Is she hardcore? Well, you could try asking…

A study was published in the Journal of Communication and has gotten significant media attention, at least in blog form (the paper is available for free here). Like many studies that end up in articles, particularly on somewhat controversial subjects, the conclusions being reported are not strongly supported by the study. On the other hand, it is interesting to watch what aspects of the study are appealing to bloggers and which are ignored, as well as which articles were clearly based off other summaries of the article, rather than reading the original. The researchers themselves appear to have an agenda, though, of emphasizing the “hardcore”-ness of female players, and in doing so perpetuate some of the value judgments described in Diamonds in the Rough and Those Other Girls: Conflicts Between Female Gamers

The article ignored anyone with non-binary gender or sex; I will be speaking here about the people surveyed the way they self-identified on the survey, though some of them may have only chosen to identify as male or female because of a lack of other options.


The article itself used data from a survey of Everquest II players and data about their playtime from Sony (which was accessed with the player’s permission). The researchers were Communications professors. Only one of them, Mia Consalvo Ph.D, appears to have a background in gender studies. She taught Women’s Studies at the University of Iowa. Many of her publications have be about gender or sexuality in games and gaming, though her degrees are in Communication. The data reported in this paper asked about gender, sexuality, income, happiness, loneliness, exercise, BMI, playtime, motivations for playing and whether or not people played with a spouse or partner. It appeared that they ignored people who did not report gender, and didn’t report data about race, disability or whether people considered themselves gamers, and conflated weight (classified by BMI) with health. It interpreted all this data through an extremely limited and simplistic gender role lens. They had a moment when they conflated gender identity and sexuality, speculating on “the potential androgyny that may be driving the bisexual players.” Within their limited, kyriarchal description of gender roles this might make sense, but they ignore much simpler explanations that rely only on sexual identity (like that these particular bisexual women may enjoy looking at the sexualized women in Everquest II) and feels unnecessary, particularly since the speculation is uncited and has no data behind it.

The finding that has been widely reported was that women played more hours than men and were less likely to be thinking about quitting, from which the researchers assert: “Our data suggest that female players—not males—are the real ‘‘hardcore’’ MMO players.” Of course, women also had significantly lower yearly salaries than men (despite higher education) and they didn’t appear to control for employment, so the extra time they play *could* be explained by more women not working in paid employment outside the home, or working part-time, and having more time to squeeze in an hour or two in the virtual world. It could also be explained by motivations; more women reported that their motivation involved socializing with other people in game, which presumably is more open-ended than progressive game content (I haven’t played Everquest II, but this is certainly true of WoW at the moment). Women were also likely to have played fewer other games, and men might have spent more time per week playing other games instead. That women were less likely to be thinking about quitting could be the result of women quitting more quickly when they did think about it, having begun playing more recently, enjoying the game more, or being more committed to the game than men. The researchers don’t seem to have made any particular effort to entertain other possible explanations for their findings, which I think is lazy science. Anyway, I digress, but my point is beyond the data they’ve collected I don’t hold this paper in particularly high regard. However, probably because it’s a paper that includes information about female gamers, it’s gotten a reasonable amount of digital ink.

How does “women play, on average, 17% more than men”, “women who play the game less likely to be considering quitting” and “Everquest II players underestimate their playtime” turn into a headline? (Warning: some of these articles are accompanied by borderline work-unsafe pictures of women playing video games in their underwear):
Study Finds Girl Gamers Are Seriously Hardcore, from Escapist Magazine
Girl Gamers Lowball Their Geekiness, from Edge
Study: Female Everquest II Players Still Logging More Hours, Still Not Honest About It, from Kotaku
Gal Gamers Geekier Than Guys, from Scientific America

So the finding that women played slightly more hours, described by the researchers as them being “hardcore”, also makes them as “geeky”. The fact that they “lied” by underestimating the hours played also appears to make a good headline (playing into the stereotype of women as deceitful, or ashamed of being geeky), despite the fact that men also underestimated time played. Many of the actual posts make this clear, but it’s presumably less eye catching than the implication that women are unwilling to admit that they are “really” hard-core geeks. The articles don’t always make it clear that this definition of “geeky” is simply how many hours per week someone remains logged into Everquest II and whether they are ambivalent about continuing to play.

Are there women who are also hardcore gamers? Of course! Absolutely! No doubt! The fact that that seems to be news to some people is unfortunate, but not unexpected. However, this study isn’t talking only about hardcore gamers, especially since no one involved seems to define the word (which the researchers get around by consistently using quotes around it, though bloggers and reporters don’t.)  I would have been more interested if the researchers investigated whether there was another group of committed women who play games many hours a day and don’t intend to quit.  Perhaps members of this group don’t think of themselves as “hardcore” and might “only” play because their romantic partner does. These are women who don’t have blogs, don’t wear t-shirts proclaiming their love of the game, and are ignored or discounted by the gaming press, by game companies and by gamers. They may, though, be playing just as much and with at least as much enthusiasm as visible gamers.

This data doesn’t go far enough to come to that conclusion, however, and the researchers don’t seem to ask that question. It almost seems like they are so excited to demonstrate the existence of a group of women who play Everquest II that they erase the possibility of heterogeneity within that group, particularly since bimodal distributions within their data might muck up some of their other conclusions if they are comparing hardcore apples to committed, enthusiastic, non-geek oranges. It’s nice to see data being collected on people who play games. The reports on that data, both published and in the blogosphere, leave something to be desired and overgeneralize egregiously.

This entry was posted in MMORPGs and tagged , , , , , , . Bookmark the permalink.

12 Responses to Is she hardcore? Well, you could try asking…

  1. That’s an original Xbox controller in that picture, not a Xbox 360 controller.

  2. Rachel Walmsley says:

    I’ve just skimmed through the paper and I’m not terribly impressed either. There seem to be a few bits and pieces in there which seem to merit further investigation, but nothing that I saw that I’d rank as more than that.

    Like you mentioned, the lack of any attempt to control for other variables is troubling. One thing I’d be interested in seeing would be a comparison between females and males who have comparable MMO experience and play other games a comparable amount. My suspicion is that if they did that they’d find that hours played were pretty much the same.

    I also noticed that there seems to be a discrepancy in the information they collected about playing time. From the way they talk about it, they seem to have asked players to estimate how many hours per week they play, whereas their data for actual number of hours played is a historical average since account creation. Of course, I can’t check this because they haven’t included the actual survey questions as an appendix that I can see, but if this is the case then that’s a potentially huge difference.

    I also note that they don’t even try to address (again, that I saw in my skim read) the issue of sample bias brought on by the incentive used to have people take their survey (the reward of a unique and apparently desirable item). I can imagine that that sort of incentive would definitely bias their sample. It’s difficult to eliminate that bias, but it would have been nice to see it at least addressed.

  3. Thefremen says:

    It’s funny that in a survey about sexual identity they would only have binary options. Also, the data was skewed by the fact they chose Everquest 2. Anyone who plays Everquest 2 is automatically a Super Duper Ultra God Tier Hardcore MMO Player. It takes something like 40 hours of killing rats to get to lvl 10 in that game, compared to 30 minutes of exploring/question in WoW. (note: those times are somewhat exaggerated. Only slightly though.)

    • Blake says:

      This is actually very common; almost no survey-style studies I’ve seen outside of specific queer studies involve non-binary gender options.

      It’s unfortunate, and is a sign of the disconnect between internet-based discussions of gender and academic discussions of everything else. It may also rely on the particular audience a paper is targeted at: even if the researchers are aware of non-binary gender, asexuality and other people who don’t fit into check-boxes provided, they may not want to take responsibility for introducing those concepts to a less-aware audience.

      • Thefremen says:

        Still, I don’t see the harm in including “Decline to state” or “Other” as a checkbox…

  4. Brinstar says:

    Great critical analysis of this study. When I posted about it last month here on TBH, I had hoped that one of our bloggers would really dig into it like you did (as I didn’t have time myself to do that). Thank you for pointing out those issues with the study. It certainly seems as if the study neglected some key issues. I agree with Thefremen: it seems weird that a study on gender wouldn’t provide options beyond a binary.

  5. oliemoon says:

    Wow, the Kotaku headline and summary are really telling. Out of all the conclusions that the study drew, the only one they found worthy of reporting was that women lie. GO FIGURE. Kotaku, +10 to misogyny.

  6. Great analysis! It makes me cringe when people–even feminists–take a few stats and figures and run with it, extrapolating way beyond what the data actually suggests. Thanks for writing this up.

    • Blake says:

      It is always important to distinguish between what a paper shows, generally the actual numbers they report, and the discussion and interpretation of that data contributed by the researcher. I usually see the discussion is most useful as a place to find things to prove or disprove with future research, rather than something I’d be willing to cite with confidence.

      I try to see academia as a dialog, where given the evidence currently available, these people are making these points. It’s sort of their job to weave a larger picture, but if taken out of the discussion and considered as an individual statement of How The World Is many papers don’t fare well.

  7. Mia Consalvo says:

    hi there,

    A friend pointed me to your post about my article, so I thought I’d respond to a few of your concerns about it. I’d agree that after doing a study, there’s no telling what the press will choose to emphasize or ignore, despite the actual findings. Actually, several articles even left my name off the study- fairly ironic given I’m the only female among the co-authors. So I’m not always that thrilled w/the coverage either. :)

    Anyway, to your questions. We chose EQ2 mainly because Sony would let us have access to their players. The vast majority of player studies of MMOGs rely on self-selected responses from recruitment via forums, blogs and friends, and the like. Thus there is no representativeness in the findings–its mostly the fan players who read the forums/blogs, who then respond. Thus, Sony letting us recruit via the game itself was HUGE, and they should be lauded for doing so. We got an enthusiastic response from players too. So while Rachel asks about the representativeness of those who wanted the ‘unique item,’ we were still getting a much wider swath of the game player base than past studies have.

    Another key aspect of the study was that we asked players about their play time, and then also had access to their actual play time. So basically, we could compare what they said, versus what they did. It’s fairly common with media use for people to under-report what they do, but we thought it was interesting that women did so more greatly than men did.

    Regarding why women play more than men and reasons for that, we were of course speculating based on past theories and findings, and that is mentioned in the paper. However, even if women are under-employed compared to men, they could still choose to do other things rather than play EQ2. Why they do, though, is something we are still working on.

    Unfortunately, our survey was designed to study a large number of things (this paper is just one piece of the overall findings) so we had to be somewhat economical in terms of the questions asked. I’d love to see more research in this area, addressing some of the questions you mentioned above. Please let me know if you have other questions I can answer for you about the paper. I don’t want to hijack the comments. :)

    • Blake says:

      Thanks for visiting the site and taking the time to participate.

      No study can possibly be everything to everyone, and it would be unreasonable to expect it to. I highlighted the shortcomings I saw with the paper in part to hopefully inspire or guide further research, and in part to counteract some of the other blog coverage I’ve seen so that future research might be ever-so-slightly less likely to be subject to the same simplistic, sensationalist reporting. This is one of the difficulties with having few voices talking about a subject (or at least, with quantitative numbers behind their conclusions); real-world limitations of that particular data set can end up swamped with over-determined expectations.

      I think it is great that you had access to that data and took the opportunity to investigate gender, even if I respectfully disagree with some of the discussion in the paper.

Comments are closed.