An approach to teaching based on a variant of the Keller contingency management method of instruction was investigated. Two matched sections of introductory psychology students were taught the same material by the same instructor. One group received the Keller method, while the other group was taught using a traditional lecture approach. Students taught by the Keller method performed significantly better on a post-course quiz than did those receiving the traditional lecture approach. Students taught by the Keller approach spent significantly greater time in preparation, enjoyed the text material more, and felt they learned significantly more than those receiving the traditional lecture approach. On a long-term retention test given 5 months after course termination, those receiving the Keller method retained significantly more material than did those receiving the traditional approach.