Open up to research integrity

Is a lack of integrity a problem in universities?

Inge Lerouge: I wouldn't define it as a problem, but something we need to be aware of. Research integrity must be taken seriously. Reports tell us that only a very small percentage of researchers are committing serious fraud, such as fabrication or falsification of results, but we have to keep raising awareness about integrity with our researchers, and insisting that integrity needs to be a part of the culture at our universities. This requires continuous work, since new staff arrive all the time and we host researchers from all over the world, often for a short amount of time.

What is at stake if universities do not get this right?

Ton Hol: Trust in all kinds of institutions, including universities, is no longer self-evident, so it is more important than ever to pay attention to the issue of integrity and show that the outcomes of research can be trusted. If there is a serious case of fraud, it can damage the image of science. 

Inge Lerouge: Trust in other researchers is also very important, especially in international and interdisciplinary collaborations. If researchers cannot trust their research partners in other locations, or from other disciplines, then we have a problem in science. 

So how can universities demonstrate they are trustworthy?

Ton Hol: First of all, it's very important to show we have integrity by achieving good research outcomes. Then, if there are serious cases, we should to be open about it. Transparency is very important. You have to show that you take any complaints seriously, and act when something is wrong.

Inge Lerouge: The shift towards open science is also important. Sharing your data, as long as there are no ethical or legal restrictions, can make the research process more transparent and also more open to scrutiny, increasing opportunities for other researchers to verify results. Universities should also be transparent about their institutional policies, so that it is clear for example who can be contacted in case of an allegation of research misconduct. And incentives should stimulate good research practice, for instance focusing on the quality of research during evaluations, not the number of publications alone.

Poor quality research is also an integrity issue in your paper. Why is that?

Ton Hol: The outcomes of research projects can easily be criticised because of sloppy science, which again undermines its trustworthiness. So this is very important.

Inge Lerouge: Questionable research practices may have a lower impact on an individual basis than serious and deliberate misconduct, such as faking results, but many small instances of sloppy science or questionable research practices can add up to become a much larger systemic problem. So these practices should be improved through education and training.

Are students and young researchers being taught enough about integrity?

Inge Lerouge: In our paper we say that integrity training should be mandatory for young researchers, such as PhD candidates, but a bigger challenge is to reach more senior researchers. Our PhDs often used to ask why they have to have training on research integrity while their supervisors did not, and they were right. It's very important that supervisors give the right example and provide good mentorship, and that we give a consistent message on integrity. Therefore, we have introduced a compulsory training module for all new professors at KU Leuven since last year.

Are there differences between the disciplines when it comes to integrity?

Inge Lerouge: The challenges may vary between disciplines. For instance, image manipulation is much more of an issue in the life sciences than in other disciplines. More generally, the size of a research group is often larger in the sciences than in the humanities, making it more challenging for a principal investigator to keep effective oversight. 

Ton Hol: It's also important to acknowledge that there are differences between disciplines with regard to standards of integrity, but we have to see if these differences are justified or not. 

Inge Lerouge: One well-known example is the sequence of authorship on a research paper. In the biomedical field the first and the last authors are the people who contributed most, while in other disciplines the contribution level decreases with the position of the author. And in mathematics alphabetic listing is standard. The important thing is to be transparent about the contribution each author made to the publication, and many journals now ask for a precise description of that contribution.

How could universities better handle allegations of misconduct?

Ton Hol: In our paper we leave the decision about how to handle allegations to individual countries or universities. However, we think that standing committees have an advantage, because their members build up experience and knowledge for handling this kind of complaint. The idea of a fair trial is also very important, because you can easily damage people if you make mistakes or when things go wrong in handling a case. 

Inge Lerouge: Learning from each other is important, to get the experience that Ton just mentioned. And sharing experience and good practice is something we do in the LERU network.

If a university has a case of fraud, can it still be trusted?

Ton Hol: Yes, a university can still be trusted, but it has to handle the case well and have good training to prevent further cases from happening. If it doesn't, this can damage the reputation of the institution and trust in science.

Inge Lerouge: Research misconduct has happened in the past, and it will happen in the future, simply because researchers are only human. In our paper we call for universities to be open and transparent about the number and type of allegations, and this should be applauded rather than criticised.

Read the LERU paper 'Towards a Research Integrity Culture at Universities'

Text by Ian Mundell - Pictures by KU Leuven/Rob Stevens and Utrecht University

© LERU