Saturday, May 28, 2016

About that Twitter harassment study that claimed that women send 50% of misogynistic tweets...

The Twitter logo.

As many of you know there was a study that was released recently that said that 50% of "misogynistic" tweets are sent by women. It was done by a British think tank called Demos, and the results haven't been fully announced yet. But it did say that misogyny is a huge problem on Twitter. You can find the outline of the study here, but I wasn't able to find the full study even though it was supposed to have been released on May 26th

I've got some real problems with their conclusions though. Before anything else I will be using some offensive language from this point on. I don't know if I need that warning on this subject or not, but no harm in putting it out there. I am going to have to repeat a bunch of bad language, which, for some reason, people find more offensive then all the descriptions of ISIS atrocities  I post on this blog. 

My first instinct is that they are correct, saying "sexist" things against women is an equal opportunity sport. It's not surprising to me at all that women send mean tweets at each other and it should not be surprising if these results are correct. Especially when it comes to the terms "slut" or "whore". Anecdotally, when I have heard those words as a pejorative in my personal life, it has almost always come from a woman describing another woman. The rest of the time it was either used as a synonym for "hot" or "sexy" from a man and only occasionally did I hear a man say it as an insult about a woman. Indeed, I wouldn't be surprised if follow up studies found the balance to be shifted towards women sending these kinds of tweets. 

That being said, I do have to say that I question this study. First of all, I don't have any idea what their methodology was. All they said is that they counted up the number of people that tweeted the words "slut" and "whore" at women. They did say that they filtered out tweets that used the words non-aggressively but I wonder how effective they were at doing so. It also doesn't say how those tweets were chosen other then that they were from the UK, which isn't an accurate sample size for anything outside of that country. 

I think the most obvious thing that is missing is context. Obviously it can be upsetting if someone tweets you something offensive but it really depends on a lot of things. For someone with a small account like me (currently around 70 followers) getting these kinds of tweets would have much more of an impact then someone with 10,000 or even 100,000 followers. I am also guessing that many of these tweets are "drive-by" attacks and not part of a campaign of harassment. Many more are also the results of fights, which happen on twitter all the time. It's not harassment when both sides of an argument are attacking each other.

Twitter also has a major problem with fake users. I wonder how many of the people cited in this study are fake users. People often create bots with female avatars that could gather abuse and it is certainly possible that bots could be used to harass others. It's a huge problem on twitter and could completely skew any data if it wasn't taken into the account. I also wonder how they determined what sex people are. Just because someone has a female name or avatar doesn't mean that they actually are a woman. Plus many people have gender neutral accounts. The number males with anime avatars and names alone could skew this study completely, and I don't know how you would control for it. 

I also think that the Demos organization seems to have an obvious agenda. If their first reaction after posting this study is trying to make it trend with a hashtag then it seems obvious to me that they aren't doing science for science's sake. Supporting the #Reclaimtheinternet hashtag shows me that they are clearly feminists and therefore biased. Admitting that they found that 50% of these tweets come from women helps a bit but it also seems like they are advocating for censorship as well, so I don't know if I can trust their study. They very well could have exaggerated things or otherwise misled us. 

Speaking of bias, it's also clear that they completely left out male users in this study. I doubt many males get called "slut" or "whore", though more then you would think, but I am sure they get called terms that are sexist against men. Getting called "dick" or "cocksucker" or even the ubiquitous "faggot" is just as annoying and I would bet that it happens just as much, if not more, than women getting called anti-women terms.

Of course all of this is irrelevant. I really don't think that this kind of language is all that harmful. Sure, it is annoying as hell to have people send you mean tweets but that is why there is a block function. If the harassment doesn't stop, you can always go private, or even quit twitter for a bit. Twitter is obviously a confrontational medium, if you don't like the idiots, then you don't have to post there. What really is harmful is when people gang up on you, or keep after you long term. Getting called a bad word by random people every once in awhile should be treated as noise, but getting attacked by multiple users or the same person over a long term is where the real problem is. These trolls are common and are a bigger threat then any random person tweeting an occasional insult. 

I also would like to point out that getting trolled on the internet is an universal experience, regardless of gender. I know that I, as a male, have faced the same kind of harassment that women have, and most of the time it's completely random. Trolls live on picking fights with people and they will use any weapon they can to get under your skin. For women, especially feminist women on Twitter, the best way to get under their skin is to say something they will perceive as sexist. But every other group has their own hot button issues so we shouldn't act like women alone face this kind of thing.

So what should be done about Twitter harassment? Not much I'm afraid. I think the worst thing you could do is try to control it from the top down. You can't avoid bias that way and I guarantee if you get a bunch of people trying to control harassment then free speech on the platform will be damaged. Given that twitter is more like a postal service then a company at this point, cracking down on harassment, real or perceived could silence a lot of people that disagree politically. 

What I do think would be helpful is if there was more education about how to deal with trolls. People need to know that they can walk away from the internet if things get too tough. They also need to know that engaging with trolls is always the worst mistake that you can make. Ignoring them makes them go away, feeding them makes them comeback forever. Again, the best thing you can ever do if you get a tweet from a jerk is to totally ignore it. Even blocking them can encourage them, but if you ignore then they can't do much to you. 

And people also need to realize that Twitter is a public platform. If you say something offensive or stupid people are going to notice and probably respond in kind. Remember when I said that the context of the tweets was missing in this study? How much do you want to bet that some of the women that got sent sexist things got them because they posted something like #killallmen or something else equally offensive to others? 

You really do have a right to free speech on Twitter, but that doesn't mean that you can act like a victim when you poke the hornets nest. I don't think that anyone should stop from posting anything just because they know that someone could be offended, and that in no way justifies threats or attacks, but some common sense is needed. If you really feel the need to tweet something like #killallmen, go ahead and do it, but just keep in mind that some battles aren't worth fighting. I'd argue that every battle on Twitter isn't worth fighting. 

Finally, some Twitter harassment is not harassment at all. Though the study focused on bad language, I have seen quite a few people complain when people are honestly disagreeing with others on the platform. If you post something wrong or controversial and another person tries to correct or argue with you, that is not harassment. It's only harassment if they are threatening you, but all too often people use anti-harassment tools to shut down legitimate criticism. 

No comments:

Post a Comment