Research institutions ‘pay lip service’ to integrity rule

While thousands of Australian researchers say they are hamstrung by unethical practices in their workplaces, many are unaware of a network of helpers – including some of the helpers.

A “census” of “research integrity advisors” (RIAs) across 99 universities and medical research institutes has found that many are advisors in name only, with 32 per cent never assisting students and 22 per cent never assisting anybody. The typical advisor devotes half a day a month to the role, helping one staff member and one student over the course of a year.

Thirteen percent of RIAs have never been trained and some only realised that they had been appointed advisors when they were contacted by the census team. Almost two thirds of institutions offer no information about their RIAs on their websites – including those that nominate RIAs as the “first point of contact” for outsiders concerned about research integrity.

The findings have been reported on the Open Science Framework preprint repository. Lead author Adrian Barnett, a statistician at Queensland University of Technology, said it was astonishingly difficult to find information about research integrity on some institutions’ websites.

He said the organisations covered by the study undertook research involving community participants, some of whom would have found it “very hard” to report any integrity concerns. “You need websites that clearly show you’re an institution that takes research integrity seriously and, if anybody has any issues, who they should contact,” he added.

Under the Australian Code for the Responsible Conduct of Research, universities and other research organisations have been required to employ RIAs for the past 16 years. Their task is to advise and support colleagues before integrity issues escalate.

But the study found that their contribution was often undermined by a “tick-and-flick” mentality, with some institutions “not providing training, not updating their advisors’ details, not informing researchers that they are an advisor [or] having no advisors at all”.

Such a laissez-faire approach stands at odds with staff concerns. Recent Australian studies have found that poor research practice is commonplace, with 56 per cent of respondents acknowledging a “significant” reproducibility crisis in a 2019 survey by the National Health and Medical Research Council (NHMRC). 

Another study found that bad research practices affected 47 per cent of early career researchers, with internal complaints about research misconduct or harassment “actively ignored or swept under the rug”.

Professor Barnett said perceptions of research misconduct were widespread, yet there was only a “very small” disciplinary caseload: “The big concern is that a lot of this stuff currently is not getting detected.”

He said the first thing that researchers learned was “do the experiment and listen to the data. It’s not do the experiment and then change those results because you don’t like them. If that’s becoming part of day-to-day practice for a lot of people, that’s a real concern.”

The recent Australian Research Council (ARC) review recommended amendments to the council’s act to provide a “legislative basis” for additional functions such as upholding research integrity.

The review found that existing arrangements – whereby individual researchers and institutions take responsibility for research integrity, with their activities monitored by the Australian Research Integrity Committee of the ARC and NHMRC – were failing in “limited instances”. Legislative change could support “other arrangements that have a broader remit”, the review found.

john.ross@timeshighereducation.com