For most people, April 26, 1949, was a wholly unremarkable day in rural Kansas. For social scientists Roger Barker and Herbert Wright, it was one of the most important days of their career. On that spring day, Barker, Wright and a team of eight observers meticulously documented every single twist and turn—every decision made and behavior displayed—by a seven-year old boy named Raymond Birch over the course of the 14 hours he was awake.
From today's perspective this exercise might seem ludicrous—not to mention absurdly cumbersome—but Barker and Wright turned their observations into a 435-page report that spelled out, to the best of their interpretation, exactly how real-life children behave. It was a seminal undertaking in that it demonstrated the limits of simple survey- or questionnaire-based social research: in other words, it established that there's a difference between what we say we like, do, and think, and what we actually like, do, and think.
The report became a book, entitled "One Boy's Day", and Sandra Matz, Associate Professor of Business at Columbia Business School, admits that it's one of her favorites. In part, she's fascinated by it because it established that integral principle: that our actions speak louder than our words and that our behaviors can't always be captured by process of question and answer. But she also appreciates it for underscoring just how far we've come in the field of social psychology.
“Today, we don't need eight research assistants any more to determine who we are, what we like and what we do,” she says, speaking at an event at CBS—hosted at the Dear Mama cafe. The event forms part of a series on Business, AI and Democracy (BAID) under the umbrella of CBS's The Hub think tank that explores some of the most pressing issues shaping the world today, and how scholars, practitioners, policymakers, and the broader public can collaborate to solve these daunting challenges. “In the morning, we pick up our phone or pay for our coffee using our credit card and the information available is already all there,” says Matz.
Matz, who hails from Germany, uses big data to study human behavior across different domains of business and life. Drawing on the academic fields of psychology and computer science, she's spent the last few years exploring the connection between people's psychological characteristics—whether they're introverts or extroverts, for example—and the digital footprints they leave.
Specifically, she says, her research has focused on three core questions: What can people's digital footprints tell us about their socio-psychological characteristics? What can people's digital footprints tell us about the real-life consequences of their unique psychological characteristics? And how can insights extracted from people's digital footprints help individuals and businesses make better decisions?
At the event in early May—facilitated by Professor Gita Johar—Matz demonstrates that she has answers to the first two of those questions. The third one, however, presents more of a challenge. “The question really is, how do you make the tradeoff between the good side and the bad side,” says Johar, who has served as Senior Vice Dean and is currently the Meyer Feldberg Professor of Business at CBS, and a core faculty lead for The Hub. “In discussions about the general topics of misinformation, disinformation and why people are losing their faith in democracy, I think it's really important to bring these different topics together for that discussion,” she adds.
The “Breadcrumbs” We Leave
Her answer to the first question about what we can learn from the digital footprints that people leave—the “digital breadcrumbs”, as she refers to them—is “a lot”.
“When we think about data traces, we think about them in isolation,” she explains. “But over a period of years, and when looking at all the data points that come together, we know that any type of digital footprint really gives us a window into someone's psychology,” she adds.
Social media, she explains, is an obvious place to start. “We can start by looking at what people talk about on social media and we can establish how that relates to their personality traits. We can fairly easily, for example, establish whether someone is an extrovert or an introvert—whether they are agreeable or disagreeable.” And crucially, Matz says, it's not complicated.
“In order to establish who someone is, we can quite simply construct a word cloud of the words they use on social media and see how many swear words they use,” she says. A tendency to swear on platforms like Facebook or Twitter, she explains, tends to suggest that someone is not agreeable—not a people pleaser and perhaps not as concerned about the social fabric of society as other people.
Indeed, she explains that one piece of research on this topic conducted by an acquaintance, established that Facebook likes provide a more accurate representation of who someone, how they behave, and what they like, than might be constructed from information provided by friends. In other words, Facebook might well know us better than even the people we love most.
Great Potential and Great Risk
So what can people's digital footprints tell us about the real-life consequences of their unique psychological characteristics?
Again, Matz has a straight answer: there are limitations, of course, but generally a considerable amount.
Working in the UK with a consumer retail brand, Matz established that targeting introverted potential clients with ads and marketing that were tailored to their particular character and behavioral traits yielded a much higher engagement rate than a one-size-fits all approach where both introverts and extroverts saw the same ads. The commercial potential is extremely promising, but Matz is accurately aware of the fact that the power of such a strategy makes it inherently risky and susceptible to abuse. Look no further than the Facebook–Cambridge Analytica data scandal.
In fact, Matz admits that she has fundamental concerns about what might happen if consumer echo chambers are created and then exploited to such an extent that individuals are no longer behaving in a way that's in their best interest. The risk of personal data derived from the digital footprints we use being weaponized against us is real and not to be underestimated, Matz warns. And in this respect, much still needs to be done to implement regulation and policy that ensures that members of the public can benefit from being known by the corporations that serve them, without having to fear adverse consequences.
Bridging a Divide
In conclusion, the final question—of how insights gleaned from people's digital footprints can help individuals and businesses to make better decisions—is an extremely difficult one to answer.
It's a balance, Matz says, between the benefits of being served content and products that you want, and losing control over personal data and information that could, under some circumstances, be compromising. “It's a trade off,” she says. “In very basic terms, you want to enable people to do the right thing—the thing that's right for them. And right now, the vast majority of people aren't equipped to understand how they can control their personal data and what the implications of not protecting it might look like.”
And then there are other considerations too; considerations that don't necessarily relate directly to policy, regulation and immediate risks of sharing too much data, but that have to do with the longer term effect of being known intimately by the corporate world.
To understand this dynamic, Matz draws on the example of going to a favorite restaurant over and over again. “We keep going back because we know we like the food there. But as a result of that, we never try somewhere else even if it's possible that we'll like the food even better at a different place,” she says. Similarly, if the data we share with corporations informs the selection of the products they offer us and the experiences they try to sell us, then we're not seeing what else is available. “That makes our worlds smaller, that makes us boring people,” she says.
Instead, Matz proposes using the information available about the echo chambers in which people reside, to actually offer those people the opportunity to temporarily “jump” into someone else's echo chamber; to see things from a different perspective. “It would be an incredible opportunity to use these powerful algorithms in a different way,” she says. “Think of what might be possible if we could see something from someone's else's perspective,” she adds. “Of course, not everyone would want to do it, but I think it could be an incredible way of creating connection and bridging divides.”