Bias in organizations may not just come from the top

Women are woefully underrepresented at the top echelons of organizations. In 2018, for instance, just 5% of Fortune 500 companies had a female CEO.

When researchers have studied the phenomenon, they have often investigated the women who fail to rise into leadership positions. What barriers might be preventing them from assuming top jobs? But that leaves out questions about what happens to women who do make it to the top. Might some of the difficulty they face be due to how others in the organization perceive their leadership?

In new research, Kellogg finance professor David Matsa and colleagues explore whether subordinates are resistant to working for a woman. The researchers focused on the dynamic between female principals and the male teachers who work for them. They focused on schools in part because of the rich datasets available in public education, and the uniqueness of employee turnover in schools, where departures are almost always the teachers’ decision and not their bosses’.

They found that male teachers are more likely to leave schools that have female principals, while female teachers display no greater rate of departure from female-led schools.

“The results are robust,” says Matsa, who teamed up with Amalia R. Miller, an economist at the University of Virginia, and Aliza N. Husain, a PhD student in education policy. “They hold true across schools, within a given school, and over time.”

The researchers point out that this is happening in a field that is predominately female.

“Male teachers opt into a female-dominated workforce,” the authors write, “implying that these individuals are likely to be more accepting of female leaders than men in other, more male-dominated professions. We would expect therefore that males’ preferences for male leaders are larger and more deep-seated in other professions.”

Teacher Departures Matter
Staff turnover anywhere can be problematic. But teacher departures have particularly far-reaching implications for the schools the teachers are leaving. Multiple studies have shown that teacher turnover hurts student achievement. And hiring and training new teachers cost districts—and thus taxpayers—precious funding at a time when many schools are strapped for cash.

“So teacher turnover is one of the metrics that school districts and principals are looking to get a handle on,” Matsa says.

Public schools also represent particularly fertile ground for research due to robust data and unique employment conditions.

“There’s a lot of data on teacher–principal pairings and relationships, which makes it easier to study,” Matsa says. Because of tenure policies and strong unions, “forced dismissals are rare, and even untenured teachers are rarely fired. So we’re able to get a clearer window into voluntary decisions workers make, including those related to the gender of leaders.”

A Matter of Principal
The researchers studied 40 years’ worth of administrative data from New York State’s public schools. That included data on almost 650,000 teachers across nearly 6,400 schools from 1970 to 2010.

The team found that male teachers are 12 percent more likely than female teachers to leave schools led by a female principal. This is true across analyses of elementary and secondary schools over time.

In contrast, female teachers display either no difference in departure rates depending on the gender of their principal, or are slightly less likely to turn over under a female principal.

These results remained consistent when the authors studied each individual school over time. “When a school has a female principal, there is more turnover than when it has a male principal,” Matsa says.

The pattern persists for both tenured and untenured teachers, suggesting it isn’t a matter of untenured teachers being forced to leave, but rather teachers exercising a preference.

Moreover, data from New York City schools show that male teachers are more likely than their female counterparts to request transfers from schools led by women to those led by men, and are more likely than female teachers to leave the state’s school system entirely when they work for a female principal.

Bias and Career Opportunity
So why do male teachers seem to prefer male principals? The researchers explore two potential explanations.

The first has to do with the possibility of bias against female leadership.

“Work in social psychology shows that leadership stereotypes can motivate people to favor male leaders over female ones,” Matsa says. For instance, stereotypes of leaders as being physically very strong may sway people to favor male leaders—“even if that’s not a relevant characteristic for the quality of a principal.”

But the researchers’ data did not include information about how male teachers regard female leaders. So, as a proxy, the team used data on how those attitudes have changed at a societal level over time, as well as how they may differ across geographies.

National surveys show that men have grown more accepting of female leaders. So the researchers divided their data into two periods—from 1970–1989 and from 1990–2010—and found that, indeed, male teachers are more likely to leave a female-led school in the first two decades than in the second two. In other words, as cultural bias against women has decreased, so have male teachers’ preferences against working for a woman.

Next, the researchers looked at Census data to identify parts of New York where women are more versus less prevalent in the workforce. In places where more women work, the researchers reason, there is likely less bias against women in the workforce. And, indeed, the pattern of male-teacher departures is stronger in parts of the state that have a lower rate of female participation in the workforce.

These two findings, though correlational, are consistent with bias playing a role.

The second explanation that the researchers explore has to do with career opportunity.

“Some male teachers may not think female principals are worse leaders, but they may see better career prospects in working for a male principal,” Matsa says. “So they might request a transfer to a male-led school to get what they see as a ‘better deal’ there.”

The researchers find some evidence consistent with this explanation, but it is quite modest.

Under female principals, male teachers have fewer additional responsibilities—such as department head or club advisor—than their female colleagues. This leads male teachers to earn 1.2 percent less and female teachers to earn 0.5 percent more under female principals, compared to male principals. However, these percentages translate to a relatively small amount of actual cash—less than $1,000 annually. So the researchers speculate this likely has little to do with teachers’ departure decisions.

Similarly, male teachers are less likely to be promoted to a principal position when serving under a female principal than under a male leader—though the likelihood of any given teacher being promoted to principal is very small.

While neither of these measures of career advancement fully explain the findings, they may be indicative of something else that male teachers are reacting to that cannot be seen in the data.

“It is possible that they hint at broader differences in the quality of the work environment, level of support, and investment in professional development that vary with principal gender,” the authors write.

Multiple Ways Forward
It’s hard to know the full set of implications of male-teacher turnover under female principals.

“I can’t tell you if the teachers who are leaving are fantastic,” Matsa says. “But they’re tenured. They’re experienced. The additional turnover might harm students’ achievement, especially if the teachers that replace these male teachers are less experienced.”

Additionally, if male teachers are leaving schools, that means fewer role models for the boys enrolled there.

Matsa points to two ways of addressing the issue. “To combat bias, schools could train teachers to recognize their implicit biases,” Matsa says, similarly to initiatives taken by the Department of Justice and companies, including Starbucks. “As more women enter leadership roles, the bias might also dissipate on its own.”

In terms of male teachers’ concerns about career opportunity, solutions could range from providing better mentorship networks to creating development programs for male teachers in female-led schools.

Beyond education, the results point to broader concerns about how female leaders are perceived by male subordinates, and how organizations might be codifying these biases. For instance, a corporate HR policy that penalizes managers when their direct reports leave the company could be problematic.

“If you use something like a 360-degree performance review and evaluate business leaders based on their subordinates’ assessment,” Matsa says, “you need to keep in mind that biases might be present. Any instrument you’re using to measure performance should be unbiased.”

[This article has been republished, with permission, from Kellogg Insight, the faculty research & ideas magazine of Kellogg School of Management at Northwestern University]