Computer Self-Efficacy and Reactions to Feedback: Reopening the Debate in an Interpretive Experiment with Overconfident Students ^†

The accuracy of self-perceptions and the maturity to handle feedback received from others are instrumental for one’s mental health, interpersonal relations, and effectiveness in the classroom and at work. Nonetheless, research on one’s computer self-efficacy (CSE) and reactions to feedback on task performance has been ambiguous in terms of quality, motivations, and results. A particularly important case involves overconfident individuals, i.e., those with unrealistically high CSE beliefs. Using the theoretical lenses of technology use effectiveness along with a mixed methodological approach of thought and lab experiments with 54 undergraduate students performing computer-aided tasks who were randomly assigned to different groups receiving feedback on task performance, we found that valence-based feedback possibly introduces unnecessary information to adjust the levels of CSE and the actual performance of overconfident students. This finding contributes to knowledge on whether feedback is important when skills and learning naturally mature across tasks, in addition to how judicious one is when processing externally motivated feedback. This study additionally offers a novel three-dimensional CSE construct, an instrument to measure the construct, and a scenario-based tool to conduct experiments with sequential decision tasks in the classroom. The practical implications include the planning of tasks and feedback in the classroom, with further insights into organizational hiring, training, and team building.