What Facebook reveals about your sexuality: Gaydar project unveiled
Posted by Sandra Cosser on 21 Sep 2009 | Tagged as: Social Media
Gaydar
With too much time on their hands, Jernigan and Mistree became interested in finding out what people unwittingly reveal about themselves on social networks. So, in 2007, they downloaded openly displayed information from Facebook (supposedly to evade accusations of privacy violations), such as gender, things of interest (which, in the mother of all assumptions, they took to denote sexuality) and friends.
They analysed the friend links of 1,544 straight, 21 openly bisexual and 33 openly gay men and found that gay men had proportionally more gay friends than straight men. They determined that they had successfully trained their programme to infer sexuality based on online friendship and went on to analyse the Facebook accounts of 947 men who kept mum on their sexual orientation.
They admit that they had no way to confirm the results with scientific rigor, but to make up for this they used their knowledge of 10 people (out of 947) to prove the accuracy of their programme. Lo and behold, they found that project Gaydar successfully predicted that all 10 men were gay. Unfortunately, it has less success when it comes to determining bisexuality or lesbianism.
To even a casual observer (myself) it doesn’t appear that the study is the resounding success that they hoped for. And the less than ‘scientific rigor’ of their testing means that they don’t have the backing of the scientific community, for whom scientific rigor is an inviolable rule. But Jernigan and Mistree, along with some professors and assorted researchers, are convinced that the results are significant.
Hal Abelson, the MIT computer science professor who taught Jernigan and Mistree says that it “pulls the rug out from a whole policy and technology perspective that the point is to give you control over your information – because you don’t have control over your information.”
Carolyn Johnson, who is a science reporter for the Boston Globe, says that it “provides a provocative warning note about privacy” because “… people may reveal information about themselves in another way, and without knowing they are making it public”.
Shockingly, we can be defined by our friends. If your friends are all of an age, you must be of a similar age, if your friends are mostly women, you must be a woman and, logically, if your friends are all gay, you must be gay by association. Kevin Bankston, senior staff attorney for the Electronic Frontier Foundation, digital rights NPO based in San Francisco, says that your online friends could even cause people to make incorrect assumptions about you, like say, that you’re gay when in fact you’re not.
Correct me if I’m wrong, but do we not also run this risk every time we step outside? When we go out with our friends who have a propensity for dark gothic clothing, won’t people assume that we too subscribe to a gothic lifestyle? If we hang out with people who have missing teeth and rusted cars resting on bricks in front of their homes, won’t people assume that we also occupy the lower rungs of the social ladder?
Why should we applaud a study that shows the same assumptions can be made online? Isn’t it blindingly obvious? Simon Axten, Facebook spokesman, seems to share my way of thinking (we must be Facebook friends), saying, “In general, it’s not too surprising that someone might make inferences about someone else without knowing that person based on who the person’s friends are. This isn’t specific to Facebook and is entirely possible in the real world as well.”
What is genuinely frightening is that proposed applications for this kind of software include identifying potential terrorists. I reckon it would take a pitifully dumb terrorist to advertise his or her status on Facebook, or to ‘friend’ similarly cognitively-challenged terror mongers.
It could also, apparently, be used to identify fat or happy people – a giant social leap forward I’m sure, especially for those who have difficulty discerning detail in profile pictures. At least advertising companies would know who to target for their ‘lose 10 centimeters in 10 days’ ads.
I get that social networks are still a relatively new phenomenon and that they represent an untapped gold mine for social scientists, psychologists and human behaviourists, but could we please focus studies on matters of real importance, such as why people join Beyonce’s fan page or follow Ashton Kutcher on Twitter?
