Racism in Virtual Worlds

Two social psychologists from Northwestern University conducted one of the first experimental field studies in a virtual, online world and found racial biases operate in much the same ways that they do in the material, offline world.   The study’s co-investigators are Northwestern’s Paul W. Eastwick, a doctoral student in psychology, and Wendi L. Gardner, associate professor of psychology and member of Northwestern’s Center for Technology and Social Behavior.  The study was conducted in There.com, which is similar to Second Life, and offers users a relatively unstructured online virtual world where people choose avatars – or human-looking graphics – to navigate and interact.

This next bit gets a little technical, so bear with me.

The experiemental study design is referred to as a “door in the face” (DITF) and it works like this:  the experimenter (in this case an avatar) first makes an unreasonably large request to which the responder is expected to say no, followed by a more moderate request.  In the past, researchers have found that people are more likely to comply with the moderate request when it was preceded by the large request than when the moderate request was presented alone, and this held true in the virtual world as well.   In the virtual world, the experiment’s moderate request was: “Would you teleport to Duda Beach with me and let me take a screenshot of you?” In the DITF condition, that request was preceded by a request of the avatar to have screenshots taken in 50 different locations — requiring about two hours of teleporting and traveling.

Still reading?  Good.  What these researchers then did was to vary the skin tone of the avatar making the request, like this:

6a0120a51c39be970b0120a61e1f19970b-320wi

What’s interesting to note is the way that the skin tone change altered the responses:


In one of the most striking findings, the effect of the DITF technique was significantly reduced when the requesting avatar was dark-toned. The white avatars in the DITF experiment received about a 20 percent increase in compliance with the moderate request; the increase for the dark-toned avatars was 8 percent.

While it may not be surprising to learn that people take their racism with them into these (supposedly) new virtual worlds, this research is still noteworthy both for its innovative methodology and because it challenges the conventional wisdom on two fronts: one that we are living in a post-racial society and that the Internet is an inherently liberatory technology that offers an escape from old hierarchies of oppression.

Comments

  1. ellen says

    I’m going to show this to some older students I tutor, and then ask them why they think subjects complied more readily with the lighter toned avatars. This should be interesting. Thanks!

  2. People do not take their racism into virtual worlds, in fact, because they aren’t as racist as you imagine, and not especially racist in a virtual world of anonymous avatars, but the politically correct *do* take their prejudices with them.

    Before you keep recycling this meme over and over again, you need to dig deeper.

    In virtual worlds, the worst thing you can do is force-port, or try to force somebody to teleport with you or make a stranger go somewhere with you. The next worst thing you could do is demand that you do that *and* take a screenshot, adding insult to injury.

    So picking that sort of negative behaviour to test *anything* in There or Second Life already creates a bias that could simply be present in all cases and randomly exhibit at a high percentage in some skin tones.

  3. distance88

    “..creates a bias that could simply be present in all cases and randomly exhibit at a high percentage..”

    Nothing is ever random, particularly when it comes to human behavior.

  4. Finny

    I did a study like this in Second LIfe 2 1/2 years ago. Same results if not more disturbing. Over the course of one week i was Asian, White and African. The White got hit on, the Asian got ignored, the African got shot down a hill with a machine gun. I am not kidding.

  5. I have not done enough research to be certain. But Prok’s mistaken on one count. Why wouldn’t a person driving an avatar bring his or her real-life preferences along to a virtual world?

    That said, the evidence I’ve seen suggests that being new in the world, or even being a nonhuman “furry” can lead more often to active discrimination than could real-world race (I have a black male and olive-skinned male avatar, btw).

    My students switch race or gender every year, for one week, when we use SL. Women get “hit on,” the men-turned-women discover. Newness seems to lead more often, in 2008 and 2009, to being ignored, not skin tone.

  6. Nquest

    So picking that sort of negative behaviour to test *anything* in There or Second Life already creates a bias that could simply be present in all cases and randomly exhibit at a high percentage in some skin tones.

    Seems like the “negative behavior” is a constant for each of the skin tones so why would the response be “random” with significantly different responses to different skin tones? One would think the responses would even out with each skin tone accumulating roughly the same kind of composite response.
    .

    In virtual worlds, the worst thing you can do is force-port, or try to force somebody to teleport with you or make a stranger go somewhere with you.

    Why would there even be any “random” difference then in the responses to avatars of different skin tones making the indecent/annoying requests?
    .

    People do not take their racism into virtual worlds

    You say that based on what? How do you know whether or not people take their racism into virtual worlds or not? Seriously…
    .

    they aren’t as racist as you imagine, and not especially racist in a virtual world of anonymous avatars

    Too funny… To say that people “aren’t as racist as you imagine” acknowledges, by definition, that people are, in fact, “racist.” The strawman quibble, then, is over HOW racist people are. This is like arguing that someone is a little pregnant — i.e. hilariously, ridiculously irrelevant.
    .
    You’re either pregnant or you’re not. So, since we’ve established just how you (Prokofy Neva) have acknowledged that people are racist — just not as racist (i.e. as pregnant… lol) as whatever you’re assuming Jessie believes they are. Exactly how you know how racist Jessie (or the researchers) think people are in virtual worlds and exactly how you know how racist people are, no one knows including you.
    .

    the politically correct *do* take their prejudices with them.

    What prejudices are those? And how exactly does that not contradict you initial claim? Simply put, “they aren’t as racist as you imagine” sounds like a “politically correct” way of saying:

    Yes! People are racist in virtual worlds BUT . . . {insert your minimizing rationale for why the racism you acknowledge as being there isn’t *AS* bad or *THAT* bad fully knowing that you are the only person who has tried to quantify the degree of racism, again, as if someone can be a little pregnant vs. real/good-n-pregnant}”

Pingbacks

  1. Tweets that mention racismreview.com » Blog Archive » Racism in Virtual Worlds -- Topsy.com
  2. uberVU - social comments
  3. Race Bias in Virtual Worlds? « Karl Bakeman
  4. Teaching Doctors to Recognize Racism :: racismreview.com

Leave a Reply