Recently, John Whitmarsh, one of my colleagues at NIGMS, pointed out a short article called “The Glass Ceiling’s Math Problem” 
from the May 31 edition of the Washington Post. It highlighted a study by researchers associated with the National Bureau of Economic Research that focused on 9,500 U.S. Air Force Academy students from 2000-2008. The findings showed that female students (but not male students) performed better and were more likely to go on in science when they were taught by female faculty members.
I wondered why the population from the Air Force Academy was chosen for this study. I learned that Air Force Academy students are assigned to faculty randomly, and they all must take math and science courses. These characteristics minimized factors of self-selection bias due to students choosing particular faculty, which would confound similar analyses at many other institutions. I encourage you to read the study and welcome any comments you may have.
This recent work aligns with an ongoing effort at NIH to encourage the advancement of women in research careers. As part of this effort, NIGMS has led an initiative to identify and support research related to understanding the factors and interventions that encourage and support the careers of women in biomedical and behavioral science and engineering. I’m pleased to say that there was a strong response to the request for applications and that we expect to make awards soon.
Instead of putting more money into studies for why there are not enough women in science- why not put money back into basic research and make it possible for them to succeed? What a novel concept!
I absolutely agree with Lee Graves.
Put money into research! More money, more possibilities of success for women (or everybody else). I think it is clear why there are not so many women is science. There is not more need to study that!