R. W. Doerge, M. Fuentes, S. C. Morton, H. J. Newton, S. G. Pantula, and A. R. Sampson
Increasingly in academia, statisticians have been appointed to university administrative positions. Pioneers include W. Allen Wallis, president of the University of Rochester, 1962–1970; Albert Bowker, chancellor at the University of California, Berkeley, 1971–1980; Gertrude Cox, director of the Institute of Statistics at the Consolidated University of North Carolina, 1945–1960; and Lowell Reed, president of The Johns Hopkins University, 1953–1956.
During the past decade, there has been an acceleration of this process. These more recently appointed academic administrators who are also statisticians can offer valuable perspectives about the current dynamic atmosphere for statistics. With this motivation, an invited panel discussion session was organized for JSM 2017 that included five distinguished deans who are renowned statisticians in their own right: Rebecca Doerge, Montserrat Fuentes, Sally Morton, H. Joseph Newton, and Sastry Pantula. The breadth of their academic experiences gave them excellent perspective on statistics and the challenges now facing statistics departments.
Panelists: The participants and their affiliations (as of August 3, 2017) in the discussion included the following (from left): Rebecca Doerge, Dean of the Mellon College of Science at Carnegie-Mellon University and earlier Head of Statistics at Purdue University; Montserrat (Montse) Fuentes, Dean of the College of Humanities and Sciences at Virginia Commonwealth University and prior to that Head of Statistics at North Carolina State University; Sally Morton, Dean of the College of Science at Virginia Tech and previously Chair of Biostatistics at the University of Pittsburgh; H. Joseph (Joe) Newton, Dean Emeritus of the College of Science at Texas A&M University and earlier Head of Statistics at Texas A&M; Sastry Pantula, Dean of the College of Science at Oregon State University and earlier Head of Statistics at North Carolina State University
The goal for the discussion was to provide insights for statisticians in the “trenches.” Allan Sampson of the University of Pittsburgh was the session organizer and moderator. To help focus the discussion, four broad areas were delineated:
- Based upon their experiences as deans, what advice would the panelists offer to individuals in statistics departments who are dealing with the many challenges facing statistics departments in our current environment (e.g., managing growth, developing interdisciplinary programs (particularly with regards to data science), marketing and publicizing departments, and developing and mentoring personnel)?
- How do the panelists’ academic backgrounds impact their relationships with the statistics departments in their setting?
- What advice would the panelists as deans wish they could give their past selves as department administrators that would have been helpful to know at that time?
- What advice would the panelist give to academic statisticians who would like to pursue a track in academic leadership, in particular, building an appropriate portfolio of experiences? Also, what are the panelists’ views about addressing the difficulties in balancing leadership responsibilities with both research and personal lives?
While the panelists explored some of these areas in depth, other areas were addressed more tangentially. What follows is a summary of their discussion with a few additional observations.
The panelists broadly observed that successful university leaders have great communication skills and are flexible in their outlook and actions. Importantly, they have a passion for what they do. Their collective view was that being a department chair or head was one of the hardest jobs on campus, but also among the best jobs for making a real impact. Their advice for department chairs, as well as other leaders, is to think one level above where they are. Faculty, for instance, might consider the various concerns of the chair when proposing an idea or request.
More specific observations were that chairs need to get along with other chairs and deans. Chairs need, for example, to make “deals” and build collaborative enterprises that are mutually beneficial to multiple university units. And chairs are recommended to avoid unnecessary participation in politics. Chairs are part of the dean’s leadership team and not just the department’s advocate. It was suggested a chair holds an unhelpful viewpoint if they find themselves saying, “I fought with the dean and here’s what I got.”
It is important for a chair to be able to make decisions. To be an effective decision-maker, panelists’ advice included “don’t perseverate too much,” “be as transparent and ethical as possible,” “be clear about the criteria for the decision,” and “be collaborative.” Making hard decisions with faculty, students, and staff is a skill chairs must master because decisions are often made hard by difficult people. Some panelists thought chairs shouldn’t necessarily assume “the university has your back” in all situations. The panel agreed among themselves that deans should empower chairs and faculty to, as one panelist noted, “figure it out.”
It was strongly recommended that leaders of departments and programs should learn about the business/financial model used at their university. Growing a department requires knowing how money is allocated and being creative within the given financial environment. It is important to understand the broad higher education landscape with particular attention paid to current demographics and technology change—older returning students and increased online education, for example. Panelists noted that as deans and provosts order ativan mastercard change at a university, policies and priorities will change, so chairs need to adapt accordingly.
In approaching deans about issues and concerns, the panelists recommended chairs should understand the constraints facing a dean and aspire to make the dean’s life easier. A way to accomplish this is to base solutions to problems upon evidence and, as a panelist observed, possibly offer three solutions to an issue—one that is free, one that is expensive, and one in between. Also suggested when scheduling a meeting with the dean is sending a written agenda and, ideally, a one-page executive summary in advance.
In regard to fundraising, for both deans and chairs, the panelists agreed philanthropy is part of the job. For deans, it is not unusual for one-third of the dean’s time to be devoted to fundraising. The panel strongly suggested it is best to focus on the donor’s passions and interests, not the chair’s, when a chair is talking to prospective donors. And as one panelist said, “It is always important to listen!”
To obtain funding from industry, a chair should be proactive in approaching a possible company by having the company identify in advance what their needs are so the statistics department can be appropriately responsive with the right faculty presenting. One panelist used this approach with local industry and noted it would often lead to research contracts. Caution was expressed about not spreading a department’s resources too thinly, as making promises that can’t be sustained is counterproductive for the department in the long run.
The panel uniformly advocated it is absolutely essential for statistics departments to be involved in the data science era. Even if the department’s sentiment is that it is too late to do so, the panel urged it is never too late and the department should just start to get involved and not necessarily start big. Moreover, they said that for a statistics department not to be part of data science disadvantages the department’s students. To facilitate statistics departments’ involvement in data science, the panelists recommended that having joint appointments both sets the stage and advances the relationships. To further these relationships, the panel agreed it is imperative that a statistics department’s faculty venture out from their offices and spend time interacting with other departments’ faculties.
Within their universities and the broader community where they are located, the panelists believed statistics departments need to expend much more effort to “market themselves.” Many colleges have good publicity people, and it should not be seen as “bragging” for a statistics department “to tell its story.”
Several panelists urged statisticians to do more than “academic papers” and work on research that “matters” and make sure it’s known. One panel member suggested a statistics chair should “empower” the more extraverted faculty to promote the department and that not every faculty member need “get onboard” to do this. Another offered that chairs should always have available three promotional “bullets,” one for departmental research, one for teaching, and one for service and outreach.
An obvious suggestion, but one that is often missed, is for departmental faculty, staff, and students to include in their self-identification that they are a “member of the department of statistics” in any professional public setting where they are introducing themselves.
Prior to discussing some of the challenges faculty face in taking on administrative responsibilities, several panelists noted two specific challenges statistics chairs and statisticians in general face in an academic environment. One is the need to “constantly explain” why statisticians should have different promotion criteria than most of the physical and social sciences. Motivating this concern is that accomplished statisticians have portfolios with both basic statistical research and high-level collaborations in a variety of settings, including with researchers, policy makers, and governmental organizations. The other challenge identified was the sense that when statistics is part of a university’s science division, there are some in the sciences who hold a view that statisticians are not “really scientists.” To counter these ideas, panelists remarked that having more national statistics awards would be excellent.
For academics taking administrative positions, the panel thought some of their colleagues criticized them for abandoning their departments when they should have been supportive of their “being an advocate of their discipline while serving in a broader leadership position.” Further disconcerting to the panel was the “perception that faculty, especially women, go into administration in mid-career because their research wasn’t successful.” Contrary to this misperception, the panel collectively agreed that, as an administrator, one should maintain one’s research career. Not only is this satisfying, it also makes it less difficult to deal with highly accomplished faculty in negotiating tough issues.
To have more statisticians move into administration, the panel reiterated that the discipline should support fellow statisticians who focus on administration, thereby benefitting statistics and other disciplines more broadly.
Overall, the panelists agreed that being a department chair or dean is both rewarding and personally gratifying. They concluded by strongly encouraging the audience to make use of their statistical backgrounds to be leaders in serving their departments and the broader academic community.
Leave a Reply