Some downtrodden elementary schools try to boost students' test scores by serving high-energy, low-nutrition lunches on test days, according to a study published online 19 November by the National Bureau of Economic Research. And regrettably, say the researchers, the idea just might work.
Standardized tests are a high-stakes business for school districts because low scores can mean reduced funding and trigger requirements that students be offered the option of attending a different school. So not surprisingly, schools have developed strategies to improve students' scores. Some approaches, such as increasing the pace of instruction and encouraging hard work, might offer lasting benefits to students, says Brian Jacob, an economist at Harvard University. But research has indicated that other schools "game the system," he says--for example, by altering discipline and suspension policies during test periods so that low-performing students will not be present for tests.
To investigate whether schools use food to their advantage on test days, economists David Figlio of the University of Florida, Gainesville, and Joshua Winicki of the American Institutes for Research, a nonprofit behavioral and social science research organization based in Washington, D.C., examined lunch menus from a random sample of 23 Virginia school districts for the 1999-2000 school year. They reasoned that because high-calorie foods heighten brain function for a bit, schools might attempt to exploit this effect to raise test scores.
The data, from the U.S. Department of Agriculture, weren't broken down by school, so Figlio and Winicki compared districts, each of which sets a menu for all of its schools. They found that in school districts with at least one school facing sanctions for poor test performance, lunches served in elementary schools during 3-day periods when tests were given contained roughly 15% more calories than lunches served during other periods. The high-energy foods also tend to contain fewer nutrients, such as vitamin A and C. There were no such discrepancies for districts in which no school faced sanctions.
Figlio says that it's unlikely schools are simply offering treats on test days as a reward for enduring stressful testing. Students' favorite foods--pizza, tacos, and cheeseburgers, which vary in caloric value--were no more likely to be served on test days than on nontest days, he says. Preliminary findings in a subset of the schools suggested that the extra calories do make a difference: School districts that increased calories the most showed the most striking improvement in test scores, especially in mathematics.
"If the schools are stooping this low to boost test scores, this is scary," says Cynthia Mears, a pediatrician at Northwestern University's Feinberg School of Medicine and a member of the American Academy of Pediatrics' Committee on School Health. The director of the school nutrition program for the Virginia Department of Education declined to comment.