Yesterday I wrote the first part of my analysis of the following “op-ed” that was published in Cell. The point of the paper is clear from its title. Click on the screenshot to read it.
Yesterday I addressed the disparity in funding rates between black and white investigators submitting grants to the National Institutes of Health (NIH)—a disparity that is quite substantial (black “principal investigators”, or “PIs”, are funded at only 55% of the rate of white ones). The article above both implicitly and explicitly attributes this disparity to racism: structural racism in science as a whole, racism in the National Institutes of Health, and racism of those people who review and score the grants and decide which ones deserve funding.
Yet the only relevant study addressing the issue of grant evaluation shows no disparity between black and white investigators (or male and female investigators) in ratings given grants. The authors of the study above ignored that finding, though they surely knew about it. They also ignored a study showing a likely reason for the disparity that has nothing to do with racism: black investigators tend to submit proposals in areas of research that are generally not well funded. (That’s often applied research.) A large part of the disparity is thus due to a disparity of research interests, not to racism.
Further, an even larger amount of the disparity is due to the past record of investigators: papers published, quality of work, types of funding, and so on. While background differences may reflect past racism, this is not an issue of discrimination by the NIH, but a “pipeline problem”: a lack of opportunity for minorities early in life, depriving their members of the ability to build good curricula vitae later in life.
As I also noted, my view has always been to favor affirmative action in hiring, both of students and professors, as well as in other professions. But I think that form of preference has to stop at some point in one’s career, and to me a logical place for scientists is when they begin their first academic job. At that point they’ve been found to be qualified to enter the “science club”, and from then on I think that fairness demands that one see the playing field as level and that preferential treatment should stop. That does not mean that minority faculty don’t deserve mentorship or consideration of special problems they may have, only that they should be given the respect of being treated identically to everyone else, regardless of ethnicity, gender, or race.
Stevens et al., however, disagree. They argue that the palpable racism of the NIH and its reviewers—a racism for which there’s no evidence—mandate profound changes in how the grants of African-Americans are reviewed. (They don’t mention other groups like Hispanics in their piece.) These grants, they argue, should be given preferential treatment. And if they’re not, the authors claim that this is itself racism.
Here’s an example of one of their recommendations and their response to how it’s addressed. The authors are requesting that the numerical scores given to NIH grants by reviewers, which are the overwhelming criterion for whether a grant is funded, take into account the racial diversity of the group of investigators applying. (Nearly all my own grants, by the way, had only myself as the researchers, with unnamed graduate and undergraduate students put into the budget since they hadn’t been hired.)
We ask: why is “diversity of the investigator team” not a scorable criterion in NIH grant review and priority for funding?Be careful with responding, because one answer is racist and the other is not.
No, the “wrong” answer isn’t automatically racist because you can argue that diversity of researchers is not a good criterion for grant quality. It is also arguable whether the mission of the NIH should include social engineering as well as promoting good science, and whether equity should have already been engineered before people apply for grants.
Here are some of the changes Stevens et al. wish to make in the NIH system of awarding grants:
- Diversity of the investigator team should be a score-driving criterion in NIH grant review. This includes race/ethnicity and other forms of diversity such as gender, sexual orientation, and disability.
- Diverse teams should be prioritized for funding. Until there is no NIH racial funding disparity, all applications from Black PIs must be discussed. These applications should be automatically slated for discussion, prior to the review meeting by an automated system or the scientific review officer (SRO)
- Program officers/program directors (POs/PDs) should be encouraged and empowered to reevaluate grants of Black PIs that score above the funding pay-line and bring these grants forward to council for funding. We calculate that an average of only ∼2 additional R01 applications from Black PIs would need to be funded per institute to achieve racial equity.
- More Black PIs should be included on study sections. NIH should institute a minimum number of Black reviewers on each panel and publish a timeline over which this number will represent the US population. We note that some, often in the majority, may voice that this may lead to an unmanageable burden on Black PIs. Yet, the “race tax” is most problematic for service that is not career enhancing. Service on an NIH panel is universally viewed as career enhancing and prestigious, and panel invitations can also be declined. The pool of Black reviewers available for each panel could also be increased if NIH adopted suggestions in the “Beyond 2020: A Vision and Pathway for NIH,” which recommends that narrowly defined organ- and disease-centric panels be replaced with panels that are broader in scope.
This seems unfair to me, though of course all forms of affirmative action can seem “unfair” to those who are passed over in favor of minorities. What I mean is that preferential treatment of investigators should stop when they are hired as faculty in the first place. Grants should be evaluated completely race-free. (One way to do this, which I mention below, is to eliminate all clues to ethnicity from grant proposals, which in principle should ensure fairness. I favor attempts to do this.)
You can argue that my point of view is wrong, but it’s the one I take when judging fairness. Giving extra attention to grants from African-Americans, including mandatory discussion of there’s a black PI but not a white one, is unfair. Likewise with the “mandatory reevaluation” of grants of black PIs but not white ones. Anonymity of proposals would also make that completely superfluous—unless you don’t want a “I don’t see color” system but an “I do see color and want to favor it” system. Stevens et al. seem to favor the latter option.
I do agree that we need more minority reviewers and panelists to ensure that the credibility of our commitment to equality of opportunity be maintained, although my experience is that race and gender are not very useful ways of bringing different views of “quality” to the table. Rather, minority reviewers show our commitment to equal opportunity without promulgating unfairness, and also help ensure that if any racism creeps into the discussions (I have to admit I haven’t seen any in my many years of NIH funding), it will be more easily called out.
In the end, Stevens et al. provide a list of things that all scientists involved in the reviewing system should do:
- Score grants of Black faculty well
- Rescue grants of Black faculty to ensure they are discussed.
- Consider diversity when scoring the investigator team and innovation.
- Learn what racism is, especially topics such as “systemic racism,” “racism,” and “antiracism.”
- Call out and stop all racist statements in review panels and elsewhere. Do not let racist comments pass.
- Include Black faculty in scientific collaborations and write papers and grants with Black faculty.
The last two deserve consideration, though in my entire scientific career I have never heard anything remotely close to a racist or sexist statement uttered on a grant review or in a panel. The panel administrators, who are scientists, are chosen for their fairness and competence, and would shut down anything like that in a second. And remember, I entered the system in the early 1980s.
Including minority faulty in collaborations (#6) is fine, as it’s a form of affirmative action that doesn’t seem unfair, since you get to choose your collaborator and can collaborate with whomever you want. But I would say “Consider including black faculty. . .” as it should not be mandatory. In fact, there are no black evolutionary biologists who work on speciation in Drosophila, so this was never an option for me.
The first three actions represent preferential treatment of grants based on race, and may be illegal. In fact, “score grants of Black faculty well” is invidious because it means “give good scores to proposals from all Black investigators, regardless of the proposal’s quality” rather than “treat these proposals fairly”. At any rate, I think grants should be treated equally, and even reviewed with the names taken off of them if that is possible (sometimes it’s easy to guess whose grant you’re reviewing). If grants were reviewed blindly at both ends, any accusations of racism would be moot.
As for learning what racism is, and reading about “systemic racism and antiracism” (#4), yes, I’ve done that, and surely all academics now should understand these terms. But I suspect Stevens et al, based on the rest of their op-ed, have a specific definition of these terms, and one that corresponds to their meaning in Critical Race Theory. And to that I object, for that’s a form of indoctrination into an ideology. All we need to know about racism and grant reviewing is this: investigators should be judged by the content of their proposals, not by the color of their skin.
_________________________
Stevens, K. R., K. S. Masters, P. I. Imoukhuede, K. A. Haynes, L. A. Setton, E. Cosgriff-Hernandez, M. A. Lediju Bell, P. Rangamani, S. E. Sakiyama-Elbert, S. D. Finley, R. K. Willits, A. N. Koppes, N. C. Chesler, K. L. Christman, J. B. Allen, J. Y. Wong, H. El-Samad, T. A. Desai, and O. Eniola-Adefeso. 2021. Fund Black scientists. Cell. DOI: https://doi.org/10.1016/j.cell.2021.01.011
“Systemic racism” = “racism that we can’t find any evidence for, but we want to assert it exists anyhow”.
“Antiracism” = “just like traditional racism, but with the pecking order reversed”.
That’s an excellent and succinct summary.
If these changes are put in place, there will be a big incentive for PIs to identify as Black. I agree with our host that these are bad ideas in many ways, but it seems likely the ideas will become policy. If they do then I sincerely wonder how this will all work: how Black would a PI have to be to qualify? Would a sincerely held belief by a PI that she has some Black ancestors be good enough, or would genetic evidence be required? Would African immigrants qualify, or only ADOS?
I think that you can just identify any way you want. There is a set of race boxes to tick when you apply for an NIH grant, and no “proof” required. It may be true that at the higher-up level, when scores are translated into money, that grants are fiddled with and the race of the investigator may help a grant get funded. But usually the program directors, who would be doing this, know the people who are applying for grants.
Rather, minority reviewers show our commitment to equal opportunity without promulgating unfairness, and also help ensure that if any racism creeps into the discussions (I have to admit I haven’t seen any in my many years of NIH funding), it will be more easily called out.
–But for the sake of the quality of the science that is funded, isn’t it important that reviewers be selected solely for their expertise and discernment? And if, as you say, you’ve never witnessed any racist discussions in the selection process, why is it important to set a system to ensure against a nonexistent–and, to me, bogus–eventuality?
Realistically, I don’t think that’s possible. I think a candidate’s field of expertise + their institute are going to be sufficient to identify who the candidate is to other experts in their field, at least a lot of the time. When a toe cancer panel of experts gets a proposal to research toe cancer from the University of California, Illustrationsville, they’re going to say “oh, that’s Bob, he’s the only toe cancer researcher there.”
One thing NIH could possibly do is allocate more funding towards less experienced PI’s or PI’s coming from less traditional ‘powerhouse’ research institutions. This would be higher risk, but it would help ‘widen’ the pipeline, letting more researchers in general get more experience.
Eric, I strongly doubt that grant reviewers know from which institution the applicant comes.
Your second paragraph seems to aim at bringing in race as an evaluative criterion through the back door.
I agree with PCC(E), at this level race should not matter at all.
I’ve seen it on at least one NSF panel, actually; they knew just by looking at the Institution (and knowing the grant subject) who it was.
There’s really no way of hiding it if the reviewers are going to have access to the PI’s experience.
Take Jerry as an example, if you get a proposal from an evolutionary biologist proposing Drosphila research, and the applicant has 30+ years of experience and is applying from the University of Chicago, how many people fit that description? Moreover, the CV’s are going to be a giveaway even with the names removed. Consider the publication list – youu can black out the author names, sure, but the paper titles are going to give it away if the reviewers have read any of them.
It would expand the pipeline by giving more grants to younger researchers. That doesn’t use race as a criteria for award, but if we take as granted that women and minority representation in science becomes less and less as experience, renown, and professional status go up, then yes, this would have the practical effect of seeing more awards go to women and minorities.
That’s a good point. This is WAAAY out of my bailiwick but it appears to me that in some science grants – at the upper end – there can be a very small pool of candidates, all known to each other. Like how PCC (E.) knows all the fruit fly guys and gals. I’m quite happy to be wrong in this, again, not my expertise. Nor is toe cancer but it sounds dreadful 😉
D.A.
NYC
I think it’s time to revive the old term “race hustling”.
Please knock off these ascerbic comments. We try to be polite here and deal with arguments.
I’m uncomfortable with affirmative action being acceptable until the recipients have become faculty members. If AA is unfair, then it is unfair the whole way up the academic or employment ladder.
It sets everyone up for problems, as anyone who might be perceived as having benefited from AA will not quite be accepted or taken as seriously as she or he might be were the process completely colorblind.
Well, those who benefit from affirmative action don’t always feel that way, and if they don’t, are you to tell them that they should? Also, not every minority student is in college or grad school because of affirmative action, so who’s to know?
“Also, not every minority student is in college or grad school because of affirmative action, so who’s to know?” That’s certainly true, and it’s wonderful.
” In fact, there are no black evolutionary biologists who work on speciation in Drosophila, so this was never an option for me.” That disparity will, of course, be taken as a sign of systemic racism—the system being the field of speciation in Drosophila, and perhaps speciation in general—or maybe even evolutionary Genetics in general. In order to properly “decolopnize” science, these sub-fields may just have to be eliminated.
I recommend two books about what happens when science is subjected to ideological standards. They are: “The Rise and fall of T. D. Lysenko” by Zhores Medvedev”; and Loren Graham’s magisterial “The Ghost of the Executed Engineer—Technology and the Fall of the Soviet Union”. The latter book goes into detail about matters that are not widely appreciated: the way the ideologizing of science has practical spillover effects on technology, and on the teaching and practice of engineering, not to mention on choices in social investment.
Back in grad school I met Joe Graves, who described himself as the only Black man (at least at that time) to earn a PhD in evolutionary biology in the USA. He made a big impression on the grad students because he had experienced so much direct racism in his life and education, and talked about it openly. He worked mostly on experimental evolution in Drosophila. But not directly on speciation so not quite in the same field. Interesting guy.
Sub
“scientists”, in particular, “s”.
In case it wasn’t pointed out. Not like readers here wouldn’t sniff it out.
“it’s the one I take when judging fairness. ”
Fairness is not the goal here. The civil rights movement in the 1960s was largely about fairness and access. Apparently that did not yield the desired results for enough people, so they are now demanding equality of outcome.
I cannot think of any downsides of equality of opportunity, either for the individuals involved or society as a whole. Equality of outcome has lots of drawbacks, from my perspective. Obviously the highly qualified person with unfavorable heritage is going to be unhappy. The less qualified person who gets the gig may well find themselves out of their depth and expected to perform beyond their abilities, instead of thriving at a performance level more suited to their abilities. The institution and the field of scholarship are going to suffer, because they are never going to produce the sort of innovations and scholarship that come from hiring the best and the brightest.
Highly qualified persons who happen to have favored heritage are going to have to deal with their peers low expectations. That is kind of unfair in itself, but pattern recognition is such a fundamental human survival trait that it is hard to put aside completely.
I just don’t understand how so many otherwise educated people could endorse such a strategy.
This is transparently a demand for special treatment for a particular group, and I hope it will be resisted. However, by current standards it is more likely to be implemented by spineless administrators who are frightened of being labelled as racists.
I was curious about Black investigators submitting grants in areas that are poorly funded, namely applied research.
What about increasing the pool of NIH money that goes to applied research? I presume that this is research that could also be funded by industry, and that perhaps industrial funding doesn’t help with tenure as much as NIH research? I’d hate to reduce funding for pure science, so instead of robbing the main NIH budget, perhaps industrial money could be raised and pooled into a new research fund (a new tax?) that would be tailored to meet the interests of researchers who might want their work to have more immediate, tangible benefits? Black and other interested researchers would of course help shape the goals of this fund.
Would this be an acceptable and fair way to help reduce racial disparities in funding?
There’s Translational Research grants – or Translational Medicine – get basic research to help people right away in the hospital or clinic.