18 Comments
Last month I blogged about the participation of women in NIH extramural programs, and I promised more information when the paper that we prepared in collaboration with the NIH Office of Research on Women’s Health came out, so here it is. You can read the full article to delve into the analytical details, but I wanted to highlight a few of the important findings, primarily focusing on the R01 program.
The paper presents two studies. The first, a cross-sectional analysis from fiscal year 2008 data, compared women’s and men’s grant application, success, and funding rates for 17 award programs (plus T32 trainees) that represent a typical “career ladder.” As shown in the figure below, which summarizes data shown in the report, the proportion of women applicants compared to men generally decreased at more advanced career stages. Similar findings appeared in a 2007 National Academies report.
The report goes on to show small differences in the success rates or funding rates between women and men for most research programs. Looking specifically at the R01 program, the sexes had equal success rates (23%); however, women had small and what appear to be persistently lower funding rates than men (26% vs. 28%), presumably because men submit more applications per person than women. When we looked separately at new (type 1) applications, the sexes had nearly equal success and funding rates but there were differences with renewal (type 2) applications where men had higher funding rates. Another bright spot is that women and men awardees both received the same percentage of requested direct costs (87.5%) and women actually received larger awards than men.
The second part of the paper goes on to show that men were more likely to have more than one concurrent R01 awards than women. But, looking at all principal investigators with two, three, or four concurrent awards, the study shows that women attain these milestones when they are younger than men. Honing in on factors that may explain the differences between male and female applicants, the paper shows that proportionately more women proposed human subjects research than men (~50% vs. ~30%), and they were equally successful on those applications. However, it was noted that many of these projects are less likely to be submitted for renewal. Also, women were less successful than men on applications that did not involve human subjects.
Our take… Overall, the authors showed that women have comparable success and funding rates to men for most award programs. And, women awardees, on average, received at least as much as men in direct costs requested per application. However, there remain areas that I think could benefit from improvement, including retaining more women in biomedical research as they climb the career ladder. Continuing to look critically at trends in funding will be important as we move forward.
I encourage you to take a look at the paper by Pohlhaus, J.R.; Jiang, H.; Wagner, R. M.; Schaffer, W. T.; Pinn, V. W., “Sex Differences in Application, Success, and Funding Rates for NIH Extramural Programs” Academic Medicine 86 (6): 759 June, 2011.
This statement: “…but there were differences with renewal (type 2) applications where men had higher funding rates.” did not sound a bright spot to me.
It seems to me that the goal of retaining more women in the biomedical sciences as they climb the career ladder requires that they are able to get their grants renewed at the same rates as men.
I agree. I don’t see it as a bright spot that women have an unspecified lower success rate with renewals. This seems to be a significant problem.
I would be very interested in the data regarding success rates for A0 vs A1 vs A2 for women in comparison to men. My experience as a reviewer suggests that more women go to the A2 before they are able to convince the reviewers that the work should be funded. If this is true, we are about to see a significant drop in women in biomedical (non-clinical) research with the loss of the A2.
First: How is doing good science possibly related in any way to making sure women, or any category of human, get equal grant funding?
Second: It is blatantly sexist to purposefully fund someone just because they are female (which is what the NSF and NIH do). Under that logic, I should get drafted into the NBA because I’m white, to balance out the obvious racism of the teams’ owners and to make for a more diverse/entertaining league for the benefit of all.
Third: Has anyone considered the apparent obvious reason for this trend (women getting even initial funding, and less renewals)?
Shouldn’t we be worried about doing good SCIENCE? Remember what that was: generating hypotheses, testing them, and communicating the results to better humankind?
It is there to counter the many biases that permit males with inferior science to succeed above more deserving women. That is how such policies improve science.
We do not need to rely on “obvious reasons” when there are studies showing biases against women scientists (I.e. review of gender ID’d vs gender unsignalled CVs).
In addition to the worry that women have a lower success rate for renewals than men, I wonder about the classification of the women who are succeeding with human subjects. And interestingly, these projects were less successful for renewal. I see a potential situation where there is a much higher success rate of female physician scientists (M.D) versus female basic scientists (Ph.D.). Has anyone done that comparison?
I am having trouble with the graph. The title is Average Age and Percentage of Female Awardees. I presume the Y axis is not the average age but rather the percentage of successful awardees.
If that is the case then at the RO1 level it appears that women have a 28% success rate compared with the 50% success rate of men.
What the graph really shows is the lack of funding for women after the early training awards.
Depending on which line you are looking at, the Y axis is either average age or percentage of successful awardees. Success rates for R01s are equal (23%) as explained in the paper. The graph shows that 27% of R01 awardees were female in 2008 and their average age was 48. (The paper shows that the percentage of female applicants for the R01 program was also 27%).
I think this means that the average age of those with an RO1 award is ~ 50. But I assume that the data are not actually longitudinal, as would normally be implied by the use of a line rather than a bar graph. (Obsessive-compulsive editing tendencies…)
@Helen Salz and @eae: You are correct in noting that differences in renewals are not a bright spot, but I do want to clarify a couple of things. If you look at the data in the paper about renewals, experienced women had success rates that were not statistically different from experienced men. It was the funding rates that differed. (Funding rates measure people, while success rates measure applications.) We noted that women tended to submit fewer applications, which could be part of the explanation.
@RMH: Regarding human subjects research, we found that women awardees were more likely to perform human subjects research as compared to men, but that their funding rates for human subjects research at renewal were not different. We did not look at the comparison based on degree type for this study, but that is a variable we have considered in other studies.
I still have concerns regarding the success rate at the A0 vs A1 vs A2 for women compared to men. We are heading for a huge gender gap if the A0 and A1 success rate for women is significantly lower than for men.
I share eae’s concerns that we are heading for a huge gender gap as funding becomes increasingly scarce. At highly competitive study sections, such as the Molecular Genetics Study Sections (MGA, MGB, MGC), a single negative, factually incorrect or biased comment can sink an application. Although all applicants are hurt by what has become an increasingly capricious review process, female applicants may be especially vulnerable, since numerous studies (including one by the National Academy of Sciences) have documented the existence of unconscious bias in evaluating women’s contributions. In this regard, I note that a search of the NIH Reporter database reveals that as of today, 24 RO1/R37 renewal applications from the three Molecular Genetics study sections have been funded in FY 2011. Amazingly, only one of these renewals (4.2%) has a female Principal Investigator. I would love to hear Dr. Rockey’s thoughts on this statistic.
With regard to the study sections you mentioned, I think it’s too early to make a verdict. There are still several months for ICs to make awards for fiscal year 2011, we don’t know how many applications in these groups were from women, and it’s ultimately the ICs that make the funding decisions, with many factors at play.
I found the article hard to understand but potentially very important (an unfortunate combination.) Two things I think I did understand were extremely concerning: (1) Women don’t do as well as men in renewing their RO1s and (2) Very few women have more than one RO1 compared to men. If true, over the course of a career, this would be absolutely deadly. So deadly, that it’s essential that NIH find out if this is true and if so why. It is pointless for women to go into basic science funded by NIH if they can’t compete for money with men. The report seems to echo what so many studies found for decades – “women start out equal but they don’t end equal” due to the small but cumulative drip drip drip of inequality they encounter in so many aspects of the job. Very important study and looking forward to the full explanation and follow up. Many thanks!
@ Sally Rockey: I looked at the data in the paper about renewals, and unfortunately your statement that women do not differ from men in success rates is incorrect. See page 761 of the article:
“Further analysis of the R01 program data showed that funding and success rates for first-time applicants were comparable for men and women, but experienced male applicants had higher funding and success rates than experienced females (Table 1). Furthermore, for experienced investigators, success and funding rates were nearly equal for new applications (women were within one percentage point of men for both measures), but men were three to four percentage points more successful than women on renewal applications, when measured either by success rates or funding rates (Table 1). “
Thanks for your comments. @Senior Women Scientist: If you look at the data for renewals in table 1, you will see that while there was a three to four percentage point difference in success and funding rates, it was only statistically significant for the funding rates.
So, yes there are differences, but the real question is why there are differences when looking at this crucial transition point. In the paper’s summary, the authors discuss some of the factors that may be causing decreased application rates (family circumstances, self-confidence, and other barriers) and funding rates for women (unconscious biases). We are currently supporting further research to examine possible causes and the effect interventions have on the career patterns of women in science and look forward to the results from those studies.
As someone who has served on many study sections, I believe that study section reviewers should be required to refer to all principal investigators in gender neutral format when these applications are discussed. The investigator could be referred to as the PI or as Dr. so and so (without first names). The NSF report noted below suggests that under situations in which data is ambiguous, unconscious biases exist and this negatively impacts women more than men. I believe this could be particularly relevant and impact funding of women at early career stages.
Hill, C., Corbett, C., St. Rose, A. (2010) Why so few? Women in science, technology, engineering and mathematics. AAUW (Wahington DC), pp. 1-109 (http://www.aauw.org/research/whysofew.cfm).