Haley Hedlin earned her PhD from The Johns Hopkins University Department of Biostatistics in 2011 after graduating with a BA in mathematics (concentrations in statistics and linguistics) from St. Olaf College in 2006. Her research interests include developing statistical methods for neuroscience data.
Postdoctoral training is becoming more common in statistics; however, postdoctoral positions with a significant statistics education component are rarer. One such position at the Five College Consortium combines research at a large research university and teaching at liberal arts colleges.
The Five College Consortium consists of four liberal arts colleges (Amherst, Hampshire, Mount Holyoke, and Smith) and a large research university (University of Massachusetts Amherst) within a 15-mile radius in western Massachusetts. The consortium is home to the Five College Statistics Program, a program created to “coordinate and integrate resources in order to better serve our statistics students” .
The close geographic proximity of the campuses and the consortium’s tight-knit statistics community provide a unique opportunity for recently minted statistics PhDs to explore careers at small and large academic institutions in a postdoctoral program.
Jeffrey Stratton and I are the first two postdoctoral associates to hold a position in this program, headed by principal investigator Michael Lavine and funded by a National Science Foundation (NSF) grant. A search is under way for a third postdoctoral associate (the application and more information can be found at mathjobs.org).
In addition to funding postdoctoral statisticians, the NSF grant supports a graduate student in statistics at the University of Massachusetts Amherst who provides statistical consulting services for undergraduate researchers and faculty during weekly visits to each of the liberal arts colleges. Both prongs of the grant benefit the colleges by injecting additional statistical expertise into their communities. Early career statisticians benefit greatly from experiencing different types of academic departments. By providing a taste of the life of a statistician at a liberal arts college, the grant serves to better inform the career choices of the pre-and postdoctoral statisticians funded on the grant.
The responsibilities of the postdoctoral associates are divided between the typical postdoctoral role of developing their research programs and, less typical of statistics postdocs, activities related to undergraduate statistics education. The educational component consists of teaching one statistics class each semester at a liberal arts college.
Teaching a single course each semester allows time to develop course materials, and teaching different courses allows me to create a corpus of course materials to use in the future. During my first semester, I co-taught an introduction to statistics and probability course with Katherine Halvorsen, a senior statistics faculty member at Smith College. Co-teaching a course with such an experienced educator provided invaluable insights into teaching.
During my second semester at Smith, I taught a statistical literacy course to students with a wide range of interests and majors, but limited mathematical background, with support from the other statisticians in the consortium. This fall, I am teaching an applied linear regression course at Mount Holyoke College. By the end of my postdoc, I will have taught at women’s colleges (Smith and Mount Holyoke), coeducational colleges (Amherst and Hampshire), and a nontraditional college in which students design their own majors and grades are written reports on their progress (Hampshire).
An entirely new experience I encountered during my first year was mentoring students enrolled in a statistics project seminar course. Master’s-level students, along with a few advanced undergraduates, were matched with one or two other students, a statistics adviser, and a researcher with a project requiring an advanced statistical method such as hierarchical linear models, classification and regression trees, or kernel smoothing. As lead statistics adviser of a group, I led students through the literature review, learning the statistical method required, applying the method, and presenting their findings. This experience taught me how to guide students learning a new method outside of the classroom, communicate what is expected of the students, and guide someone through a research project from start to finish.
In addition to the statistics education opportunities, a postdoctoral position that rotates between academic departments allows me to peek into several departments and get a sense of life at a variety of academic institutions. I am receiving mentoring by senior and junior statisticians at large and small institutions alike. I know of no other postdoctoral position that combines research at a large university and teaching at liberal arts colleges and consider myself fortunate to have found myself in this unique position.