small regional conference that provides a great opportunity for new researches to get to know the SPES and Q&P communities. Send questions by email.
The annual Joint Statistical Meetings (JSM) will be held in Vancouver, BC, Canada, this year from July 28 to August 2. SPES will host a variety of invited and contributed sessions during which you can learn about new statistical methods and applications to a variety of physical and engineering sciences. And don’t forget to attend the business meeting/mixer, which SPES usually cosponsors with other sections.
The last SPES-sponsored conference of the year is the 2018 Fall Technical Conference being held October 3–5 in West Palm Beach, Florida. This conference is cosponsored by the ASQ Statistics Division and the ASA’s SPES and Q&P sections. The theme is “Statistics and Quality: Riding the Big Data Wave.”
A continuing SPES program is the Marquardt Memorial Industrial Speakers Program, which is funded by a generous endowment made by Margaret Marquardt in memory of her late husband, Donald Marquardt, who was an ASA Fellow and former ASA president. The program’s objective is to familiarize students with the role of statisticians in industry, an application area to which students often are not exposed. The program seeks to fill this gap by bringing experienced industrial statisticians to campus to talk directly with students about their work and industrial experiences. If you would like to have a speaker visit your campus or if you would like to tell the world about life as an industrial statistician, contact the program’s chair, Vaneeta Grover.
We are also interested in new initiatives to provide services to our members. Currently, the ASA is promoting an initiative to improve mentoring in the statistical profession. If you have an interest in mentoring and would be willing to help SPES develop a mentoring program, send me a note.
SPES is open to any type of career development programs you think would be useful. If you have an idea, please contact me or any of the officers to see what we can do to help implement it. Also, if you want to add a leadership role to your résumé, please consider running for a SPES office. The jobs don’t require an excessive amount of time and you get the chance to meet some interesting and enthusiastic colleagues.
President’s Chair in Statistics (2016-)
Distinguished Professor of Liberal Arts and Sciences (2010-)
Director, Center for Statistics and Applications in Forensic Evidence (2015-)
Professor of Statistics, Iowa State University (2000–2009)
Associate Provost, Iowa State University (2000–2004)
Associate Professor, Iowa State University (1996–2000)
Assistant Professor, Iowa State University (1990–1995)
Iowa State University: PhD, Statistics/Animal Genetics (1989)
Iowa State University: MSc, Statistics (1986)
University of Illinois, Urbana: MSc, Animal Genetics (1985)
Universidad de la Republica: BS, Agricultural Engineering (1982)
Alicia Carriquiry was born in Montevideo, Uruguay, in 1957. She earned a BS in agricultural engineering from the Universidad de la Republica (at the time, the only university in Uruguay) in 1982. She truly hated her first job, and this motivated her to pursue a graduate degree in an area of interest to her. The opportunity arose soon after she graduated from college, when a professor from the University of Illinois who was visiting Uruguay offered her an RA in his animal breeding program.
Carriquiry completed an MSc in animal breeding and genetics in Urbana-Champaign in 1985. Also that year, a professor from statistics at Iowa State University approached her with an offer to join his group. Carriquiry moved to Ames and earned her doctoral degree in 1989. In Ames, she also met her husband of 30 years, Wolfgang Kliemann, who was an assistant professor in mathematics at the time. So that ISU would not lose him, Carriquiry was hired in 1990 for a tenure-track assistant professor position, half in statistics and half in agricultural economics. She was promoted to associate professor with tenure in 1996 (and moved to statistics full time), to professor in 2000, and to distinguished professor in 2010. Carriquiry was the first woman to become full professor in statistics at Iowa State and the first Latina/o to become distinguished professor at ISU. Between 2000 and 2004, Carriquiry was associate provost at ISU, in charge of research.
Carriquiry’s research has always focused on Bayesian methods and their application in various disciplines. She has worked extensively with human nutritionists on problems as diverse as survey design to measurement error modeling and density estimation. More recently, Carriquiry was principal investigator on a large award from the National Institute of Standards and Technology that helped establish the Center for Statistics and Applications in Forensic Evidence, which she directs. Carriquiry has been a prolific writer (more than 130 peer-reviewed papers) and has raised more than $35 million in research funding. Her proudest achievement is the 20 doctoral students who have already completed (or are in the process of completing) their doctoral work under her mentorship.
Carriquiry is an elected member of the National Academy of Medicine and a Fellow of the American Association for the Advancement of Science, American Statistical Association, Institute of Mathematical Statistics (IMS), International Society for Bayesian Analysis (ISBA), and International Statistical Institute. She was a member of the board of trustees of the National Institute of Statistical Sciences, president of ISBA, vice-president of the ASA, and executive secretary of IMS. She has participated in multiple national and international panels and committees and currently advises governments in Asia, South America, and North America.
Managing Director at BlackRock where she leads Data Science together with Sherry Marcus.
Columbia University: PhD, Statistics (2009)
New York University: Master’s Degree, Mathematics (2003)
Stanford University: Master’s Degree, Engineering-Economic Systems and Operations Research (1999)
University of Michigan: Bachelor’s Degree, Honors Mathematics (1997)
Rachel Schutt was the Chief Data Scientist of News Corp where she oversaw the company-wide data strategy as an executive on the senior technology leadership team. There she established the company’s first data science team for Dow Jones, the Wall Street Journal, and other brands. Schutt was named a World Economic Forum Young Global Leader in 2015 and is on the 2014 Crain’s New York Business 40 under 40 list.
She has also been at the forefront of data science education. While working at Google Research (2009–2012), she recognized an emerging skill set (hybrid software engineer-statistician) was required. This skill set was not being taught in universities because it spanned traditional departmental lines and represented a new space of engineering, computational, and statistical problems coming out of technology companies. So to train the next generation of data scientists, Schutt proposed, designed, and taught the first Introduction to Data Science course at Columbia University (and one of the first such courses in the country). This course became the basis for the book she co-authored with Cathy O’Neil, Doing Data Science, published in 2013. Other university curricula now reflect the initial structure and content of her course. She is a founding member of the Education Committee for the Data Science Institute at Columbia.
While at Google Research, Schutt was part of the machine learning group in New York and holds patents based on her work in social networks, large data sets, experimental design, and machine learning.
Schutt is on the advisory board for Harvard’s Institute for Applied Computational Science (IACS).
She was born in Boston, Massachusetts, in 1976 and grew up in Cambridge, England, and Princeton, New Jersey. She was interested in math starting at a young age and found solving problems and puzzles fun and calming. When she was five years old, her father realized the girls at her primary school were being directed to practice knitting while the boys solved math problems. He intervened and came to the class to teach set theory to the girls. Schutt’s interest in math persisted throughout college, where she studied theoretical math. She worked for several years in a variety of jobs, and then—after having a chance conversation with Andrew Gelman—she returned to earn her PhD in statistics. Gelman became her adviser at Columbia along with Regina Dolgoarshynnikh. Schutt’s thesis work was on the spread of contagious processes in networks.
Louisiana State University, Baton Rouge: BS, Mathematics (1980)
The University of North Carolina at Chapel Hill: PhD, Statistics (1989)
Susan Murphy is a professor of statistics and computer science at the Harvard John A. Paulson School of Engineering and Applied Sciences and Radcliffe Alumnae Professor at the Radcliffe Institute, Harvard University. Prior to joining Harvard in September 2017, she was the H.E. Robbins Distinguished University Professor of Statistics, professor of psychiatry, and research professor at the Institute for Social Research—all at the University of Michigan.
Susan’s present research focuses on causal inference and sequential decision-making. She works on both data analysis and design of experiments to inform the sequencing of treatments, as well as how online algorithms can be used both in experimental designs and in treatment design in mobile health.
Susan is a member of the National Academy of Sciences and National Academy of Medicine, both of the US National Academies. In 2013, she was awarded a MacArthur Fellowship for her work on experimental designs to inform sequential decision-making. She is a Fellow of the College on Problems in Drug Dependence, a member of the International Statistical Institute, a Fellow of the American Statistical Association (2000), and a Fellow of the Institute of Mathematical Statistics. She is also a former co-editor of the Annals of Statistics and delivered the IMS Wald Lectures in 2015.
Susan grew up in southern Louisiana and, as an adolescent and college student, always found mathematics to be beautiful and accessible. While attending Louisiana State University, she realized mathematics could be a career (!) and, better yet, she could use mathematics to improve our society via the field of statistics. Since then, she has never looked back!
Co-Founder and Chief Technology Officer, Data Collaboratory
Carnegie Mellon University: PhD, Statistics (2005)
Carnegie Mellon University: MS, Statistics (1996)
Harvard University: AB, Applied Math (1995)
Michelle Dunn is CTO and co-founder of Data Collaboratory, a technology company that builds data science tools. Data Collaboratory’s flagship product is GRANTED!, a tool that uses data science to guide researchers to the most appropriate funding opportunities. GRANTED! builds on Michelle’s years of experience in federal funding at the National Institutes of Health (NIH).
She started working at the National Cancer Institute (NCI), a part of the NIH, in 2009 as a program director who advocated for the funding of grants to develop statistical methodology. In this position, she advised applicants from across the country about appropriate funding opportunity announcements. GRANTED! automates this function by scraping FOAs from publicly available sources and using text-mining techniques to match them with a researcher’s interests.
Next to building a data science company from the ground up, Michelle is proudest of her contributions to developing and leading the NIH Big Data to Knowledge (BD2K) Initiative. She was instrumental in conceiving and implementing a collection of programs aimed at nurturing a biomedical workforce capable of analyzing data and developing new methods for analyzing data. The BD2K Initiative provided funding for statisticians, informaticians, and other biomedical scientists to do research and receive training in data science.
Michelle grew up in Memphis, Tennessee, and enjoyed math from an early age. She first learned about statistics on the drive from Memphis to Cambridge, Massachusetts, for the start of classes at Harvard. All freshman had to take a quantitative reasoning test based on a booklet containing essentially a “Stat 101” course. After reading that booklet, Michelle realized why math is important. At Harvard, she was fortunate to find professors and graduate students who encouraged her to pursue her studies in statistics. From Harvard, she went to Carnegie Mellon University (CMU) for a master’s, worked in government for a while, and then returned to CMU for a PhD.
Michelle’s thesis adviser, Jay Kadane, was supportive not only of her choice of dissertation topic, but also of her choice to be a stay-at-home mom following the completion of her dissertation. This choice could have ended her statistical career had it not been for an open-minded statistician and hiring manager at NCI, Brenda Edwards. Brenda strove to create family-friendly working conditions because she had witnessed the hardships women and working mothers had endured during her career.
Brenda is Michelle’s hero; she is a statistician who is dedicated to making the world a better place, not just through her work measuring the burden of cancer, but through her leadership of people and projects. Leadership requires taking calculated risks, making sometimes unpopular decisions, and communicating a vision and road map for concerted action. Michelle’s goal is to continue to put into action what she has learned about leadership from Brenda.