Women more drawn to hypothetical jobs where most employees were women

Learning that a workplace employs mostly women made women more interested in hypothetical jobs and more likely to expect that they would belong there, according to four studies published in PLOS One. The findings suggest that existing representation may encourage future representation, and the gender makeup of a workplace sometimes mattered more to women than whether a job ad stressed exceptional brilliance.

Women remain underrepresented in fields such as science, technology, politics and academia. One explanation is that some professions are associated with “brilliance” or natural intellectual ability, qualities that people often stereotype as more masculine. Previous research has found that messages emphasizing brilliance can reduce women’s interest in certain educational and professional opportunities.

Stella F. Lourenco of Emory University and colleagues examined another possible influence: how many people of one’s own gender already work in a field. Because fields seen as requiring brilliance also tend to have fewer women, it has been difficult to tell which of the two is driving women’s choices. The researchers conducted four studies involving 1,680 U.S.-based participants recruited through the online research platform Prolific.

The first study included 96 participants (average age 30.7 years; 60 women). Each participant estimated the proportion of men and women in 30 occupations drawn from a list of 60 and rated how much they felt they would belong in each one. They also rated how much each occupation required innate talent, how competitive and cutthroat its culture seemed, and how similar they felt to the people who work in it.

The second study included 194 participants (average age 36.3 years; 95 women). They read a fictional job advertisement that emphasized either brilliance or dedication. The participants then estimated the gender balance of the workplace and rated their interest in applying and their expected sense of belonging.

Free daily newsletter

The third study included 399 participants (average age 33.5 years; 200 women). This time, the advertisements explicitly stated that either 75% of employees were men or 75% were women. The advertisements also emphasized either brilliance or dedication.

The final study, the only one whose design and analyses were registered in advance, included 991 participants (average age 39.5 years; 496 women). It varied both which gender formed the majority and how large that majority was, using ratios of 60% to 40%, 75% to 25% and 90% to 10%.

In the first study, participants’ average estimates closely matched actual occupational gender distributions, suggesting that people are generally aware of which occupations are more male- or female-dominated. Women reported a stronger sense of belonging in occupations they expected to have more women, even after accounting for beliefs about brilliance, perceived workplace culture and perceived similarity to workers. Men’s sense of belonging was not linked to their estimates.

In the second study, participants assumed both jobs employed more men than women, estimating about 40% women on average. Those who expected more women in the workplace reported more interest and belonging, and once their estimates were taken into account, the type of ad no longer predicted either outcome. Expected representation still predicted interest after accounting for perceived workplace culture and similarity, but no longer predicted belonging.

The third study provided stronger evidence because gender representation was directly manipulated. Women showed more interest in female-dominated jobs than in male-dominated ones, and when a job was female-dominated, their interest matched men’s. They also reported greater belonging in female-dominated workplaces, even when the job advertisement emphasized brilliance. Men’s interest and belonging did not depend on whether the workplace was mostly men or mostly women.

Women even reported greater belonging in a female-dominated job that emphasized brilliance than in a male-dominated job that emphasized dedication, suggesting that representation could outweigh the wording used to describe the qualities needed for success. Messaging still mattered within female-dominated jobs, where women anticipated more belonging when the ad stressed dedication. In male-dominated jobs, the type of ad made no difference to women’s sense of belonging.

The fourth study showed that women’s responses generally grew stronger as the gender imbalance became larger. Women reported more interest and belonging when women formed a 75% or 90% majority than when men did, but showed little difference at 60% to 40%. When men were in the majority, women anticipated less belonging than men did, a gap that widened with the imbalance, while women’s interest and belonging in female-majority jobs matched men’s in male-majority jobs. Men showed some preference for workplaces in which men were the majority, although this pattern was less consistent.

Messages about dedication still mattered. In the fourth study, women, but not men, were more interested in jobs that stressed effort and commitment rather than natural brilliance. However, once perceived workplace culture and similarity were taken into account, the type of ad no longer predicted belonging, while the effects of gender representation held.

Across the studies, perceptions of a competitive, cutthroat workplace culture and of how similar participants felt to the employees were also strong predictors of interest and belonging. The researchers suggest that gender ratios may serve as a signal of what a workplace’s climate will be like, including how supportive or competitive it is. They argued that improving representation and changing workplace messages may be most effective when pursued together.

“Altering beliefs about brilliance for success in specific professions, though important, may be insufficient for increasing female representation in fields where women are underrepresented,” the authors concluded.

There are a few things to keep in mind. The studies used hypothetical job advertisements, so participants’ ratings may not predict real applications or career decisions. Only the fourth study was preregistered, and the effects, while consistent, were generally small to moderate in size.

The study, “Gender representation begets gender representation,” was authored by Stella F. Lourenco, Kylee Novick, Ryno Kruger and Arber Tasimi.