Interviewer: Anthony Hoffmann

Interviewee: Elizabeth Buchanan

Earlier this month, Anthony Hoffmann (University of Wisconsin Ph.D. student and Project Assistant with the Internet Research Ethics Digital Library, Resource Center, and Commons project) sat down with Dr. Elizabeth Buchanan (Associate Professor at the University of Wisconsin-Milwaukee, director of the Center for Information Policy Research, and Principal Investigator on the same IRE project) for a friendly conversation regarding the subject of Internet research ethics.

Your work is focused in the area of Internet research ethics (IRE). Can you speak briefly about the field and your experiences in it?

I’ve been working in the area of Internet research ethics, really, since 1998. When I wrote my dissertation, it was studying online interactions in a Bioethics course in a medical school, and when I went through the Institutional Review Board (IRB), I had to explain how I was collecting data when I would never sit f2f with the participants. I had to justify issues of consent, pseudonymity, online interviewing, data storage, and the like, and at that time, the IRB was not very well versed in online research. What got me really piqued was when the course I was studying was focusing on traditional research ethics issues, justice, consent, beneficience, so, I was experiencing a very meta-level engagement with the issues while grappling with them in this online setting.

What makes the issue of Internet research ethics distinct from research ethics more broadly?

When we discuss IRE, we tend to begin with those three principles—justice, consent, and beneficience—from research ethics, rightly or wrongly so. That is where much of the work has been focused in the past ten years. How do we “fit” forms of online research into a particular paradigm, namely, a paradigm that comes from the biomedical, and is codified in the United States Belmont Report. Most of us know the history of research ethics, to some extent, but it is worth reviewing briefly. The modern traditions of codified, formalized research ethics stem from the Nuremberg Code, released in 1947, as a response to atrocities committed in the name of research during World War II. From the Code, the concept of informed consent of subjects or participants emerged as a basic premise of medical or biomedical research. The World Medical Association’s Declaration of Helsinki (first adopted in 1964), which subsequently followed the Code, included the concept of informed consent along with broader notions of human dignity and safety. Both the Nuremberg Code and the Declaration of Helsinki were strongly grounded in the medical/bio-medical perspectives, while the 1948 United Nations Declaration of Human Rights spoke to a range of basic rights beyond research specificity. Nevertheless, the UN Declaration shares in common with codified research ethics a commitment to basic rights of autonomy, protection, safety, and knowledge.

In the United States, research violations of human rights of dignity and autonomy became well known through the Tuskegee experiments, which began in 1930 and lasted for 42 years. Such ethical issues as deception, respect for persons, and disclosure were raised along research lines. One major debate that emerged specifically out of the Tuskegee experiments is the balance between individual harms and greater scientific knowledge, the latter of which was used as a rationale for the conduct of the research. This debate between the greater good of the individual versus the societal underscores the diverse consequentialist and non-consequentialist ethical approaches to research ethics.

And this dichotomy also plays out in IRE…

Yes. We still use that “dichotomy” of consequentialist and non-consequentialist framework to sort through IRE issues. What is unique to Internet research has been debated for years now. Typically, we tend to identify differences in terms of identity, privacy, consent, ownership of data, longevity of data, storage, and research with minors. Basically, I argue that we cannot use the same language, the same discourse, with IRE as we do with “traditional” research. IRBs/ethics boards, and researchers must change the ways they conceptualize research and research ethics to better fit this new paradigm.

The difference between “traditional” research, as you put it, and Internet research seems to hinge on the Internet itself. How does the Internet fit into research today?

It fits into every discipline in so many ways. I am hard pressed to think of a discipline that is not, in some way, making use of the Internet. From online epidemiology to social work, they are all there, in many different methodologies. We see tons of online surveys, which I don’t particularly like, from either an ethical or methodological perspective (if you are interested in more on this, see my paper with E. Hvizdak, 2009*). We see online ethnographies, studies of virtual communities, case studies. What has been interesting to watch has been the ways in which different disciplines struggle with ethics.

How so?

In some disciplines, the “human” in the human subjects aspect of the research, is not at any risk, and the research is, perhaps, not human subjects-based at all, as is the case in discourse or content analyses of Internet materials. In others, there is authentic human subjects work, where lives can be affected, and where risks and benefits must be calculated, individual identities strongly protected. Specific research methods along this continuum range from analyses of data sets, aggregated and decontextualized, merely representing some facet of experience, devoid of connection with individual people, to the use of online surveys, in which a range of ethical issues can be found, sometimes to the surprise of the researcher and researched, for instance, where individuals may or may not be readily identifiable, and where data may or may not be in the full control of the researcher, depending on the tool to participant observations of lists and the people who inhabit them; to complex, in-depth ethnographies of people and their communities online, with the potential to harm individuals if their real-life (or on-screen life) identities were exposed.

What sorts of agents – groups, individuals, or professionals – are relevant to these issues?

Many professional associations are indeed interested. The Association of Internet Researchers is a prominent player in the IRE field, and as I said, they’ve been working on this for many years. But, in recent years, I’ve given talks on IRE to groups as diverse as PRIMR (Public Responsibility in Medicine and Research) to HICCS (Hawaiian International Conference on Systems Science) and to the Computers and Philosophy (CAP) in India.  But, most recently, in the US, the federal bodies are very interested—the National Science Foundation has funded us generously to study IRE in the context of ethics boards, and in the context of web 2.0; the Office of Human Research Protections (OHRP) has invited me to speak and this July, Michael Zimmer and I from the IRE Commons will present to the Secretary’s Advisory Committee to OHRP. That is very important that on the federal level, our work is making a difference and being heard.

What, in your opinion, are the most pressing IRE issues today?

I am very interested in cloud computing, databanks, and forms of bot or agent research. These are all areas that completely break down the model of traditional research regulation. They simply don’t fit our previous paradigm. For instance, in traditional research, we would tell our subjects where and for how long data would be stored. We cannot actually make those ascertains to the same degree of certainly with cloud computing. With shared data banks, the idea of consent is fundamentally changed. When data is pooled and reused, researchers must work according to a different ethic. And, we are not yet in agreement as to what that is. Another hot area—do bots or avatars “fit” as a “human subject?” It may sound like a strange question, but when we look more closely, it may not be so odd. We have bots and agents acting autonomously, making decisions, scraping data. According to what—if any—research ethics are they held?

The hardest part of IRE work is the moving target. Just when we understand one thing, for instance, online survey tools, another app or venue pops up to confound us. Think of Twitter. Now that these are archived by the Library of Congress we have issues of consent and privacy facing us.

It is a great time to be studying IRE, and I am fortunate to have been here through the years.

For more information on this subject, visit the Internet Research Ethics Digital Library, Resource Center, and Commons.

* Buchanan, E. and Hvizdak, E. (2009). Online Survey Tools: Ethical and Methodological Concerns of Human Research Ethics Committees. Journal of Empirical Research on Human Research Ethics (JERHRE) 4(2), 37-48.