Both sexes continue to believe men are better at maths than women and so more suited to science-related professions, an American study has found.
When scientists set up a mock job selection experiment and asked interviewers to select candidates they felt would perform best at a maths task, men were twice as likely as women to be chosen when the selection was based solely on appearance.
The bias against women occurred regardless of whether the interviewer was male or female.
The discrimination against women was reduced but not eliminated when interviewers were given more information about the candidates' abilities, such as how well they performed in the task, before making their selection.
Despite no evidence that men were better at maths than women, the study reported a "pervasive" perception that women were inferior at these tasks.
"This stereotype can lead to a decreased demand for women in STEM [science, technology, engineering and mathematics] fields and/or a reduction in the number of women choosing to specialise in this field," lead researcher and finance professor Luigi Zingales, from the University of Chicago, said.
In the US, women outnumbered men in undergraduate enrolments but were much less likely to major in mathematics or science, said Professor Zingales, whose findings were published in the journal Proceedings of the National Academy of Sciences.
In Australia, similar numbers of both sexes study science at university, but women were more likely to choose a biology and chemistry degree than an IT or engineering course. Men also dominate senior or leadership roles.
Sharon Bell, the interim vice-chancellor of Charles Darwin University, said while there were clearly hurdles for women working in science-related fields, her and other research had found these challenges appeared to be more complex than just direct discrimination against woman.
"We've just done a very large survey of colleagues in chemical and biological sciences and there is very little evidence of direct discrimination," she said.
Instead, the accumulation of small advantages that favoured men but aversely affected women, which were based on an organisation's dynamics and decision making, were to blame for a gender imbalance in science fields, Professor Bell said.
Lyn Yates, a professor of education at the University of Melbourne, said while stereotypes around the capabilities of one sex over another did exist, the study's methodology did not reflect how scientists were selected for jobs in the real world.
"By the time people are getting into careers in science their results and achievements in maths and science are clear," she said.
Although many people would describe themselves as unbiased, the US study showed incorrect perceptions still persisted.
"It can be hard to show people that they are biased. Most people would say they employ on merit."