The primary question to be answered as a result of the research was: Do native English-speaking undergraduate students learn CD-ROM database search skills more effectively through the use of conventional instruction method or by training confined to a written manual self-instructional method? Thirty-seven Oregon State University students were randomly assigned to two groups and participated in the study. The study utilized an experimental pretest-posttest control group design. The length of the study was six weeks. All participants were pre and post-measured using the CD-ROM Database Competency Test which was validated using the Delphi method. The subjects were trained using Wilsondisc and SilverPlatter databases. ANCOVA and t-tests were the main statistics utilized in the analysis. The results of the study were as follows: 1) Subjects in both treatment groups experienced significant test score gains between pretest and posttest; 2) Those subjects which were assigned to the conventional instruction group had significantly higher scores than did subjects assigned to the self-instruction group; 3) Subject gender was determined not to be a factor when differences were considered from test results; 4) Interaction did not play a significant role in influencing the data results; 5) Students rated formal training as being very helpful in improving the effectiveness of their CD-ROM database searches. The major conclusion drawn from the study was that the native English-speaking students at Oregon State University learned CD-ROM searching skills more effectively through the conventional instruction method than those students who were trained by self-instruction. The results of the study substantiated the need to undertake further efforts to improve the efficiency of learning CD-ROM access instructional methods in higher educational settings. / Graduation date: 1994