Neuroscientist Eve Marder urged universities not to cap biomedical PhD numbers. Credit: Mike Lovett/Brandeis University

Biology PhDs are struggling to find academic jobs, and some say it is because there are too many graduate students. So when a senior scientist wrote that universities should resist the urge to stem the flow of new PhD students into biomedical programmes, the online world took notice. Scientists on social media are also talking about a rarity: a published study that failed to replicate the author’s own work.

Eve Marder, a neuroscientist at Brandeis University in Waltham, Massachusetts, argued in the journal eLife that it is hard to predict who will excel in science, so any attempt to limit access to PhD programmes will inevitably exclude potential stars1. The reaction was mixed. “Reduce the number of admitted graduate students? Agree with Eve Marder: not the greatest idea,” tweeted Sergey Kryazhimskiy, an evolutionary biologist at Harvard University in Cambridge, Massachusetts.

But Mike White, a geneticist and instructor at Washington University School of Medicine in St. Louis, Missouri, argued in a blog post that Marder was “perpetuating the PhD pyramid scheme”.

Based on data from Altmetric is supported by Macmillan Science and Education, which owns Nature Publishing Group.

In her article, Marder said that both sides win when universities keep the PhD pipeline running at a high volume. The scientific talent pool stays full, and students gain valuable training that can help in whatever career they eventually find. “Society would be enriched if more of the people making decisions in industry, law, medicine, education and politics had lived through the rigors of a PhD program,” she wrote.

White, who says he has struggled to find a job that uses all of his training and skills, disagreed with Marder about the value of a PhD on a group blog called The Finch and Pea. “A life science PhD leaves you poorly prepared to get a job doing anything else,” he wrote. In an interview, he added that science as a whole suffers when too many new researchers are added to the mix. “The field becomes corrosively hypercompetitive,” he said. Those people who Marder doesn’t want to exclude will lose out anyway, he says.

Summer Allen, who has just finished a neuroscience postdoc at Brown University in Providence, Rhode Island, and is now a science writer, tweeted that she agreed with Marder’s main points, adding:

Reached for further comment, Allen said that “the glut is forcing lots of good researchers out of science, and there’s research showing that women and minorities are more likely to leave the pipeline.”

Speaking to Nature, Marder said that many scientists have thanked her for essay: “I have e-mails from department heads saying that they would have had trouble getting into graduate school today.” She adds that she sympathizes with young biologists but that life-science grad students are typically paid to learn, “and they’re free to leave at any time”.


Failure to replicate

When psychologists Sarah Brown-Schmidt and William Horton published a study2 that cast doubt on the findings of a previous one, it wasn’t especially shocking, except for one thing: the author of that now-refuted paper was none other than William Horton himself. In 2007, Horton, of Northwestern University in Evanston, Illinois, reported that people could more quickly recall the names of objects if they were in the presence of a partner. The latest study, using a nearly identical protocol but with more participants, found no sign of this partner effect on memory. The authors wrote that “this failure to replicate suggests that the originally-reported finding, if real, may be of limited generalizability or of smaller magnitude than originally estimated.”

Online commenters praised the paper for its honesty and good scientific practice. Marianne Guenot, a postdoc at Imperial College London, tweeted: