Committee on Career Development Holds Virtual Office Hours with Experts

Claire McKay Bowen is lead data scientist, privacy and data security, at the Urban Institute, where she researches methods of data privacy and confidentiality.

Elizabeth Mannshardt is the associate director of the Information Access and Analytic Services Division of the US Environmental Protection Agency (EPA). She is also an adjunct associate professor in the department of statistics at North Carolina State University and past chair of the ASA’s Government Statistics Section. Prior to joining the EPA, Mannshardt completed two postdoctoral fellowships in statistics.

The ASA Committee on Career Development (CCD) recently held its first two rounds of virtual office hours (VOH) for 2020, focusing on postdoctoral positions and featuring statisticians who were once postdocs along with professors with advice and/or open positions at their institutions. Questions focused on the interview process, finding a postdoc mentor, pros and cons of choosing the postdoc path, and how and where to find postdoc positions.

Murali Haran of Penn State, Ofer Harel of the University of Connecticut, and Yawen Guan of the University of Nebraska were guests during the first round, with Elizabeth Mannshardt of the US Environmental Protection Agency serving as moderator. The second round showcased Brian Reich of North Carolina State University (NCSU), Veronica Berrocal of the University of California, Irvine, and Howard Chang of Emory University. Claire McKay Bowen of the Urban Institute moderated that round.

More than 30 participants submitted questions in advance and live during the session. Recordings for both sessions and panelist bios are available online, along with information about future CCD offerings.

The sessions started with panelist introductions and initial thoughts. The general theme was to have a plan for your research path and corresponding papers, as well as to branch out with your areas of interest and professional connections. Haran said, “Have a plan, one that hopefully includes easy papers to get out.” Reich agreed. “Time is short and papers are crucial, so start off with a few concrete projects,” he said.

Haran also suggested participants expand beyond their current mentors. Harel echoed this when he said, “Connections are so important.” He also emphasized that postdocs should aim to have fun.

Guan agreed and said, “Come up with a detailed research plan and timeline with [your] postdoc mentor. Early postdoc is a good time to broaden research areas, start new projects, and establish your own brand.”

Chang emphasized searching for a good research fit, stating this as the reason he chose his postdoc position at the Statistical and Applied Mathematical Sciences Institute (SAMSI).

Reasons for selecting particular postdoc positions were varied. Reich chose his postdoc at NCSU for the chance to work with a great mentor and the opportunity to teach. To work with an exciting data set was a key reason Harel chose his postdoc at the University of Washington. Both Guan and Haran saw their postdocs at SAMSI and the National Institute of Statistical Science, respectively, as a chance to gain experience prior to a faculty position. Berrocal also pointed out a key reason for choosing her postdoc were the other high-level institutions in the area that would offer additional opportunities for learning and collaboration.

Identifying a Good Fit and Mentor

A postdoc interview and preceding conversations will help determine what a good fit is for you and your career. Harel says it is a chance for you to “gauge the postdoc position’s ability to make an impact and give input.”

Guan suggested trying to determine what percentage of your time would be allocated to postdoc work versus establishing your own research and finishing your existing projects, encouraging “new but not completely new projects.” Reich agreed and added, “Sometimes great ideas come from taking ideas that are well developed in one area and applying them to solve an open problem in another.”

Chang says to investigate what the project entails, including deliverables, and to consider how this may fit with your ultimate career goals.

Mentoring is a crucial component of a postdoc position. All the panelists recommended talking to other students and postdocs and asking questions during the interview process. Reich encourages asking about the work environment the adviser envisions.

The guests also offered frank insights into potential drawbacks of a postdoc position. Haran pointed out that academics “might treat you as if you [have] more experience than you do—a postdoc is ‘two’ years, but given the academic calendar, it may really be 16 months.” Chang commented that it can be intimidating to teach 100 students. Berrocal stressed the importance of gaining grant-writing experience. All former postdocs commented that postdocs can be isolating, stressing the importance of networking and creating a social network.

Where, When, and How

There were several questions about finding a postdoc position. Reich pointed people to online resources such as the University of Florida website for jobs in statistics and encouraged students to talk with their dissertation advisers and ask around their field. Chang encouraged students to email research groups they might be interested in. “You never know who has ‘half’ a postdoc position,” he said. As for when, Reich said it is never too early, and funding is variable each year.

Berrocal encouraged becoming a member of the ASA, along with sections and chapters, and pointed out the NSF often has postdoc funding.

Student participants asked how they can be competitive in their own postdoc search. The question of the ‘magic’ number of needed papers came up. The answer: There is no magic number. Haran offered guidelines to consider such as “maybe two submitted or somewhere in the pipeline to show that you can produce work.”

Chang also encouraged having some form of completed research to show, such as a couple of papers and professional presentations.

Reich agreed it is a competitive market and, when asked what stands out in an applicant, he said he looks for students who are smart and ambitious. Chang echoed this, including “ability to work independently” as a key trait.

Should You Do a Postdoc?

When considering the benefits of a postdoc versus going straight into a faculty position, Haran stressed both paths work. “There are lots of personal considerations and lots of randomness year to year in [the] job market,” he said.

A postdoc is a chance to accumulate research ideas and collaborations. Harel pointed out that when you start a faculty position, the tenure clock starts to tick, so you should think about where you want to be in 10 years.

Experts pointed to components of their postdoc that helped them prepare for their faculty positions, such as how to identify important projects, work with PhD students, multitask, write, be a professional researcher, and see the big picture.

Postdoc positions can be valuable for future careers in industry and government. Moderators Mannshardt and Bowen have both completed postdocs and are currently in the government and nonprofit sectors, respectively. Bowen found completing her postdoc was an opportunity to explore research areas beyond her dissertation expertise, which prepared her well for her current position. Mannshardt pointed out that you have a short window of opportunity to do a postdoc. “I personally made the decision as it kept options open and was a great opportunity to learn more about both statistics and my personal path,” she said.

Postdoc Experts

Murali Haran, professor of statistics and department head at Penn State University, completed a postdoc at the National Institute of Statistical Science, where he worked on random forests for applications in software engineering research. Haran also spent time as a research fellow at SAMSI in 2009 for the program “Space-Time Analysis for Environmental Mapping, Epidemiology, and Climate Change” and has continued to collaborate with SAMSI fellows. Haran’s Penn State research has involved statistical methods for studying climate change, modeling the dynamics of infectious diseases, and studying the spread of invasive plant species. 

Ofer Harel, professor of statistics and director of graduate admissions at the University of Connecticut, is a former University of Washington biostatistics postdoc. He earned his doctorate in statistics in 2003 from Penn State, where he developed his methodological expertise in missing data techniques, diagnostic tests, longitudinal studies, Bayesian methods, sampling techniques, mixture models, latent class analysis, and statistical consulting. Harel has served as a biostatistical consultant nationally and internationally since 1997. Through his collaborative consulting, he has been involved with a variety of research fields, including Alzheimer’s, diabetes, nutrition, HIV/AIDS, and alcohol and drug abuse prevention.

Yawen Guan, assistant professor in the University of Nebraska at Lincoln Department of Statistics, is a former SAMSI postdoc.Prior to joining UNL, she spent two years at SAMSI/NC State University as a postdoctoral fellow developing spatiotemporal methods for air quality and emulation-calibration methods for Artic sea ice. 

 

Brian Reich, distinguished professor of statistics at North Carolina State University, is a former NCSU postdoc. Reich completed his PhD in biostatistics in 2005 at the University of Minnesota. After graduation, he joined NCSU first as a postdoc and then as a member of the statistics faculty in 2008. His research interests include spatial statistics, extreme value analysis, variable selection, and dimension reduction. Reich applies these methods to environmental areas such as ecology, air pollution, and climate change, as well as data from the physical and materials sciences. 

Veronica Berrocal, associate professor of statistics in the University of California, Irvine Department of Statistics, is a former SAMSI postdoc. She joined UC Irvine in the fall of 2019, after spending nine years on the faculty at the University of Michigan in the department of biostatistics. Berrocal earned her PhD in statistics from the University of Washington in 2007 and spent three years in the Research Triangle Park area as a National Research Council postdoc and Duke University/SAMSI postdoc. Her research interests are in spatial and spatio-temporal statistics, with a focus on applications in atmospheric/geophysical sciences, environmental health, social determinants of health, health disparities, and imaging.

Howard Chang, associate professor of biostatistics and bioinformatics at Emory University, is a former SAMSI postdoc. He joined Emory in 2011 after completing a two-year postdoctoral fellowship. Chang works on developing statistical methods for the analysis of complex spatial-temporal environmental exposure and health data.