In this research, I attempted to show that increasing a person's computer self-efficacy is a viable route to making that person more likely to use a computer tool when given the opportunity to do so. I also show the strong link between a person's computer self-efficacy and intentions to use a computer. Computer self-efficacy is a person's belief about his or her ability to successfully use a computer. Subjects in this field experiment were 193 students in eight introductory computer classes at two colleges. I attempted to increase their computer self-efficacy through their participation in a self-efficacy-increasing strategy that used goal-setting, self-reinforcement, and feedback. Half of the subjects randomly received this strategy, and half received a placebo strategy, which was not expected to have any influence on the subjects' computer self-efficacy. Due to the subjects’ insufficient use of the strategies, I was unable to increase the computer self-efficacy of the subjects through this strategy. Although the subjects’ computer self-efficacy was not increased, I was still able to see the effect of computer self-efficacy on a person's intentions to use a computer. Of all the variables tested in this study, computer self-efficacy was the best predictor of intentions to use a computer—better than computer experience, computer ownership, academic major (computer-related vs. not computer-related), instrumentality beliefs about computers, and others. This finding showed that computer self-efficacy is a construct that people should pay attention to, particularly people who design computer tools, those who choose which tools to use, and those who train others on new computer tools. These people should particularly understand the needs of users with low computer self-efficacy and how to best meet those needs. Additionally, I tried to find out why the subjects did not participate in the use of the strategy. It appears that the main problem in my implementation of the self-efficacy strategy was not providing the subjects the proper amount of time, encouragement, and incentive for learning and practicing the strategy. My results also showed that those who used the strategy had higher computer self-efficacy than those who didn't use the strategy; that is, those who could benefit most from the self-efficacy strategy (those with low computer self-efficacy) were less likely to use it. This finding underscores the importance of providing a supportive environment for people to learn and practice the strategy, so those with low computer self-efficacy will be able to use the strategy successfully, rather than avoid the strategy before they even get a chance to learn or use it. / Master of Science