BOSTON—A theme that came up repeatedly during a symposium on improving the representation of women and minorities in the science pipeline was what sociologist and civil rights activist W. E. B. Du Bois called "twoness": the idea that whites and minorities perceive and experience two different worlds, two different sets of expectations and circumstances. Achieving ethnic, racial, and gender parity in the science pipeline will require measures that help white and minority graduate students, and their mentors, better understand the world they each live in, the panelists said.

Rashawn Ray, a social psychologist at the University of Maryland, College Park, noted a "systemic and numerical disadvantage of women in higher education," especially women of color. Although women outnumber men at the undergraduate level and earn more bachelor's degrees than men do, the situation reverses sharply further along in higher education and in tenure-track jobs in academia. While outright racist suppression is relatively rare, Ray said, women of color face a "chilly climate" in graduate school.

One of the chilliest factors, he said, is the difficulty that women of color have in locating mentors who share their experiences and who can help them navigate the academic world. Ray challenged himself to figure out what mentoring support women of color are lacking.

Ray and his colleagues broke down mentorship into six advising styles:

  1. Intellectual: contributes to research goals and not much else.
  2. Instrumental: teaches "survival skills" such as grant writing and tips for getting published, helps build professional relationships, and serves as an advocate for the student.
  3. Available: is always around to listen to concerns and needs.
  4. Affective: tends to socioemotional needs.
  5. Exploitative: uses students as lab labor and not much more.
  6. Respectful: treats students' ideas with respect and consideration.

Ray surveyed graduate students to find out which areas were missing. Exploitative advisers aren't common among any demographic, Ray said. But the survey showed that women of color are less likely than white men, white women, and minority men to report having primary advisers who embody both the instrumental and respectful advising styles. In other words, while their advisers mostly satisfy their needs for research guidance, availability, and emotional support, they—the advisers—lagged behind other mentors in teaching academic survival skills and respecting the ideas of their protégés. White men are about 1.5 times more likely than minority women to report that their mentors respect their ideas. Because white men, white women, and black men don't report feeling slighted when it comes to respect and instrumental support, they build much more robust relationships with their primary mentors than women of color do, giving them an advantage in career advancement, Ray said. What that amounts to, Ray said, was a "mentoring glass ceiling."

However, Ray said, women of color are more likely than other students to report having secondary advisers who respect their ideas and teach them how to navigate academia. Although it's good that women of color find such support in their secondary mentors, these professors are often less influential in their field than their primary mentors.

To improve the mentorship experience for women of color, Ray said, departments need to change their cultures to encourage all mentors to offer the whole range of positive advising styles. He added that while minority students often attend workshops to help understand how to interact with their mostly white colleagues, white scientists and students rarely receive information on how to interact with minority colleagues; such back-and-forth could improve relationships across the racial spectrum, Ray noted.

Another speaker—Crystal Bedley, a sociologist at Rutgers University in Piscataway, New Jersey—discussed a National Science Foundation–supported program at her university called the Women of Color Scholars Initiative that for 4 years has been offering mentorship and support to minority faculty members in the form of tenure and promotion workshops, town hall meetings, writing workshops, and teaching training.

"We're seeing increases in the number of faculty of color, but they're pretty modest increases," Bedley said. Other positive effects may be difficult to measure. One reason, she said, is that more faculty members are declining to report their race so the percentage of scientists of "unknown" race is growing. That's a confounding variable that sociologists and other researchers who follow issues with minority scientists will increasingly contend with, she said.

Michael Price is a staff writer for Science Careers.

10.1126/science.caredit.a1300022