A multiple-baseline design across math-fact sets was used to evaluate the effects of a taped-problems intervention on students' performance with addition facts and their inverses in an intact, rural, second-grade classroom. Results suggested that the procedure was effective in improving fluency on math facts as well as their inverses. Across 3 sets of math facts, 15 students increased their fluency (i.e., average increases of 5 digits correct per min or more) on math-problem sets and these improvements were maintained over time. Additionally, results provide evidence that these increases generalized to inverse-problem sets. However, 4 students showed little gains (i.e., average increases of fewer than 5 digits correct per min) on math problems. Discussion focuses on evaluating academic interventions for generalization and directions for future research.