The goal of this study was to explore how knowledge-check questions in video lectures affected learning. In a quasi-experimental study, six courses (n=84) were assigned to one of three groups: a control group and two treatment groups. The three groups saw the same video and knowledge-check questions. The three groups were evaluated with different video length segments with questions embedded. All participants received immediate feedback on their responses and were allowed many attempts to correct wrong answers. Instructions, learning activities, and assessments were integrated into a self-paced, self-regulated unit. Even though there were no significant differences between the three groups' post-test performance, relationships were found between knowledge-check scores and length of the videos. There were also correlations between the total number of incorrectly answered knowledge-check questions and post-test scores. Students who watched shorter videos, with the questions placed closer to the content, scored better and were less likely to guess.