What Happy People Buy: Psychology and Social Media Data

Our joint Psychology, Computer Science and Business workshop on ‘Predicting Psychology from Social Media Data’ in the Business insights Lab joined up digital innovation across sectors and departments.

Guest speaker Dr. David Stillwell is a Lecturer in Big Data Analytics and Quantitative Social Science at Judge Business School in the University of Cambridge. He is also Director of the Computational Behavioural Science Unit and the Deputy Director of The Psychometrics Centre. David studies the links between online behaviour and psychology, applying machine learning methods from textual and image analysis to predict psychometric questionnaire results. In 2007 he started the myPersonality Facebook app that enabled 6 million users to take a Big Five personality questionnaire and get feedback on their results.

Turns out, people spend more on products that match their personality type, which leads in turn to a higher sense of satisfaction with their lives – ‘Happy People’. If a business matched their products more precisely with their customer’s ‘type’, it would more directly meet the customer’s psychological needs.

The Big 5 indicators of personality type are:

  • Openness;
  • Conscientiousness;
  • Agreeableness;
  • Extraversion/introversion; and
  • Neuroticism.

These have been shown to reliably predict personality type and behaviour – much more accurately than, for example, factors like geography, age and gender. This data is of use to almost any organisation that needs to assess and predict behaviour and risk – for example, a car insurance company.

But how do we gather and then apply the personality data?

This is where Facebook comes in.

Clearly, an increasing proportion of our lives is mediated through digital devices, and the data records our decisions, preferences and associations, and so is psychologically very meaningful. Using the myPersonality.org database of 6 million users’ responses to a personality questionnaire, we can use machine learning methods to predict intimate traits like personality and IQ from Facebook Likes, status updates, and even profile pictures. In some instances, personality predictions from social media are more valid than self-report — the computer knows you better than you know yourself.

Following David’s talk, we heard from Surrey Business School/CoDE’s Travis Street on ‘Why Is Social Media Actually Influential?’, from Dr Shujun Li of the Computer Science Department/Surrey Centre for Cyber Security on connecting social media to security, and from the Psychology Department’s Dr Patrice Rusconi on trust, credibility, and fake social media accounts.

Social media is clearly a hotbed of research possibilities and practical applications. The workshop was a great example of how at Surrey we are able to connect the dots between disciplines, departments and faculties – with Digital being the great equaliser. If you’d like to attend the next workshop, please get in touch.

Photo of Dr David Stillwell


Share this Post!

About the Author : Kris Henley

Communications and Outreach for Surrey Business School's Centre for the Digital Economy, a newly-founded research centre to explore the implications of the Digital Economy for business, government and society.