但是该论文已经引起了学术界关于性别和指导的更广泛的讨论。对于宾夕法尼亚大学生物工程学教授丹妮尔·巴塞特（Danielle Bassett）而言，促使论文被撤回的方法论上的担忧远非其最严重的罪过。她本人已经研究了引文的作法，发现神经科学领域，与男性资深作者的论文被引用的比例异常高-主要是因为其他男性科学家优先引用它们。她认为，建议年轻女性尝试与男性合写论文是一个严重的错误。她说：“这是指责的问题。” “我们有责任创造一种科学文化，让学生选择适合自己的导师。”
研究人员已经证明了对选择共享人口统计资料的导师的科学家的好处，尽管这些好处可能比学术上的情感更诱人。一项研究于2017年发表《美国国家科学院院刊》（National Academy of Sciences）上，显示工程学中有女性导师的女性更有可能留该领域，并拥有更大的归属感，尽管她们的成绩并不比同班同学的成绩更好与男性导师。 2011年发表《社会问题杂志》上的另一项研究发现，STEM领域的同性别和同种族导师对成绩没有影响，但学生们认为拥有与他们相似的顾问很重要。
匹兹堡大学企业管理，心理学以及公共和国际事务教授奥黛丽·穆雷尔（Audrey Murrell）说：“指导不仅仅是打开大门。” “这是要让人们感到宾至如归。这是关于发展他们，是为整个人提供食物。”
不难想象为什么女性会偏爱女性导师，以及为什么女性导师会帮助她们留自己的领域。科学界的妇女面临着男人很少遇到的障碍：性骚扰，产妇歧视以及基于性别的能力丧失，仅举几例。自己经历过这些障碍的顾问可能最有能力为年轻的科学家提供帮助。 UNC教堂山心理学和神经科学副教授克里斯汀·林德奎斯特（Kristen Lindquist）说：“很难知道您没有要跨越什么障碍。”去年秋天，他对《自然通讯》的论文发表了回应。 （她和她的合著者认为，研究人员没有发现关于引文模式的新知识，并且他们忽略了同性别指导下众所周知的好处。）
同样，男人可以提供出色的支持。 “男人不应该出于对妇女将无法做好工作的误导而害怕向妇女提供自己的潜导师，”医学生物伦理学和社会科学中心主任Reshma Jagsi说。密歇根大学。
这种负担可能会使少数族裔科学家更难以管理他们的其他承诺。妇女已经承担了学术部门的大部分服务工作，例如担任委员会委员和担任行政职务。 Lindquist说，这样做的部分原因是性别定型观念。她说：“作为一个女人，我的同事们可能希望我更加关怀，成为一个支持者，等等。” “因此他们可能更有可能让我担任这些职务。”例如，如果要求一个系中唯一的女教授为每位女研究生提供指导，这个问题只会变得更糟。
When a group of researchers at NYU Abu Dhabi published a paper in Nature Communications last fall suggesting that young women scientists should seek out men as mentors, the backlash was swift and vociferous. Countless scientists, many of them women, registered their indignation on Twitter—some even penning open letters and their own preprints in response. The original paper had found that female junior scientists who authored papers with male senior scientists saw their papers cited at higher rates. But a number of critics contested the assertion that this result established a link between male mentors and career performance. Scientists routinely coauthor articles with people who are not their mentors, they argued, and citation rates are just one metric of achievement. In response to these criticisms, the authors eventually retracted their paper. (They declined to comment to WIRED.)
But the paper had already stirred up a broader discussion about gender and mentorship in academia. For Danielle Bassett, a professor of bioengineering at the University of Pennsylvania, the methodological concerns that prompted the paper’s retraction were far from its worst sin. She herself has researched citation practices and found that, in neuroscience, papers with male senior authors are cited at a disproportionately high rate—primarily because other male scientists preferentially cite them. To suggest that young women should therefore try to author papers with men is, she believes, a grave error. “That was a problem in assigning blame,” she says. “The onus is on us to create a scientific culture that lets students choose a mentor that’s right for them.”
Creating such a culture is no easy task. Men dominate the upper echelons of the sciences. Even in fields like psychology, where women make up the majority of students in undergraduate and graduate programs, men held two-thirds of full professorships as of 2014. In engineering, that number rises to 88 percent. So young women and other minority scientists face a conundrum that most men never need to consider—should I work with a mentor who looks like me, or work with a mentor who has a big name?
Researchers have demonstrated benefits for scientists who choose mentors who share their demographics—though those benefits may be more emotional than academic. One study, published in Proceedings of the National Academy of Sciences in 2017, showed that women in engineering who had female mentors were more likely to remain in the field and felt a greater sense of belonging, although their grades were no better than those of classmates with male mentors. Another study, published in Journal of Social Issues in 2011, found that same-gender and same-race mentors in STEM fields had no effect on grades, but that students felt that having an adviser who was similar to them was important.
“Mentoring is not just about opening the door,” says Audrey Murrell, a professor of business administration, psychology, and public and international affairs at the University of Pittsburgh. “It’s about making people feel welcome. It’s about developing them, it’s about providing for the whole person.”
It’s not difficult to imagine why women might prefer female mentors, and why female mentors might help them remain in their fields. Women in science face obstacles that men rarely do—sexual harassment, maternity discrimination, and dismissal of their abilities on the basis of their gender, to name a few. An adviser who has herself experienced these obstacles is probably best positioned to support a young scientist as she navigates them. “It’s really hard to know what hurdles you didn’t have to jump over,” says Kristen Lindquist, an associate professor of psychology and neuroscience at UNC Chapel Hill who authored a response last fall to the Nature Communications paper. (She and her coauthors argued that the researchers had discovered nothing new about citation patterns, and that they had neglected the well-known benefits of same-gender mentorship.)
And it can be invaluable to have a supporter “in whose shoes you walk, or are likely to walk,” says Nilanjana Dasgupta, director of the Institute of Diversity Sciences at the University of Massachusetts Amherst. In her research, Dasgupta has found that same-gender peer mentors, who are close in age and career stage to their mentees, can have particularly salutary effects, perhaps because it’s easier for mentees to recognize themselves in those advisers. “Peer mentors who are just a couple of years more senior than the women they mentor are particularly inspiring, because their success seems more achievable,” she says.
Murrell says that her peer mentors have proven to be her most influential advisers; having progressed through their careers in tandem, they have all been essential sources of support for one another. “We move together throughout our career,” she says. “We’re sharing resources, sharing opportunities, sharing information, and providing support as we move.”
That said, a woman scientist is not necessarily a good mentor for a young woman solely by virtue of her gender. Bassett recalls that on one occasion, she was asked by another woman scientist to name the man responsible for her success. On another, a female PhD program interviewer asked her whether she planned to have children—and told her that, if so, it wouldn’t be worth the money to fund her. “I’m pretty sure that’s illegal,” Bassett says.
And, by the same token, men can provide excellent support. “Men should not be afraid of offering themselves as potential mentors to women out of a misguided sense that they won’t be able to do as good a job,” says Reshma Jagsi, director of the Center for Bioethics and Social Sciences in Medicine at the University of Michigan.
Granted, it will require effort on the part of a male professor to learn about the issues affecting a woman in his lab. But Lindquist thinks that men can, and should, take on that responsibility—perhaps by electing to work with minority mentors themselves. “It’s just as useful for a male student to work with a female mentor, or a white male student to work with a Black male mentor, so they can learn how power-based differentials in science impact different identities,” she says. “In turn, they’ll maybe understand how to adjust their mentorship to account for those challenges in their own students someday.”
Bassett, too, points out that gender isn’t the only thing that matters in adviser relationships. “[The] goodness, humanity, humility, respectfulness of the person, generosity of the person, are more important,” she says. But she emphasizes that the preferences people express for finding an adviser who is like themselves are real and important. “There’s clear evidence that, for many graduate students, having a shared gender with their mentor is something that’s important to them and allows them to succeed in ways that they couldn’t otherwise, because they have a role model,” she says. “They have someone they can talk to about what the biases are and how to respond to them.”
But supporting women and minorities in the sciences can’t be as simple as making sure that everyone has a mentor who is a demographic match. As underrepresented as women still are as students in many STEM fields, the situation is far worse for professors. That imbalance creates a math problem. “If every woman needs a woman to be their mentor, and there’s only one senior woman in the department, she ends up having to be a mentor to half a dozen people,” Jagsi says.
And white women are still more likely to find a match in academia than women of color, who are sometimes surrounded by only white or male colleagues. The situation is similar, if not worse, for young trans scientists. “Being the only one is really much more toxic than being one of a few,” Dasgupta says.
Even if a young female scientist is able to find a supportive woman to advise her, she will face challenges that her male colleagues won’t necessarily have to contend with. Women—and especially women of color—receive less grant funding than their male counterparts and are less likely to hold prestigious positions. Even at Ivy League schools, female professors receive 25 percent less funding from the National Institutes of Health than men. So a student who intentionally chooses a female mentor could be missing out on the benefits of working in the lab of someone who is well funded and connected. Her white male classmates, if they choose to work with mentors who look like them, can benefit from “falling money from the trees, because they’re near a tree that’s white and male,” as Bassett puts it.
Bassett and others are quick to point out that funding and professional connections don’t only arise from traditional mentorship relationships, though they often can. “It’s basking in reflected glory,” Dasgupta says—a symptom of proximity to a powerful scientist. Jagsi and Murrell prefer to call this sort of relationship “sponsorship,” to clarify that the support is more financial and instrumental than emotional or pastoral.
Traditionally, graduate students in the sciences will have one official mentor—their PhD adviser—in whose lab they will complete their research. In this model, a young scientist receives mentorship and sponsorship from one person. But it’s also possible to obtain both kinds of support from a variety of sources: postdoctoral fellows, other professors, and peers, to name a few. So Jagsi advocates for what she terms “mentor networks,” in which students can benefit from the resources afforded by all of these individuals.
Expecting one person to serve all of a student’s needs, says Murrell, might never have been realistic. “Mentoring is not just one thing—it’s a whole range and portfolio of different types of relationships,” she says. And, she continues, success isn’t just a function of good grades and highly cited articles. “You can’t just look at mentor quality as access to a publication,” she says. “You have to look at the extent to which that person is doing well, and is thriving, and feels included, and feels welcome.”
Since sponsorship doesn’t necessarily take much time—it can be as simple as facilitating a connection between a student and a potential collaborator—sponsors can be easily incorporated into mentor networks. And with the options afforded by a network, a woman need not choose between someone who looks like her and someone who can help her make advantageous connections—she can have multiple supporters. But on the flip side, the networks could pose problems for those who provide the more time-consuming forms of aid. Sponsorship may be easy, but helping someone navigate discrimination is not—and, since women and people of color are underrepresented among science faculty, there are fewer people to share that burden.
And that burden could make it more difficult for minority scientists to manage their other commitments. Women already shoulder the majority of the service work in academic departments, like serving on committees and taking on administrative roles. Part of the reason for this, says Lindquist, is gender stereotyping. “As a woman, my colleagues might expect me to be more caring, to be more of a support figure, and so on,” she says. “So they might be more likely to put me in those roles.” This problem would only get worse if, for example, the sole female professor in a department were called on to provide mentorship for every woman graduate student.
So mentor networks are not enough. It’s also crucial to spread academic service work more equitably among faculty members. “That may be as simple as just keeping a list of who’s doing what,” says Lindquist. And, she says, that time spent mentoring and supporting the department—which takes away from a professor’s own research, the chief determinant of whether or not they get tenure—should be more highly rewarded, so that mentorship doesn’t reduce women’s career prospects and keep them out of the highest ranks of academia.