In Brief

From an employer's perspective, two heads are often better than one. But for women in the work force, the ever-more-ubiquitous group project may have a serious downside. A study in September's Journal of Applied Psychology (Vol. 90, No. 5) finds that when men and women work together on a task-particularly a stereotypically "male" task involving leadership and decisiveness-both male and female outside observers devalue the women's contribution relative to the men's.

"These results don't make us particularly happy," says study author Madeline Heilman, PhD, a social psychologist at New York University. "They demonstrate another way in which stereotypes against women can work in a subtle way to hinder success."

In the study, Heilman asked 60 college students -35 women and 25 men-to read what the participants thought were detailed notes taken by an observer watching two people work together to develop an investment portfolio. The participants also viewed photos of the fictional workers and reviewed background information about the workers' jobs and educational backgrounds. For half the participants, the first worker was clearly female and the second worker was clearly male; for the other half of the participants all information remained the same, but the workers' genders were reversed.

Later, Heilman asked the participants to evaluate one of the workers' individual contributions to the project. On average, participants rated the female worker as less competent than the male by more than 1.5 points on a nine-point scale, and less influential by more than one point. Both male and female participants showed this bias: Heilman eventually collapsed the two groups' data together when she found no significant differences in their responses.

The study is not all bad news for women, though. In two follow-up experiments, Heilman found ways to counteract the intrinsic sex bias. In the second experiment, the participants were told that each of the workers had a very specific task to complete-in this case, researching the tax laws of two different states-and it was clear that both tasks were necessary to complete the project. In the third experiment, participants read very specific information about both workers' past successes on the types of tasks at hand.

In both these studies, the extra information was enough to counter the participants' biases, and they rated the male and female workers as equally competent.

"It seems that whenever there is any ambiguity about the performance of an individual, these biases come into play," Heilman says. "But if it's made very clear that the woman is competent, then those negative expectations are undercut from the start."

-L. Winerman