This study disaggregates posttest, pretest, and value-added or difference scores in economics into four types of economic learning: positive, retained, negative, and zero. The types are derived from patterns of student responses to individual items on a multiple-choice test. The micro and macro data from the Test of Understanding in College Economics (TUCE) are used to show how aggregate scores can be reinterpreted based on their learning components. The regression analysis shows the relative contribution from learning components to aggregate scores. A value-added or difference score has a potential problem because it is a mixture of positive and negative learning.  A better alternative would be to use the positive learning scores to assess improvement in economic understanding.