Publication ethics issues in the social sciences
Background
The history of research ethics in general and publication ethics more specifically was initially and primarily grounded in the biomedical sciences. As concern over issues of animal care, human participant protection and research integrity developed throughout the latter part of the 20th century, some members of the Social Sciences and Humanities communities raised concerns that the cultures in their particular disciplines differed and that a ‘one model fits all’ was inadequate to evaluate research ethics and publication ethics.
Social Sciences
Most quantitative Social Science research involves research methodology that is similar in scope and nature to that found in the natural and biomedical sciences, including hypotheses testing, statistical analysis and is nested within long recognized scientific norms of investigation. This is also true with respect to quantitative studies with human subjects and participants. In many countries, the approval of research using either animal or human participants requires the researcher to obtain prior institutional research ethics approval before undertaking this type of research. Some researchers disagree with the requirement of prior ethics approval saying that their research is low risk and needs no review. These researchers sometimes do their work under the auspices of community groups and avoid institutional ethics review.
As well, there are disciplines which object to all such requirements or find them problematic to their particular fields of study. Many of the complaints concerning the need to recognize difference come from fields including: Ethnography, Anthropology, Sociology, Criminology, and some sub-disciplines in Psychology, Education and Social Work. This list is illustrative rather than exhaustive.
Questions
- Some Social Scientists have objected to institutional research boards on the grounds that such approval is heavy-handed and inappropriate as a tool to evaluate their research. They conduct their research through community groups and avoid institutional approval. How should editors handle such studies and manuscripts?
- What should editors do when they receive manuscripts from Social Scientists who have completed the research without institutional ethics approval?
- What guidelines do editors require to assess manuscripts where authors argue that their fields should be evaluated using very different criteria than is standard in terms of research ethics and research integrity than other fields of inquiry?
- Should publishers develop policy to deal with the claim by researchers that Social Science is different from other sciences and needs to be evaluated without ethics review?
- Should university exemptions be sufficient for editors to consider publishing manuscripts which have been deemed exempt from ethics review?
- How should editors determine whether to publish research on illegal activities and illegal behaviours?
References
Eysenbach, G. and Till, J. (2001). Ethical Issues in Qualitative Research on Internet Communities. BMJ, 323 (7321), 1103-1105.
Hoeyer, K., Dahlager, L. and Lynoe, N. (2005). Conflicting Notion of Research Ethics: The Mutually Challenging Traditions of Social Scientists and Medical Researchers. Social Science & Medicine, 61, 1741-1749.
Katz, J. (2006) Ethical Escape Routes for Underground Ethnographers. American Ethnologist, 33 (4), 499-506.
Kitchin, H. (2007). Research Ethics and the Internet. (Fernwood Press: Nova Scotia).
Stanton, J. and Rogelberg, S. (2001). Using Internet/Intranet Web Pages to Collect Organizational Research Data. Organizational Research Methods, 4 (3), 200-217.
Comments
Our approach is to require all authors to include an ethics statement in the body of their submission. This will be published as part of the paper if it is accepted. If there is no ethics approval, or the authors have not observed a recognised set of ethical guidelines then they need to say so.
If I or the reviewers felt that the research was unethical then I would exercise my perogative to reject the submission.
to post comments
Personally, I do not think it should be within the purview of editors of social science journals to require evidence of institutional ethics approval - their exclusive concern (and that of reviewers) should be whether the research was conducted ethically based on what they ascertain from the manuscript itself. In my view, the expansion of journal editors' mandates to include requirements for evidence of institutional ethics approvals is problematic in three respects:
1) It encourages a box-ticking approach to ethics, whereby editors and reviewers focus not on the ethics of the research itself but on evidence of formal approvals;
2) It assumes that the line between what requires institutional ethics review and what doesn't is consistent across and within countries (it isn't) and that the boundaries between these categories are clear cut (they aren't);
3) It is promoting ethics creep. For example, in Canada, there is anecdotal evidence that academics wanting to publish straightforward program evaluations (which don't require institutional ethics approval under the national guidelines) are submitting them for institutional ethics approval anyway, because they assume they can't publish them otherwise.
It's worth bearing in mind that various countries are now considering ways of scaling back the institutional ethics oversight of social science research (e.g. the proposed changes to the US system) and the COPE requirements make this process much more complicated and difficult.
to post comments
My experience with IRBs (both submitting projects for approval and sitting on IRBs as a faculty member) has been that the model is not suited to all kinds of research. Most are very comfortable with the biomedical model and understand what has to be in an informed consent, what are appropriate and feasible methods, who are protected populations, etc. as long as it has something to do with a medical treatment, drug, or device. As a sociologist, trying to get approval for ethnographic studies or focus groups, the IRB didn't want to hear about it. First, you get the argument that sociology isn't a science anyway, so it doesn't matter what you do. Then, they tell you how you could do the study "more scientifically" (i.e., a quantitative analysis). Admittedly, my experiences improved over the years and now there are some really good university IRB websites that try to address the concerns of all researchers.
That said, in my role as editor, I've had submissions from researchers and faculty who claim that their projects met IRB exemption guidelines because it was a program evaluation or an economic analysis; yet, the data were clearly protected data (Federal government, Tribal Health Service database, or personal electronic health records) that have their own requirements for safeguarding privacy. In some cases, publishing a paper with these types of data could get the authors, the editor, and the publisher in a lot of trouble.
These questions could be answered at a discipline level and by the appropriate journals in that discipline. The problem comes when authors submit to journals outside their discipline and when in fact, the findings could cross into and be meaningful for other disciplines. Do we want to stifle that cross-pollination of ideas? I don't.
So far, I've dealt with these issues on a case-by-case basis and tried to use common sense. I also go to the top journals in certain fields, particularly sociology, for guidelines on requirements for submission of research manuscripts. I would be happy to see editors of journals in social sciences contribute thoughtful explanations of what to consider in evaluating research in their fields. Even if IRB approval is not required and the authors make a good case for their claim, I would probably ask the authors if they had clearly informed administration they would be publishing their findings. I have found that some institutions say that evaluation or QI research does not require IRB approval because it is not meant to be disseminated. The findings are meant to inform change in a specific location and it may not have occurred to the administration that their shortcomings will be aired in a public forum such as an article. In these cases, some kind of oversight is required in my opinion.
to post comments
I find Charon's IRB experience quite alarming and something I came across to a slight extent in UK NHS ethics committees some time ago. The developing awareness of the differences between the social and the natural sciences has been good to see and dependent to some extent on health professionals working increasingly with a social science perspective.
But I would agree with others and want to counsel caution about insisting on formal REC approvals given the wide range of funding sources and methodological approaches in the social sciences. I convene an ethics forum for the UK Social Research Association and we often get independent researchers conducting commissioned work but with nowhere to go to gain 'formal' approval. At the very least they receive advice and guidance from our Forum and can announce this on seeking publication. I also do ethics review for the European Commission and we have supplied several documents advising reviewers about the special characteristics of ethnographic and anthropological approaches - the prime one being that since it is hard to anticipate ethical dilemmas given the emergent and developmental nature of such work (even when both qualitative and quantitative data are being collected), the one-off review for formal approval can miss the spontaneous problems that can emerge when researchers are 'in the field'.
As Editor of 'Quality in Ageing and Older Adults' I looked for 'ethical awareness' in the submissions I received as much as formal review. Indeed there were many of the service evaluation form and they are by no means devoid of ethical implications.
to post comments