Study: Minorities underrepresented in games

USC researcher surveys 150 best-selling games, finds zero Hispanic or Native American protagonists; women, children, and the elderly also disproportionately rare.


Recent years have seen the explosive growth of games into previously untapped demographics. And while publishers have pursued these new markets by focusing on accessibility and games in nontraditional genres, they're missing an opportunity to open the field up even further, according to one University of Southern California researcher.

Black characters are underrepresented outside the sports genre, with Williams saying games often portray them as
Black characters are underrepresented outside the sports genre, with Williams saying games often portray them as "gangsters and street people."

A study published last month in the journal New Media & Society surveyed the 150 best-selling games from March 2005 through February 2006 (which accounted for more than 95 percent of game sales over that span) and tallied up the race, gender, and ages of the characters within. The study, conducted by Dmitri Williams, an assistant professor at USC's Annenberg School for Communication, found that adults, whites, and men are disproportionately represented in games, to the detriment of nearly all other demographics.

"The results show a systematic overrepresentation of males, white and adults and a systematic underrepresentation of females, Hispanics, Native Americans, children and the elderly," Williams concluded.

As might have been expected, men were grossly overrepresented in games. While they make up slightly more than half the US population, men account for more than 85 percent of game characters. As far as age groups go, adults were overrepresented (nearly 87 percent of game characters compared to 59 percent of the US population), at the expense of children (less than 4 percent of game characters but 21 percent of the country) and the elderly (less than 2 percent in games, but more than 12 percent of the population).

The study also found that white people make up about 80 percent of game characters, compared to 75 percent of the US population. Asian/Pacific populations were the only other race category to be overrepresented (5 percent in games compared to 4 percent in the US), with Hispanics and Native Americans both significantly underrepresented.

Hispanics make up less than 3 percent of game characters, but more than 12 percent of the population, while Native Americans represent less than 1 percent of the country and make up less than 0.1 percent of game characters. On top of that, the two groups appeared solely as supporting characters in games, with the researchers recording no Hispanic or Native American protagonists in their study.

Williams said the underrepresentation could be interpreted by players as a message that the missing minorities are "relatively unimportant and powerless compared to more heavily present groups." He went on to express concern that Latinos and other groups could be turned off to games by their lack of representation, and "subsequently, they may have less interest in technology and its opportunities for class advancement. Ironically, they would be less likely to become game makers themselves, helping to perpetuate the cycle."

Interestingly, the breakdown of gender and race representation in games closely mirrors that of US game developers as a whole, with one exception. Williams said black characters are overrepresented compared to the number of black developers in the workforce.

Portrayals of Native Americans are rare in games. Prey and Brave have come out since the study was conducted, but examples are still few and far between.
Portrayals of Native Americans are rare in games. Prey and Brave have come out since the study was conducted, but examples are still few and far between.

The researchers looked at the numbers for player-controlled characters, as well as supporting characters, and weighted the results according to sales. As explained in the article, "when the most popular game (Madden 06) sells over 6 million copies and the least popular (game #150 in the present study's sampling frame, BeyBlade) 15,000 copies, it is safe to assume that one game will be played significantly more than another. Thus, if the goal is to measure what the public is actually consuming, content from the two should not be given equal weight in the analysis."

There are other factors acknowledged by Williams that should be taken into account. The researchers only considered human characters in compiling their numbers. Data was obtained for quasihuman and nonhuman characters as well, but not incorporated in the paper on age, race, and gender portrayal. The popularity of sports games influenced the survey's gender results, as there are an abundance of games like Madden NFL and the WWE SmackDown titles, but no games for women's professional sports.

Finally, Williams said there was a "striking similarity" between his findings and similar analyses performed on television programs. He suggested that Latinos and Native Americans are systematically underrepresented in many forms of media and expressed surprise that the situation persists despite the growing Latino population in the country.

Williams told GameSpot that he has another article that will soon be published looking at body shapes of game characters versus the body shapes of real people. He also previously studied the amount of profanity in games (and determined there wasn't a whole lot of it).

Watch live streams, videos, and more from GameSpot’s summer event. Check it out

Got a news tip or want to contact us directly? Email

Join the conversation
There are 193 comments about this story