What constitutes racism?  

If you have a pool of applicants for a research grant and applications with good scores were likely to be funded, regardless of race or ethnicity, can there be racism?  

A recent survey of NIH R01 applications found that applications from black investigators were 13.2 % less likely to be awarded than whites while Asian investigators were 3.9 % less likely to have their work funded.   This correlates to the number of applicants as well; blacks are only 1.4% of total applications while Asians are 16.2% and whites are almost 70%.

NIH funding is not arbitrary.   In fact, it is downright lacking in its ability to be creative at all, given the number of restrictions it operates under, including diversity programs at the federal level.  NIH even has training fellowships which can serve as a good stepping stone to the arcane R01 grant level.    They sponsor fellowships, traineeships and career development awards and people who participate in those were more likely to apply and get grants;  69% of Asians, 54% of blacks, 62% of Hispanics and 62% of whites had participation in NIH training programs.  They do what they can to help inexperienced researchers.

What does this study mean?  Well, nothing.  Race and ethnicity information is not given to the review committees so for there to be alarm we would have to assume dozens of reviewers are all secretly agreeing to penalize someone from a 'black college' or by their name and yet none of them have raised the alarm about the secret racism.  It is Area 51-type conspiracy kookiness to think dozens of reviews on hundreds of panels are all secretly underscoring minorities - though only the ones without very good applications.

But alarm bells are going off in Washington, D.C. just the same.   Since the good applications were funded regardless of race, is anyone claiming the panels that review the grant applications were discriminating against black applicants?  No.   Is there any evidence the period sampled just had weaker applications?   No there either, that would have taken real research.   It's easier just to run a model and make a conclusion, and so Donna K. Ginther, the professor of economics at the University of Kansas who led the study, does just that.

“It is striking and very disconcerting.  It was very unexpected to find this big of a gap that couldn’t be explained,” she told Kenneth Chang of the New York Times.

Dr. Francis Collins, the director of the National Institutes of Health, reacted quickly and maybe a little of the mark, saying, “This situation is not acceptable.  This is not one of those reports that we will look at and then put aside.”

Probability of NIH R01 award by race and ethnicity, FY 2000 to FY 2006 (N = 83,188). Based on data from NIH IMPAC II, DRF, and AAMC Faculty Roster. ‡, P < .001; **, P < .01; *, P < .05.

You're probably as curious about which reports they do look at and then just put aside but he doesn't even seem to have read this study.   Nothing in this says the NIH or the process is racist.  If anything, it says the NIH funds too many large universities.   Is there any reason Johns Hopkins should be getting $1 billion per year in taxpayer money?

What the study does show is that there were fewer total applications from blacks at institutions receiving the most NIH funding and working at an institution that already got a lot of money from the NIH was correlated to a much greater chance of getting a grant.  The NIH clearly uses school name recognition a lot, though those schools will argue they hire the best people.  What you believe in that regard is largely a matter of whether or not you are in that business.

Previous applicants also had a greater chance of being funded and the study found that blacks who were rejected did not re-apply as much as whites and Asians. The NIH additionally seems to be biased toward number of citations.

Overall, blacks graduate high school less than whites, they are less likely to go to college than whites and they are less likely to major in science than whites - that is well established.  But once blacks get a Ph.D. they are equally likely to be tenured in academia or research institutions as anyone.

So why the disparity in NIH awards?  Is it much ado about nothing, the kind of thing the New York Times lives to write about?   Women and men have no difference in NIH awards, for example, but a subset of women insist academia is sexist. 

The answer seems to be a confluence of minor things, not racism in award panels.  As outlined above, there are various slight advantages that make the difference in NIH funding.  If those add up it may lead to a difference in race of awards even if it doesn't mean racism.

Since there was no racial or ethnic disparity for good applications, the answer may be more good applications rather than the inevitable call for quotas.   The NIH is scoring more heavily based on the school and citations, even if they claim the criteria are primarily significance, innovation, and approach of the grant application.  Clearly the investigator, being their citations, and the research environment, being the school name, are too much of a focus, right?

But no, even a quality scientist like Collins who runs what is arguably the most ethically fair funding group in the world, gets fuzzy and says “Even today, in 2011, in our society, there is still an unconscious, insidious form of bias that subtly influences people’s opinions.”

So he's running the group and rolling his review panels under the bus, despite any evidence anyone did anything wrong.  That will be great for morale.

Look for someone in the social sciences to have a study out by next week claiming minorities don't write stronger grant applications because they are worried if they get denied it will reflect badly on their race or ethnicity.


Donna K. Ginther, Walter T. Schaffer, Joshua Schnell, Beth Masimore, Faye Liu, Laurel L. Haak, Raynard Kington, 'Race, Ethnicity, and NIH Research Awards', Science 19 August 2011: Vol. 333 no. 6045 pp. 1015-1019 DOI: 10.1126/science.1196783