Guilty Pleasures

Based on research by Utpal Dholakia, Scott W. Davis and Kelly L. Haws

• A lack of self-control in one area, such as overeating, doesn’t necessarily correspond to a lack of self-control in other areas, such as overspending.
• To understand why people make bad choices, researchers need to look at specific activities rather than broad concepts such as “self-control.”
• Public policy researchers armed with precise data have a better shot at creating behavioral interventions that work.

Is indulging in a lavish dessert you shouldn’t eat the same problem as buying an expensive bag you shouldn’t buy? Past research has tended to treat all forms of potentially harmful self-indulgent behavior, from smoking to procrastinating, as shades of the same overarching issue: a lack of self-control. Lumping these behaviors together leads to the conclusion that self-control is a universal trait — something you either have or you don’t. But recent Rice research suggests that not all overindulgences are alike. Some people are stoic in the face of a Kate Spade sale but unable to resist a tempting tiramisu, and vice versa.

Rice Business professor Utpal M. Dholakia and former postdoctoral fellow Scott W. Davis collaborated with Vanderbilt University associate professor of marketing Kelly L. Haws on a study that found no evidence that people fail — or succeed — equally in all aspects of self-control. An accurate measure of self-control in eating, the team found, might not apply equally to other domains such as spending and saving. Correctly understanding the nature of self-control is crucially important for researchers, since curbing social problems such as obesity and consumer debt are of vital concern for the people they affect and the societies they undermine.

A great deal of time and money has gone into studying self-control — and its absence. But in the past, researchers have taken a one-size-fits-all approach to measuring self-control. They’ve used something called the general self-control scale, which assumes that all forms of self-control tap into the same reservoir and that, therefore, low self-control in one domain will predict similarly low self-control in others. So, for example, a compulsive buyer should also be a bing eater.

Research using this broad concept of self-control has shaped public policy. But Davis, Dholakia and Haws argue that to get more accurate, and more useful, findings, researchers need to refine their measurement tools to apply specifically to the object of their study. To gauge the potential effectiveness of a tax on sugary drinks in lowering soda consumption, for example, researchers should measure self-control in the face of fizzy, bad-for-you beverages, and not extrapolate from unhealthy habits in other areas.

Dholakia and his colleagues based their conclusions on five studies that measured self-control in shopping and eating, comparing them to levels of general self-control. They concluded that “domain-specific” measures were better at predicting behavior than general measures. People who demonstrated low levels of self-control in eating, for example, could be expected to binge eat in the future, regardless of their general self-control level.

In one study, participants were told they were testing a new eating and exercise smartphone app. They were asked to add a Snickers bar to the list of foods they planned to eat that day. For some participants, the app produced a picture of a Snickers bar and its nutritional information; for others, the app produced a picture of feet walking and the comment, “You must walk 65 minutes to burn off that Snickers bar.” Then they were asked to rate the likelihood that they would actually eat the Snickers bar.

For people with high “eating self-control,” the likelihood of indulging in the candy bar stayed roughly the same across both scenarios. But for people with low eating self-control, the likelihood of eating the Snickers was significantly higher when they were given the nutritional information instead of the exercise equivalent. “Our findings … raise the possibility that providing nutritional information may actually enhance the appeal of the Snickers bar, leading to greater desire that those lower in self-control are less equipped to handle,” the authors wrote.

These findings could help design interventions to prevent people from overeating. Current policies designed to curb sugar consumption rarely have the intended effect — and that failure, the authors suggest, could stem from a lack of understanding about what causes people to choose short-term pleasure over long-term health. The research stopped short of investigating how to actually stop people from eating too much sugar (or spending too much, gambling too much, smoking too much, etc.), but Dholakia’s team concluded that asking more targeted questions was a fundamental step in the right direction. The takeaway? Call a Kate Spade bag a Kate Spade bag, not a cupcake. Overindulging in one doesn’t mean you’ll overindulge in the other.

Utpal M. Dholakia is the George R. Brown Professor of Marketing at Jones Graduate School of Business at Rice University.

To learn more, please see: Haws, K. L., Davis, S. W., & Dholakia, U. M. (2016). Control over what? Individual differences in general versus eating and spending self-control. Journal of Public Policy and Marketing, 35(1), 37-57.