Aspects of design and procedures used in the evaluation of a performance contract are discussed, and the project and findings are briefly summarized. The Virginia performance contract in reading was carried out in seven school districts. Evaluation of the project showed that although instructional changes that utilized desirable individualized approaches were used in the experimental program, the experimental group achieved neither less nor more than the control group on ordinary standardized test measures. However, the experimental group had more positive attitudes toward reading than did the control, and in addition mastered the majority of the instructional objectives prescribed. Eleven observations resulting from the evaluation, presented for the guidance of those who contemplate performance contracting, are presented. (DB)