Psst…did you hear? The effect of gossip on misconduct

If scientists are hesitant to formally report their colleagues when they suspect them of misconduct, can simply gossiping about their concerns in informal settings – at meetings, conferences, etc – clean up the literature? That’s a question Brandon Vaidyanathan and his colleagues tried to answer in “Gossip as Social Control: Informal Sanctions on Ethical Violations in Scientific Workplaces,” published last month in Social Problems. We spoke with Vaidyanathan, now the director of research at The H.E. Butt Family Foundation and Public Policy Fellow at the University of Notre Dame, about how scientists use gossip to warn others of potential misconduct – and whether it works.

Retraction Watch: What prompted you to discuss the role gossip can play in scientific misconduct?

Brandon Vaidyanathan: We didn’t set out to study gossip initially. It emerged unexpectedly as a theme during our interviews with scientists across national contexts, when we asked them about their encounters with misconduct.

A number of scholars have begun to stress the importance of looking beyond egregious forms of misconduct for which there are formal sanctions—fabrication, falsification, or plagiarism. We were interested in how scientists respond when they encounter “normal” misbehavior.

Our data come from two separate research projects we conducted. The first, funded by the National Science Foundation, was a qualitative study on ethics among physicists in China, the UK, and the US. The second project, funded by the Templeton World Charities Foundation, was on the social contexts of science in eight regions (France, Hong Kong, India, Italy, Taiwan, Turkey, the UK, and the US). Here we conducted nationally representative surveys in addition to interviews.

What spurred us to investigate the role of gossip was a paradox we found early on in our interviews. Scientists we talked to expressed great confidence in the self-policing power of the scientific community through the reputational impact of gossip. On the other hand, they also claimed that several successful scientists in their field (i.e., who were publishing frequently in high-profile journals) had bad reputations—colleagues didn’t trust the quality of their work, or warned others not to work with them. So we began to systematically investigate this phenomenon of gossip to understand when and how it worked.

RW: Can you explain how you tested scientists’ willingness to take formal action against alleged misconduct?

BV: During our interviews, we asked scientists to describe an incident in which they had encountered unethical behavior on the part of a scientist. We then probed them further to discuss what about the situation they found unethical, and how they responded. If they said they hadn’t personally encountered unethical behavior, we asked how they would respond to a hypothetical situation.

In cases of egregious misconduct, scientists were often quick to give examples of formal actions taken. These were cases where the misconduct was clear, easy to prove, and there was knowledge about channels through which one could respond.

But we also asked respondents to talk about any ethically questionable behavior, beyond simply egregious misconduct. Most of them could quickly come up with examples, and in such cases, they were often unsure about (1) whether the incident really counted as unethical, (2) how one could prove that the perpetrator had committed it, or (3) how one should respond.

For instance, you submit a grant proposal. The review is inexplicably stalled, and then you see a publication with an identical experiment to the one you proposed. You suspect, but cannot prove, that the author reviewed your proposal.

In such cases, respondents didn’t think formal action was either possible or worth the effort; it would be difficult to prove that a grant reviewer intentionally delayed responding to a proposal because he or she wanted to incorporate insights from the proposal into their own publication.

Retraction Watch: Instead of lodging formal complaints, you found researchers resort to “informal gossip” to warn people of scientists who are behaving badly. Can you provide examples of what that informal gossip can look like?

BV: Yes, in many cases, scientists believed that addressing the transgression formally would threaten their careers or could gain them a reputation as a rabble-rouser. As a result, they resorted to gossip to informally warn their colleagues.

We found two main categories of such gossip. The first is about the quality of work. Gossip circulates about scientists who are known for “sloppy” work. This allows them to publish faster, but their work gets treated with some suspicion. Respondents are unsure about whether such sloppy work should be considered unethical, but they recognize that it produces an abundance of scholarship that is difficult to replicate and build on, which hinders scientific progress.

The second category of gossip is about persons to avoid. For instance, junior scientists and graduate students warned one another about senior faculty who mistreated them, and senior scholars warned each other about potential hires whose work they considered untrustworthy.

The scientists we spoke to believe that these problems are produced by an intensely competitive culture as well as perverse incentives in scientific institutions that value networking and fundraising over producing meticulous, high-quality scientific work.

RW: You note there are three consequences of gossip about scientists’ behavior. What are they?

BV: For those who participate in gossip, we found three main consequences: it leads them to (1) avoid taking the transgressor’s work seriously, (2) avoid collaboration with the transgressor, and (3) avoid recommending the transgressor for positions. Gossip can crucially shape trust, a central factor when making decisions about collaborations—trust in the quality of the person’s work and in whether one’s colleagues in the field consider that person trustworthy as a collaborator.

RW: What are the potential flaws of using gossip to reduce misconduct?

BV: Yes, gossip can cause undue harm as well. Some scientists we talked to found themselves victims of false rumors by competitive colleagues. For instance, we came across one case in which a scientist saw a colleague spread false rumors about a potential hire whom he saw as a threat. Our respondent tried to intervene to correct these rumors, but it was too late.

RW: Based on your data, how effective do you think gossip is in controlling scientific misconduct?

BV: We conclude that scientists are unduly confident about the self-policing power of the discipline. The reputational impact of gossip seems to work when the targets of gossip are of lower status than the gossipers. It also seems to work better in smaller sub-fields where people know one another better.

But it doesn’t work when the targets of gossip are of higher status. Those in junior positions have little recourse but to warn one another informally, but such gossip has little corrective power over the transgressor’s behavior. Status and hierarchy severely limit the effectiveness of gossip as a means of controlling misconduct.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.