If you're testing a program that calculates student grades, which type of test data would be most appropriate to verify that it correctly handles the lowest possible score (0) and the highest possible score (100)?