Late last year we talked about the challenges the National School Lunch Program faces in supplying healthy school lunches. It is only half the battle though. Once healthy food is in the schools, students still need to make the choice to eat it. Researchers and schools are trying many creative ways to nudge students into making healthier choices. Dr. Elton Mykerezi from the University of Minnesota Applied Economics program along with Marla Reicks, Joseph Redden, Traci Mann and Zata Vickers did just that in a new study that put pictures of vegetables on lunch trays.
The article Photographs in Lunch Tray Compartments and Vegetable Consumption Among Children in Elementary School Cafeterias was recently published in The Journal of the American Medical Association. We 'sat down' with Dr. Mykerezi to talk about the study which was conducted with about 800 students at a Richfield, MN elementary school.
*Questions from The Food Thought Blog in bold.
Thank you for the opportunity to discuss our study with you.
What was the motivation for your study? How was it different than what people have done before?
The motivation was to increase vegetable consumption in school cafeterias. The idea, to make subtle changes in the choice environment in order to guide choices, is both old and new. I say it is old because it has been long known to behavioral scientists that, very often, individuals make choices in an automated or semi-automated manner. In many cases this is a positive thing, as life is filled with millions of little choices, and investing significant cognitive resources into each of these can be exhausting. Consider, for instance, the way in which you open doors. On your way in, you'd likely encounter a handle and you'd pull. A flat surface on your way out signals a push. The average person would likely not be able to recall the last time they stopped in front of a door and pondered on exactly how to get it open, yet we go through dozens of doors each day. Well-designed doors and our automated responses to them, save us considerable time and trouble.
I think what lies at the heart of behavioral economics is the insight that automated decisions are made daily and we need to seriously think about how to isolate instances when humans are most likely to make choices in a less than fully rational and deliberate manner and to incorporate these into our models of human behavior. Research has shown that choices related to food (what, when and how much we eat) are near the top of this list.
The team is composed of faculty in Food Science, Nutrition, Marketing, Psychology and myself (an Applied Economist). We were funded by a USDA seed grant intended to create collaboration between interdisciplinary research teams and schools to investigate how nudges can be used to improve choices in school cafeterias. So the novelty in our project was in finding ways to collaborate with the school staff in order to identify changes in the cafeteria choice environment that are feasible, unobjectionable and effective (in altering behaviors and costs of implementation). These efforts put some formal research behind choices that the cafeteria staff has to make anyway, but with the explicit purpose of increasing vegetable consumption.
We tested several strategies, but I think using images of vegetables was particularly elegant. Improving nutrition in cafeterias is an on-going battle. However, reform involving menu changes has multiple stake holders and, not surprisingly, has resulted in heated debates (sometimes even on how to define vegetables). Placing images of vegetables in the tray compartments that are intended for vegetables seems rather natural. I think it is extremely unlikely to spur any debate or objection, yet it affected choices in a meaningful way. One additional point to make is that national data show that despite the fact that the overwhelming majority of schools serve vegetables daily, only half of the children consume one on any given day. The school we worked with faired much worse; only about 10% or the children took vegetables, on average. So this strategy supplements other efforts to improve school nutrition rather nicely!
How did the results compare to other similar 'nudges'? Did you find them to be more or less successful?
Actual field experiments using 'nudges' in cafeteria settings are relatively rare, especially in elementary schools, so it is difficult to say how our results compare to those of other researchers. Traditional education-based approaches to increasing vegetable consumption in schools have produced mixed results. Ours were comparable to a number of these (but the strategy is likely less costly). There are some examples of nudges causing increases in fruit and vegetable consumption among children and college-aged young adults. I would say we found pretty substantial effects, but our findings are quite consistent with most studies. The fact that we were working in an environment where very few children took vegetables (so there is much room for increases) may also be a noteworthy factor in interpreting our findings.
You say in the article that the experiment needs to be repeated to learn of its sustaining effects. Do you think this is a tactic that could 'stand the 'test of time'?
I will start by stating that there is no substitute for a well-designed long-term study to provide an answer to this question. But I will indulge in some educated guessing and say that I honestly expect the tactic to stand the test of time. I base this on the following facts. First, whether or not to pick up a vegetable from the lunch line is a decision that is made within a few seconds, often while also looking over all the other items that are available and perhaps chatting with peers and thinking about where to sit once one is through the line. This setting has all the ingredients of a "nudge-friendly" choice environment. Pictures of vegetables on trays likely suggest a "social norm", i.e. "people put vegetables here". In instances of quick decisions such as the present we (children and adults) tend to rely quite heavily on social norms, i.e. on the implied behavior of others, to determine what the right course of action is. So there is good theoretical reason to expect that the strategy will stand the test of time. Secondly, I think two of our empirical findings supply hints. Loosely speaking, think of three groups of children; the definite takers (likely the roughly 12% of children who took carrots even without the nudge), the definite non-takers (likely the 63% of the children who didn't take carrots with or without the nudge), and the "marginal choosers", those who could take it or leave it on any given day. The 26% who were affected were likely to be marginal choosers. So the strategy didn't "fool" people into taking something they'd never eat, it just nudged the ones who were "on the fence". Indeed they ate what they took. Even among the children who only took vegetables when nudged there was virtually no increase in waste in green beans and a minimal increase in carrots! So while I think there may have been some "novelty effect" (a reaction to a change in the way the tray looks) captured in this pilot experiment, I expect the bulk of the effect to persist.
Did you or the other researchers find anything that surprised you?
Honestly, the absence of a change in waste per child who took vegetables was a bit of a pleasant surprise to me.
Anything else? Are you working on any other ideas to get kids eating healthier?
One of our other experiments involved making vegetables available first, as children waited to enter the lunch line. The pilot produced substantial increases in the amounts of vegetables eaten, and we are currently applying for funding to further study the generalizability and long-term effects of this strategy.