Perhaps it was her training as a chemist that led Donna Nelson to insist on filling out every space in her table for identifying the gender and ethnicity of each person in the top 50 departments nationwide for 14 academic disciplines. Or, maybe it was her Native American heritage, which emphasizes visual ways of perceiving patterns and understanding the world. Either way, her thoroughness brought some key information to light, proving that women and minorities are dramatically underrepresented on the science faculties of some of the most prestigious schools in the country.

"The remarkable thing Donna did was simply collect the simplest data ... to answer a very simple and very important question ... . When people saw [her results] they were shocked," said Nancy Hopkins, a professor of biology at the Massachusetts Institute of Technology who chaired a committee examining possible gender biases at MIT. Nelson's data had "an enormous impact," Hopkins said.

Nelson had posted a chart on her office door that illustrated the very small representation of women as faculty among the top chemistry departments. Students asked her whether she had any stats for minorities.

So she used her salary to hire a student to collect demographic information about the 50 US chemistry departments with the highest research expenditures. Next, between 2001 and 2004, Nelson and her students surveyed faculty race or ethnicity, gender, and rank within the "top 50" departments for physics, math, chemical engineering, civil engineering, electrical engineering, mechanical engineering, computer science, political science, sociology, economics, biological sciences, and psychology.

Nelson then compared her data to statistics compiled by the National Science Foundation on PhD and BS attainment. She found that fewer women and minorities were represented on faculties than would have been expected if assistant professors - presumably, those most recently hired - represented the diversity of the doctoral pool. People were taken aback.

In biological sciences, for example, women comprised 44.7% of doctoral recipients between 1993 and 2002, but only 30.2% of the assistant professor hires. White males, meanwhile, garnered 43.2% of the doctorates and 55.4% of the jobs.

Nelson observed that some disciplines, such as chemistry, have a high representation of women among BS recipients and tend to lose many women both after they get their BS and after they get their PhD. Conversely, other disciplines such as engineering have low participation in undergraduate programs but good retention afterwards. These different disciplines, therefore, need to do better outreach at different levels. "Programs need to be data driven," Nelson says. "It's amazing how well data can help you formulate a good plan."

Nelson was born in Eufaula, Okla., the capitol of the Creek nation. After her doctoral work at the University of Texas and postdoctoral studies at Purdue, she joined the University of Oklahoma (her alma mater) in 1983, where she is an associate professor of chemistry. She has won several awards, including a Guggenheim in 2003, for her work with addition reactions of alkenes. She studies the ways in which different groups bonded to alkenes can increase or decrease the rates of reactions; she is now looking at similar reactions involving single-walled carbon nanotubes.

As a young scientist and mother, as well as the only female and Native American in the department, Nelson could not find any living role models to help her balance competing demands in her life. Somewhat wistfully, she reports turning to Marie Curie. Nelson's deep desire is that minority students will not have to turn to history for role models.

Nelson is currently writing up the results from a second survey, and she foresees tracking the information perhaps one more time. "Bench chemistry is my job and diversity is my hobby," she explains, "and I think my hobby is taking over my life!"