In this study, we examined the effect of video-based feedback designed to highlight a peer engaging in effective thinking processes on self-efficacy beliefs and learning outcomes (performance on a delayed quiz). Students in an introductory statistics course participated in an online learning activity where they received feedback in one of three randomly assigned conditions: a video of a peer demonstrating the process of arriving at a correct answer (mastery condition), a peer making mistakes then self-correcting those errors before arriving at a correct answer (coping condition), or a screenshot of a peer's correct worked example (as a control). Results indicated that students who watched the mastery videos, but not the coping videos, rated their self-efficacy higher and scored higher on a class quiz taken more than a day after the feedback intervention than students who viewed a worked example. However, students in the two video conditions did not significantly differ in terms of either self-efficacy and quiz performance. The results of this study, although modest in scope, illustrate how the design of feedback could lead to noticeable differences in student learning.