This project was designed to develop research strategies and hypotheses for evaluating the effectiveness of exhibits. An exhibit on the role of the Federal Government in science and technology was used as the subject matter. Two basic groups of viewers were used, casual viewers and paid experimental viewers. Both were tested on knowledge gained and on interest and attitude levels. Control data were obtained for both groups. The paid subjects, obtained from high school, college, and adult populations, were divided by age, sex, education, and science and nonscience background. Time was the variable in the experiment, with one group of paid subjects having unlimited viewing time and one group being limited to one-half hour. Casual viewers were observed for comparison with the experimental groups and to determine the relative attractiveness of the exhibit elements. A mockup or small-scale replication of the exhibit was built to determine the feasibility of using such devices for prevalidation and experimental purposes. Results indicate that an exhibit can be analyzed using a wide variety of techniques and that studies of this type can make a major contribution to improved exhibit design and effectiveness. The mockup study also proved successful. A theory of exhibit effectiveness is described. (HW)