The full explanation is below, but we value your time so if you already know the context and background, all you need is the magic sauce.
There is so much hullabaloo right now about the abuse of Facebook user data for delivering targeted ads and engineering elections. The mess has spawned a #deletefacebook movement. The problem is not really Facebook here. The activity of users on all modern social media platforms can be potentially analysed and converted into psychographic profiles. Now psychographic profiling in its current form has been around since the 1980s, but it was social media that allowed researchers and organisations to create a large number of psychographic profiles at scale. The basis of the current use of Facebook for psychographic profiling goes back to 2012.
Traditionally, advertisers use another kind of profiling for creating advertisements. This is called demographic profiling, and most of the advertisements we see in traditional media channels such as television or newspapers, use this approach. The approach has been carried over to online campaigns as well. The underlying assumption is that all middle aged men, young adults, pregnant women, or children will react to certain messages in certain ways. The advertisements therefore try to speak to the people most likely to buy the product.
Psychographic profiling goes a step further, and allows advertisers to fine tune their messages according to how an individual would react. Not all people in a target demographic profile would react in a particular way. If you can know the fears and motivations of a person down to the individual level, you can create messages that are more likely to get the reaction that you want. The idea is to use the “likes” that you freely dole out on Facebook, to get an understanding of your personality. Researchers from Cambridge came up with an algorithm in 2012 that was pretty accurate at predicting your race, your gender, your age, and personality traits according to the OCEAN model.
The model gives you a score on what is known as the “big five” personality traits which are openness, conscientiousness, extraversion, agreeableness and neuroticism. It can also be used to predict your Jungian personality type. An article posted on Motherboard in January 2017, soon after the 2016 US Presidential election, provides an explanation of exactly how good this profiling is. At 70 likes the profile knows more about you than your friends, at 150, it knows more about you than your parents, at 300 likes, more than your partners. Beyond that, the profiles can know more about you than yourself. Although students at Cambridge University came up with the algorithm, it’s use for providing laser focused marketing campaigns is unrelated to the university.
Apart from Facebook likes, the profiles can be generated through Facebook and Twitter posts as well. The accuracy varies from person to person. This is the psychographic profile of Digit Geek. Our online behaviour is like that of someone between the age of 25 to 29. The social media use indicates that Digit Geek is the “epitome of masculinity”. The Jungian personality type of Digit Geek is introverted, intuitive, thinking and perceiving, or (INTP). This is the OCEAN model for Digit Geek.
To find out your own psychographic profile, all you need to do is check out the demo by a team of researchers at Cambridge. This was the research group that originally came up with the method to generate the psychographic profiles from social media use. Researchers and developers can apply for access to the API.
The profiles may not be very accurate for Indian audiences. These profiles are not inherently a nefarious thing. It is trivially easy to identify advertisements on Facebook, as the platform identifies them as such. If you care to shatter the echo chambers around you, go ahead and like a bunch of pages that you would not otherwise like. This might mess up your timeline, but it is also likely to confuse any profiles. In any case, it is good idea to be sceptical when acting on calls to action in social media. Learning about cognitive biases is a good place to start.
The main takeaway here is not the need for data protection laws, or the endemic problems in social media platforms. It is about how to protect oneself from the abuse of digital footprints that we leave around everywhere on the internet.