By Margaret Harris
Women in science are more likely than men to have smaller families than they would like because of their demanding academic careers – but men are more likely to be unhappy about it. This is the striking conclusion of a new study that also demonstrated a strong link between concerns about family size and a desire among junior researchers of both sexes to leave science altogether.
The study, which was conducted by two Texas-based sociologists, Elaine Ecklund and Anne Lincoln, is unusual in that it examines the effect of scientific careers on men as well as women. Some of the similarities they found are as intriguing as the differences. For example, although scientists who have children work fewer hours per week than those who do not, the mothers in the data set were working as long as the fathers, averaging 54.5 and 53.9 hours per week respectively. Male and female faculty members were also equally likely (16.6% vs 17.1%) to report being somewhat or strongly dissatisfied with their lives outside work. Among graduate students and postdocs, Ecklund and Lincoln found no significant gender differences in respondents’ career satisfaction or the number of children they had.
However, some of their other results make sobering reading for those concerned about the “leaky pipeline” for women in science. Although male and female grad students and postdocs reported similar levels of career satisfaction, and were almost equally likely to seek jobs in industry or as research scientists, a gender gap opened up in the numbers seeking a tenure-track academic position. While 66.5% of male students said they wanted a faculty role, only 60.1% of women agreed; for postdocs, the gap was larger, 84% to 69.2%. And among survey respondents who had already made it to the top of the academic pyramid, women were somewhat more likely than men (15.% to 11.5%) to report being dissatisfied with their working lives.
To uncover some of the reasons behind these differences, Ecklund and Lincoln used a statistical technique called logistic regression to estimate the degree to which one variable (such as being female) predicts another (such as life satisfaction), while adjusting for other influences like age, race, income, marital status and publications. This part of their study focused on faculty members, and perhaps unsurprisingly, they found that career satisfaction is a good predictor for life satisfaction (and vice versa). Women, older people and those who were married were likewise more likely to report being happy with their non-work lives.
One of the strongest predictors of life dissatisfaction for both men and women was feeling that they’d had fewer children than they’d wanted to because of their scientific careers. At first glance, it might seem like this would hurt women more, since the study found that female faculty members had significantly fewer children, on average, than their male counterparts. Women were also almost twice as likely as men (45.4% to 24.5%) to link their smaller families to the demands of their careers, and to be unhappy about it.
However, since women were still, overall, more likely to be satisfied with their lives, Ecklund and Lincoln came to the striking conclusion that “having fewer children than wanted has a more pronounced effect on life satisfaction for male scientists”.
That’s an important result, because as the authors note in their abstract, it provides evidence that “the impact of science on family life is not just a woman’s problem”. It’s also a problem for men – and for society as a whole, since concerns about family size also appear to be driving early-career scientists out of the field. When they examined data gathered from grad students and postdocs, Ecklund and Lincoln found that those who had fewer children than they wanted were respectively 21% and 29% more likely than their peers to be considering careers outside science. Moreover, concerns about family size trumped several other factors – including the fear that a science career would preclude having a family at all – in predicting interest in non-science careers.
To obtain their data, Ecklund and Lincoln sent surveys to a random sample of 3455 US-based grad students, postdocs and faculty who were associated with the country’s leading astronomy, physics and biology PhD programmes. Their study was based on responses from 2503 people, with a roughly even balance between disciplines and career stages. More details can be found in their paper “Scientists want more children”, which is published in the open-access journal PLoS ONE.