A code to keep away judges, juries and MPs

Social scientists must regulate the ethics of their research or risk having rules imposed on them, says Ron Iphofen

The ethics of scientists have gained increasing prominence in the past few years, and Lord Winston's recent statements on public engagement with science only extend the debate about the consequences of their work. Surprisingly, social scientists appear to have had less to say about their engagement with the public than their colleagues in the physical sciences. But, over the past year or so, this has been changing.

Without the intensity of the debates that have beset biomedicine - such as those over genetic modification, the measles, mumps and rubella vaccine or organ retention - it may be only a matter of time before a social researcher makes an error of ethical judgement that treads sufficiently on a respondent's rights to raise public concern. In an era of increasing litigation, such an error could cost a research organisation dear and could have a big impact on which research gets funding. Moreover, the ensuing "contamination of the field" could damage the whole research community irreparably.

Some problems have already emerged. One case that could have major consequences for the research that university departments might consider "safe" and, therefore, permissible is that of Russel Ogden, the Exeter University PhD student who was recently compensated financially after departmental ethical approval for him to conduct research on assisted suicide and euthanasia was rescinded for fear of legal consequences. Some criminologists have also expressed concern about the government's censorship of research that questions the effectiveness of its policies.

And internet research that blurs the boundary between public and private behaviour - such as the use of discussion group archives as data for naturalistic observation studies - raises a host of ethical complexities.

Social scientists have also become more aware of the need to consider the potential threats to vulnerable research participants such as the physically or intellectually disabled, the ill, the old and the young - all sectors of the population subjected to increasing research attention. The recent national conference of Ireland's state-run National Disability Authority was devoted to establishing ethical guidelines, and it gave special attention to the problems of "over-researching" a vulnerable community.

There is concern about maintaining standards in social science throughout the European Union. One EU-funded project, Respect, has emerged to meet a perceived need to "professionalise" socioeconomic research across the region. It is essential that existing standards are disseminated through and practised in the new states that join the EU.

In the UK alone, the legislative minefield of human rights, data protection and intellectual property is one in which all must tread warily. Few of the laws have been tested, and one only has to be aware of cases concerning the protection of journalistic sources, such as the David Kelly affair, to realise the potential explosiveness of what lies just beneath the surface.

The Human Rights Act, which raises special concerns for social research conducted by public bodies, has led to a series of reviews of ethical guidelines. Both the Cabinet Office and the Scottish Executive have had internal discussions about ethical issues in government social research.

The Department of Work and Pensions has been receiving feedback on draft ethical guidelines for consultation, and professional associations such as the British Sociological Association have updated their guidelines in the past few years. The Office for National Statistics instituted an internal review some time ago. More recently, the Economic and Social Research Council has been funding a review of research ethics that will have an impact on the awarding of grants.

Even some of the independent research companies have been considering setting up institutional review boards. The Nuffield and the Joseph Rowntree foundations will require that independent ethical review be conducted before grants are awarded and funds released.

Until now, a system of research governance that is open to public scrutiny has been seen as the best means of ensuring high standards of ethical behaviour. Under this system, ethical review committees take the role of gatekeepers of social research, ensuring that the quality and independence essential for good research is balanced against the rights of, and obligations to, those being studied. If, however, decisions are influenced by institutional fear of litigation, research independence will become increasingly threatened.

Social science researchers have been particularly loath to being subsumed under the now highly systematic ethical review procedure established by the Department of Health and the National Health Service. This is not just to do with being "precious" about the profession of social science. The overly bureaucratic health ethics committees have never been particularly well informed about social science research designs and methodologies. More important, they are caught between the protection of the NHS organisations they are acting for, the disparate interests that they represent and the maintenance of general principles that protect human rights. Their insistence on the need for risk assessment of research "interventions" suggests that the balance of their consideration is towards minimising the potential for litigation within the NHS.

It is vital, however, that the social science community learns from the experience of the health ethics committees, which grew without any effective central national coordination. Projects rejected by one committee could be approved by another. Because their voluntary membership rotated, there was no collective memory of decisions taken. There was little systematic guidance for understanding the many research methods available for health research other than uncritical faith in the "gold standard" of the randomised controlled trial. The establishment of a Central Office for Research Ethics seems to have remedied all of that.

There is a danger that the plethora of attempts to create ethical codes, institutional review procedures and professional standards in social science will repeat the errors of the early NHS system. Each group producing the proposed systems is meeting the requirements of its own commissioning bodies and is liaising in an ad hoc way. This means that wheels are being reinvented, and there is a danger that we will end up with a confusing multilayered process of ethical review that inhibits and perhaps even prevents the progress of good social science.

We need a centrally coordinated UK-wide system - such models already exist in Norway and Canada. This would also help with standardisation of ethical research procedures across Europe and would reduce duplication of effort and of error. Researchers across the government, public and private sectors should have no problem in signing up to it if led by a recognised independent body such as, say, the Academy of Social Sciences.

Behaving ethically while producing social research of a high standard requires constant awareness of the ethical consequences of one's work. This must be embedded in initial research training and reinforced in continuous professional development. There is a wealth of experience on the setting of ethical research standards in social science, but it is being applied disparately. It would be better if social science research regulates itself properly to prevent inadequate regulation being imposed by external agencies or, worse still, in response to case law.

Ron Iphofen is a senior lecturer in the faculty of health at the University of Wales Bangor. He led the review of the Social Research Association's ethical guidelines and is scientific consultant on Respect ( www.respectproject.org ). The Association of Research Centres in the Social Sciences will hold a seminar on ethics with the SRA ( www.the-sra.org ) at the Policy Studies Institute on January 23.