The purposes of this study were to develop a coordinated set of seven Computer Assisted Instruction (CAI) modules for a one-semester, introductory, college-level nutrition course and to compare the effectiveness of the CAI Modules with traditional lectures. A quasi-Solomon four-group experimental design was used. Two instruments were used to collect pretest and posttest data: a nutrition knowledge test and a Likert-type attitude instrument containing three scales designed to assess students' attitudes toward computers. After finishing each of the seven CAI modules, students in the experimental groups completed a semantic differential attitude instrument containing four scales designed to evaluate student attitudes toward each CAI module. The results of analysis of covariance procedures indicate that the adjusted mean post scores of the experimental group on the knowledge test and Likert-type attitude scales Computers Help Me Learn and I Am Capable of Using Computers differed significantly from those of the control group. Mean scores on the semantic differential instrument scales tended to rise steadily for the first four modules assigned, then decline to a plateau. The findings indicate that CAI can improve nutrition knowledge and that students enjoy using CAI; thus, CAI modules may be a valuable and effective educational tool for nutrition educators.