In this lecture, I will discuss some of the challenges and opportunities facing graduate education in statistics and biostatistics, in light of the very high demand for data science across all fields, and both in and out of academia. I will touch on some of the following topics:
(1) Most PhD programs in statistics and biostatistics require extensive mathematical pre-requisites, and extensively cover statistical theory. Such training is certainly very useful for a career spent as faculty in an academic setting. However, there is very high demand for well-trained (bio)statisticians outside of academia, and increasingly, many of our PhD alumni do not pursue academic positions after graduation. Does this suggest the need to broaden the scope (and perhaps decrease the depth) of PhD training? (2) There is increasing interest among undergraduate majors from a wide variety of fields in graduate work in (bio)statistics, and an increasing number of universities are beginning to offer undergraduate degrees in statistics and data science. However, these students may not be well-suited for the extensive mathematical and theoretical training in many (bio)statistics PhD curricula. Such a disconnect between undergraduate training and PhD pre-requisites is bizarre: for instance, an undergraduate chemistry major is well-prepared for a PhD in chemistry; why isn't the same true in statistics? (3) Finally, if we broaden the scope (bio)statistical PhD training, then what will PhD dissertation research look like for these non-classically-trained PhD students, and how will this research advance the careers of the students' faculty mentors? I will ask some of these questions, and will suggest some of my own answers.