Cracking the code of ethical research practices

Publishing practices in the biomedical and social sciences commonly fails to conform to Australian codes of practice. AAP

In February, Minister for Innovation, Industry, Science and Research Kim Carr said Australian taxpayers could be confident the research activities they fund meet the “highest ethical and moral standards”.

Why was the Minister so confident? Because the Gillard Labour Government has established an Australian Research Integrity Committee (ARIC).

The establishment of the Committee was welcome news. For too long, Australia has pretended that our research practice is sound.

In recent years, several high profile cases acted momentarily to shake this confidence: allegations of misconduct, including guest authorship practices, against UNSW Professor Bruce Hall in 2002; the resignation of Monash Professor David Robinson after accusations of plagiarism; and allegations that a Sydney endocrinologist might be one of many who fronted journal articles ghost authored by Wyeth pharmaceuticals.

Only one of these allegations was ultimately upheld: however, a qualitative study (published in February 2010 in Social Science and Medicine) carried out in two Australian universities, but reflecting a wider experience, would suggest that there are systemic problems at least in the publication of research, as does anecdotal report from research leaders at the 2011 Quality in Postgraduate Research Conference.

We would suggest that publishing practice, particularly with respect to authorship, which does not conform with the Australian Code of Responsible Conduct of Research is common in the biomedical and social sciences in Australia, although this is far from a uniquely Australian issue: similar behaviour has been recorded internationally.

Specific cases of misconduct which have reached public scrutiny have been few and those that have done so have been through media publicity.

Whistleblowers do poorly and it is possible that most researchers walk away from such confrontations but it is also not in the interest of institutions to publicise such issues.

Judging the state of research ethics in Australia on the basis of identifiable cases is almost certainly illusory but it is difficult to gauge the extent of the problem without better evidence.

For some years now, researchers and institutions have been under pressure to ‘perform’ and to be accountable for the public funding provided to them for research.

Performance measurement, in many disciplinary areas, is built around assessment of ‘quality’ of journals, the number of publications, and the citation rate of those publications relative to the ‘norm’ within the discipline.

The introduction of the research quality assessment framework, Excellence in Research in Australia (including recent amendments to it) has shifted emphasis particularly to the last of these and reinforced the need for authors to pursue strategic publishing practices to maximise bibliometric success.

The ways in which research performance is measured can undermine ethical behaviour in research publication.

At the very least, in the present ERA climate, there are incentives to pursue strategic publishing practices which shift the focus of publication away from research dissemination and towards bibliometric factors.

In particular, there are incentives under the ERA framework to maximise the number of citations. Self-citation is counted in the analysis, so that for researchers and institutions to self-cite would seem sensible as would reciprocal ‘cite-your-mate’ approaches.

It is possible that coalitions may form to cross-cite papers. The Australian Research Council has already considered this matter and ruled that self-citation and ‘cite-your-mate’ practices are not significant enough to influence overall citation outcomes and therefore will not be excluded, although it is not clear on what basis this decision was made.

These measures are part of the ‘game’ and each time the ‘game’ changes in terms of the expectations of the performance framework, researchers scramble to adapt.

Kim Carr’s recent Ministerial Statement acknowledged this adaptive practice and that the use of journal rankings within the ERA had acted to distort publication practice.

The introduction of the ERA seems unlikely to change the norms of behaviour in authorship practice. Without effective intervention strategies, authorship will continue to be used as a bartering chip in the ‘commerce’ of research and in many cases will continue to reflect the exercise of ambition and power differentials in academia.

Distortions in publication practice such as citation gaming and inappropriate attribution of authorship are part of a continuum of ‘normal misbehaviour’ which, at its extremes, include fabrication, falsification and plagiarism.

The Australian Code for the Responsible Conduct of Research has been an attempt to regulate some of the more egregious excesses and it does provide disempowered junior researchers with some leverage in publication conundrums.

Yet, monitoring and enforcement is rare and, four years on from publication of the Code, many institutions have yet to comply fully, particularly with respect to staff ethics training, ethical authorship practice and storage of research data.

The Code is, in part, an acknowledgement that there are considerable shortcomings to any system which places pressure on researchers to perform and sets up output targets when there are clear imbalances in power and opportunities for individuals and institutions to misuse that power.

Carr has suggested that it is an institutional responsibility to regulate and inspire: that institutions “must provide the education to ensure their staff behave ethically and do not foster ‘negative research cultures”.

It is self-evident that we need a strong system of research governance to counteract the pressures and perverse incentives placed on researchers but where is the incentive for researchers and institutions to act?

Beyond the Code we need a supported dialogue about the ways in which institutions can work to create cultures of integrity and it is possible that the Code itself needs further work.

For example, there may be room for auditing systems to support good practice, probably with associated incentives and sanctions.

Good practice could also be supported through independent institutional advisors in research ethics, improved researcher training in research ethics and strong statements about unacceptable research practice.

The Australian Research Integrity Committee can provide an avenue for the prosecution of egregious unethical behaviour, however, Mr Carr should understand that it cannot guarantee that the research activities the public funds “meet the highest ethical and moral standards”.

We will need to do far more before we can say that.