This research investigates whether usage and satisfaction measures are appropriate surrogates for decision performance by examining the relationships among three commonly used measures of information system (IS) effectiveness: decision performance. usage. and satisfaction. Decision performance, the system's ability to support managerial decision making. is the most direct of the three measures. Because it is often difficult to measure decision performance, system usage and user satisfaction are often used as substitute measures for decision performance in research and practice. A laboratory study was conducted in which subjects used a computerized accounting information system to perform a series of inventory management tasks under varying situational factors (feedback and complexity). Results suggest that although usage and satisfaction are closely related to decision performance, the relationships arc not sufficiently strong to warrant their use as surrogates. Further, situational factors differentially affected the three measures. These findings are important for both researchers and practitioners. A large body of research examining factors associated with IS effectiveness depends on the adequacy of surrogate measures. Evaluations of effectiveness by practitioners are often similarly dependent on these surrogates. The results of this study suggest that conclusions about IS effectiveness based on satisfaction and usage measures as substitutes for decision performance should be interpreted with caution.