Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 23, 2014
arrowPress Releases
October 23, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

The Gender Cocktail. Part I: Learning to Sample
by Dmitri Williams on 03/03/14 02:14:00 pm   Expert Blogs   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Sex. Gender. Gaming.

Historically, this hasn’t been the industry’s best cocktail. Even if we put aside for the moment that the US has some serious issues with female sexuality--we’re far more OK with eviscerated organs than breasts--gaming hasn’t exactly been at the forefront of the progressive movement.

As a kid playing video games I didn’t give this any thought. There was a princess, she was hot, and she needed saving. Game on. As a parent (with a daughter), it’s hard to not grow up a little. And like the violence debate, the great gender and gaming debate rears its head at pretty regular intervals--usually when a game does something truly egregious or on the border between cute and creepy.

Cards on the table, I’m moderate to slightly liberal, and I’m ready to be offended on behalf of the women in my life. Yet I’m also a researcher who wants systematic answers and hard data to base conclusions on. And on that score, there is evidence that hypersexualized portrayals of women leads to greater tolerance of sexual violence and even rape. But linking representations to attitudes (note that you can’t do a study to link to behaviors due to ethics) presupposes that we actually know how people are represented in the first place. So:

What is the state of representations in gaming?

Is it getting better or worse?

What fundamentally drives it?

It’s easy, and very tempting, to fire off answers to these questions based on the games you play and the people you know. After all, what could be less sexy, yet more important in all of this than sampling? I mean sampling, who even does that? Let’s start with what it is.

Sampling is the process of taking a few selected cases and then extrapolating up to a larger population. It’s how political pollsters can take 1,000 people and say they know what “America” thinks about something, within some margin of error. How can they do that? They know a lot about America thanks to the US Census. So, they know they need to get X amount of women, Y amount of Latinos, Z amount of people aged 30-35, etc. in order to get a representative sample.

Tackling issues of representation in video games causes a similar kind of exercise, only here it’s much, much harder. I’m going to cover the very simplest question first: What are the demographics of game characters?

Consider the following basic elements of that smallest question:

Do we care about playable characters, background characters, or both?

Do we count all games equally?

There are zillions of games. Do we pick them all?

How do we do this in an unbiased way so the results are beyond reproach?

The good news is that we actually have tackled this. I’m going to spell out the results, but you can find the full research paper here. The bad news is that we did it nearly nine years ago and games have changed a lot since then. Players, platforms, mechanics and business models especially have been all over the place. So what follows is essentially the “before” and there is no “after” yet. Aside: I run a game analytics company that’s starting up with its first customers right now. We measure characters, so at this time next year we can start providing more regular benchmarking.

Step 1: Find the list of what games were sold in a given time period. We used NPD data to track a single year of games sold across nine different platforms. We picked the top 150 games sold by retail, which covered slightly more than half of known dollars spent. Remember, back then this really was mostly a retail industry, so you could indeed track these things before the explosion of the App Store and microtransaction games.

Step 2: Don’t treat all games equally. Clearly Madden should carry more weight than Beyblade if for no other reason than that it sold a zillion times more copies. As a proxy for how much exposure content has, we weighted the results by sales. So, a game selling 7 million copies counts 7 times as much as one selling 1 million copies. Not perfect, but not terrible.

Step 3: Pick the characters. Hire a bunch of coders to view the games and record every single character in them. We looked at the first 30 minutes of every title and had at least two coders take independent records. We made a note of playable vs. non-playable. Our two coders agreed on the categories around 93% of the time. We covered over 5,000 characters.

Step 4: Throw it all into a database and look at the graphs.

So, what did we find? First, we decided to pick a baseline for comparison. You can argue what the game world should or should not be compared to and there is no perfect answer. We picked the US Census because we were looking at US sales only and because it stands to reason that the “real” world is a decent baseline. This lets us state that if a group shows up more or less often than the Census baseline, that group can be said to be over- or under-represented.

OK, results.

It may not be terribly surprising to see that there are more male characters than female, but the results were more skewed than we expected. First of all, the “real” world of the United States is 50.9% female, 49.1% male. If we were to see the “real” gender world show up in games, that’s what we would see in them. Overall, characters in games are heavily skewed male: 85/15.

But what about playable characters vs. the eye candy of the scenery? Here the skew is actually stronger, at almost 90/10.

You can make an argument that in recent years games have allowed for more character control. It’s often easier to customize your character now, and that can sometimes mean the gender. Unfortunately, in the wilds of the current marketplace it’s very difficult to give a systematic answer to that.

Does any of this matter? Here we have to be clear that we don’t have data on the effects of these patterns, just theories and research from other media. It’s pretty clear that the simple absence of a group is generally speaking not good for that group. Think about Blacks, Asians, Hispanics or Gays on TV. Before they were on, it was like they didn’t exist (the fancy academic term is symbolic annihilation), and then they had to go through a phase as villains or sidekicks before being able to be primary characters.

So no, I’m not claiming that the lack of female characters proves anything, but I suspect it probably doesn’t help much. Consider the young female player playing game X and thinking “where the heck are we?” At the time of the study, females were 38% of the playing audience, yet only 10% of playable characters. Is that related at all to the giant shortage of women entering computer science? Again, there’s no causal link in this study, but it probably doesn’t help much.

In Part II, I’ll go from simple numbers to actual body shapes and add some stats from the industry.

Related Jobs

DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer
Nix Hydra
Nix Hydra — Los Angeles, California, United States

Art Director
Avalanche Studios
Avalanche Studios — New York, New York, United States

UI Artist/Designer
Nuclear Division
Nuclear Division — Sherman Oaks, California, United States

Director of Product Management


Innes McNiel
profile image
This isn't surprising to me. I've long stated that if you go to most stores and pick random titles off the shelves, most of them probably won't have a female character in them at all, let alone a female lead.

Mark Velthuis
profile image
I think Step 2 might need some refinement. Madden should not carry more weight than beyblade just because it sold more. I would argue that it should carry less or perhaps even not at all. Madden is a game based on a sport played and watched mostly by males, with characters based on real life athletes, and I've personally never seen a female american football player (tho this in itself might indicate a problem). The same goes for football (soccer for americans), Fifa is another popular franchise based on a male dominated sport. While there certainly are female football players, they are the minority. I don't think these games that are based on real life sports should be "blamed" for not having female characters when that actual sport barely has any.

Now I'm not trying to deny there's a problem here. But I think we need to be carefull with these kind of statistics. I don't think anybody is going to benefit if we were to start putting female characters in these games just to make statistics look better.

A while ago I've read 2 disturbing reports about related cases.
*Publishers didn't want to have a female character on the cover of Bioshock Infinite
*Publishers didn't want to have a female protagonist in the game "Remember me".
I think someone needs to tell these publishers that digital females don't carry cooties.

Kaitlyn Kaid
profile image
I think if we were to start saying "X games don't count, and Y games don't count..." it will very quickly degrade the sample to uselessness.

Games like Madden that sell extremely well should count for more than "Barbie's fun house" that sells 1/100th the numbers because they represent more accurately what games people are buying.

Mark Velthuis
profile image
There's a couple of issues regarding that statement.

How would you account for the fact that many sports, a very popular genre in gaming, are extremely male dominated ? How would you add a statistically fair ammount of female characters to a game about a sport with allmost no female athletes ? Is this what female gamers would actually want ? Or should we compensate with female-only games ? And what if those don't become as popular ? Should we then conclude that female characters aren't as wanted as we thought ?

Keeping only sales into account sais nothing unless you add a bunch of other statistics, like : Who is it bought by? Is it bought by those people because or despite the domination of male characters? Would adding female characters have helped, hurt or have no effect on sales?

And then comes the question : Is it actually fair to claim "The real world is about 50/50 male/female, so games should too" ? There's researchers out there who claim that while the ammount of gamers is close to 50/50, they still have very different playing habbits. Looking at those habbits we might conclude we should be aiming at an optimal 70/30 spread. Keep in mind there are a lot of character-less games and games where gender can be selected. Not counting these is quite unfair aswell.

What I realy mean to say is, these statistics by themselves aren't realy saying much about what realy needs to be done. That's even beside the fact that they are nearly 9 years old.

Katy Smith
profile image
The "NFL is watched mostly by males" thing is a pet peeve of mine:

"Women represent approximately 45 percent of the NFL fan base, according to Scarborough Research, and approximately 33 percent of the NFL viewing audience based on Nielsen data."

"In terms of female fans, the NFL trails only college sports, according to data from The ESPN Sports Poll and the U.S. Census, with league officials saying 44 percent of all football fans are now women."


Mark Velthuis
profile image
I stand corrected on the watching part then. But the sport is played by mostly (only?) males still, right? Wouldn't that still make it very odd to have playable female characters in these games ?

Tyler King
profile image
There should be 2 sets of numbers. One that has all games, and another that has no sports games as those are going to heavily skew the results.

Katy Smith
profile image
You are correct, there are no female NFL athletes (although Holley Mangold totally should have tried). Sports sim games should reflect the sport they are simulating, so it wouldn't make sense to add female NFL players just because. However, there should probably be more female athletes in MMA or the Tiger Woods / PGA games, since they are currently underrepresented in relation to the real world.

Dmitri Williams
profile image
So the study doesn't claim anything about "should." It really only is trying to measure what is.

As for the sample of games and the presence of sports, the reality is that sports games are heavily played and so generate a large amount of impressions. Would I like to measure by something more accurate than sales? Yes, I would. If I could get every login session and tie it to the demographics of every user, I could make a stronger claim. That's just not possible to do, though, and so I acknowledge that this sample is good, but not perfect.

Shea Rutsatz
profile image
Mildy interesting sidenote: NHL 2012/11/something had a female created character by default.

Randall Stevens
profile image
I agree that a sports simulation should accurately reflect the sport, but how would MMA or golf games be handled?

MMA leagues do not mix genders, and I can't imagine them opening that up.

Golf is largely segregated, though a few times a woman has qualified for a men's tour, but at the bottom and with a poor showing in the tour. To accurately simulate the sport of golf (which I think a woman could do very well at, it just hasn't happened yet) would you be handicapped just by picking a female?

There could be a strictly WNBA game made, but I can't imagine a basketball game being made where NBA and WNBA teams competed without upsetting someone.

I have no problem with this being included in the data, but if you are then using this data to draw conclusions about the games industry you are not accurately representing this sampling. If (when) this is used to make a statement about the games industry being sexist, it will be misrepresenting the industry. The creator of a football game isn't sexist because there are no female football players in the game, just as the creator of a military game based on the navy seals isn't sexist because it accurately reflects that there are no female soldiers in those units.

I'm not saying there aren't problems, but this sampling doesn't accurately reflect the issues.

Julian Cram
profile image
Actually for soccer, the percentage of women players in the USA is 40% to males 60%.

In Australia, the amount of girls who play soccer is equal to the amount of boys in primary schools. Australia's women soccer team consistently do better than the male soccer teams in international tournaments and competitions.

So I think you're absolutely wrong - these games should be blamed for not having females represented, because it's not a true representation of the sport, particularly in the case of soccer.

Thor Rienguard
profile image
Julian these games are not trying to represent a sport. They are trying to represent real life leagues and real life teams. Teams and leagues that do not have a single female player playing for them. If these leagues and teams allowed women, and they were made up of 50% women you would see these games be made up of 50% female characters simply because these games are trying to mimic reality. Right now they mimic reality.

It is disingenuous to include such titles into such a study.

Mark Velthuis
profile image
Look, perhaps I chose some wrong examples, but all I'm trying to say is that if a game reflects a real life activity, and that real life activity does not have an equal gender representation, the game should not be blamed for not having an equal gender representation. And this should be taken into account when making statistics like these aswell.

btw, a while ago EA asked the fans if they should include women in FIFA games.
es-should-be-included-in-the-future/ in case anyone wants to add a vote.

Merc Hoffner
profile image
This is a great study and something I've long been wondering about! I'd love to see these stats re-collated over time. I had this suspicion over the last generation or two that the rate of female protagonists actually decreased significantly from a higher point, but this could be my imagination - which is why figures charting over time would be such a boon!

If it would be possible, I'd also love to see the breakdown of abstract/genderless/non-human protagonists also over time. Not to be difficult or play down the significance of gender representation, but also to chart the move of the industry regarding 'realistic' characterisation. I believe the early days of gaming were much more free to invent weird stuff, so a protagonist could be a tribble, or a unicycle, but the expectation now is much more anthropomorphic.

Sometimes I feel like we're Star Trek in reverse: all the aliens in Star Trek were humanoid because the production designers didn't have the budget or technology to do anything more unhuman than a guy in makeup or a suit, but they gradually got weirder over time as prosthetics improved (old vs new Klingons!). Meanwhile early games were totally limited in representing humans and started out fairly abstract. But as the technology has improved, gaming has anachronistically focussed on increasingly realistic representation of mundane humanism (both physical and moral!). Perhaps that's only my view. Perhaps there's more variation with the resurgence of 2D in the burgeoning indie market (which would actually support the trend - a step back from AAA tech is a step forward for abstraction?). The point is, the numbers would tell us the trends!

Dmitri Williams
profile image
I'd love to see this over time as well. That'd take effort and funding, and this was done out of personal interest, not driven by a foundation. If the ESA or some agency were interested, I'm sure researchers would love to make it an annual benchmarking study.

On the human/alien part, I'll note that we removed any non-humans from the analysis. Originally there were around 9k characters and 5k humans. We have the data on the non-humans but didn't do the analysis on it because we had no clear research goals for aliens . . .

Merc Hoffner
profile image
Lol, well even knowing that ratio (9k to 5k) is interesting. I wonder how pokemon factors!

I don't suppose there's a way that you could get the crowd to update the stats for you? Like a Wiki or something like it, with a public database recording character compositions for new games. Then let users apply their own analytics criteria to the dataset? Complicated I suppose. But it would be interesting!

Alfa Etizado
profile image
Eager to see part 2, specially if you analyzed the range of of designs that exist for male characters and compare it against female characters. Male characters can be anything, from Raziel to Drake, while female characters are usually stuck somewhere between a non sexualized design, but still young and attractive (think Alyx or Jade), to a super sexualized design that makes T&A the primary concern.

Dmitri Williams
profile image
Part II will show literal wire-frame body shapes, but the focus is on averages, not the full range.

Satish Shewhorak
profile image
Hi Dmitri,

Interesting article. I recently did a quick analysis of the gender split in lead/ playable characters of 421 games released in 2013 (listed by Wikipedia). It wasn't weighted by sales but I think you might find it quite interesting?:

According to ESA men make up 55% of gamers and women 45%. So pretty even.
However 52% of games had a male lead and only 5% had a female lead (43% had mixed/ unspecified leads).

The gender split of gamers is almost 1:1, whereas the ratio of specified gender leads is closer to 10:1 male to female leads, which is a big disparity.

My list of games did not include casual apps, which you might hypothesise that women play more than 'hardcore' games, but a quick skim of the Top 100 games in the appstore will find a similar 10:1 imbalance, possibly with more mixed/ unspecified.

Tab 2 shows a geographical breakdown but it is not weighted by population or games released per country.
However America released the most games with female leads (9), whereas the UK and Germany only released one game with a female lead, Velocity Ultra and Giana Sisters respectively.

I did not breakdown ethnic representation as the figures would have been negligible, which is a shame in itself.

Fan Zhang
profile image
Ha, Sorry to wandering off the subject somehow, but you need to know that Japan, aka probably the "most hated country by western feminists in gaming world", created games with female leads almost a hundred times more than any western country, like, every year. And surprisingly some of them are not treating female characters like some kind of objects, not limited to Touhou Project doujin works, there are many otome games designed for female market (if there's a male lead in it, it must be a boy's love story), or niche visual novels stick on yuri (lesbian) subject, 0/10 would care for gross otaku eye candy.

And some western gaming media still think Gone Home is somehow more sophisticated or advanced, blaming Japanese gaming industry for blah blah almost everyday.

Satish Shewhorak
profile image
I wouldn't say that there is a consensus from western feminists that Japan is the "most hated country in the gaming world". Where does that quote come from or do you have any other references to back this up?

My analysis (Tab 2) shows Japanese developers released the 2nd largest quantity of games with a female lead to the Western market in 2013.

I do not have access to Japanese game release data but if it is as diverse as manga comics then you may be right.

However there are some games that hypersexualise/ fetishize (Senran Kagura Burst) moreso than western games, which still try to hide behind the defense of satire (Duke Nukem Forever).

evan c
profile image
Maybe those surveys should also include what are the games those women are playing.

Dmitri Williams
profile image
Satish, this is really cool to see, and I encourage others to click the link. You might consider writing this up to share more formally here. You'll get some pushback on sampling/weighting, but that's unavoidable and it's way, way better than nothing.

My takeaway point is that the results from 9 years ago to today aren't very different.

Mike Jenkins
profile image
"First of all, the “real” world of the United States is 50.9% female, 49.1% male. If we were to see the “real” gender world show up in games, that’s what we would see in them. Overall, characters in games are heavily skewed male: 85/15."

Would we? What percentage of games involve violent combat? Wouldn't this be far more relevant than 51/49 in those games? Is the fact that these numbers match up almost perfectly just a coincidence?

Katy Smith
profile image
Apologies if this was answered in the full paper, I only skimmed it. :)
How did you count playable characters in sporting games like Madden? Did you count all 1,696 players on active rosters, include free agents, or did you only count the starting players? I can see the analysis getting tricky when including sporting titles.

Dmitri Williams
profile image
We had an expert player play each title for 30 minutes. We recorded this 30 minutes of play and that was the sample material. So, it's whatever characters showed up in those 30 minutes. That means the sample doesn't include typical endgame content and is skewed toward the early parts of play. This is the most consumed part of the content, though.

Ben Larkin
profile image
Although I can understand the need to simplify the methodology of a study in order to generate results that are consistent and readable, simply using retail sales as a criterion for choosing the games may skew the results. Certain genres of games may be more popular (and thus sell more) because of their patriarchal or male emphasis, since (as your stats demonstrate) more games are being played by a male audience. So the results you get for male/ female presence in games will be biased before they've even been tabulated.

That being said, as a general debate about a male-dominated gaming sphere, I agree with most of what has been said. I quite like Satish Shewhorak's analysis, as well. Shows us that despite female presence in the gaming world increasing, presence in-game has yet to change significantly. Perhaps the narrative structure of games--in conjunction with commercial motives they deem to "work"--has stagnated the industry. Just speculation, of course.

Razvan Pricope
profile image
I understand your point of view, but a huge amount of videogames are violent (or about some war). It is awkward to see a videogame about the first crusade in which the main character is a Bloils black female soldier because it has nothing to do with the reality. It is also awkward to have 2 female aces in a 4 people Luftwaffe squad (Sid Meier's Ace Patrol) when in reality very few women fought in dogfights (if I'm not mistaken the only female ace was a russian girl who died at 21). If you bring elements that are out of place in a historical event, you are just slaughtering that event.
Same with the men-predominant sports - is there any women in Manchester's football team? No! Do you think that only men are watching football? If the men/ women viewers % is 50-50 do you think that they should put 50% men and 50% women on the field?
This being said, I think that games about a sport or about a historically event should also be excluded from your charts.
The real problem is not that "most" of the games are about white men, but that people don't want to risk and to exploit other events and activities; instead they stick with whatever was already successful in the past and with naked women.

Wes Jurica
profile image
This females/minorities in games thing reminds me of the theme tackled in Star Trek: Deep Space 9 episode "Far Beyond the Stars" ( Publishers and developers go with what is perceived to be their audience's wants/needs/preferences for fear of shaking things up. I find it really hard to believe that someone would skip out on a game if it had a black, female lead (or whatever race and gender they are not). Look at all the black, latino, female, gay, or non-white, male gamers that still play games.

Kaitlyn Kaid
profile image
I find it totally believable. I've seen it done in books, no reason to think it doesn't happen in games.

Dmitri Williams
profile image
Some of it is perception, and some I think is due to content creators simply making content that reflects themselves and their interests. I see nothing sinister in this. I think it's systemic, though. More on this in part II.

Ben Larkin
profile image
That's one of my favourite episodes, actually. Star Trek was able to tackle serious political and ideological issues by using species as proxies for human traits which were particularly unfavourable--Klingons for aggression, Romulans/ Vulcans and an excess of logic and reason, Cardassians for oppressive political systems and a misuse of ambition; grander issues such as territorial dispute, religious pluralism and so on. And, as you say, race and gender politics (Kira's character using a pseudonym, as well as Sisko's). A little off-topic, but still.

I think that the profile of people playing games differs significantly from the expected, appropriate and accepted stereotype of the protagonist or indeed any character in a game setting. Commercial motives may have driven this, but it's almost a given--which, of course, if dangerous. Assumptions are what often lead to dispute. Game designers are certainly trying to innovate in that regard, particularly on the independent side, but so many ideologies surrounding every stage of game design--from narrative, to characters, to environments--are still being influenced by what I call the easy stuff (formulae for money, images of women in society writ into gaming logic, men as heros, saviours, rarely undergoing any kind of emotional catharsis). I could go on forever, really! Such is life and culture. We're not in any better or worse a position in terms of actualising ourselves as designers, just a different one. So we have to just embrace the changes as they come.

Christopher Landry
profile image
It's been mentioned that the method for weighting games solely by sales is causing fairly obvious issues for the sports games in favor of male representation in the samples.

To put it simply, the sports represented in the games (NFL football, American Soccer, etc.) are typically 100% male players, so those games SHOULD have 100% male players to be true to the real life counterpart. If those games are then given 7x weighting bias compared to a non-sports game, as the example given in the article for Madden vs. Beyblade… I think it should be obvious how that can skew the results grossly in favor of male representation.

I want to point out that this logic also applies to many other game types. For instance, military shooters. I served in the Marine Corps for several years, and I can assure you that there is no military branch that has a 51/49 split like the overall US Census data. I would be surprised if it's not more like 85/15.

The samples need to be divided into categories of representative attributes and compared to the real life counterparts for those specific categories alone. Then we can point to one category or another and specifically say that one is a problem or not.

For the sports games category, I suspect the results would show that our games are nearly identical to the real life counterpart.

For the military shooters category, I suspect the results would show that our games are close to real life, but could use some small improvement. For instance, if the games are 99/1 and real life is 90/10, there's a small improvement to be had, but it might be difficult to implement fairly. Would it make that much of a difference if the game had the option of selecting 1 of 10 character models for the playable character and one of them is a woman rather than none? Would it be a good thing to make 1/10 soldiers on the ally and enemy teams women? Is that something we want to do (making players shoot women)?

I'm not sure it would be an improvement,. Maybe it’s just some chivalry in me that balks a little at shooting women. And since we’re talking about “making games more representative or real life,” should we go the next step and have our Middle East military shooters put guns in the hands of children and have the players make the hard choice of shooting the child before it kills the player? That happens in real life, so should it happen in our games as well? I really can’t find a justification for that, personally. Maybe it’s better if all of the enemies are faceless nobodies, since that’s what the real life military tries to teach soldiers to view their enemy as.

And if someone wants to say that we’re not talking about children with firearms here, I want to point out that “fair representation to the real life counterpart” implies exactly that when taken to its logical conclusion. Pretending that this only applies to male/female representation is an idealistic fantasy.

I suspect that fantasy games would need to be their own category, and if any category can, it should be the closest to representing the overall US Census data, except where specific exceptions exist. By this, I mean things like a species of all males/females, like the Asari in the Mass Effect franchise. I find it interesting that the category of fantasy, the least like real life, may actually be the closest we get to matching the overall Census data.

Real Time Strategy games that are focused on military-like actions, such as Starcraft, Command and Conquer, and Company of Heroes, should probably be included in the broader “military game” category along with military shooters. Again, if we’re representing the military, even a fantasy military as in Starcraft, it should not be like more evenly split between male/female representation than we would expect to see in our real world military.

I can't guess what other categories should be used right now, but without separating the samples into categories, we're getting some gross misrepresentation in the results.

Kaitlyn Kaid
profile image
If this was measuring the demographics of the games being made, then yes I would agree with a 1:1 comparison between the games... however it is looking at what is being *played*. If one game is being played by seven times as many players it should be worth seven times as much, regardless of what genre it is.

When you start saying "this type of game shouldn't count, and this type of game shouldn't count" that is when you REALLY start twisting the data into something else.

Dave Bellinger
profile image

I understand what you're getting at, but it's a tricky situation:

Having sports simulation games considered using the given metric will always result in the predictable outcome of being heavily skewed male; as of yet, there's no resolution to improving this (such as creating female players for next year's Madden -- unless a female player does actually compete in the NFL)

But by utilizing sports simulation as a data point, the research is effectively 'weakened': The BioWare/Saint's Row/Bethesda players that knows a (generally) gender balanced gaming environment doesn't look at these numbers as 'eye-opening', they look at them as inaccurate. If the research is not taken seriously, how can it have the desired impact?

Obviously the best-case scenario is that the research is just taken seriously: but if we're disregarding games with no gender-assigned characters (Tetris, Marble Madness, etc.,) why would we consider games with gender-accurate characters?

Christopher Landry
profile image
I am not advocating that any game shouldn't count. I am saying that there should be separate categories for the different types of games based on what we would expect the real world counterpart to look like.

Sports and the military are predominately male populated in real life, so they should not be in the same graphs/samples as, say, Diablo III, which is pure fantasy and could conceivably be as balanced as the developer wants it to be (i.e. as balanced as the overall US census data, if the developer wants it to be).

My argument is toward separation, not exclusion. I want to promote understanding, not ignorance. The composite (everything in one group) graphs/samples in the post do not represent any reality I am aware of, and so they are greatly misleading.

Dave Bellinger
profile image

To take a look at the other side of the coin, I'd have to say that separation in itself would in effect be exclusion, wouldn't it?

Given the current topic, if the data were to be separated into a different graph, why would the sports simulation and military graphs even be considered if it's already a forgone conclusion? Given the circumstances, that data would still be considered 'excluded' I'd think.

Dmitri Williams
profile image
The study goal was to look at the characters that players see. Players play sports games, in addition to FPS, RPGs, etc. I wouldn't exclude sports. They're part of the universe of consumption. They aren't skewing the results--they're an important part of the sample frame. Remember a content analysis like this has no judgment per se. It's simply reporting proportions of what is. In this case, what is played includes a lot of sports, which means a lot of male characters. You don't remove it because it seems too anything. It is what it is, part of the results.

Now I can see interest in seeing results by genre or ESRB ratings, etc. You get some of that in the full paper, but we didn't dig that deep.

Christopher Landry
profile image
I agree that all of it is part of "the universe of consumption," that is why I am saying that nothing should be "excluded," only separated.

What I am saying is "Have more graphs!" Show me all of the truths in the research, not just the single comprehensive truth.

I want to see things like this:
- A graph that shows the whole of the results in one big group.
- A graph that shows the breakdown for realistic military-esque games
- A graph that shows the breakdown for realistic sports games
- A graph that shows the breakdown for fantasy/self-realistic games.
- A graph that shows the breakdown for puzzle games that include at least one humanoid character portrait that interacts with others or the player.
- A graph that shows the breakdown for cow-clickers that include at least one humanoid character portrait that interacts with others or the player.
- A graph that shows the breakdown for playable characters across all groups.
- A graph that shows the breakdown for playable characters in military-esque games.
- Etc. through the list same groups.
Some of these groups will overlap with each other, a game may be in more than one group, and that is perfectly ok. The idea is to provide lots of information and let the reader come to their own conclusions.

With a slew of graphs/samples like this, I can point to one group and conclude whether or not it is successfully portraying the limited reality it claims to portray. That is really the issue I am getting at: Each game portrays a limited reality, not the whole. Comparing Everything to Everything is a good start, but there are few games that try to be Everything. Maybe the MMO’s can be classified as Everything games, as they usually create an entire working world to play in. Most games are only trying to be one small piece of Everything, and if we don’t respect that in our sample groups, we are not relating the truth that those games are trying to portray.

If the goal of the post was simply to shout “Hey, it’s imbalanced, y’all!” then it was successful and nothing should be added or changed. However, if the goal is to inform with the intent to spur change and improvement, I have to conclude that it falls short of being helpful. Will the upcoming posts in the series be more helpful? I’ll definitely stick around to find out.

Johnathon Tieman
profile image
@Christopher Landry:

As an interesting parallel to your argument, the same sort of arguments are being made about the existence of the gender wage gap. When you compare all men to all women, it looks like there is some sort of problem, but when it is broken down by career field and other factors, the wage gap vanishes.

This is one of the big issues with doing these sorts of studies - collecting the statistics and producing the graphs are certainly scientific, but deciding *how* to break down the statistics isn't scientific. In many cases, how to break down the statistics can end up revealing personal biases the author(s) may have (for the record, I'm not saying that is what happened here). In the wage gap example, the way the statistics were originally broken down showed the author(s) had a bias for a socialistic economic system (i.e. since 51% of the population was women, they should control 51% of the resources). Since America has a capitalistic system instead, adding in controls for career field, years of work, and other factors that impact things in a capitalistic economic system caused the wage gap to vanish.

Unfortunately, our social system is vastly more complex than our economic system, which makes determining valid breakdowns even harder. Given that there isn't any sort of consensus as to what sort of society should exist, it is pretty much impossible to find anything that would arguably be a "correct" breakdown, which is why I find your suggestion for breaking down things as much as possible as one of the most vital things that needs to happen for studies such as these. Hopefully Dmitri Williams will release the raw data he acquired, so even if he lacks the time to do so, others can continue breaking down the data in other ways to see if any useful results appear.

Cordero W
profile image
Can we get an article on the representation of dogs in video games? I don't think there's enough of them ever. And what about velociraptors? When's the last time you saw one?

Katy Smith
profile image
Thanks for comparing me to a dog -_-

When dogs and velociraptors start playing and making games, we can start worrying about their representation as playable characters.

Cordero W
profile image
I didn't compare you to a dog. I just think this article is silly and a waste of time. I liked when we wrote about video game mechanics and implementation.

Katy Smith
profile image

Then why comment? There are plenty of other articles on Gamasutra that cover lots of different topics. In fact, that's why I like Gamasutra so much. It's people who are in the industry acting with people who aren't in the industry having awesome discussions about anything relating to video games. Personally, I'm bored with the "what is a game" argument, but I understand it's important to some people. I don't go jumping into the comments section and poo poo any discussion on the matter.

And I would argue that diversity issues are very important to both the Art & Business of making games (see what I did there?)

Article after article comes out talking about closing studios, ballooning budgets, and "why do we keep making Call of Gears of Duty of Warcraft?" risk-averse game development. One of the theories is that the 18-35 year old cis-white-male demographic is not getting bigger. It's the same size it was 10 years ago. Diversifying games, whether that be through playable characters or exploring other topics is one way to help the industry grow again.

Now, I've seen arguments on other articles that can basically be summed up as "Lara Croft, so your argument is invalid." Dmitri Williams decides to actually do something about that and examine, scientifically, what the breakdown is. This is exciting stuff! Is there really an overwhelming disparity between male and female characters in games, or is it the wrong place to look? How can you claim that this is silly and a waste of time?

And you are right, you didn't compare me to a dog. In three sentences you attempted to shut down a discussion you weren't a part of because you didn't like it. You compared the daily frustration I feel as a game maker and game player to the obliviousness a dog feels about not having the same opportunities as a human.

It appears that there are at least 40 other people here who don't feel that this is silly or a waste of time. I like to encourage discussion about video games, not shut them down.

Masaru Wada
profile image
While I agree Cordero's argument is weak based on the fact that none of those demographics actually play games, he also didn't call you or compare you or women to dogs.

Thor Rienguard
profile image
"How many in-game characters are women? A glance at the stats"
That is the title of this article that Gamasutra posted on their main page. This is an inaccurate take on the survey at hand.

The survey is looking at the characters of games being played. Not the characters as a whole. Unless you are trying to tell marketers that gamers prefer games that are male dominated, which is a chicken or the egg situation, this survey is useless to the industry.

If all you are trying to say is how many characters are women you have to look at only that. By putting weight towards certain games based on sales you are not taking an accurate look into how many in-game characters are women. You are creating biases towards games that are more popular, and not showing an accurate look at the industry as a whole.

As has already been brought up, when look at sports games these games are typically based off of real world leagues, teams, and players. They aren't making a conscious or sub-conscious decision to not include women, they simply cannot include women because it isn't accurate to what the games are based off of. Adding women would be akin to recreating my family in video game form, then adding a third child who is black in an otherwise all white family. If you are trying to make an accurate representation of my family it makes no logical sense to do that if you are trying to stay true to the source material.

A lot of times in these kinds of topics subtlety and context are forgotten or ignored. When in reality it is something that needs to be taken into consideration. Take out the the weighting of titles, then show us the data again. Only then will we really have accurate information that accurate opinions can be formed from. You can't control what games are bought. If consumers are not buying the games that have female characters that is on the consumer, not the industry.

Justin Kovac
profile image
What if a game is 60/40 m/w, but you only see it as 90/10 in the first 30 minutes?

Bernardo Lazo
profile image
I think it would be very interesting to additionally show the data (% of characters) without the number of units sold in the equation (weighting I guess is called). The difference could give us a clue about whether it is a creativity thing or a market dominance one.

Most of the top selling games are a legacy of times when there were not may female players:
In AAA titles, producers/developers have 50-100 million reasons to not change the formula at all, pretty much like a Hollywood franchise. So, most of the unit volume sold nowadays (AAA) is a creative legacy of a time when female players were no more than, say, 10% of the market

So, could it be that the majority of the market buys what is out there, creating a chicken and egg problem?

PS. Indies to the rescue: Our current production, Enigmarella, has a female lead. if you are interested.

Stephen Phillips
profile image
I look forward to having regular updates on these statistics from now on. I love the annual Salary Report and think we should be keeping track of much more information in this industry that quickly changes.