« Visit from Greg Wadley | Main | Avatar Survey: Age »
July 21, 2008
Avatar Survey: Gender Demographics
After several months of data collection we have begun analyzing the responses to our survey on avatar customization. A big thank you goes to the 180 participants who took the time to go through the questionnaire and provided us with some very interesting information! In the coming weeks we will be posting regularly about what emerges from our analyses.
In this post, we will start with some basic demographic data about our participants, in particular, their gender. While this is not directly related to avatar customization per se, it gives us a sense of how representative our sample is. Interestingly, our participants data broadly confirms the player profiles that have been reported by others in previous research and in industry data. Read on to learn more.
During three months of data collection, 180 participants (89 for World of Warcraft, 50 from Second Life, and 41 from Maple Story) have taken part in our online survey. The gender distribution of participants was 115 (64%) male and 63 (35%) female. There were two participants that didn’t specify their gender.
This distribution varied across virtual worlds: the male/female ratio of both WoW and Maple Story were approximately 70/30; in Second Life, it was 50/50. Our gender distribution for game-based environments like WoW and Maple Story is broadly similar to data reported earlier by other researchers; in SL, it looks as if the gender distribution is more balanced.
In addition to the gender distribution of users, an analysis was performed to understand these users’ gender preference when they are living in a virtual world. The avatar gender of each user was recognized by analyzing the avatar screenshot that they uploaded to our server. Within the 157 valid screenshots, there are 89(57%) female characters and 68(43%) male characters found across the three games. In WoW 43(53%) of the players were using a female avatar and 37(46%) players chose to have a male avatar. In the 36 valid samples from Maple Story, 20(55%) players have a male avatar and 16(44%) players use a female avatar. The avatar gender balance in Second Life is significantly different to the other two games. Within the 41 valid Second Life samples, 30 (73%) users played using a female avatar but only 11 (26%) users played using a male avatar.
We then analyzed gender swapping and found that 38 (24%) users played a different type of gender (across these three games) compared to their actual gender. Specifically, 34 male players use a female character and 4 female players use a male character in their virtual world of choice.
Posted at July 21, 2008 11:40 AM
Comments
I find the language you use at the conclusion of your fourth paragraph odd. You say, "...whereas a social-based virtual environment like Second Life can attract both male and female users." A casual reader or a journalist looking for an inflammatory quote might draw the inference that a "game-based virtual environment" is unable to attract both male and female users. I find it odd because the ethos of the rest of the post is quite academic, and avoids drawing any stealthy conclusions.
Turning to the graphs, I find them very helpful at summarizing the data. I think your presentation would be improved if you swapped the Second Life and Maple Story graphs, so that WoW and Maple Story (the game-based VEs) were adjacent, instead of sandwitching Second Life (the social VE.) This would allow a visual learner to discern that there are major differences in gender decisions between game-based and social VEs.
A question on significant digits: with your sample size of 178, do you really need three significant digits, or does it impair comprehension? Consider just dropping everything after the decimal point.
Are you able to get information from the VE providers about what their overall gender makeup is like? IE, what is the percentage breakdown between male and female avatars? This would help identify how much of a self-selection bias your 180 participants exhibit.
I look forward to the future posts about your survey data.
Posted by: Notelrac at July 23, 2008 08:26 AM
@Notelrac: thanks a lot for your constructive comments! I have updated the fourth paragraph to (hopefully) avoid any kind of value judgement, this is not our purpose here. We'll redo the graphs too but this will take a bit more time, stay tuned :)
Posted by: Nicolas Ducheneaut at July 25, 2008 09:26 AM
One curiosity: did your study consider 'alts'? I assume it's less-common in Second Life, but many people in WoW may have more than one avatar. How would that effect this data?
Posted by: Amanda at July 29, 2008 08:40 AM
@Amanda: I am not one of the researchers, but I took the survey. The only questions about alts that I remember were about how many you have and how much time you allocate to playing them. The rest of the survey asked you to consider your main avatar. I imagine it would be too difficult to consider multiple alts because they would introduce too many variables.
It might provide some interesting data though, especially when considering the gender question. Personally, my alts are always female, but I know many people that have both male and female alt/mains. I would bet that the number of gender swappers indicated in this study would be very different if alts were also taken into account.
Posted by: Cindy at October 9, 2008 11:14 AM