Virtual Research Ethics

The flood of virtual world research has become a torrent with academic institutions finally turned on to some of the interesting social phenomena which occur as a result of real-live human beings interacting in online space (shock horror). To date, however, it’s been rather difficult to get professional and scholarly journals interested in publishing such work. Finally, there’s an indication that traditional science is taking this area seriously, with coverage of some of the more prominent studies in top flight magazine Nature.

But while this article discusses the opportunities virtual worlds offer social scientists for “manipulating society” for its own ends, /. points to another Nature article which highlights some of the ethical conundrums which emerge when bringing old school ideas (and experiments) to the digital realm.

/. reports that Stanley Milgram‘s classic compliance shock study – in which participants were asked to increase the power of electricity given to a “person” in another room for incorrect responses to the point of suspected serious harm – has been replicated in an immersive virtual environment. From the paper, on PLoS One:

Stanley Milgram’s 1960s experimental findings that people would administer apparently lethal electric shocks to a stranger at the behest of an authority figure remain critical for understanding obedience. Yet, due to the ethical controversy that his experiments ignited, it is nowadays impossible to carry out direct experimental studies in this area. In the study reported in this paper, we have used a similar paradigm to the one used by Milgram within an immersive virtual environment. Our objective has not been the study of obedience in itself, but of the extent to which participants would respond to such an extreme social situation as if it were real in spite of their knowledge that no real events were taking place.

The results, as reported in the paper above, indicate that respondents to the current test felt similar distress as those who took part in the initial (offline) experiments. In other words, it doesn’t matter if the environment is virtual or not; what happens online affects the offline person.

There are countless papers on this crossover; Julian Dibbell’s A Rape in Cyberspace is the most readable. What this and other examples of human response to virtual action demonstrate is a need for ethical protection for participants who may be distressed by what happens to them in online research. To date there is no overarching ethics protocol for internet scholars (each study must go through the ethics processes required by the university or research institute which it is allied with, and each country has its own ethics philosophies), although the Association of Internet Researchers have developed a broad-scale guide which is quickly becoming standard.

Virtual worlds in particular pose interesting questions for researchers as they replicate (to an extent) the offline social space (economies, stratifications, hierarchies, community processes) and therefore users/players’ experiences online are often extrapolated to non-game social phenomena (for more on this, read up on the topic at Terra Nova). But rather than assuming that a population of an online game or an immersive virtual environment is made up of a bunch of isolated digital guinea pigs, researchers must remember that, as the Milgram replication reminds us, there are people behind the orcs. And they deserve our respect.

Note: I completely forgot to post this gamesblog article here, where it’s far more relevant to my own reserach. I published it there on 8 January 2006.

For the record, the above is a complete cut and paste. If you’re interested in the topic, I’d recommend going to Guardian Unlimited to immerse yourself in the great discussion that resulted from the post. But feel free to comment here as well!


~ by aleks on January 31, 2007.

6 Responses to “Virtual Research Ethics”

  1. Aleks

    Dibbell’s piece is a fascinating (and rather disturbing) read, and while I have no doubt that harm can be inflicted in virtual spaces, do you worry that your own research is capable of it? The sociologist has less interaction with their subjects than almost any other scientist, but is also just as much a part of society.

    Do you feel you have a duty to step back from affecting the inhabitants of Second Life, simply because you’re studying them? Seems a shame.

  2. Hi Ian,

    The thing I worry most about my research is that it will negatively affect the community because I feel that it is the most important thing in Second Life and other virtual spaces.

    When I first started in SL in 2004, knowing full well that I was going to be researching the social networks of the space, I knew I was entering an already thriving community (with only 24,000 accounts!) made up of autonomous individuals with rights. I knew as well that the community had previously been “burned” by researchers who had ignored peoples’ rights to privacy by posting analyses and transcripts of conversations – without changing the names of Residents – to public fora and to external blogs. Seeing the response of the community to these indiscretions paralysed me and I read everything I could on virtual research ethics (for those with accounts in SL, you can find many of them on the shelves at the SSRL, a research lab facility I established specifically because I wanted to give something back in return for the generosity of Residents for participating in my research).

    I was also aware of the fallout from mass media attention to goings on in another virtual community, The Sims Online, when a community public figure (who was also a researcher) wrote about some of the nefarious activities happening in that space. While there was, I believe, little anger directed at the researcher, his reports in the web-paper The Alphaville Herald affected the community profoundly. I didn’t want to do that to SL.

    We worked extremely hard to create research questions and tools which we felt would cause no harm to the community (and indeed, we believe the results will benefit it, by helping to identify those areas of social life which enhance interpersonal cohesion, thus informing future designs of social spaces within this and other online worlds), but the sociometric survey technique we have implemented has received criticism and concern from a few Residents. This has caused me to re-think my (and other researchers’) affects on the community.

    Interestingly, it’s not caused me to pull back formally, just personally. My actions in the virtual space as Mynci Gorky are now more in line with the role of Professional Researcher, certainly to the scores of people I contacted directly for their participation and to a lesser extent to those I know informally. I am there to study, analyse and provide information.

    My Researcher tag (which I wear when “researching,” as recommended by Robin Linden when she invited me to the Researchers group) affects social interaction as much as my offline Journalist tag affects the content of informal chats with people I know only professionally. Man, if only I could have a Press tag floating over my head in meatspace when I’m on the job, like I wear a Researcher tag in SL, so people could know when I’m not working. I don’t know how other VW researchers in SL respond to this duality, but I’ll tell you what, the place is flooded with ‘em.

    My space in the social life of SL is part of this research. I can’t avoid that. If you removed me from the data, you’d lose the connections between people/groups that I or my immediate group of friends bridge. So in that way, I am inextricably affecting the relationships in the community.

    I feel it’s only right to get involved and to give something to the space in which I am, by all accounts, participating, rather than swooping in and taking what I need with no consideration for the other people involved. This musn’t affect my objectivity however, which may explain why I have pulled by personally from Second Life.

    On an interesting side note, on 13 January Linden Lab announced that they no longer require researchers interested in using SL as a giant petri dish to sign a formal Researcher’s Agreement. Nor do they have an official Linden Lab Ethics Policy anymore. The only ethical considerations for reearchers entering this virtual world come from their home institutions (if applicable).

    Some links of interest:

    Part I of Tony Walsh’s coverage of the Research Fallout from November 2004:
    Part II of Tony Walsh’s coverage of the Research Fallout from November 2004:
    Tony’s interview with me, defining my role as a researcher, for the public of SL at large:
    BBC coverage of the Sims Online fallout:
    The SSRL in Second Life:

  3. I get slightly peeved by those who take the ‘switch off’ stance in response to potential abuse online. It underestimates the degree to which intimate and emotionally compelling relationships and experiences can develop in cyberspace and it is in this context, in my view, that most potential problems lie. As a potential PhD student, to be partly concerned with SL, I think the important thing is to be sensitive to people’s personal diverse experiences of virtual and online worlds. That raises all sorts of issues with regards to engagement (how deeply should I get involved with this/them) and social grouping (being friends with people who see SL differently).



  4. I agree Yakoub, but also think there are different levels of engagement that academic disciplines ascribe to which affect how much they are allowed to integrate with the community. Then again, researchers should arguably be familiar with the norms and rules of an environment before they choose to begin analysis and every online research method book or paper I’ve read talks about the importance of community in spaces like SL.


  5. Aleks,

    In light of your observations here, I would like to hear what your thoughts are on a recent post by Professor Castronova at the Synthetic World News. In it, Castronova outlines the possibility of “experimental ethnography” returning anthropology, where rampant postmodernism has, in the recent decades, repeatedly challenged the field’s alleged ability to make observations of cultures. Castronova envisions two parallel communities inhabiting an MMO environment, where a variable is altered in one and the subsequent changes documented and compared to the control community. It’s a short post, but as you can see in my reply as Vincent, I’m definitely interested in hearing what the ethical implications of “experimental ethnography” in virtual worlds are. Your thoughts?


  6. Hi entropyrising, thanks for pointing out Ted’s post. I think you bring up a very good point – these are established (or establishing) societies and the danger of so-called “tampering” with them is greatest for those involved. It would take a looong time to ensure that community is established in such a way as to be relevant for ethnographers or social scientists (unless we’re aiming to understand the emergence of virtual community norms?).

    I think even more important is understanding how the various participants may conceive of ethics. I know from reading the AoIR’s ethical guidelines that this is an important consideration; in virtuality we’re dealing with consumers who come from all over the world, with conceptions of personal ethics that vary according to each region, nation, state and even city. The authors of that document do very well in outlining some of the basic differences, and I’d direct you to that for their take.

    Ted does point out that the Arden project has already passed the University of Indiana’s internal research ethics board which at the minute is the only thing we can lean on when it comes to regulation; here in the UK we don’t have IRBs, but we do have institutional ethics committees who may make their suggestions based upon what they deem as ethical within the constructs which they have considered before.

    To summarise a rambling reply, I do agree with you that the “societies” which emerge in these worlds are as precious as those which exist offline, and any potential harm affects the populations within them. I also think that the ethics very much depend upon the research questions that are being posed. Is Ted’s Arden project looking to “experiment” with established communities or is it hoping to control factors based upon mutual community emergence in similar (but obviously slightly different) virtual environments? Then again, surely this is what people who’s been studying virtual communities over the decades have had to contend with when they’ve decided which virtual community to explore? Second Life is different, for example, from World of Warcraft, is different from The Sims Online, is different from MySpace etc).

    Finally, at the moment what we have to rely upon for ethical correctness (in terms of research design) are the boards who grant us permission to go forward and experiment (if this is what’s appropriate), who also grant us these permissions for offline research designs.

    Rambling, tangential, but I hope I’ve explained my standpoint!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: