December DEI Journal Club – Differential retention contributes to racial/ethnic disparity in U.S. academia
For December, Eli McDonald presented Differential retention contributes to racial/ethnic disparity in U.S. academia for the DEI journal club. The paper sought to elaborate the well known but poorly understood “leaky pipeline”, a phrase describing the prevalent underrepresentation of women and minorities at higher level academic positions. The group hypothesized that poor retention rates contributes to this lack of representation, and that these rates vary between groups, resulting in barriers at specific stages. To explore this, the group generated a model using simple differential equations to express rates of change at the transition to different academic levels. Data was taken from the NSF Science and Engineering Indicators, the NSF WMPD Annual Reports, and the NCSES Graduate Student and Post-doc survey. Using basic assumptions about exit rates, the group compared expected retention rates to the actual number of individuals retained from different ethnic groups.
The group found that Asians are over-represented at all levels, while black students are massively underrepresented across the board. The rate of loss steepened from undergraduate to the professoriate, with fewer and fewer African Americans retained than expected as one climbs the academic ladder. Hispanic populations were similarly underrepresented, while white populations were overrepresented at all levels, but especially at the level of tenured professorship. With this information in hand, the group argued that quantification of retention reveals specific, actionable information that institutions could focus on at different hiring stages.
The journal club discussion noted that it was positive to see that undergraduate to graduate transitions meet expectations across the board, indicating that recruitment at the base of the academic ladder is doing something right. A possible solution would then be to apply the tactics for graduate school recruitment. However, it was pointed out that professorships can’t use the same marketing materials because they are fundamentally two different positions. Also, the problem of the leaky pipelines is fundamentally one of retention, not recruitment. It’s not enough to get people in the door – they have to want to stay.
The discussion further expressed interest in learning exactly what factors influence whether or not someone decides to stay. Several possibilities were brought up. As some people put it, “We can get them in the door, but their experiences once they’re in aren’t the same.” Different treatment by other members, inherent biases, and other relational factors may get in the way of retaining diverse talent. Outside factors may also have differential impacts on different groups, such as location and cultural expectations. Different proprieties may make it harder for someone to relocate far away from family, especially if there are elderly family members in the picture who need care. These priorities may also come into conflict with the way science is rewarded, as working long hours or weekends to generate seminal results may not be feasible based on family or community expectations.
These considerations led the discussion to ask, “Do people leave because they want to, because they can’t get the job in this space, or because of external factors?” The truth is, it’s hard to know, but it would be well worth it to find out. Because the more we can support people of all ages, races, and backgrounds, the better science will become.
Leave a Response