Yes, you read that title right. A study recently published in PLOS One examined skin covering/exposure behaviour by examining a Star Wars roleplaying community in the online game Second Life. If you don’t know what Second Life is, I’ll invite you to first come out from under the rock you’ve been under for the past decade, and then tell you that it’s an online virtual world where you can create a completely customised avatar and interact with other virtual people.
Here is their ground-breaking result: “Our results indicate that virtual females reveal considerably more skin than virtual males”.
And here’s my main problem: how the hell do you know if the user who created the avatar is male or female? The answer is they didn’t know, because they wanted the study to be purely observational and random (meaning that participants in the study didn’t choose to take part in order to be contacted about additional information). But they assumed that less than 25% of the users created an avatar of the opposite sex because some other study estimated this across all online users. Not good enough, especially considering that this study examined a specific subset community.
And here’s why it’s a problem: because we don’t know how many of the virtual females were created by IRL males and how many were created by IRL females, this research tells us absolutely nothing about sexual behaviour. Except maybe that people, regardless of gender, find naked skin on females more attractive than naked skin on males. Hardly earth-moving. To be honest, it doesn’t even tell us that. It only tells us that people prefer naked skin on virtual females. Useful stuff!
And here’s another thing: It was only examined in a specific community of Star Wars role-players. I’m as big a Star Wars nerd as you get, but how much can we really generalise based on this specific subset of people?
I like the reseachers’ creative approach and embrace of technology to think outside the box when it comes to addressing their research questions, but they can’t really think that this method is coming close to answering a research question that is relevant for real people.
I don’t usually pan research studies like this, but I couldn’t pass this one up. Normally I believe that if something has been peer-reviewed, there must be some scientific value in it no matter how small, but I’m afraid the novelty factor was the only thing that got this one published. However, if anyone would like to read the paper and point out something interesting/useful from it, I am all ears:
Personally, I smell a PhD student wanting to play Star Wars on Second Life instead of getting out and collecting real data!