Hey, folks. Things have been quiet the last several days because I’ve been working on another three-parter. (What can I say? I’m a glutton for punishment.) I’ve been picking on video games a lot lately and was feeling an itch to go back to pen and paper RPGs. In the past, I’ve looked at the D&D 4E core books as well as the D&D press kit so I thought I’d take a look at a year’s worth of Dragon Magazines and see how they stack up against the sources I’ve already looked at. (Get it? Stack up? Magazines? …oh never mind.)
Numbers
It shouldn’t come as a surprise to anyone that the art for Dragon Magazine Issues 383-394 displays clear sexist trends:
Yep. Women are underrepresented, more likely to be found in neutral poses, comprise the minority of fully-clothed characters, and are far more likely to be suggestively attired. And of course, their chances of being depicted as a fighter are pretty slim when compared to their male counterparts. Again, nothing new or suprising here. I’ll grant, as always, that at least D&D does better in terms of numbers of female depictions when compared to other gaming sources. But women are still consistently under-represented.
What’s interesting is when you take a look at how these numbers compare to the numbers for the 4E core books:
The interesting thing is that while the numbers are pretty much the same, the numbers for Dragon Magazine are just slightly worse across most categories – suggestive depictions being a notable exception. There are slightly fewer active women, and slightly fewer women overall. They are a little less fully clothed, a little less likely to be fighters and a little more likely to be thieves.
The suggestive depictions pose an interesting wrinkle. About 70% of all suggestive figures are women, down from 80% of all suggestive figures in the core books. However, a little less than one half of women in the core books are depicted as suggestive while almost three quarters of women in Dragon Magazine are depicted as suggestively attired. So while the number of suggestive male figures has increased, it doesn’t seem to have kept pace with the increase in suggestive female figures.
I’m still working on the other stuff
As mentioned, this will be a three-parter. Next time I’ll do an images post picking out some points of interest. I’ll also be doing an entire post about Shelly Mazzanoble, who will take up too much space to cram into this post.
Since I realize that today’s post is a bit light on content, here for your amusement…
Caesary sinks to new depths
Caesary is a browser-based game owned by the same folks who publish Evony, so it shouldn’t come as a surprise that they use the same tactics in their advertising. Still, this is pretty ridiculous, even for them:
(You can go here to see the page it came from, complete with nifty animations.) I mean, wow. “Real Men”? They do know that the game has absolutely nothing to do with actual women, right?
Go Make Me a Sandwich 
That Caesary ad is… something else. o_0 “Your only reason to live besides women! Play to become a real ladies man!” Seriously? And then they say the game is “for mature adult” — isn’t it a top-down strategy game?
I would be for any proposition that involved kicking anyone who was convinced to sign up by that monstrosity out of the gene pool entirely. =P
isn’t it a top-down strategy game?
Yep! Clearly they’re going after customers with no brains.
Clearly they like their ladies with top’s down?
Give them some credit – they managed a picture of a woman in a pose that humans with skeletons and internal organs can assume.
But wow – that’s just bad.
You say that like they actually are responsible for any element of that image.
They probably just Googled something about seductive-looking women and randomly went with whatever they found. Because this is exactly how they assembled the Evony ads.
I really like your blog. I think you are doing great work quantifying and cataloging the kinds of sexism prevalent in gaming media. It is great to have these posts as a resource to show people who would like to just wave this problem away or ignore it.
I do have a question about your graphs that has bugged me since I first saw them. Why do all of the values in the graph have % signs next to them?
You have the % scale on the left and set up the graphs so that it is easy to follow all the way across. In addition to that there are % signs next to the values. Those seem to be incorrect (unless I am misunderstanding the graphs). For instance, in the first graph today under “Neutral poses” the numbers read 49% vs. 60%. That would add up to 109% and the values conflict with the scale to the left. The next column is “Fully Clothed” which is 48% vs. 28%. That only adds up to 76% and the values listed also conflicts with the scale.
When reading the numbers, you have to read across. I.E. – Male Active + Male Neutral = 100% of all male poses. So 49% and 60% are the respective numbers of neutral poses for each gender, but they don’t add up to 100%. That’s why I went with the stacked bar graph so that you can see what percentage of each category each gender accounts for. Like, it’s useful knowing that women are ~40% more likely to be suggestively clothed than men, but it’s also useful knowing that women account for more than 70% of all suggestive depictions.
As for Suggestive and Fully-Clothed – those numbers don’t add up to 100% because they’re not mutually exclusive. A character can be fully clothed and still suggestive if they have a cleavage window. Or they can be neither fully clothed nor suggestively attired, say, if everything but their arms is covered.
Thanks for the explanation.
After your answer it makes sense now.
if this is the same thing I think it is? Then did it not use to feature a toga-wearing big breasted blonde with a concealed push up bra? If it did this is better, but it is still brainless crap.
Sometimes I have to hand it to the Evony company, they really know how to sell their shitty Age of Empires II knockoff, don’t they? Guess there are enough gullible people out there for this to work, barf.
At this point in time, I’m still surprised Dragon magazine still exists. It feels so.. outdated(?) I don’t really have any hope that they’ll cut it out with the cheesecake
I assume this is the version of Dragon that’s been around since ’08 or so…the online-only thing.
Not that the paper magazine (published by Paizo, who now does the Pathfinder books) was much better about suggestive/scantily-clad women (that I recall), but it was still incredibly good if you regularly did D&D games. Online-only publications like that will never match something you could actually have in-hand.
Yes, it’s being published in PDF now. I have to say I prefer print magazines too but given the cost of printing and distributing magazines it makes more sense to go digital – especially since they’re trying to consume people to consume D&D content on mobile platforms.
Wow, I can’t believe they managed to top the Evony advertising for offensiveness. I’d like to hope it will fail due to how badly they are insulting their audience, but I know better. All gaming media is insulting to men(ie “we know you’ll buy any junk we put out there as long we put tits on the box”), and we seem to lap it up.
If I get energetic, I’ll go through my game cases later and run some numbers to see if it’s actually true. Hopefully I don’t dissappoint myself.
Alright, I decided to make it easy on myself and just do my Xbox 360 games. I have a total of 43(when the heck did I buy all of these?!?!?! There’s at least 5 I haven’t even put in the console yet!), so it’s a pretty decent sample size.
Here’s my results:
# with no characters on the box: 7
# with only male characters on the box: 18
# with only female characters on the box: 3
# with both male and female characters on the box: 15
The rest of these are going to be a bit more subjective.
# where the women on the box were objectified: 4
# where the men on the box were objectified: 1
My standards on what is and isn’t objectification don’t necessarily match up with everybody else here. Generally I labelled it objectification if there is a combination of alot of flesh showing, if the breasts are unrealistically enormous and/or gravity defying, or the way the box is designed draws your eyes towards sexual body parts(I work in photography, so I have some rudimentary knowledge on this). A bit of cleavage wasn’t enough to meet my standards(unless in combination with another of the above mentioned requirements).
It was interesting to take a look at the numbers, and they turned out to be around what I expected. I’m pretty disinclined to buy a game that uses female sexualization as it’s primary selling point. Unfortunately though, even good games do it at times.
I decided to do the same thing, but with PS3 games.
Both Genders: 3
Unknown Gender: 1 (Fallout 3, the person is wearing full power armor)
Male Only: 8
Female Only: 1
No People: 3 (Oblivion, it has the Daedric “O” (the Oht) on the cover, Dead Space – a severed arm doesn’t count as a person! -, and Jericho (neither does an evil, twisted, undead, goblin looking thing!))
—-
Females Objectified: 4 (Maybe 3. It’s fairly subjective.)
Males Objectified: 0
Please note that this is based only on the front cover, and all numbers refer to the number of games.
Just… wut?! @ that Caesary ad. It’s so ridiculous to try and use sex to sell a top down strategy game, it just doesn’t even make sense. I’m so confused o.0
Anyway I mainly wanted to post to say I’ve been enjoying your blog for a while now, it’s always nice to see a different perspective and in some cases open my mind to things I have internalized as “normal” or things I simply wouldn’t have thought to notice. I also wanted to say I REALLY enjoyed your Mass Effect posts!! I’m a huge Mass Effect fan and while I don’t necessarily agree 100% with everything you say (like I’ve played through with manShep a few times and the voice acting for him is still quite excellent) I’ve found your posts very interesting. Currently I’m playing through ME1 with femShep and it is indeed awesome!
the only thing more ridiculous than than ad is the concept that someone would have a year of Dragon magazines. I assume you torrented? If not, why? did you run out of useful or interesting things to spend it on? Are you a secret, easily bored trillionare? I have never seen a human being read one of those in my entire life.
But back to that ad. Pretty much everything from Evony mockery applies here and need not be said twice, but an extra side of “What in the fuck?” has been provided. Let’s see: “Only reason to live besides women?” So, if I’m reading this right, these people assume that 1. gay, bi, and pan men don’t exist (or do exist, and killed themselves. Not sure which is less flattering) and 2. Women were the center of the male universe before this game existed. Nothing else. Art, family, work.. drugs even? Nope. All about the ladies. Fuck, even the strippers here give men more credit than that. Can you just imagine how vile a picture these marketers have of the human race for this to be considered a decent investment?
It’s amazing how easily misogyny turns into general misanthropy.
Actually I got a friend to send me their copies. Believe it or not, there are people who are sufficiently into D&D that they find the gameplay-related content valuable. Or at least interesting. (Admittedly, I’m not one of them.)
Yeah, why would anyone pay someone money for supporting the hobby they enjoy, dumb asses.
I wish I had a webcam Sam just so you could see me and gaze at my trillionares mansion….I mean thats how I afford the £5 a month for D&D insider 😉
Um how do you stack up digital magazines( Mazed wrote the magazines you talk about here are digital)? Ok so the magazines do not offer the same level of active poses to women they offer to men, that is bad, but the whole only 28 % of women shown were fighters is not something I get fully. Here is why: you did not write how much of the fully clothed women were not fighters( both values are 28 %), also I have had the experience of growing up with some quite eager to fight girls in my elementary school and almost all off them lost the mentality needed for a fighter in high school, and no the eagerness to fight does not a fighter mentality make in my book, but almost all of them lost any interest in melee combat, one of them is a sports type likes PE and brawls and she was the only one who showed any interest in fighting, we used to spar between classes, the rest of them did the usual girl and/or nerd stuff. So my question is this: what % of combat ready women in your opinion should be fighters? Also I agree with you thieves are not mainly women and should not be depicted as such, it is a profession for both genders and thus the percentages for both male and female character should be the same. Now about the suggestive appearances of women compared to men, as I’ve mentioned before I do not mind suggestive appearances that are realistic and within character for both men and women, but I agree with you women are put into suggestive appearances way more often than it is real and needed to be seen for any reason.
It was a stupid pun. I happen to enjoy incredibly stupid puns. I was just hanging a lampshade on it to acknowledge that it was stupid. Of course you can’t stack PDFs. That’s part of what makes it so stupid.
And honestly, the reason it bugs me that women aren’t depicted as fighters as much is because it plays into the whole stereotype of women not being involved in real combat. I get so. fucking. tired. of the woman = mage = sex object cliche in fantasy/D&D art. And in a universe where – according to all of the setting material – a woman isn’t considered a social deviant for picking up a sword and going adventuring, there should be more women hitting things with swords. Sure they might be less suited to wearing a million tons of plate mail, but not all fighters subscribe to the full-plate bash-you-in-the-face of fighting anyway.
And what? I can not ask an equally stupid question to continue the bad pun? (The above asked question was asked wide-eyed and with a deranged look with just a hint of crazy and fear :))
Ok I get what you are getting at with the fact that it bugs you that women are depicted as only the mind/boobs combo heavy on the boobs, but what i was talking about when I mentioned my elementary school experience was that most females lost interest in physical altercations in elementary school while most men lose that in high school, so again my question to you is: Of those women who are ready and/or eager to go to a battlefield how much percentage of the armies fighters should be women? And about the sex object thing, if you have the time then please compare this: http://eu.leagueoflegends.com/board/attachment.php?attachmentid=24107&d=1302650464 to this: http://leagueoflegends.wikia.com/wiki/File:Kayle_OriginalSkin.jpg . Why? Because the first is the newest skin( an optional premium currency purchase that changes the in-game model of the character) for Kayle and the second is the original one you get for free when you buy the character(with premium or with in-game free currency).
I’m not wundergeek but my percentage of melee/fighter types would be.. just as many or close to just as many as the men. Equal representations of women as brutal warriors and mages, pretty much, instead of the imbalance right now. Also, the whole women not wanting to get involved in physical altercations thing more has to do with social conditioning than anything. Women are discouraged (through narratives, strict gender role reinforcement, stereotyping, pigeonholing, shaming, pressure) from being aggressive/physical quite a bit.
Oh, trust me: The only thing girls are actually discouraged from is blatant violence. High school girls are very good at jumping people, and they fight dirty. Like, and speaking form firsthand experience here, waiting for the target to get drunk at prom and then going at her in a group of ten with hockey sticks and fucking with the car’s engine so she can’t get away. (This might seem like an exaggeration, but it’s actually just Vancouver)
You don’t really see that sort of stuff in games. Most of it is just going into a room and fighting things inside it. Fighting “like a girl,” as it were, isn’t really facilitated in most ‘big’ games. The closest you get to the sort of cunning women exhibit is the rogue class, which is why so many rogues are female.
There are exceptions though: Vampire hit it big primarily through hooking women. Amber is still beloved after 20 years without a new edition. Nobilis, if rpg.net counts as “big,” is unmistakably targeted at women and is the sort of game where “kill the immortal giant” is a random encounter. Houses of the Blooded, agian if “bestseller for ‘Indie Prss Revolution’ counts, was designed consciously as a combination of Tragic Opera, Victorian romances, Elric novels, and Dune; you’d best believe that the majority of it’s players are women who do terrible things to each other’s characters.
Those are raw sample numbers, so you can’t really draw any conclusions from them (you haven’t accounted for random distribution). So, for curiosity’s sake, I did some tests you might find interesting. All of these are against a 95% critical level, so Z tests outside of the -1.96 to 1.96 range indicate that the difference you see is NOT random but due to intentional choice or bias.
The hypothesis that population proportion of male/female ratio = 50% nets a Z test result of 3.7, meaning that underrepresentation of females in the art is not due to random chance. However, I suspect from your next article where you point out you counted monsters like goblins that you may find that if you counted male and female “PCs” and male vs female “foes” you may find that male vs female PCs are much closer to even and that almost all “foes” are male, as it’s never cool, no matter who’s standards you’re using to show adventurers beating the piss out of women.
The hypothesis that 50% of males should be in each active and neutral poses, nets a Z test result of .26, meaning that there’s no evidence to reject the assumption that the difference is random.
Whereas, the same hypothesis for females nets a Z test result of 2.08, which means there is evidence to reject the hypothesis: ie the abundance of females in neutral poses is not accidental.
Regarding fully clothed and suggestive, I wasn’t sure what to test. I suspect that your ideal is that 100% of characters should be fully clothed and 0% are suggestive, but that that proportion isn’t supported seems to be a forgone conclusion. Instead I tested vs 50%. ie: are half the males fully dressed, are half the females suggestive, etc…
Unsurprisingly, only fully vs half dressed males returned a non significant test score (-.51) indicating that if males were portrayed fully dressed or not was random, whereas half vs fully dressed females returned a Z score of -5 indicating a clear non random factor for uncovered females. Both males and females had significant Z scores for suggestiveness, (-4.62 and 5.28 respectively) indicating that the greater than 50% proportion of suggestive females and less than 50% proportion of suggestive males were not random.
For the class mix you note, I ran tests vs an ideal percentage of 33% as a third of both males and females should be warriors, rogues or mages. Ironically, the number of female fighters and male rogues ARE statistically random (scores of -1.14 and 1.32 respectively) which is one thing you specifically cited as being biased. However, males were significantly overrepresented as fighters and underrepresented as rogues (scores of 3.58 and -8.25 respectively) and females were underrepresented as rogues and over represented as mages (-2.71 and 3.47 respectively). The underrepresentation of rogues however, may be due to fewer rogue type classes as opposed to fighter or mage classes in the game. It might be interesting follow up to test to see if the distribution of genders matches the distribution of classes.
Oops, in that last paragraph, I meant to say “Ironically, the number of female fighters and male MAGES(not rogues) ARE statistically random (scores of -1.14 and 1.32 respectively)”. Typo. Sorry.
Anyway, the above stats should add some extra punch to your evidence, and shows a few interesting follow up avenues. Plus it was some good stats practice for me.
Um. Thanks for that? I only followed about half of that, but the half I did follow was interesting.
And just to clarify, I’m not asking for 0% suggestive figures. What I want is equal distribution of suggestive figures across genders.
You’re welcome. I mainly did it for me because I was curious, but once I did it, there’s no reason not to share, and I thought it might be useful/interesting to you. Let me see if I can make it a bit more “englishy”. One of the basic precepts of stat is to make it understandable to the layman, so I’ve obviously failed at that to some degree if you only caught about half of it.
A Z score is a measure of how unlikely it is to observe a certain result purely randomly. The further the Z score from 0, the more unlikely a result is, so if we assume in the population of Dragon Magazine art, 50% of males are fully dressed (and the other 50% not) and we do our math and get a Z score of -.51 (which we did) that indicates that if the population is actually 50%, we’d have a very good chance of observing he proportions we did from a sample of it’s art. On the other hand, if we assume that 50% of females are fully dressed (and the other 50% are not) and we do our math and get a Z score of -5 (which we did) this indicates that if the population proportion is actually 50%, we have almost no chance of observing the numbers we did by random chance. Since we’re assuming you counted and evaluated the numbers correctly without bias, we then reject the assumption that the population proportion is 50-50 fully dressed to not.
The arbitrary breakpoint I chose for Z score was -1.96 and 1.96. At those two scores, we have a 5% chance of randomly observing what we did given the assumption we test. A score closer to zero means there was a better chance to randomly observe that proportion, a score further from zero indicates less of a chance. 95% is an arbitrary value (we could have just as easily used any number) but it’s a fairly standard one, and it gives us only a small chance of rejecting an assumption when it’s actually true.
So, to sum up, the data you collected REJECTS the following hypotheses (because the Z scores are too far from zero):
There is an even representation of male and female figures.
The proportion of active and neutral female poses are 50-50.
Females have a 50-50 fully clothed ratio.
Females have a 50-50 suggestively posed ratio.
Males have a 50-50 suggestively posed ratio
A third of the males portrayed are fighters.
A third of the males portrayed are rogues.
A third of the females portrayed are rogues.
A third of the females portrayed are mages.
However, the data does NOT reject the following hypotheses (because the Z scores are close enough to 0):
Males have a 50-50 active to neutral pose ratio.
Males have a 50-50 fully clothed ratio.
A third of females portrayed are fighters.
A third of males portrayed are mages.
Since you mentioned the same proportion of suggestive males and females, there is a different test that compares two proportions and finds the probability that the population proportion for the two is the same and the observed difference is due simply to random chance. It also produces a Z score, and the criteria for rejection/non-rejection is the same. The further the Z score is from 0, the less likely we are to randomly observe the proportions we did if the population proportions are actually equal.
Performing this test on your male/female data for each category, we are able to REJECT the following hypotheses:
The population proportions of fully clothed males and females is the same. (score of 2.74)
The population proportions of suggestive males and females is the same (score of -5.08)
The population proportions of male and female fighters is the same. (score of 2.62)
The population proportion of male and female rogues is the same. (score of 2.06)
However we CANNOT reject the following hypotheses:
The population proportion of females and males in active poses is the same. (score 1.42)
The population proportion of females and males in neutral poses is the same. (score -1.35)
The population proportion of female and male mages are the same. (score -1.6)
Hope that is further useful to you (or if you’d like me to drop it cause it’s giving you a headache and you don’t really care, just say so. I won’t cry… much. :p )
[…] with the content. Anna has supported her assertions that the gaming culture is sexist with analysis and often hilarious comparisons of how male and female characters are dressed and portrayed in […]
The racier they are the more they sell.