Online virtual worlds promise an escape from mundane everyday environments and exempt users from the normal laws of time, space, and gravity. However, the laws of social influence may not be as easily dodged. In the virtual world of There.com we tested two robust real-world compliance tactics (foot-in-the-door, door-in-the-face) with avatar ‘‘race’’ as a moderator. Results revealed success for both techniques, suggesting that avatars are sensitive to influence tactics targeting both self-perception and reciprocity norms. Additionally, the race of the avatar requesting help impacted the success of the door-in-the-face compliance technique, raising the specter that real-world racial biases may also emerge in virtual environments.
The details are found in this link. (recommended before reading the rest of my post)
Now I’ve only skimmed the paper because it came out so suddenly and I’ll fill in details that might be worthy to know.
Let’s see the game used is there.com and the experimental avatars’ clothes were identical, the skin colours were clearly white and black, physical features were the same, except that the black avatar’s face looked more like an African-American.
Participants were 416 There.com users, 218 went through the control condition. 97 through the Foot-in-the-door condition and 101 through the Door-in-the-face condition. Quite sufficient numbers to see differences, I say.
Results: Looking at the article’ graph. It seems that there were no differences in the control or Foot-in-the-door condition with an average compliance rate of ~50% and ~75%, respectively. The kicker is in the Door-in-the-face condition where the white avatar got an ~80% compliance rate whereas the black avatar got a ~60% compliance rate, so a white avatar would have 20% increase in getting a request.
I’d like to address an issue that some commentators (especially DarthM.) had been stressing on gamepolitics.com. Is the problem of using a game that has crappy graphics really an issue in scientific research? I believe there was research that participants who know only a person by his/her name can influence their social behaviours, especially if the name is related to an ethnicity. The fact they found behavioural differences in a crappy graphics game means it doesn’t take much to influence our behaviours, even if we think we’re not biased or consciously trying to be an egalitarian. It can be hard to self-monitor our behaviours.
Can this study generalized to other online game, I think yes on some conditions. First, the skin colour and race should be directly related to us humans. Elf wouldn’t cut it out, people would have different reactions to an elf or orc depending on which game.
Why not use World of Warcraft? The problem is that the researchers would have to pose more hypotheses they can bite, logistical, cultural and ethical problems that comes along with researching WoW players. First, as Brokenscope mentioned, you’ll have to take into account the level-based social system, races other than human, then you’ll have to find a visibly black or white avatar in a world full of armour suits, the cultural and social baggage that WoW players possess when helping others and not to mention that players are busy people doing something. In addition, the researchers are not looking at gamers’ behaviours, just regular people who happen to play in an online environment, much like in Second Life.
Another commentator I think from kotaku or maybe both think it’s a waste of money and time because it’s common knowledge to online players. Sure, people in antiquity thought that lightning was caused by the gods. But, you have to test any assumptions to see if it’s true because it can lead to surprising results and change our perspective on things.
In any case, my take of this study is that it’s same old, same old regardless of where we are, virtually or reality.
Eastwick, P. W., & Gardner, W. L. (2008). Is it a game? Evidence for social influence in the virtual world. Social Influence,