This paper presents the findings of an educational intervention designed to examine the effects of self-monitoring actualized by videotaping students’ English role-play performance in compulsory English classes on the improvement of students’ self-efficacy toward speaking in English. The participants in this study were 70 lower-intermediate second year private university students in Japan who majored in welfare. The participants belonged to four compulsory English classes (class A, B, C, & D) and were divided into a video self-monitoring group (class A & B) and a control group (class C & D). Students attended required English classes once a week in a 90-minute class for fourteen weeks. All of these classes focused on language output especially on speaking, used the same textbook and followed the same curriculum. The educational intervention conducted in those classes includes making a dialog collaborating with other student(s) and performing a role-play using the dialog they made in front of the class, employing self and peer-to-peer evaluation on the performance and its preparation procedure, and keeping reflection journals. The role-play projects were conducted three times during the semester. The video self-monitoring group videotaped their performance with their smart phone and watched their videotaped performance prior to self and peer to peer evaluation on their performance. The control group did not videotape their performance. Other than that, all the classes received the same educational intervention. The data was collected qualitatively by using a questionnaire adapted from Bandura (1990) at the beginning and at the end of the semester to assess the students’ self-efficacy toward speaking in English. Changes in scores: the differences between pre-test and post-test scores, served as the measures of development of students’ self-efficacy toward speaking in English. The data was examined from the perspectives of general tendency and individual differences.
