Gaming and Gender

Williams, Consalvo, Caplan, and Yee have produced a piece (pre-pub version) that touches on the gaming and gender issue. As you may recall, Williams, Caplan, and Yee published in July 2008 an article that attempts to debunk the gamer gender stereotype. In other words, the image of a young boys playing video games alone in their parents' basement is no longer an accurate one. The latest article, which has been accepted to the Journal of Communication takes a more obvious gender approach due to its use of gender role theory as a lens with which to view the data. The data used in this analysis came from a large survey dataset and unobtrusive behavior data from one year of game play. The focus of this work was on the massively multiplayer online game (MMOG) EverQuest 2 (EQ2), and the primary research question centered around the basic differences between male and female EQ2 players.

Starting with some of the demographics, 80% of the sample were male (N=2,006) and 20% were female (N=434). Williams et al. found that the females were older, less likely to be students, less likely to be employed, but played more hours than their male counterparts. Females played for social reasons, but the males played to beat the game. Also, the females typically played with a romantic partner, whereas the males played alone. In cases where the males and females played together, the males were often less happy than the females. Why? Maybe the males felt like their space was being invaded? Maybe they didn't like their female partners interacting with other male players? Maybe the females were as skilled (if not more so) than the males, and they didn't like being shown up by their partners? Questions, lot of questions.

There's more. Both genders underreport the amount of time they play video games. However, the female players underreported their gameplay activities at a substantially higher rate than the males. In terms of the players health, the males thought they were healthy and fit. Interestingly, even though the females played for longer periods of time, which means they were engaged in prolonged periods of sedentary activities, the females were actually healthier than the males. The health and fitness levels of the males were aligned with the assumptions but the female levels were not. Could this be the result of inaccurate self-reports by the females? The authors claim that more research is needed to further investigate this issue.

One demographic question the authors asked was related to sexual identification, which produced, according to Williams and his colleagues, unpredicted results. The analysis found that female players were more more likely to report being bisexual; in contrast, males were less likely to report being homosexual. These findings were surprising in that they did not match the general population estimates. Almost 4% of the males and approximately 7% of the females declined to respond to the question put forth by Williams et al., so the difference between the EQ2 players and the general population estimates could be even more pronounced.

Gender differences among individuals who venture into virtual worlds is a rather untapped area. Thus, research like the Williams et al. study is important to further our understanding of gameplay activity. This is a first step, and more research is definitely needed to investigate the plethora of questions that remain. Are EQ2 players unique or do they share characteristics with World of Warcraft (WoW) players or even Second Life (SL) residents? Longitudinal research is needed as well to determine whether these characteristics represent a snapshot in time or whether they persist over extended periods of gameplay in virtual environments that continue to evolve.

Adults, Video Games, and Contradictions

Pew just released a new report that focuses on adults and their video game playing activities. The data were collected October - December 2007 as part of a gadget survey. Information on video games is very popular, and it is surprising that it took Pew almost a year to process these results. This is a rapidly changing area, so I wonder whether the figures present an accurate assessment of the current population?

Regardless, the findings do represent at the minimum a snapshot in time of video games and the adults who play them. Here are some of the numbers:

• 53% of adults (age 18+) play video games. In fact, the 18-29 age bracket includes the largest percentage of game players.
• While the younger generations are more likely to play video games than older groups, the length of game play time increases with age.
• A slightly higher percentage of males play video games than females, 55% to 50% respectively.
• A higher percentage of students play game in comparison to non-students - 76% vs. 49%.
• Younger people prefer to play video games on consoles, whereas older individuals prefer games on a computer.
• Parents are more likely to play video games than non-parents.

Overall, the largest percentage of video game players are between the ages of 18-29, their estimated salary range is between $50,000 and $74,999, they have completed some college or are a college graduate, they are more likely to live in urban areas and be Internet users.

Questions: One question I have is related to the findings about virtual worlds. According to this Pew report, only 2% of gamers say they visit virtual worlds like Second Life (SL); 11% of teen game players responded that they have visited a virtual world. However, in an earlier Pew report on video games and teens, the percentage of teens visiting these worlds was 10%. A rounding error? The other numbers mentioned in the two reports match.

Next, Williams, Yee, and Caplan (2008) recently surveyed EverQuest (EQ) players. Their data from 7,000 participants suggests that the average of of game players is a little over 31 years of age. Stated another way, these researchers found that more players were over 30 than in their teens or in the college age range. So, are the game players younger or older? Like the current Pew study, however, Williams and his colleagues found that game play increases with age.

Another question deals with gender. The survey conducted by Williams et al. found that a large percentage of game players are male - 81% male in contrast to 19% female. The Pew findings were more balanced at 55% male and 50% female. Who's right? Or, is there a right answer? Is it possible to get an accurate assessment of video gamers given that it's difficult, if not impossible, to survey the entire population?

Will the Real Gamers Please Stand Up

When many people think about video gamers, the first image that comes to mind is the pasty teenage boy sitting in his parents' basement. But how accurate is that? Williams, Yee, and Caplan (2008) conducted a survey of 7,000 EverQuest 2 (EQ2) players. With the blessing of Sony Online Entertainment, the game operator, the authors were also able to collect in-game behavioral data. This is quite an accomplishment given that game operators are typically reluctant to provide gamer data to researchers. The goal was to combine the self-reports with the in-game data collected by the game operator to examine player motivation, as well as the mental and physical health of the players. Demographic data was collected via the survey instrument. The research approach taken by the group led by Williams differs from the common tactic. Typically, this type of research involves single-player computer lab sessions and self-report data.

Prior research suggests that there are physical and mental health risks associated with video game play. For example, a study by Anderson et al. released last week found a connection between aggressive physical behavior and video game playing among children. Williams and his colleagues wanted to determine the accuracy of some of the earlier findings about video game play and assess the characteristics of the players themselves.

For the most part, the Williams et al. findings contradict earlier research.

• Play time: ~26 hours/week (among all players)
• Average age: 31 years old [older players played more than younger; there were more older players in general]
• Gender: 81% male; 19% female [however, adult females logged in more hours]
• Race: Whites and Native Americans played more video games
• Income and education: Players were wealthier and more educated
• Religion: Players were less likely to be religious
• Media Use: Players spend less time watching television and reading newspapers
• Physical health: EQ2 players were slightly overweight but less so than the national average
  
• Mental health: Higher levels of depression, especially among the female players; higher levels of substance abuse; lower levels of anxiety
• Role playing: A fringe activity

Because many of the findings put forth by Williams and his colleagues do not match the rhetoric found in the literature, they conclude with a list of questions (i.e., areas for future research):

• Why are there inaccurate stereotypes?
• Why are older females playing at higher rates than younger males?
• Why are EQ2 players healthier than the general population?
• Why do minorities play video games at lower rates than other groups?
• Why are players less religious than non-players?
• Why are mental health problems more prevalent in players than in non-players.

The introductory paragraphs indicate that in-game behavioral data were collected by the authors. Based on this article, it isn't clear what information was collected and how it supported or refuted the survey findings. In the "Sampling and Procedure" section, Williams et al. note that there was a link between the survey participants and the in-game data. However, it appears that the main measure collected in-game was playing time. While this could be used to support the self-reports associated with playing time, it doesn't do much to inform the other variables. One point the authors emphasize is that much of the past research is not generalizable because it is difficult to get access to the players. These authors were able to get access to the players, but the findings still appeared to rely heavily on the self-reports of players, which can be inaccurate.