Editor’s Note: For decades, bias and prejudice stalled progress for women in science, technology, engineering and mathematics (STEM). In recent years female scientists, engineers and mathematicians have caught up to their male colleagues in many respects although gender discrepancies in the number of tenured faculty and senior scientists persist. Here researchers Stephen J. Ceci, Donna K. Ginther, Shulamit Kahn and Wendy M. Williams explore the possible reasons for a gender gap in salary where it exists.

Gender differences in salaries for academic jobs in science and engineering could lead women to leave the sector. A pay gap is not present throughout academic science, however. In 2010 in only six of 24 fields were salaries of males significantly greater than those of females: assistant and full professors in economics, life science assistant professors, associate and full professors in engineering and the physical sciences, and full professors in geoscience.

Economists have many competing explanations for sex differences in salaries. Women’s decision to spend time in child care may be directly related to their salaries. Hundreds of studies have identified a “child salary penalty” for women in the labor market as a whole, at the same time identifying a marriage and child premium for men. Using data from the National Science Foundation’s Survey of Doctorate Recipients through 2001, one of us (Ginther) examined the economic explanations for gender differences in academic salaries in science and engineering fields and found that children and marriage did not explain the substantial gender salary gap present at that time. Nor did including measures of productivity appreciably reduce the gap. Other work published in 2012, however, suggests that marriage-based salary penalties in science, engineering and math are explained by married women publishing fewer papers.

Children also make a difference in academic biomedicine, explaining half of the male–female salary gap, according to a 2012 paper by two of us (Kahn and Ginther). A recent study by other authors also showed that stopping the tenure clock had no significant effect on the probability of being promoted at a large public university but it did significantly reduce salaries. Taken as a whole this literature suggests that when children lower publication output or cause gaps in careers, they seem to negatively affect women’s salaries.

In some cases even single, childless women continue to earn less than men in the same field and rank and the gender gap increases as careers unfold. This leaves open the possibility that bias may be playing some role in the remaining gender salary gap found in some fields.

In her 2014 presidential address to the American Economic Association economist Claudia Goldin of Harvard University showed that women disproportionately place a premium on flexible work conditions, which results in lower wages and slower promotion. Her analyses suggest that women’s status in science may be the result of personal choices and time-flexibility preferences as opposed to gender differences in human capital and sex-based salary discrimination. (The premium for flexibility, however, is relatively less in science than in the corporate and legal worlds, she argues.) And in a 2009 study educational psychologist Camilla Benbow at Vanderbilt University showed that men and women in top graduate programs in STEM fields expressed the same preferences for flexibility in their future work schedules, but by their mid-30s, the women with children were much less like men (with or without children) and women without children with respect to preferring to avoid long hours and wanting flexible schedules. Time flexibility is thus a priority for at least a portion of women who excel in science.