We like to argue over the lunch hour. Not in the sense that anyone gets upset or angry, but in the more positive sense of “argue”: presenting and debating the evidence and merits behind different points of view. Since we both enjoy trying to see things from multiple perspectives, whatever position either of us first proposes about the topic of the day, they are almost certain to have a ready opponent. One recent topic of debate was especially thorny: equity and diversity targets in the Canada Research Chairs (CRC) program.
The CRC diversity targets represent just one example of many efforts across the global scientific community to reduce and hopefully eliminate discrimination based on gender, race, sexual orientation, or any other feature of an individual that is irrelevant to their ability to contribute to the scientific enterprise. A recent column by a well-known rabble-rouser at the Globe and Mail took aim at this policy (provided here), and sci-Twitter and others took issue with the column (for instance here, here and here). And so the topic came up at lunchtime. Despite vigorous rebuttals of the premises of the original column, the motivating questions seem to have gone unaddressed: are quantitative targets a good way to tackle equity and diversity issues, and if so, what are appropriate targets? We were thus motivated to write this post to find out what people think.
Equity and diversity targets have a contentious history in higher education. Just last year, one American university had to deny using targets when they were sued for discrimination on the basis of race by Asian American applicants. In contrast, frustration in many quarters over the slowness of progress* has increased support for this type of measure: one article has argued that any proven systemic and self-reproducing biases in hiring procedures provides grounds for diversity targets, with some following through in a big way.
What brought the CRC equity and diversity targets into the news recently was the announcement that the nature of the targets was going to change. Before describing the nature of these changes, it’s important to note that the targets – however defined – are consequential. From the CRC website:
“Institutions must establish equity and diversity targets using the methodology below to ensure individuals from the four designated groups (women, Indigenous peoples, persons with disabilities and members of visible minorities) participate in the program. The program monitors the institutions’ progress toward meeting their established targets. An institution’s failure to participate in the target-setting exercise will result in the suspension of payments.”
No words strike fear into university administrators more than “suspension of payments”, so the incentive to reach the targets is very, very strong. Indeed there have already been increases in the proportional representation of women, visible minorities, people with disabilities, and indigenous people in the few years since targets were introduced. But how to define the targets?
When initially introduced the idea was to use the “availability approach”, under which availability “is determined by estimating the representation of a designated group within the pool of potential nominees”…“Active university researchers represent the pool of potential nominees”. In the specific case of the CRC program, active researchers have been defined as anyone participating in competitions for grants from one of the major Canadian funding agencies. The recently announced change redefined the targets by representation in the general population. So, if the target for women was formerly 21% (as it was for the Natural Sciences and Engineering Research Council**), now it will be ~50%. This is a huge difference, and one that seems worth discussing (with civility).
To be clear, we are not questioning whether excellence and diversity are conflicting goals (they are not), and we fully agree that measures are needed to counter unequal challenges and barriers facing different groups of people in science. That is, we are not aiming to stimulate debate on whether there is a problem to begin with (there is). Rather, we are hoping to stimulate discussion that goes beyond the question of being for or against action in general, focusing on legitimate questions about whether particular actions will help achieve stated goals and how precisely to go about setting targets.
Any specific action or policy comes with pros and cons (relative to alternatives). We can imagine a number of different reasons to justify using a specific set of targets. The CRC program’s initial targets (described above) had both supporters and detractors, the latter arguing that the targets create discrimination against individuals from overrepresented groups. In contrast, expanding the “availability” pool to the “qualified” pool (e.g., all Ph.D. holders in the relevant discipline) could address one important step of the leaky pipeline by targeting groups who, having earned PhDs, tend not to pursue academic careers. Using a still broader population, a set of targets might seek to optimize mentorship benefits – e.g., all students enrolled in science would reflect the population likely to most benefit from CRC mentors. Finally, as newly announced, a set of targets based on the entire population reflects the diversity makeup of anyone who could potentially want to pursue both higher education and academic research.
So we’re inviting you to have lunch with us: we’d love to have your thoughts, or at least your standardized and anonymized opinion. Our questions were sparked by thinking about CRCs, but we have phrased them so as to apply to any type of prestigious research position. Once we have data we will make them available to all, and subsequently write some thoughts on the results.
** CRCs are awarded through three funding agencies, NSERC, CIHR (Canadian Institutes of Health Research), and SSHRC (Social Sciences and Humanities Research Council)