That NIH race thing is still a big deal, and completely forgettable

Dec 05 2011 Published by under Uncategorized

Unless you’ve been living under a rock recently you’ve heard that the NIH came out with a report of a possible racial bias. Specifically the report concluded that African-American scientists were less likely to be granted funding from NIH, even when controlling for various relevant factors. There was a brief brouhaha, responses from NIH, and commentary from scientists (more links at the bottom).

Reactions ranged from “Well this couldn’t be true. I know people at NIH and they are very nice.” to “The whole grant system is rotten, so this isn’t surprising”, also “Maybe those people just aren’t good scientists” and “Actually I bet the opposite is true: a positive bias” Essentially, the typical distribution of responses for the possibility of such a bias.

Perhaps there’s no systemic bias on the part of the NIH grant system. There are a variety of reasons the NIH numbers could be low. The NIH is just one part of the a much larger system, there’s undergraduate education, graduate education, faculty hiring and such that feed into those numbers. Some have suggested that this is really just a reflection of biases in resource allotment at the home institutions. The obvious counterpoint is to take a look at the numbers for NSF, which aren’t perfect, but much closer to equal. I can't add much to the analysis of the actual report. I’m no statistics expert. The study sounded reasonable, though I’m sure there area some fair criticisms. All I can say is that it is entirely plausible that there is a systemic bias.

If the study results are taken at face value, it implies that NIH, an organization near the heart of the scientific endeavor in the United States, is exacerbating the problem; not improving things, or even being neutral. That is shocking. The stats that I have looked over show that the  proportion of black scientists reduces at every level of the academic ladder (figure 1). It is difficult to think of a more effective way of shedding black scientists than to throw a wrench in the NIH-R01 mechanism. The possible existence of such a powerful deterrent should give everyone pause when reading those ridiculous “we encourage diversity” statements that many institutions have written up for job ads and mission statements. Nice words I suppose. Who knows what the actual actions or results may be. Right now it's not looking so great for NIH.

 

Fig1. Percentages of African-Americans in my general field, which is one of the highest. Bachelors (18%), PhD (10%), Tenure Track (7%).

Given this possibility, what’s a black scientist to do?

Switch to that other granting agency down the road? Make a note in your tenure file? Complain?

NIH has promised swift action. Meanwhile, black scientists are going up for tenure, applying for jobs, submitting papers, and applying for more grants. From the perspective of the individual scientists there's not much to do if there were bias in any of these areas. So, while this seems like a big deal, it is rather forgettable. What’s going to be done by NIH? How would this change the situation for current or future black scientists? The report did push me to examine other sources of funding, which I had already been considering. Like other cases of systemic bias, there doesn't seem to be  much recourse for the individual. Were some African-American scientists negatively affected? Maybe. If so how could we even identify them for sure? If things are “fixed” how long will that take? There’s a possibility that things will be a little closer to fair for the folks coming up 5, 10 years from now?

I can see why a black scientist might read over this report and just shrug. Same shit, different day. Even if we all agreed that there bias in the grant system, the wants and needs of scientists with regard to obtaining grants remains the same. Even "proof" of past or current bias wouldn't change many people's situation one bit. Thus, even though it would seem that this story could not be more relevant for an early career African-American scientist, there doesn't seem to be much to do with it. Better to spend time working on writing those papers and writing those grants, even if it is the case that they might look at you funny.

Links:
Bashir
DrugMonkey
New York Times
Science
NIH blog
NIMH blog

8 responses so far

  • BikeMonkey says:

    I was at a meeting recently where Tom Insel polled a room of scientists as to how many had heard of Ginther et al. Maybe 5 hands went up out of ~ 200 folks. This was in a session about NIH doings so presumably it was already selected for those who pay *some* attention to NIH business. It was not totally unexpected....but still rather surprising.

  • epj says:

    The problem with science is the same that with the rest of world society: economy, money, and for some strange reason the inability to appreciate life for what it is (though is the issue under inquiry).

    The day the economy is simplified and truly adjusted to the population needs that day science blooms to yield an advanced society that can be called advanced. But now the features of the economy work to keep most people chained and a few pseudo 'enlightened'

  • Joe G says:

    When applying for a grant does the application contain the race of the scientist?

    The point being is how does the NIH know the person sending the request is of what race?

    Another issue would be the projects- as in it could be the project was rejected and not the person (yeah unlikely but a possibility).

    One more thing to consider is the number of applications and the races of the applicants- meaning if non-African American scientists and their applications greatly outnumber the AA side then that would reflect in the results.

    (I am just hoping that race bias doesn't exist in science because THAT would just be too much for me to even want to understand)

    • bashir says:

      There are a many ways to gleam information about the applicant. Grants are not reviewed by random scientists. They are reviewed by those with some knowledge of the relevant area. Science is a small world. People know who people are, especially after some time in a particular field. That and, what, no one ever googles anyone? Never any clues on the CV? Rasheed Jenkins got his PhD at Howard and no one is making any assumptions?

      The person, and institution are often explicitly part of the evaluation. I myself have been dinged on that.

      Not sure what the raw numbers has to do with it. The relevant numbers are the percentages.

      • Joe G says:

        OK perhaps the NIH should also start sending the reason(s) why the grant application was rejected and also ensure their review board is race- neutral, meaning all races are equally represented (if possible).

        NIH:

        To receive NIH funding, applications are evaluated by a peer-review process that considers the significance, innovation, and approach of the grant application, the investigator(s), and the research environment. Applications determined to be meritorious are discussed in detail and scored. About half of all applications are scored. Among those that are scored, relative merit score, budgets and NIH institute priorities, which vary by year and by institute, determine which applications are funded.

        So tell the scientists that are not scored or who do not get a grant the reason(s).

        Maybe they should also do a study on the appearance of the scientist- maybe the reviewers just think those who were denied were just too ugly to receive a grant (joking but given this race issue it could be true).

  • drugmonkey says:

    CSR rules demand racial and ethnic representation on review panels. It would be interesting to see how this works out across the CSR in practice but my experience suggests African American scientists are if anything overrepresented.

    They only have one vote, however. So unless they are disproportionately assigned grants to review that have African American PIs, this has little direct effect. The major driver in the grant score is the opinion of the ~3 reviewers assigned to do the primary reviewing duties on each app.

    The NIH does "send the reason" by means of the written critique of these assigned individuals. Are you imagining that "because the PI is black, we're marking this shit down" has ever appeared in a written criticism? I am very confident this has never occurred. Do try to be serious.

  • Affinity says:

    I'm not easily impessred but you've done it with that posting.

  • drugmonkey says:

    The NIH is seeking input

    to help inform the development of recommendations to present to the ACD and the NIH Director on actions the NIH can take to increase the diversity of the biomedical research workforce.