Diversity—Looking at Different Fields of Science

Posted

Following up on last week’s post, I want to look at another issue addressed in the Advisory Committee to the NIH Director (ACD) Working Group on Diversity in the Biomedical Research Workforce report.

The working group was interested to see if there were differences among general fields of science in applicant representation based on race, ethnicity, or disability status. The working group assessed this question by looking at the breakdown of applications from 2000 to 2008 in three broad fields of science—basic, applied and clinical, and behavioral. As you can see from the data in figure 1 below, the number of applications submitted by African-American or Black, American Indian or Alaska Native, or Native Hawaiian or Other Pacific Islander applicants also made up a small fraction of the whole compared to the number of applications submitted by White applicants, and this was true across the different fields of science. In the basic sciences, 1% of the applications were submitted by African-American or Black applicants compared to 64.6% submitted by White applicants. While the percentage of applications from African-American or Black applicants was somewhat higher in the clinical (1.5%) and behavioral sciences (3.1%), that proportion was still dramatically lower than that of White applicants (65.2% and 77.5%, respectively). Note that the overall number of applications in behavioral research is lower than in the other fields, so even though the percentage of applications from African-American or Black principal investigators (PIs) is higher, the absolute numbers remain very small. These differences notwithstanding, it is evident that African-Americans or Blacks, as well as American Indians or Alaska Natives, and Native Hawaiians or Other Pacific Islanders are underrepresented as PIs in all areas of science that NIH supports, and any efforts to address applicant diversity should be aimed at all fields.

This figure shows the distribution of applicants by race of PI and field of science.

Additionally, although not contained in the report, the data on Hispanic applicants in figure 2 below show similar trends with low numbers of applicants overall and a slightly larger percentage in behavioral research. 

NIH does not collect data on disability status. Recommendation #1 of the working group specifically asks NIH to enhance its data collection capabilities for all these populations.

This figure shows the distribution of type 1 RPG applications by field of science and ethnicity of PI for fiscal years 2000-2010

Overall, these data and the data presented last week show that we have a very small pool of underrepresented groups applying as PIs. These data serve to underscore the further need for NIH and academia to partner and improve the diversity of the biomedical research workforce.

7 Comments

  1. The positive thing about the numbers being so small is that a moderate sized working group could almost go application by application in a focused effort to try to generate testable hypotheses. What do the summary statements say? Are there appearances that they are discouraging re-submissions (a major factor IIRC)? What about the near-miss scores? Are POs failing to pick those up with exception funding because they don’t know the applicants?

    Thought experiments to generate the hypotheses are all well and good but discovery efforts are useful when the people doing the hypothesis generation are established to be part of the problem in the first place.

  2. First – I agree with mat wholeheartedly.

    Second – I have been actively engaged for several years in helping a former post-doc, who is African-American, get his first R01 as well as tenure at a major university. After twice going through the humiliating maw of what is now called peer review, he improved his score from “unscored” to something that was merely below the payline. The summary statement was full of vague microscopic insecurities that (a) this former postdoc was in a better position to judge than his reviewers, (b) he had insufficient room in a 12-page app to fully discuss, and (c) were entirely new issues not raised in the summary statement of his original application. Despite effusive praise for innovation and record of accomplishment to date, the most damning criticism was that he did not provide sufficient assurance that this line of investigation would lead to a cause of XYZ disease. How in hell can he guarantee the results of an experiment? What’s the point of proposing an investigation if you already know where it leads? If that was the primary weakness of the proposal, why not raise that criticism about the original application so he could answer it?

    So now, current rules preclude him from submitting any kind of amended version of this grant, and he feels that he was set up to fail. I don’t think he is being unreasonable in his resentment of the system, and don’t ask him to encourage anyone to pursue a PhD in biomedical science, let alone an underrepresented minority. He’s actually looking at investment banking, where his contemporaries are thriving.

    And now the workforce study group recommends shorter times-to-degree? What are they thinking – let’s accelerate failure?

  3. Basic science has never been a particularly lucrative career path. At present, it is also filled with persistent insecurity for the 99% at least. For those trying to elevate thier position in life or that of their family, this would not an attractive career path.

  4. I fully agree with Matt. However, I do not think NIH has any programs to enhance the number of RO1 funded underrepresented minorities, PO’s should be encouraged to look for this. The equation is easy in academic science: no RO1, no promotion/tenure, no underrepresented PI’s in science (true for all). While industry has recognized for a long time that maintaining a diverse workforce is essential to continued growth (meaning the next 40 years) and innovation, NIH has not addressed this in a systematic manner.
    NIH should perform studies to determine bias in the peer-review system. This includes:
    a) systematic review on how new RO1 submissions are assigned to reviewers (new and established invest.) based on ethnicity, in particular for A1 submissions (if an applicant gets the same primary and secondary reviewers and have addressed 100% of the critique chances of getting a higher score are better than with different reviewers on the A1),
    b) which is the rate of RO1 renewal, and first RO1 award for underrepresented minorities compared to the mean (15%) and to other groups.
    The bias in peer review becomes apparent when the payline is low as it is now. A small bias makes the difference between 12 and 15 percentile. The first one advances, the second one struggles or drops-out, yet the scientific merit of the applications was not statistically different, and both applications were outstanding.

  5. I have had one disabled and 2 Hispanic persons in my lab over my career. I have managed to get various types of diversity supplements for them. However, the older disabled person was asked on a job interview “what’s the matter with the way you walk? are you OK?” and then they rejected her. She finished a 5 year postdoc with me and took a part time job at the community college. The Hispanic graduate student decided to leave the pipeline and took a full time job at the community college. My current Hispanic postdoc is supported by a diversity supplement, having had her NRSA triaged. The problem with this approach is that she won’t have a record of independent NIH funding going forward. We will try the NRSA again. She is nearing the deadline for a K award. The only relevant institute with a K award for diversity (K22) is NINDS. It would be helpful if all institutes offered K22 awards for PhDs. Also, the deadline of 5 years past the PhD, even for revision, is very difficult. I know the K awards are aimed at the hottest young scientists around, but they seem particularly discouraging for minorities.

    What I am asking is whether minorities should get a break? Is affirmative action dead? Or is it something that should be available through an education and then not available at the level of grant competition?

    I concur with Bert. These folks start out disadvantaged in many cases, work really hard to catch up, get a PhD, and do a postdoc and then are asked to compete with the top 5% of research scientists in the country with an overall NIH budget funding huge clinical trials and centers. Can center grants get diversity supplements for a new minority faculty member?

    I am near retirement and the last 35 years have been wonderful. I feel discouraged for all young investigators trying to get a research career started these days. They have such drive and stamina and such incredible educations!!

  6. just one illustration of a factor in the decline of the importance of the university system as a credible instrument of education and as the final word on scientific review. i predict that non- university education will gradually replace and displace the current institutions. universities descended from guilds, and never gave up the goal of creating an exclusive repository of knowledge. say goodnight, universities.

Before submitting your comment, please review our blog comment policies.

Leave a Reply

Your email address will not be published. Required fields are marked *