Under the hashtag #pleasedontstealmywork, dozens of Danish PhD students shared their experiences last spring concerning powerful researchers who use their position to gain co-authorships on papers to which they have not made a significant contribution.

As part of the INTEGRITY project, we together with colleagues from four other EU countries conducted a large-scale study investigating exactly this type of questionable authorship practices. The study has just been published in PloS One, and it shows that the #pleasedontstealmywork-stories are only the tip of the iceberg.

The study was based on a survey of 1336 PhD students. The participants were located in Denmark, Hungary, Ireland, Portugal, and Switzerland and came from a wide range of research fields. In the survey participants were asked if they had granted at least one co-authorship to a person in power even though the person had not made a significant contribution to the study. More than one third of the participants said yes.

The study also revealed major differences across faculties. The problem with guest authorships is biggest in the medical sciences, where 49% of the Ph.D. students had granted a guest authorship to a person in power. In the natural and technical sciences (STEM) it was 42%. In the other faculties, it was much less.

PhD students who had granted a guest authorship to a person in power were also asked why they had done it. To this, 49% responded that they had been told to do so by the person in power as at least a partial reason (14% gave this as the only reason). Furthermore, 46% of the participants had granted a guest authorship because they wanted to maintain a good relationship with the person in power, and 39% gave “everyone else in my field does it” as a (at least partial) reason.

A change in culture is required

There is no fast or easy solution to the problem. The study suggests that the problematic authorship practices are part of a deeply engrained culture. When we have discussed the problem with influential researchers, the responses are either that the PhD students don’t know how much the senior researchers are contributing, or that the courses in responsible conduct of research that are being offered to PhD students, post docs and supervisors at more and more universities will solve the problem over time. Some even argue that it will affect their ability to compete with peers if fewer guest authorships are granted at their institution than other universities. As we see it, responses such as these clearly show that a change of culture is required if the problem is to be solved.

Cultural changes are not easy to bring about. Although courses in research integrity may gradually change things over time, a more fundamental change in the incentive structure of the research community may also be needed. Currently, the sheer number of authorships plays a key role in the assessment of researchers, but there are other and perhaps better ways to organize assessments. The Swedish foundation Riksbankens Jubilæumsfond for instance has introduced the rule that applicants are not allowed to give their H-index, full publication lists or other quantitative measures of their research in applications to the foundation. Applicants are only allowed to send a project application, and their five most relevant publications.

This procedure incentivizes quality rather than quantity; researchers are assessed on their best publications, not the number of publications. If this kind of assessments become more widely used, they might help to create a healthier authorship culture. As it is now, the incentive to have a long publication list and a high H-index creates imbalances and undesirable side effects. Researchers who have managed to get many guest authorships will appear to have produced more than they really have, and they will be assessed more favorable than researchers without guest authorships. This means that in the current system, people who behave in an ethically problematic way get a head start in the competition. As we see it, this needs to change.

(This post is in part based on a previously published press release for the study)

Authors: Mads Paludan Goddiksen, Mikkel Willum Johansen, Thomas Bøker Lund, Peter Sandøe (all from University of Copenhagen).