<p>In an effort to determine the most efficient method for training new teachers to implement behavior-analytic instruction, we tested the use of video-monitoring and video self-monitoring on the subsequent accuracy of instructional delivery, across two experiments using a delayed multiple probe across participants design. In the first experiment, using prerecorded videos of other teachers, we taught three participants to monitor the accuracy of others delivering instructional trials that met the definition of learn units using the Teacher Performance Rate and Accuracy (TPRA) observation measure. In the second experiment, we taught a different set of participants to conduct TPRAs on prerecorded videos of themselves delivering learn unit instruction. Prior to and following the interventions, the researchers used the TPRA to measure the accuracy of each participant’s own delivery of instruction. If participants did not meet accuracy criteria for delivering learn unit instruction following either intervention, we then implemented TPRAs in situ with immediate feedback to teach participants to deliver accurate learn units and counted the number of sessions required by each participant to achieve mastery of learn unit instruction. The results showed that only participants in experiment 1 (video monitoring) required further training. Specifically, none of the participants in experiment 2 required in situ TPRAs with feedback following the video self-monitoring TPRA intervention as they had already mastered delivering accurate learn units.</p>