In a recent study published in the well respected journal Proceedings of the National Academy of Sciences of the United States of America (PNAS), Kosinksi et al. show that “easily accessible digital records of behavior, Facebook Likes, can be used to automatically and accurately predict a range of highly sensitive personal attributes including: sexual orientation, ethnicity, religious and political views, personality traits, intelligence, happiness, use of addictive substances, parental separation, age, and gender.”
The One Click Personality Test
To prove their point, two of the authors have set up the website YouAreWhatYouLike.com where anyone can test their personality, simply by signing in with Facebook. Results of The One Click Personality Test are solely based on your Facebook Likes. Below is my result, which I personally think is pretty accurate, but definitely not 100%.
The predictions rely on data from over 58,000 volunteers, who shared their Facebook Likes, provided information for demographic profiles, and completed personality tests as part of the study. These data allowed the researchers to extract a list of most predictive Facebook pages. An individual’s Facebook Likes are compared against this list to draw conclusions on the subject’s personality. Most indicative Likes of my profile for example were for HostelBookers.com and Facebook Site Governance.
The researchers categorized predictive Likes for 18 different traits. For each extreme of the trait (e.g. high vs. low IQ or male vs. female gender), they use 10 indicative Facebook pages. For example if you Like Biology and Jennifer Lopez, you most likely to have many friends; but if you like The Dark Knight and Minecraft, you most likely to have few friends.
A full list of most predictive Likes are provided in a PDF document via PNAS.
Since this test is not based on actual answers to a standard personality test, but merely on random Facebook Likes, it is not nearly as accurate as a real test, which themselves are not 100% accurate. However, it does come close. Interestingly, the prediction for the personality trait Openness is the most accurate and almost as good as that of a standard personality test.
Reflection
To be honest, in part my results reveal more how I would like to be perceived, rather than how I really am. And this raises some interesting questions.
Would you use the Like feature differently if you were aware that your Likes are visible to others? Of course! At least if you’re smart. In fact, I’m very careful with what I Like. Moreover, I have Liked very few pages because I don’t want my News Feed to be cluttered up and I don’t want to give Facebook more data than I absolutely have to. I think that my Like profile doesn’t reflect the real me.
On the other hand, the authors do caution that it doesn’t necessarily take no-brainer Likes to predict a character trait like your sexuality. In other words, it isn’t easy to influence the test and hide a specific trait. Either way, the data aren’t very reliable. However, they don’t come with such a disclaimer and a wrong result could actually hurt you more than a right one.
Finally, if you only Like what you think would be good to Like, do you gradually become more like the person you desire to be? Assuming you are what you eat, will you become more healthy if you only eat what is supposed to make you healthy? The answer to the latter is a definite Yes, but I’m not sure about the former. The question is, can you change your personality by pretending to be different, like you can lift your mood by faking a smile?
Conclusion
The study is a reminder that habitual data collected online, including Facebook Likes, browsing histories, search queries, or online purchases, can reveal a lot about us. The researchers draw a positive conclusion and say that these data can be used to automatically customize and thus improve services, marketing, and product recommendations. However, they also caution that the data could easily be used without the user’s consent and without them noticing.
The authors write: “One can imagine situations in which such predictions, even if incorrect, could pose a threat to an individual’s well-being, freedom, or even life.” Indeed, it does become difficult to stay on top of your own digital data and a loss of trust in digital service is a possible consequence. Hence the authors conclude: “It is our hope, however, that the trust and goodwill among parties interacting in the digital environment can be maintained by providing users with transparency and control over their information, leading to an individually controlled balance between the promises and perils of the Digital Age.”
The full paper covers only four pages and can be viewed directly on the PNAS website.
If you don’t want people snooping around your Likes, hide your personal information on Facebook, especially in the view of Facebook’s Graph Search. Also have a look at our (Very) Unofficial Facebook Privacy Guide.
What did your personality test reveal about you? Please share your insights in the comments!
Image credit: You Are What You Eat via Shutterstock
The post You Are What You Like on Facebook [Weekly Facebook Tips] appeared first on MakeUseOf.
No comments:
Post a Comment