A logo created by the American Geophysical Union to promote its efforts to prevent sexual harassment.

American Geophysical Union

Geophysics society hopes to define sexual harassment as scientific misconduct

A major U.S.-based scientific society is on the verge of expanding its definition of research misconduct to include sexual harassment.

The American Geophysical Union (AGU) is making the change to emphasize the serious threat that harassment and other forms of discrimination pose to the scientific enterprise, say society officials. But some experts on research ethics question the wisdom of merging two categories of undesirable behavior and worry that it could complicate efforts to promote ethical conduct. AGU’s pending move could also revive a bitter debate in the 1990s that led to the slimmed-down definition of research misconduct currently used by U.S. government agencies and universities.

Harassment “damages the scientific enterprise, and that’s why we went the step of defining [it] as scientific misconduct,” says Michael McPhaden, a former AGU president and chair of the society’s ethics committee, which has drafted the society’s new guidelines. AGU’s current code of ethics does not cover harassment and other behavior, he notes, and members’ reactions to several recent cases of sexual harassment by prominent scientists in other fields forced the society’s hand. Members have “let us know loud and clear” that AGU should be explicit on its expectations for behavior, adds McPhaden, a senior scientist at the National Oceanic and Atmospheric Administration’s Pacific Marine Environmental Laboratory in Seattle, Washington.

Sexual harassment and other forms of discrimination should not be tolerated, agrees Mark Frankel, former longtime head of the scientific responsibility program at AAAS (which publishes ScienceInsider) who is based in Gaithersburg, Maryland. But he thinks such behavior shouldn’t be lumped in with scientific misconduct. Misconduct relates to science as a profession, he says, whereas sexual harassment, bullying, and other behaviors are not only inappropriate in science, but in the larger society. “I think there is a line between them,” Frankel explains, “and I would prefer seeing each of them on one side of the line or the other rather than seeing them incorporated.”

Alan Price, a former federal misconduct investigator, also sees a practical reason for keeping the two forms of misbehavior separate. The expertise needed to investigate research misconduct cases is very different than investigating sexual harassment cases, says Price, who spent 17 years at the Office of Research Integrity, which handles misconduct by federally funded biomedical researchers. By putting the two in the same category, says Price, now a consultant in Austin, “my concern would be that [the investigation] wouldn’t be done well.”

A fourth category

The current federal definition of research misconduct covers fabrication and falsification of data and plagiarism. For many years, however, the National Science Foundation included a fourth category, “other serious deviations from accepted practices,” that allowed it to investigation a range of inappropriate activities.

Many scientists objected to that fourth category, however, arguing that it was too vague and could also suppress novel approaches to research that went against the grain. A 1992 report by the U.S. National Academies of Sciences, Engineering, and Medicine recommended a narrower definition, and in 2000 the U.S. government adopted that approach.

The draft revision to AGU’s current policy on scientific integrity and professional ethics explains why harassment should be grafted onto the conventional definition of misconduct. “Scientific misconduct also includes unethical and biased treatment of people. … These actions violate AGU’s commitment to a safe and professional environmental required to learn, conduct, and communicate science …”

Under AGU’s guidelines, anyone can file a complaint. Any allegation that cannot be resolved by its staff would go to a committee created to investigate the complaint. If federally funded research is involved, AGU will notify the alleged perpetrator’s employer and determine whether AGU or the institution should carry out the investigation.

AGU plans to follow the same process in investigating sexual harassment that it uses when questions are raised about possible research misconduct stemming from articles published in its journals. But officials recognize that people filing sexual harassment complaints may need an extra layer of protection.

“After filing a complaint with AGU or with their home institution, a complainant may request that AGU provide protections from harassment, discrimination, or bullying at AGU activities,” the draft policy declares. “Such actions may include, but are not limited to: barring the respondent from a complainant's talk, barring a respondent from an AGU activity, or providing the complainant with an escort during AGU activities. If the complaint goes to a full investigation at AGU or at the home institutions, AGU may consider further actions.”

At the same time, the investigative process would require both the complainant and defendant to participate in a meeting (held in person, by phone, or through the internet) in which the allegations are presented and discussed. But some misconduct experts expressed concerns about that arrangement. They say that such a confrontational approach, combined with the lack of anonymity for the complainant, is quite unusual and could deter many alleged victims from filing a complaint.

A misconduct investigation is “not a trial, it’s an effort to get at the truth,” says Lauran Qualkenbush, research integrity officer for Northwestern University in Evanston, Illinois, who emphasizes that she is not familiar with the AGU guidelines and has no experience investigating sexual harassment. “The act of bringing forward an allegation [of research misconduct] can be very intimidating, and there’s often a power differential between the two parties as well. So we do everything possible to keep them apart.”

Qualkenbush, who is also vice president of the fledgling Association of Research Integrity Officers, notes that universities have different mechanisms for handling the two types of misconduct. “If a society were to come to me and say, ‘We have an allegation of research misconduct involving one of your employees,’ my first question would be, ‘Does it meet our definition of research misconduct?’ And if it involved sexual harassment, my response would be, ‘That should be handled by the sexual harassment officer at the university, not me.’”

AGU officials admit that the society lacks a track record of investigating allegations of sexual harassment and other forms of discrimination. But that doesn’t mean such behavior is not occurring.

AGU received four complaints about improper conduct during its 2016 annual meeting last December in San Francisco, California, McPhaden says. AGU officials declined to provide details, citing concerns about confidentiality. AGU is also “in litigation” on a case involving a member playing an active role within the society, CEO Christine McEntee said last month in testimony before a panel of the National Academies that is examining sexual harassment in academia.

AGU members have until 28 April to submit comments on the proposed changes. AGU expects the changes to be finalized in September and go into effect immediately.

Updated, 4/7/207, 11:46 a.m.: This story has been updated to clarify that AGU is based in the United States. The group also has international members.