Eating Your Way to Higher Test Scores

04/01/2003
Summary of working paper 9319
Featured in print Digest

School districts that increased calories on test days experienced increases in 5th grade pass rates of 11, 6, and 6 percent respectively on the mathematics, English, and history/social studies tests.

Now that public schools can lose federal funding as a result of poor student performance on standardized tests, they have begun paying more attention to test scores. Although the hope was that schools would focus solely on raising test scores by improving student achievement, school officials have responded in other ways as well. Among the known adaptations are removing potentially poor performers from the test pool by reclassifying them as "disabled" and providing students with answers to test questions.

In Food For Thought: The Effects of School Accountability Plans on School Nutrition (NBER Working Paper No. 9319), authors David Figlio and Joshua Winicki examine whether schools exploit a more subtle method to increase test scores: changing their lunch menus. Several studies have suggested that consuming glucose before taking tests may increase scores. Under the Department of Agriculture School Meals Initiative for Healthy Children, schools must meet nutritional guidelines over a one-week period. This gives menu planners the flexibility to alter meals from day to day. Given the software available for school menu planning and nutrient analysis, food service directors also have the tools to fine tune the menu.

Using information from a random sample of 23 Virginia school districts, Figlio and Winicki compare the nutritional and caloric content for school meals over the testing cycle for the Virginia Standards of Learning school accounting system. They find that the schools most likely to increase the caloric content of their lunches are those in districts with threatened schools. In those districts, school lunches averaged 863 calories during testing periods, 761 calories before, and 745 calories after. Though calories increased, nutrients did not. Nor was the calorie increase a result of serving students their favorite meals -- pizza, cheeseburgers, and tacos, as measured by sales data -- on test days.

School districts that increased calories on test days experienced increases in 5th grade pass rates of 11, 6, and 6 percent respectively on the mathematics, English, and history/social studies tests. Although the authors caution that their results are to be treated with caution because of small sample size, they suggest "that test score gains associated with accountability systems may in part be artifacts of manipulation rather than improved efficiency, particularly for schools on the margin."

-- Linda Gorman