Dr Simon Caton, programme director for NCI’s MSc in Fintech, one of the authors of the paper published in PLOSONE.

We’ve all long-suspected that our friends don’t tell the whole truth on Facebook – never a dull moment, literally!

Subconsciously, all of us respond to the audience we’re sharing with and represent ourselves as we would prefer them to perceive us rather than we actually are. In academic circles, this tendency to put our best face forward has cast doubt on the reliability of research carried out using social media as an information source – can the results of a study conducted using Facebook posts be relied upon?

Dr Margeret Hall, of University of Nebraska at Omaha, and Dr Simon Caton, of National College of Ireland (NCI), decided it was important to find out.

Drawing on a pool of over 500 people who first completed psychometric testing and then gave permission for their Facebook feeds to be analysed, Dr Caton and Dr Hall were able to compare personality type as defined by testing against personality traits revealed in online posts. This allowed them to examine the validity of current social media personality studies and propose a method to mitigate for self-representation.

 “We were looking for the ‘Big 5’: openness to experience, conscientiousness, extraversion, agreeableness and neuroticism. We found that we could predict where a person fell on the scale for each trait with a high level of accuracy.” Dr. Hall continued, “posts about TV programmes were particularly revealing.”

What does this mean for Facebook users?

“What it means in the first place is that if an academic study, using our proposed research methods, observes a societal trend or political movement on social media, the conclusions it draws are largely dependable, even given the fact that people are careful how they represent themselves on Facebook. From an individual user’s point of view? If someone can access your Facebook page, perhaps a potential employer, they can possibly observe your personality. Without you knowingly sitting a psychometric test, they can discern whether or not you fit the role they’re recruiting for,” said Dr. Caton, who is the programme director for NCI’s MSc in Fintech.

You can read the full paper, “Am I Who I Say I am? Unobtrusive Self-Representation and Personality Recognition on Facebook”, in The Public Library of Science ONE (PLOSONE)

In the meantime, maybe set your Facebook timeline to private… your posts about Game of Thrones could reveal a lot more than just spoilers for those who are lagging behind!