Most would acknowledge that women and minorities already face more hurdles in academia than their white, male peers. A lack of mentors, occasionally overt discrimination and the academy’s poor work-life balance, are well-documented issues. But now a study has suggested that these groups may be at a disadvantage even before the starting whistle sounds.

A study published on 22 April (and currently under review) looked at how likely faculty were to respond to a request to meet with a student to informally discuss potential research opportunities — a scenario picked as a proxy for the many informal events that could boost an academic career and which fall outside institutions’ formal checks and balances. They found — overwhelmingly — that professors of all groups were more likely to respond to white men than women and black, Hispanic, Indian or Chinese students. Academics at private universities and in subjects that pay more on average were the most unresponsive.

Katherine Milkman at the University of Pennsylvania in Philadelphia, along with colleagues Modupe Akinola of Columbia University in New York and Dolly Chugh of New York University, sent fake e-mails to 6,548 professors at 259 US institutions, pretending to be students wanting to discuss research opportunities before applying to a doctoral programme. The messages were identical, bar their fictional authors, whose names were picked for being recognizable by gender and ethnicity — ‘Steven Smith’ representing a white male, for example, and ‘Latoya Brown’ for a black female.

White men were more likely than women and minorities to receive a reply in every discipline except the fine arts, where the bias was reversed (see ‘Biased teachings’ above). Business showed the greatest disparity, with 87% of white males receiving a response compared to just 62% of female and minority students. In the sciences, faculty in engineering and computer sciences, life sciences and natural, physical sciences and maths all showed significant biases against minorities and women.

Broken down by group, the results were more nuanced. Asian students experienced the greatest bias, despite research showing that stereotypes about Asians in academia are generally positive, says Milkman. Among private university faculty the response rate for white men was 29 percentage points higher than for Chinese woman — the greatest disparity observed. Meanwhile in the natural and physical sciences and maths there was a small, though not statistically significant, bias in favour of Hispanic women.

The study found no relationship between representation of any group among faculty in a given discipline and the degree of bias that students faced when trying to interact with them. This means the findings cannot be attributed to the largely white, male academy preferring to associate with others like them, says Milkman. “One of our hypotheses was that more diverse departments would be less biased and we just don’t see it,” she adds. The only exception was among Chinese faculty, who were less likely than other faculty to discriminate against Chinese students.

Curt Rice, a professor at the University of Tromsø in Norway and head of Norway’s Committee for Gender Balance in Research, says that the result that women and minorities are as biased as white men is not surprising. They mirror a 2012 study, by researchers at Yale University in New Haven, Connecticut, which showed that science faculty of both sexes show unconscious biases against women in hiring and pay decisions. The problem — says Rice — is implicit rather than explicit bias. “We’re talking about the absorbed effect of cultural stereotypes that lead to the formation of biases,” says Rice. “It’s no surprise they’re held by all of us because they’re subconscious and the result of cultural stereotypes that we’re all exposed to.”

Comparing results across disciplines, the team found more intriguing effects. The more highly paid faculty are on average (by subject), the greater the difference in response rate between white male and other students. “For every US$13,000 increase in salary, we see a drop of 5 percentage points in the response rate when compared to Caucasian males,” says Milkman. She links the finding to studies that recently found that wealthy, high-status people tended to be less empathetic and more self-focused. Biases were also more prevalent in private institutions than public ones, she adds.

Although the study looks at only one tiny step in the path to a successful academic career, Rice thinks the compound effect of many situations like it could well help explain why we find so few women and people from minority backgrounds at professor level. Milkman agrees: “This is a small moment — it’s one time someone’s reaching out and looking for guidance and encouragement. But if every time you do this happens to you, that’s going to add up.”