Student and test characteristics were examined by multiple regression analysis and discriminant function analysis to explain why 171 political science undergraduates scored differently on essay versus objective final examinations. Student characteristics included: (1) patterns of creative, crystallized, and fluid abilities as measured by the Cattell Culture Fair Intelligence Test, the McGraw-Hill Basic Skills System Reading Test, and the TORRANCE Tests of Creative Thinking; and (2) previous experience with the subject matter and test format. Test items on the final examination were characterized by the intellectual process required. The essay test was designed to measure creative or fluid (abstract) ability; half of the objective items required fluid ability and half required crystallized (concrete) ability. Results indicated that student characteristics failed to predict total score differences on the final examination, but test characteristics did--students who scored higher on objective tests performed better on abstract items and students who scored higher on essays performed better on concrete objective items. (CP)