Women with astronomy PhDs are leaving the field before landing a faculty job at a rate three to four times higher than are their male counterparts, a study of crowdsourced hiring data in the United States has revealed.
The results support evidence that women in the field experience systemic hurdles such as hiring biases and harassment, and also confirms anecdotal reports that female astronomers leave the field more frequently than their male peers. The work1, by astronomer Kevin Flaherty at Williams College in Williamstown, Massachusetts, was posted on the arXiv preprint server on 2 October.
“Women leaving the field faster than men is something that has been discussed anecdotally for some time, but it is always useful to have it corroborated with a study such as this one,” says Natalie Gosnell, an astronomer at Colorado College in Colorado Springs.
Flaherty collected hiring data from the Astrophysics Jobs Rumor Mill, a website where astronomers can anonymously aggregate information about the status of open fellowships and faculty positions, including who is being shortlisted or receiving offers.
Looking at hires from 2010 to 2017, he found 245 reports of tenure-track faculty positions in US universities that went to 157 men and 88 women. By Google-searching the year each astronomer received their PhD, he found that the women landed their faculty jobs, on average, 4.2 years after graduation — significantly shorter than the 5.3 years, on average, that it took men. Yet a 2014 survey found that female academic astronomers make up just 26% of the field.
To try to explain what was happening, Flaherty created a model of the labour pool with hires and departures. In an attempt to recreate the gendered differences in the crowdsourced hiring data from 2010 to 2017, he ran three scenarios: one, that more women were receiving astronomy PhDs over time; two, that female astronomers were more likely than men to be hired; or three, that women were leaving the field at a higher rate than men.
He found that the third model best explained the data: more specifically, one in which women were leaving the labour market at a rate three to four times higher than men. In this scenario, because more women are likely to leave the field as the job hunt drags on, those who are hired quickly are over-represented in the data, and so the average length for those who do get hired becomes shorter.
His simulation ruled out the hypothesis that women are more likely to be hired faster than men because institutions are trying to encourage diversity. To recreate the gap, a woman would have to be 10 times more likely to be hired than a man — and if that were true, 70% of astronomy assistant professors should have been women by 2013.
All bases covered
Gosnell says she was impressed that the paper included — and then conclusively disproved — that diversification model. Yamila Miguel, an astrophysicist at Leiden Observatory in the Netherlands, said that the result is very important in reassuring newly hired female faculty members. “It is not true that you got the job because of gender issues — you got it because you are good at it.”
At the same time, the fact that women leave the field at a much higher rate than men points strongly towards the systemic bias against women in the field, given evidence shows that women and men are equally skilled in these areas. Flaherty’s analysis suggests that structural barriers combine to drive women out of academic jobs — including reported biases in hiring and awarding telescope time, higher workloads and sexual harassment.
There are some limitations to the study, says Anna Watts, an astrophysicist at the University of Amsterdam, including its US-centric focus. The Rumor Mill often does not include interdisciplinary jobs, such as those in astroparticle physics, she says. But despite these caveats, she found the study’s conclusions convincing.
They also match her personal experience, she adds: the toll of chasing postdoc and faculty positions around the world — while also living with a partner and wanting a family — can become too great.
Gosnell points out that women of colour face extra hurdles not captured by the study. Focusing only on gender can be problematic because it ignores the experiences of these women in astronomy and astrophysics, she says.
Flaherty acknowledges those limitations, saying his paper is a small contribution to a much larger body of work that has been built up by many others focusing on under-represented and marginalized groups in astronomy.
The paper also notes that the analysis did not consider race or ethnicity because the number of astronomers of colour in the data was so small that they would no longer be anonymous.