A new report on gender differences in granting decisions has uncovered large gaps in recordkeeping at several US science agencies. Advocates of gender equality are calling for better data, saying poor records make it impossible to determine whether agencies are complying with antidiscrimination laws.

Money matters: In the past decade, NIH research grants to women have remained at about 80% of the size of research grants to men. Credit: Source: US National Institutes of Health

The report, commissioned by the US Congress and released on 14 September, found that the Department of Defense and the Department of Energy collect little information on applicants' gender. The records are so poor that the RAND Corporation, a nonprofit research organization that led the study, could not analyze the agencies' decisions, according to principal investigator Susan Hosek. “We concluded this was going to be a nightmare of very questionable value so we just gave up,” she says.

The National Science Foundation and the US Department of Agriculture had more complete data. At those agencies, the report found, women and men requested and received on average the same amount of money.

There is more than ample evidence that there is a problem here. Jocelyn Samuels, National Women's Law Center

The group's analysis of the US National Institutes of Health (NIH) yielded some surprises. Based on grant applications from 2001–2003, women applicants received on average 83% of the funds awarded to male applicants, controlling for factors such as age and academic degree—and excluding the top one percent of large awards. Women received only 13% of those large awards. The report also revealed that women are less likely to reapply after being rejected their first year.

In response to the RAND report, the NIH on 7 October released its own analysis, which suggests that men and women are about equally likely to receive a grant in response to their application. In 2004, the success rates for grant applications were 23.9% for women and 25.2% for men. But the data suggest that women ask for less money—which could explain why they get less, says Robert F. Moore, a consultant for the NIH and former director of the agency's division of statistics and analysis.

Ultimately, neither the RAND report nor the NIH gets to the heart of the matter, says Donna Dean, president-elect of the Association for Women in Science. For instance, she says, it is unclear why so few women are in charge of large awards, such as multicenter endeavors. Do fewer apply, and if so, how can the NIH address that? Why do fewer women reapply after their grants are rejected? The data should also be broken down by specialty, she says.

The US National Academies is set to release a separate report at the end of the year that will address gender differences in areas such as faculty hiring, promotion and allocation of laboratory space.

Lack of data on gender of applicants makes it difficult to track compliance with Title IX, a US law that bars discrimination in education (Nat. Med. 11, 462; 2005), says Jocelyn Samuels, a Title IX expert at the National Women's Law Center in Washington, DC. Nondiscrimination policies may also compel agencies to address the issue of better recordkeeping, she says. “There is more than ample evidence that there is a problem here.”