TR Member Perks!

It is that time again, time for yet another study supposedly proving that video games are causing the downfall of society where the message they want you to take is distinctly different from the actual outcomes they hope you will never read. There are so many studies about video games out now, you’d think another one would just fall on the pile forgotten. Of course, if it seems to be supporting a preconceived conclusion that fits with what people want to hear, then it might get some attention right? Even as we see an increase in honest studies finding that maybe academics have been wrong about video games, and maybe they aren’t all bad, there are still those who will do anything to prove the terrors of video games. Even if it means putting their academic integrity on hold. 

Grand Theft Auto Online Valentine's Day

According to studies, Grand Theft Auto may be the only violent video game that actually exists.

Acting Like a Tough Guy” is the actual name of the latest study (based in Italy) out to prove that violent games are corrupting the youth, and as is the cliche, they’re going after Grand Theft Auto. This study’s unique twist is that they did bother to differentiate between types of violent games, namely GTA and Half-Life. Everything else is a familiar tactic found in most studies of this nature: procedures that are not complete, statistics that measure subjective variables, and the narrative written around the conclusion that the authors wanted rather than the one they actually received, all while blatantly misunderstanding the medium. This one in particular though is gaining some traction, likely because it is fairly well-written and presented. Under the surface things are not so simple.

The conclusion this study makes is that “violent-sexist” video games, such as Grand Theft Auto (because it has strippers and prostitutes in it seems to be the bulk of this argument), lead to an increase in masculine beliefs and a lack of empathy for female victims of physical abuse. How terrible, right? Perhaps, but there is an issue here: what are “masculine beliefs”? Because it is listed next to “lack of empathy for female victims” your mind (unless you’re used to this nonsense by now) will likely assume a negative connotation. This is not on accident, and as we’ll later find out, is pretty much the only way this study works. In fact, “masculine beliefs” are very poorly defined in this study. They don’t completely gloss over this, but rather than a full list of how they measured it (which was based on a Kinsey survey), they give a broad and editorialized definition in the introduction. This requires a very distinct definition though, because what this study demonstrates is not that “violent-sexist” video games are correlated with a lack of empathy. In fact, there was actually no direct correlation found between that at all. What it finds is that “violent-sexist” video games are correlated with “masculine beliefs.” And that “masculine beliefs” were correlated with a lack of empathy for female victims of assault. So, obviously, “violent-sexist” video games are correlated with a lack of empathy. Clearly, of course, what’s the problem? 

The third ANOVA found no significant effect for type of video game played on empathy for female violence victims (p = .31)

Well, since they demonstrated absolutely no direct connection between the games and empathy, it leads to the question: how does this indirect connection work? Even in the comments of this piece, another gentlemen (who is far kinder in his analysis than I) points out that taking a broad generalization from this indirect correlation may not be the right step. Even with direct correlations, researchers are cautioned to take care with the conclusions they make. If there is an indirect correlation though, and the direct correlation is non-existent, this shouldn’t be taken as a sign that the connection exists and the direct correlation should be ignored. It should be taken as a sign that there are other factors at play. 

There is a lot missing from this study that could’ve been easily incorporated. In measuring empathy, why did they not include controls for situations where there is no attacker (the subject was in a faultless accident) and situations for where the oppressor is female? Why did they not measure those “masculine beliefs” and empathy prior to exposing participants to the games, to see if there was an actual correlation and not just a coincidental one? The only important direct correlation found in the entire study were between those “masculine beliefs” and empathy after all—why is there no consideration for the other conclusions that could be brought in? Why did they not use games with female or player-created protagonists—they could have accomplished this by using Saints Row III as their “violent-sexist” game, as it has similar themes and settings, but you can play as male or female because of the character creation. 

video games and empathy table

Table of the correlations calculated in the study. Asterisks denote significant relationships.

Even the section of Grand Theft Auto they use is questionable. They state that players were directed specifically to complete a mission against a rival gang, which happens to take place around a club. However, especially because you are involved in a mission, you’re likely not interacting with the ladies in the club. You have other priorities. They’re not your targets or enemies—in the midst of a mission, pretty much all bystanders become background noise (as most people who play through GTA would tell you). So how would this even impact empathy in regards to female victims? Would the impact on empathy in regards to male victims be greater? We don’t know—the researchers didn’t think about this. 

Perhaps consider this as a possible alternative explanation. Even in players who have masculine beliefs, video games actually make them more empathetic. Meaning those “masculine beliefs” you claim are toxic, even though players seem to keep them through gameplay, are altered by the gameplay, making those players more empathetic. This conclusion makes just as much sense, in fact moreso possibly. Since there is no negative correlation between empathy and the game, even though there is a small correlation between “masculine beliefs” and the game, that means that the game may very well be impacting that, but in a positive way. Otherwise, why wouldn’t those “masculine beliefs” carry over to a direct correlation? 

Or alternatively these factors are all random and have no true connection at all, because people play games to have fun, especially ones as bombastically insane as Grand Theft Auto. This would make more sense as well, since those “masculine beliefs” connected to the games were only found between the “violent-sexist” type and the “neutral” type (the neutral type being Q.U.B.E). But there was no difference in those beliefs between the “violent-sexist” and the regular “violent.” How can we be certain those beliefs are just a result of the participants preconceived ideas and upbringing, and that the games, in fact, made those participants more empathetic, at least towards female victims? 

Had the researchers considered this angle, we may know. However since the study was tailored towards getting a specific conclusion, rather than asking a question, there isn’t enough data here to even find out for ourselves. As always, the study would have to be repeated. Unfortunately, it likely won’t be, as the researchers seem to pat themselves on the back and say “See we found a significant correlation! If you played the games more, you’d be even more sexist!” This is the biggest hint that the authors went into this already assuming they were correct—there is not an ounce of humility or admission of fault or wrongness, even though there was no direct correlation found at all between the games and empathy. 

Although it is impressive that we were able to obtain significant effects after such a brief exposure, we do not know what the consequences would be for longer exposures. If the effects occur after only 25 minutes of play in a laboratory experiment, they are probably magnified after longer periods of play outside the lab.

It is so tiresome to watch academia continue this crusade against video games, even while never taking the time to actually understand the medium. They are so caught up in this decades old idea of video games causing violence that it seems like they can’t possibly view it from any other lens. It is my hope that one day researchers can band together to do a study on video games for what they truly are—one that is nuanced and interesting and seeks only to ask questions, not solely to find the answer you want to support your views. 


Kindra Pring

Staff Writer

Teacher's aid by day. Gamer by night. And by day, because I play my DS on my lunch break. Ask me about how bad my aim is.



  • webkilla

    Oh it gets better

    Read the details on their questionnaires:

    They put statements like “men should be strong” and “it’s ok for men to rape women” into one category to measure. So if you gave 7 to “men should be strong” and 1 to “boys can rape girls”, you score 4 in a set of believes that includes “boys can rape girls”.

    can you say bias? Can you say incredibly poorly done science?

    Holy hell – this is the most leading and biased questions ever. Full of bullshit

  • Albedo The Chemist

    Not for the lack of trying eh?
    I’m astounded theyd think “sexist” games would have a real-life effect on sexism when it clearly doesn’t on violent behavior.

  • Miarla

    See I would trip up there, because I would ‘strongly agree’ or whatever metric they are using with the statement ‘Boys *can* rape girls’. I would however strongly disagree with the statement ‘Boys *should* rape girls’. I definately would be interested in seeing those questions analyzed more closely.

  • webkilla

    and that’s how they twist their numbers

    hell, depending on how you want to misinterpret that – then you could argue, that anyone who doesn’t give “boys can rape girls” a high score are under the delusion that rape just doesn’t exist

    …or they could say that it confirms rape culture, since clearly it reveals that people do not acknowledge that rape is taking place anymore, confirming that it has become so prevelant and common that it is simply no recognized anymore

    I know that both of these interpretations are far out – but considering the BS that this study is spewing, and how vague and generic some of these questions, then it really wouldn’t be impossible

  • Miarla

    Take the least justifiable, most outrageous interpretation of the data that you can, and then just add water for your brand new SJW ‘study’.

  • webkilla

    Its not even that

    Look, I have a masters degree in UX – part of that includes extensive studies in how to craft and manage user testing. A huge part of that is things like questionaries – and how you have to be really freaking careful not to introduce bias in your questions and the way you want people to answer them.

    Its the difference between asking:

    “Rape is bad – yes or no”

    or

    “boys can rape girls – rate from 1 to 7, with 7 with being the most positive response to this question”

    You’ll notice that the second one, the one used in this here study, is really fucking vague. Like, what exactly are you answering? Seriously, think about it?

    Boys can rape girls. Yes, that is a factual statement. If you toss a stone, it will eventually fall down, unless you’re in space. That is also a factual statement.

    How can you rate a factual statement from 1 to 7?

    Or was that not what the researchers wanted to ask? Did they mean “Boys should rape girls – rate from 1 to, with7 with being the most positive, how much you agree with this statement”

    See, now you’re asking for an opinion – and thus getting an entirely different set of data out of the question.

    Its the same way that you can pass out a questionaire going “is it a good thing if your government fixes up the infrastructure and starts a lot of good projects to create work for the unemployed” – and when people say yes, because why not, then you conclude that people support the policies of the nazi government, because that’s what they did.

    If you don’t tell people exactly what you’re asking – and if you don’t tell people the context in which their answer will be interpreted – then… fun times, of the fuzzy logic sort.

    Sometimes deception in studies can be useful – like the placebo tests in medical trials, or when pretotyping, or when looking for spontaneous reactions instead of stuff that people are prepared for – but the kind of study in this article? Being deceptive, and using vague questions that are quite frankly very much open to interpretation…

    It means that the researcher can go “ok, so everyone scores that ‘boys can rape girls’ one low, so if we interpret that in this here way, we can still make them look sexist as fuck”

  • Smoky_the_Bear

    Sadly I don’t think things will change anytime soon. The fusty old Jack Thompson types that were crusading against video games because they didn’t understand them have now been replaced by the SJW types who DO understand them but just don’t like them for whatever reason. These types have infested academia at this point and will use any means necessary to prove that video games in it’s current form is somehow “harmful”, because it’s not the magical “safe space” that they want. They will go to any means, including these academic studies designed to fabricate the truth to fit the results they want or even outright lie about it, in order to “bad press” companies in to not making these games that they don’t like.

  • Smoky_the_Bear

    They don’t think it does, they don’t care, they are on a crusade to prove that it does so these sorts of games that they don’t like, stop getting made.
    They are smug self-superior idiots that think their way of thinking is inherently right and anything they don’t like must somehow be bad for society so they will produce nonsense like this to try and prove their point through biased weighted surveys and whatnot.

  • Gamer2002
  • Gamer2002

    The exact reference quote is “It is OK for a guy to use any and all means to
    ‘convince’ a girl to have sex”. I simplified it, though I’m not native english speaker, so I didn’t think that “boys can rape girls” is more about ability than allowance.

    Those statements were “strength is good” and “rape is good”, pretty much.

  • webkilla

    yup :p

  • webkilla

    oh, ok – I thought it was direct quotes – but even then, the wording of that is shady as fuck and be interpreted to mean tons of different things

  • Seansong1

    I can hear Anita screeching in the distance ..

  • utera

    Social studies which claim to measure “empathy” tend to be problematic by default, the metrics tend to be just made up.

  • Miarla

    My boyfriend convinced me to have sex by
    a) nuzzling my neck while I was busy watching a film
    b) talking to me in a suggestive manner
    c) rubbing my feet and then ‘taking advantage’ by moving his hands further up

    etc etc etc

    So many, many different acts that I would say are not only perfectly allowable, but damned sexy. Most of which would fall afoul of the definitions and descriptions used to define ‘any and all means to ‘convince’ a girl to have sex’.

    That’s before we get into the number of times I have jumped his bones after a drink or two. Did I have sex while I was intoxicated? Yes. Was it consensual? Erm… I initiated it so if you answer anything other than yes… screw you.

    As @webkilla and I discussed above. It is necessary to be exactingly careful when writing the questions for surveys. And to make sure that in later collating the data you don’t assume meaning into other people’s answers.

    The question you listed above uses ‘any and all’ I couldn’t answer that without there being room to misunderstand my answer.

    Can a person use persuasion and enticement to convince someone to have sex with them?

    Can a person use force or threat of force… etc?

    Can a person use social or other (non physical) threats… etc?

    Thus not lumping permissable and non permissable acts in together.

    MUCH clearer, much less likely to result in a false positive or negative. Of course that’s off the top of my head, and I am sure other people could make much clearer questions than I.