Home » Additional Features, News and Announcements

Department Chairs Gather at ASA to Discuss Pressing Education Issues

1 November 2016 1,466 views No Comment

With support from a National Science Foundation (NSF) award, the ASA hosted a chairs workshop at its headquarters in July. Attended by more than 40 department chairs from more than 25 states, the workshop featured themes on workforce demands and needs, data science education and research, mentoring and junior faculty, and federal funding.

Organizing Committee

David Hunter, Penn State
Varghese George, Augusta University
Xihong Lin, Harvard School of Public Health
Sally Morton, Virginia Tech University
Sarah Nusser, Iowa State University
Jean Opsomer, Colorado State University
Bruno Sanso, University of California, Santa Cruz
Arnold Stromberg, University of Kentucky

Intended to complement the annual half-day workshop at the Joint Statistical Meetings, the two-day workshop was organized for participants to have more time for discussion and for chairs to speak with each other and take advantage of the many outside speakers in the Washington, DC, region. Indeed, speakers included BLS Commissioner Erica Groshen and four leaders from various business sectors, four NIH and NSF officials, two leaders from the statistical community speaking about mentoring and junior faculty issues, and various data science research and education speakers.

In the session on workforce demands and needs, the four industry leaders affirmed that the demand for statisticians continues to be strong though the titles have evolved to include data scientist, quantitative analyst, and others. They also spoke highly of the unique contributions statisticians bring to the workplace, including the ability to handle complicated data structures and quantify and communicate uncertainty. In addition, they uniformly emphasized the importance of today’s statisticians being able to do the following:

  • Communicate well, including the ability to explain and interpret technical results and understand business/scientific context of analysis
  • Work effectively as part of a group/team with diverse backgrounds
  • Program and use modern collaboration 
software
More than 40 department chairs from more than 25 states attended a workshop at the ASA headquarters.

First row: Merlise Clyde, Allan Sampson, Paul Stephenson, Karen Messer, Sarah Nusser
2nd Row: Asaph Young Chun, Kendra Schmid, Magdalena Niewiadomska-Bugaj, Xihong Lin, Leslie McClure, Jeff Buzas, Madhuri Mulekar, Hao Zhang, Yazhen Wang, Olusegun George, Varghese George, Galin Jones, Shunpu Zhang, Xinping Cui
3rd Row: Ron Wasserstein, Yinglei Lai, Donna LaLonde, John Fu, Randall Pruim, Chris Malone, Lynne Stokes, Elizabeth Garrett-Mayer, Jessica Utts, Sarah Abramowitz, Donald White, Chris Wiggins, Erik Erhardt, Dave Hunter
4th Row: Steve Pierson, Daniel Mundfrom, Arny Stromberg, Sally Morton, Jean Opsomer, Ron Fricker, John Bailer, William Rosenberger, Bruno Sanso, George Luta

When asked about the challenges of educating for both breadth (e.g., communication, teamwork) and depth, the industry panelists emphasized the importance of both and noted the desirability of the “T-shaped” researcher, where the stem of the “T” represents the depth of knowledge in statistics and the top bar represents, among other things, the ability to integrate and apply that knowledge to a broader context/setting. It was also noted how the data science era had expanded the “T-shaped” researcher into a “Pi-shaped” researcher, where the second stem represents depth in a domain specialty.

Emphasizing the importance of being able to work in teams, Erik Andrejko said it was in the multidisciplinary team settings that the “magic happens.”

In addition to wearing his Westat hat, David Morganstein also spoke from his perspective as the 2015 ASA president, sharing the ASA’s efforts to convey the importance of communication, teamwork, and leadership skills.

The data science education panel was perhaps the most diverse, with a computer science dean, the NY Times chief data scientist, the founder of a business analytics master’s program, a statistician serving as university administrator, and a statistician leading an undergraduate data science program. The need to understand employer needs was a strong theme, as was the need to encourage collaboration across disciplines.

Michael Rappa spoke to the advantage of designing courses from scratch. Andrew Moore spoke passionately about the importance of computer scientists working together with statisticians, emphasizing that he views Carnegie Mellon’s successful data science initiatives to be the result of “faculty solving big problems together.” He also discussed the importance of looking at the data science life cycle, from data creation/capture/collection and data processing to analysis/modeling and decision making to data preservation/sharing/protection.

Chris Malone provided insights into the challenges of moving from a course-based perspective to a skills and knowledge perspective in the development of an undergraduate data science program. He also mentioned their Midwest Undergraduate Data Analytics Competition to provide experience for students and connect them with potential employers. His take-away message was “undergraduates can do data science.”

To make the case for the important role statisticians can play in enabling open science and open data sharing, Sarah Nusser described the myriad issues involved with open science. We are in the era when data sets must be made more accessible, she noted, yet their proliferation and, in many cases, sheer size call for smarter digital curation systems. Such systems require effective communication of the information shared, she added, and this hinges on good statistical design and documentation, starting at the beginning of a research project. One place for statisticians to engage is to become familiar with the move toward open data and share good practices in designing and documenting studies via courses and collaborations between students and researchers.

In the junior faculty and mentoring session, both Genevera Allen and Sally Morton described the changing academic environment in which junior faculty members are becoming more interdisciplinary, team-based research more prevalent, and professional outlets more diverse. Software, nontraditional journals, and online presence were among the examples cited of elements that will increasingly comprise promotion dossiers in the future. In terms of mentoring junior faculty, both speakers made the case for a long-term perspective that values an impactful career, which might include the earning of tenure and promotions, but only as milestones along a lifelong path. The benefits of mentors from inside and outside the department, and the number of mentors and where help is most useful (e.g., networking, grant writing) were also discussed.

During the funding agency panel, Michael Lauer spoke to the NIH reproducibility efforts of the offices and the many places more statistical capacity is critical. Michelle Dunn described the creation of the NIH data science efforts and its current activities and offerings. Michael Vogelius and Chaitan Baru spoke to ongoing statistics and data science programs in their respective NSF directorates and an emerging collaboration on data science, the details of which should become public in the coming year.

The panels included a large number of nonstatisticians by design. Indeed, the format of the workshop, based on short presentations from panelists and copious time for questions and discussion, ensured that the conversation went in both directions by allowing the department chairs to communicate their thoughts on many issues to the diverse group of panelists.

Workshop participants praised the organizers for the value and enjoyment of the speakers and discussions. Making this workshop a periodic event was highly recommended and the officers of the Caucus of Academic Representatives will discuss whether it will be annual or biannual and how to synchronize it with the half-day JSM workshop.

Read the full agenda. The organizing committee is working on a whitepaper summarizing workshop take-aways and expects to distribute it during the academic year.

Workshop Themes and Speakers

Professional Workforce: Demand and Skills

Erica Groshen, Commissioner, Bureau of Labor Statistics
Erik Andrejko, Vice President of Science, Head of Data Science, Monsanto/The Climate Corporation
David Morganstein, Vice President and Director of Statistical Staff, Westat
Christopher Peterson, Vice President of Data Science and Assistant Chief Model Risk Officer, Capital One
Aarti Shah, Senior Vice President and Chief Information Officer, Eli Lilly and Company

Federal Research Funding

Chaitan Baru, Senior Adviser for Data Science, NSF Computer and Information Science and Engineering (CISE) Directorate
Michelle Dunn, Senior Adviser for Data Science Training, Diversity, and Outreach, NIH Office of the Associate Director for Data Science
Michael Lauer, Director of Extramural Research, NIH
Michael Vogelius, NSF Division of Mathematical Sciences (DMS), Mathematical and Physical Sciences (MPS) Directorate

Driving the Conversation: Education in the Age of Data Science

Andrew Moore, Dean, School of Computer Science, Carnegie Mellon University
Chris Wiggins, Chief Data Scientist, The New York Times; Columbia University
Michael Rappa, Institute for Advanced Analytics, North Carolina State University
Chris Malone, Department of Mathematics and Statistics, Winona State University
Sarah Nusser, Vice President for Research, Iowa State University

Supporting and Mentoring Junior Faculty

Genevera Allen, Rice University
Sally Morton, Dean, College of Science, Virginia Tech University

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading...

Comments are closed.