Superstitions can be comforting, offering the illusion that everything is under control. This is how Professor Gita Johar, whose research explores often automatic and largely unconscious influences on choice, explains the uptick in psychic consultations reported at the onset of the financial crisis.
Most consumer decisions have some rational basis, where cause clearly bears some relation to effect — the way buying a certain car leads to feeling safe or comfortable. “But a good deal of anecdotal evidence suggests that consumers choose some brands to try to influence outcomes over which they have no control, and where there is no basis to link the choice with a desired outcome,” Johar says.
Johar, working with Eric Hamerman, a former PhD student now at Tulane University, explored superstition, control, and consumer decisions in a series of three experiments. In the first, subjects believed they were helping the researchers choose whether to use blue or green background computer screens in their online lab, and overwhelmingly expressed a preference for blue screens. The researchers then gave everyone IQ and social skills tests, alternating screen colors over different stages of the tests, leading some subjects to believe they were performing better when taking the tests on the less-preferred green screens but giving no feedback to others (although there was no actual difference in performance between screens). Offered the choice of background color for additional rounds of tests, subjects who received the positive feedback while using a green screen opted to take the remaining sections of the tests on green screens at a much higher rate than those who got no feedback.
“We showed that if subjects feel they can control the outcome of something that is important to them — these were Columbia undergraduates who presumably care if they do well on IQ tests — then they will use a less desirable screen at a much higher rate than a simple positive association would predict.”
In another experiment, the researchers asked subjects to watch fellow university students play a fictitious competitive game involving the construction of stock portfolios. The idea was to ensure subjects (who believed the online game was real) had some stake in the outcome, and to show that they would give up well-established preferences (in this case, a favorite color). While it is notoriously difficult to get people to give up established preferences, that’s just what happened: Each subject was given a stress ball — which researchers associated with the competition by describing as an item that stock traders often use when making decisive trades — in their favorite color. After watching a part of the game while holding the stress ball, and seeing their team perform poorly, subjects (who believed the game remained in progress) were more likely to switch to stress ball in a different, less-preferred color when offered one to take home.
The researchers also found that even a well-established product can be imbued with superstitious power. Participants were given Snickers just prior to watching their college compete in a simulated quiz bowl game in which their team started off poorly. As the home team’s score improved, the researchers distributed Snickers during simulated refreshment breaks. When the experiment ended — with the game still in progress — the researchers offered subjects a Kit Kat or Snickers. “You would expect most people to choose Kit Kat since they’ve already had so many Snickers. But about half the time, people chose Snickers. Moreover, when asked to predict who would win the game, those who chose Snickers rated the likelihood of Columbia winning very high. Importantly, those who chose Snickers after making the prediction rated the likelihood of Columbia winning lower than this group,” Johar says. “They believed that their superstitious choice strategy would work!”
Some businesses could have some fun — and profit — from this insight. “It would be easy for a bar to charge a premium for Guinness if, for example, fans could associate that brand with a Yankees victory,” Johar says. “Or to do dynamic pricing in a stadium based on building up associations between products in the stadium and the home team.”
The researchers also found, through a personality assessment administered toward the end of each experiment, that only people experiencing low self-efficacy gravitate toward superstitious behavior. “Trying to control an outcome is one way that people try to make themselves feel better,” Johar says.
Notably, Johar says, people employ these strategies unconsciously. “When we asked people why they made these choices, they didn’t tell us that they were trying to influence the outcome. Instead, they told us about attributes of the products: ‘I liked the green screen,’ or, ‘I liked the other color stress ball,’” Johar says. “And when we ask them to make predictions, they seemed to assume that their having done the superstitious thing would affect the outcome.”
Gita Johar is the Meyer Feldberg Professor of Business in the Marketing Division, a senior scholar at the Jerome A. Chazen Institute for International Business, and vice dean for research at Columbia Business School.
Read the Research
"Conditioned Superstition: Desire for Control and Consumer Brand Preferences"