The Ethics of Studying Social Media and Online Communities

After the horrible nightmare that was Loyola professor Dr. David Myers’ study in the City of Heroes online community, I started thinking about the ethical challenges that face researchers wanting to do studies using virtual environments – massively-multiplayer games, forums, social networking spaces like Facebook or Twitter, even blogs.

TwixtThey are very tempting environments for research on anything from human interactions to social structures to even the nature of fandoms or group imagination, but can you take the existing rules and apply them to these very non-traditional spaces?

The crux of the matter really is that rules of ethics are supposed to protect human subjects from harm. Subjects don’t become any less human or prone to harm because they are online. This is really the issue where researchers seem to get stuck most often (see Dr. Myers again). This isn’t to say that there aren’t many, many well-intentioned, ethical researchers out there looking at social media, just that it’s a new and dicey area and not every researcher handles it well.

Informed consent is required for most research as well as reasonable assurance that the research won’t cause harm. This means that if a researcher is going to be interacting with people in a forum or in a game, they have the right to be told, given the chance to opt-out of the study and be assured they won’t come to harm.

But the space belonged to the users before the study came along, so what happens if someone doesn’t want to participate? Does that mean they have to leave the game or forum or conversation? It shouldn’t, but how can the researcher’s desire to study the space be reconciled with the subject’s right to refuse? I have no answers on this one, but it’s definitely something to take pretty seriously.

Harm is thorny too. When you’re interacting with people exclusively online, you don’t know how something really affects them. Going back to Dr. Myers for a moment, he assumed that everything was fine because it was in a game and because of the people who talked to him largely just said how annoying it was, but most people won’t say anything (especially when it really starts to affect them strongly). I wouldn’t have said anything to him, but I likely would have been crying in real life and not played CoH again for a long time, if ever. There have been people who were pushed over the edge and committed suicide because of things that happened to them online. That’s real psychological harm. So how do you make sure you don’t hurt someone when you’ll probably never know if it happens? Again, I don’t know, but it’s something worth thinking about.

A big focus in the ethical rules for sociological and psychological studies is being very careful of “vulnerable populations” (children, the disabled, the mentally unstable, etc.), but online it’s hard to know who falls into one of these and who doesn’t. There are specific sites geared to children (so you know that’s who you’re studying) where a researcher really wanting to set up a serious study could likely find a way to work with the site designers to arrange for consent from parents, but what about everywhere else? My guess is that in studies on the internet, we’re going to have to start affording everyone the protections given to “vulnerable populations” simply because we don’t know who falls into them and who doesn’t.

Confidentiality might seem like an easy one, since nobody knows anything about anybody else most of the time, but I actually think it’s really important when dealing with online spaces. Blurring out identity is still important, and when dealing with social media there might be a great deal of personal information at the researcher’s fingertips. But even information that isn’t obviously personal might be. Screennames aren’t generic – they have become as much a part of people’s identities as anything else. Gamer Tags and forum screennames are chosen with care because they identify who you are. Often people carry the same screenname for years – my husband and I have both used ours since high school (and at over ten years now, that’s about how long such things have been commonly around – now even my grandmother has one).

Finally, there’s the ethics surrounding the perceptions of the researchers themselves. There are a ton of preconceptions about the internet, social networking sites, forums, online games and people who use and play them. It’s really important that if they are going to be examined they be examined with the unbiased positions we expect of laboratory studies. If a psychologist is so convinced that people who play World of Warcraft are mentally ill that s/he can’t make objective observations, even if that idea is contradicted, s/he probably shouldn’t be studying World of Warcraft. On the same note, if the researcher’s spouse is a marketer at a big company that regularly uses Twitter to market their products, it might be a conflict of interest for that researcher to study the difference in effectiveness of Twitter campaigns from large and small companies (at the very least, the connection should be disclosed in the paper).

Online spaces are great and as they are used more and more for social, political, recreational, economic and other purposes they have the potential to tell us a lot about ourselves as a society and as people, but the research on that needs to be carefully gathered or we might end up learning more about ourselves through how the research was done than what it’s results were. I think that it’s important that these issues be discussed and I would really like to see the organizations that govern research ethics discussing them. Fundamentally, human beings are still human beings no matter where you interact with them, but the logistics of the internet are different than that of a lab and the guidelines for making sure subjects are protected should be carefully clarified before another debacle like Dr. Myers’ Twixt study happens.


3 thoughts on “The Ethics of Studying Social Media and Online Communities

  1. Pingback: Pixiepalace » Blog Archive » Check out The Learned Fangirl

  2. Hi Rosepixie,

    Just saw this. Since no one has left a comment, I will.

    1. The notion that online spaces “belong” to their users may be appealing to those who like to own and control things, but it really doesn’t have much basis in fact. In fact, in the case of CoH, as you probably know, NCsoft owns and controls the rules of conduct within the game as well as all game avatars and, importantly, all avatar names. This is beneficial to users in some respects — it protects them from copyright violations, for instance — but when push comes to shove, I think you will agree that the NCsoft EULA is more likely to determine “belonging” than user wishes/desires.

    2. An “experiment” normally has a treatment group and a control group. Further, the experimentalist maintains control over the environment in which the experiment takes place, the subjects of the experiment, and the treatment variable(s) involved. In the study to which you refer, there was neither a control group nor “treatment” variables other than the rules of the game to which all player voluntarily subscribed. I had no control over the environment in which play took place, and I had no control over who entered the RV zone and played the pvp game voluntarily therein. Perhaps you intended the term “experiment” in your post to mean “an innovative act or procedure.” If so, then okay. Otherwise, your use of that term needs qualification.

    3. In 1987, I published a study called, in part, “Anonymity is part of the magic.” In that study, I concluded that online anonymity allowed online actors to more fully explore behavior and activities that were oppressed and suppressed offline. Currently, my impression is that online anonymity — cherished and protected by those who have come to the online party a little later, such as yourself — is more often used as a means to oppress than as as means to avoid oppression. While anonymity and user privacy remain valuable and important, there is now evidence to believe that online anonymity and privacy should be tempered with a strong measure of social responsibility.

    Some of the evidence:

    There’s this story in the Huffington Post:
    And this result in London :
    And this from Australia:
    And, of course, what happened/is happening to Twixt/me.

    Regardless of the benefits and freedoms of anonymity, there are boundaries that the anonymous should acknowledge and respect. It may be useful, therefore, to discuss ethical issues in broader and more realistic and relevant contexts than those you describe. Like you, I would like to see this discussion take place — but only if this discussion can equally and openly include all parties involved.

    • Dr. Myers, I wanted to respond to a few of the points you made. You are right in that the spaces are OWNED by the companies that run and maintain them (in the case of CoH, NCSoft). However, the very nature of social spaces is that they are inhabited by people and rules and structures very separate from those imposed by the developers tend to develop. This is usually part of the point of social spaces – for them to be at least somewhat player-managed or player-governed. You yourself found that out when you tried to upset the player society in CoH and it was the players who banded together, from both sides of game’s drawn line, to object to your actions and not the developers who did anything about it. So, yes, I do maintain that in a very real sense the players own their spaces. At least, they own them every bit as much as you, a researcher does, and they have the right to enjoy their spaces without being bullied in the name of research which should be at the very least looking out for their safety.

      I wasn’t disputing the common definition of experiment. I’m well aware of it. What I was discussing was how that definition doesn’t really work in online spaces and how, if researchers continue to want to study them (which I think they should), new ethical guidelines need to be laid out. This needs to happen both because the spaces are different and because the nature of the experimental model will have to be different. New spaces bring new challenges. Perhaps I should have stuck to the word “study”, but I’ve found that at least in psychology texts the word “experiment” is often used for research that doesn’t exactly fit the standard experimental model and since they are studying these spaces as much as sociologists like yourself, their practices need to be taken into account in the new rules as well.

      Your paragraph on online anonymity and oppression is confusing to me. First of all, I’m not sure what oppressing or not oppressing online anonymity inherently has to do with this. I think the issue is subject privacy, the same as in a regular study where you do know who you are studying. I’m not suggesting that if there was a good reason to reveal someone’s identity it shouldn’t be done, but I question whether that is a material point in this discussion. In studying humans, subjects have the right to remain anonymous – that’s a basic tenant of research ethics. What I’m suggesting is that this basic right laid out in every ethical rule set I’ve ever seen, be it for humanities or hard science research, should be extended to subjects of studies done online and that researchers need a better understanding of what it means to keep someone anonymous given things like screen names and avatars, which are less anonymous than people often imagine them to be.

      I would love to see “all parties” discussing this issue, but I rarely see researchers discuss this topic (It happens occasionally, but rarely), and you seem to miss the point entirely. Still, thank you for commenting and allowing me to further clarify a few points that you seem to be persistently confused about.

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