Using several criteria, we show that computers’ judgments of people’s personalities based on their digital footprints are more accurate...
This is right in my wheelhouse in terms of what I teach and study. The obvious problem is how are they measuring people's "true" personalities. In most cases, researchers measure a person's personality by having him take a personality test. This does not show us the person's true personality, only the personality he perceives himself to be. Most people are not well enough in touch with themselves to know who they really are. Just as an example, do you suppose if we measured OPapaDoc's personality based on a test he took himself, it would reveal his narcissism and deep-seated anger? No way.
Therefore, it is not surprising that a personality profile based on an individual's Facebook page, which he created himself, is a better predictor of the results of a test he took himself, as opposed to what people who know him say. Unfortunately, this is what passes as science today. As a social scientist myself, however, I love it because the endless and pointless navel contemplation results in full and lasting employment for me.