Two recent studies co-authored by Dr David Stillwell, Deputy Director of the Psychometrics Centre at Cambridge Judge Business School, have generated global attention for their use of “Big Data” to unearth telling detail about people’s personalities and everyday lives.
Dr David Stillwell
In one study, a questionnaire filled out by 86,220 volunteers found that computers’ judgments of people’s personalities based on their digital footprints such as Facebook “likes” are more accurate than judgments made by close friends and family. The other study, based on nearly 76,863 actual banking transactions, found that money really can buy happiness when people spend on things that fit their personalities. Both studies were based on techniques and expertise at the Psychometrics Centre, which this year became part of the Executive Education division of Cambridge Judge, underlining how Big Data is of increasing importance to business.
Dr Stillwell, University Lecturer in Big Data Analytics & Quantitative Social Science at Cambridge Judge, talked about how the work of the Psychometrics Centre is used by business to better understand their employees, customers and the world around them.
Your Facebook study showed the accuracy of computers to assess someone’s personality based on their everyday digital activity. How did the work of the Psychometrics Centre help make those findings?
Those findings resulted from a mix of expertise including psychometrics, computer science and people who really know how to use Big Data, especially what’s known as “messy data” – things like text messages that don’t fit neatly on Excel spreadsheets but tell us a lot about someone’s personality.
The money and happiness story relies on spending transactions as recorded by a major bank. Was that unusual?
I think the actual transaction data is what made the study special, and we were really pleased to get that data. I think that companies are keen to help business schools in these types of studies, because they know that we want to make a difference in understanding how business works.
What’s the advantage for the test-taker in doing modern psychometric tests compared to the previous model?
The key advantage is that a test using Big Data means that the person doesn’t actually have to take the test, because the results are based on things that have already occurred are are occurring. It means that people don’t have to spend a couple of hours doing the test.
What about for the company or organisation giving the test?
By reflecting things that have actually taken place, psychometric testing is a lot more accurate than relying on what people remember about what they’ve done over a long period of time. Also, because we are developing automatically generated test items, people are taking new and unique tests – and that gets around the problem of a test being leaked and compromised.
In traditional testing, are there certain sorts of questions that tend to elicit inaccurate replies from test-takers?
Definitely. If people are asked “do you follow a schedule” or “do you plan for the future” or “are you open-minded,” everyone will reply yes – so this is basically only an intelligence test. But with Big Data you can assess the kind of words a person uses to determine whether they are in fact someone who maps their future – words like “schedule” or “plan” rather than more whimsical words.
Are there right and wrong ways for companies to use data?
The research we’ve done at the Psychometrics Centre shows that people are more keen for their data to be used by companies in a partnership with their customers, rather than as a gatekeeping mechanism to say no to someone. So if an insurance company can use data to show that someone is a lower risk and therefore lower that person’s insurance premium, that’s seen by people as using data in a constructive collaboration.
How can companies best use your study’s findings on spending and happiness?
We think there’s a big application for online commerce, because Internet merchants can build deeper relationships with customers by offering them products or services that make them happier. There is of course a risk that companies can see that someone is impulsive and just try to sell them all sorts of things, but that’s a very short-term way of thinking – and something that customers will see through. It’s easier to work with existing customers than find new ones, so we think that companies will take the longer view and try to improve their customers’ lives through a better understanding of spending and happiness.
How does the Psychometrics Centre and Big Data fit into Executive Education?
We think that psychometrics should be used not only for recruitment by companies and other organisations, but also as part of a broad process to better understand who their employees are, what makes a good team, and how best to bring out a person’s abilities. Psychometrics is bringing about a big change in how businesses are thinking about their employees, and that ties right into Executive Education.