A new study from leading research firm RTI International seems to contradict the stereotypical assertion that people with athletic avatars are probably overweight and sedentary in real life. In a perfectly-titled paper, "Does this Avatar Make Me Look Fat?", RTI researchers found nearly the exact opposite to be the case. According to lead analyst Elizabeth Dean, "[I]t seems likely that virtual reality users may adjust their identity to be consistent with that of their avatars." In a survey of 29 Second Life Residents, Dean and team found that 80 percent who said their avatar was very active in-world said they were also physically active in the real world. Just as intriguingly, Residents' perception of their avatar would change based on the interviewer's avatar: When the researcher's avatar was overweight, the interview subject would describe their own avatar as overweight, and vice versa.
These are very interesting results, albeit gathered from a very a small sample; I hope to read the full paper and chat with Dr. Dean when I can. The results do seem consistent with the "Proteus Effect" described by Stanford researchers; at the same time, I wonder how much the results were shaped by the survey questions. If you had a buff, bodysurfing avatar, and an academic asked if your real life self was also in good shape, wouldn't you feel a bit embarrassed to answer No? Mixed reality file photo from my profile of Laura Oh. Hat tip: Botgirl Questi.
"80 percent who said their avatar was very active in-world said they were also physically active in the real world"
conclusion: people with athletic avatars are more likely to lie about their RL physical shape.
Posted by: magggnnus | Monday, August 03, 2009 at 12:32 PM
But 99.9% of SL avatars are "athletic"..
Posted by: Prad Prathivi | Monday, August 03, 2009 at 12:38 PM
And slutty avatars?
Posted by: Maggie Darwin | Monday, August 03, 2009 at 12:44 PM
"In a survey of 29 Second Life Residents"
They are drawing conclusions about a population of tens of thousands (if not hundreds of thousands) of active users from the responses of only 29 people? A population that is highly variable in terms of age, place of origin, education level, and so forth?
Let's hope the it was at least a random survey, and all 29 people aren't members of one subset of SL.
Posted by: Sioban McMahon | Monday, August 03, 2009 at 12:52 PM
Do they get money for this kind of thing? I wish I was capable enough with words to convince people that the tiny survey I took asking barely related, and frankly nonsensical (wtf is a "healthy physical activity" for an avatar?) questions somehow led to the conclusions published.
Posted by: Ananda | Monday, August 03, 2009 at 12:58 PM
I'm willing to buy it if we see 29 hawt Mixed Media (no feet or arms) submissions.
Speaking of which - kudos to young Arwyn for stepping up while us old larger people hide.
Posted by: Adric Antfarm | Monday, August 03, 2009 at 01:35 PM
I doubt they'll get any money for this. The only thing this kind of survey would get, is a D- in a high school level marketing class. The sample population is unusable. Basic rule is you can never make any conclusions out of groups smaller than 30. And that's the bare minimum, it becomes bigger based on the size of the total population and diversity. personally I wouldn't trust any SL survey taken with less than a 1000+ population size.
Posted by: Eirik Haefnir | Monday, August 03, 2009 at 01:38 PM
29 is awfully small. Was it the same timezone for all? I am sure cultural differences account for something as well.
For the record my avatar is nothing like me - apart from gender. Aside from Wii use, my rl self is not sporty at all :)
Posted by: Toxic Menges | Monday, August 03, 2009 at 03:19 PM
hehhehehehheheee... this might be the funniest post i have read this week... I'l just survey my self and my alts to see if we agree
Posted by: soror nishi | Monday, August 03, 2009 at 03:21 PM
Ya'll go easy on them-thar profussers at Stanford.
It am true...being a fake Hillbilly in SL dun made me change for the bettur IRL. I now drinks mo' Shine, goes to jail more, wrecks my pick-em-up truck all the time, an' chases floozies every chance I gits IRL.
We calls it The Procrustes Effect in Enoch Holler--and I are in bed a hole lot more than before I dun run into SL.
I reckons the methodolergy am rite good in this here study. I dun run the study too! Mah sample size am only 2: me and my good friend in SL, Beeble.
He begun as a Furry, a big ol' racoon-critter.
Now IRL he runs about in a fuzzy-wuz outfit and am rite happy with the change.
Posted by: Pappy Enoch | Monday, August 03, 2009 at 06:28 PM
A study last year of 7000 Everquest II players found that their body mass index was three points lower than the U.S. average, a finding utterly contrary to what the researchers were expecting.
But yes, if someone is serious about corrolating avatar appearance or level of activity with that of players, it would probably help their credibility to do some actual measuring of actual players.
Posted by: Arcadia Codesmith | Tuesday, August 04, 2009 at 06:32 AM
29 of 5000000?
Seems more faith than statistics.
Posted by: Impalah Shenzhou | Tuesday, August 04, 2009 at 07:53 AM
When I read about this, it seemed to me that they were discussing preliminary results and wrote this as a research note. I wasn't under the impression they were trying to draw hard and fast conclusions, but were exploring a new avenue of research.
Posted by: Sadie | Tuesday, August 04, 2009 at 08:06 AM