This is really cool. The announcement comes from Joe Cummins:
The University of California at Riverside is hiring 4 open rank positions in Design-Based Statistical Inference in the Social Sciences. I [Cummins] think this is a really exciting opportunity for researchers doing all kinds of applied social science statistical work, especially work that cuts across traditional disciplinary boundaries.
Relevant disciplines include, but are not limited to, Business, Economics, Education, Medicine (Epidemiology and Public Health), Political Science, Public Policy, Sociology, and Statistics/Biostatistics. We seek candidates who excel at developing, testing, and applying cutting-edge research designs and statistical methods for causal identification. Successful candidates might make theoretical and methodological contributions to causal inference, develop novel experimental designs, conduct Bayesian meta-analysis, program evaluation, applied econometrics, or political methodology, and will show an interest to work across traditional disciplines and ability to attract extramural funds. Review of the applications will begin January 8, 2016 and will continue until the position is filled. Senior application materials should be submitted to https://aprecruit.ucr.edu/apply/JPF00468. Junior application materials should be submitted to https://aprecruit.ucr.edu/apply/JPF00469.
UCR is embarking on a major new hiring initiative (http://clusterhiring.ucr.edu) that will add 300 tenured and tenure-track faculty in 33 cross-disciplinary areas and invest in research infrastructure to support their work. This initiative will build critical mass in vital and emerging fields of scholarship, foster truly cross-disciplinary work and further diversify the faculty at one of America’s most diverse research universities.
Here’s the job posting.
Any guess as to why a campus might want multiple hires specifically in “design-based statistical inference”, particularly when they are interested in such a wide range of fields?
Without putting words in the mouth of the administration or the people involved in designing the cluster, I think part of it has to do with the perceived needs of the general UCR expansion. UCR is actively trying to develop the capability and infrastructure to be at the leading edge of cross-disciplinary research, and this is likely to require the use and/or development of new statistical methods (new questions, new statistical problems). One of the aims of this cluster is to be part of the “infrastructure” (in the broadest sense) that enables that sort of interaction between faculty from different areas.
Speaking as a citizen of academia and not as an official UCR mouthpiece: it is sometimes very hard for an applied microeconomist to talk statistics with an epidemiologist or research physician. But I think that the potential for methods from some social science fields to be useful (and under-employed) in other fields is quite high (and I don’t just mean econometrics taking over).
The expansion needs to applied to the other UCs. The UC medical school at UCSF seems to lack instruction in statistics and UC Merced has no statistics department. A 4 yr degree in Math nor Stat is available. Not a research university.
I do appreciate Joe’s comment’s above and the posting about UCR. My sense is that it will be a lot harder to achieve the integration between methods and disciplines to support the cross-disciplinary work that is intended. I think Joe actually references this when he said “… it is sometimes very hard for an applied microeconomist to talk statistics with an epidemiologist or research physician….” I have had a chance to see some of this up close. At UCLA we have the Institute for Pure and Applied Mathematics that tries to bring together scholars from various disciplines like biology, chemistry, medicine, etc., with scholars from mathematics and statistics. In many of these conferences, the scholars from math and the particular discipline don’t even have a common language with which to speak with each other. Larry Wasserman once gave a talk at one of these sessions that was focused purely on translating Machine Learning words from Computer Science into Statistics words–for the math people to understand. But in other cases, I have actually seen out right antagonism as some math folks tried to use new approaches model phenomena in some other substantive field. I was frankly surprised at just how much resistance there could be to math folks trying to contribute to substantive disciplines. This kinda had a chilling effect for junior faculty and grad students (like me) who would like to see greater integration.
I think that one approach that might facilitate some of this integration follows the model of the massive online math collaborations, like the polymath blog. In that case, individuals from across professional and amateur math collaborate on large unsolved problems. I hope that we can spread this model to other subject areas where prolonged interdisciplinary collaborations can emerge.
I think my point is just that hiring faculty is only the first step in a longer process of supporting greater interdisciplinary research.