Abstract:

Background and Objectives:
This study focuses on faculty development (FD) programs that train university teachers and their applicants. It aims to develop an objective evaluation method for teaching delivery. We examine whether it is feasible to quantify lecturers’ delivery methods in real-time online classes. Microteaching is a central component of FD programs. Participants learn class design, instructional material preparation, and actual classroom practice. Microteaching serves as the focal point of our study. Traditionally, face-to-face mock classes were used. However, with the spread of COVID-19, ongoing FD programs, and the growth of online classes, more microteaching sessions now occur online. This shift has increased diversity in implementation formats. In traditional microteaching, gestures and speaking style serve as evaluation criteria. In the online format, lecturers’ appearances differ significantly. This study focuses on indicators of head and gaze movements. These can be extracted from the lecturer’s video feed and may serve as a quantitative evaluation method for microteaching.

Methods:
In this paper, we analyzed data from 59 participants in a FD program who conducted microteaching sessions in an online environment. The data included subjective evaluations by other participants and video recordings of the sessions. First, based on the results of subjective evaluations using a Likert scale administered during the mock classes, we conducted multiple regression analysis with the overall evaluation of the mock class as the dependent variable and 10 items—including goal and objective setting and class flow—as well as six items related to voice and body movements as independent variables. Next, we performed motion analysis using OpenFace on participants in the top and bottom 10% of the overall evaluation scores. In this paper, we examined the amount of eye movement and the angles of head rotation around the three axes. For each movement, we conducted a significance test to determine whether there were differences between the top and bottom groups in the subjective evaluations.

Results and Discussion:
Based on the results of a multiple regression analysis using subjective evaluation data with only six variables related to speech and physical movements as explanatory variables, facial expressions and body posture were identified as significant factors. This result revealed that, in the group with high overall ratings, vertical gaze movement was small, particularly downward. Furthermore, regarding head movements, the absolute values of all rotation angles—roll, pitch, and yaw—were small in the group with high overall ratings, suggesting that their faces were facing straight ahead during the microteaching sessions. Since this study analyzed data from actual FD programs rather than a controlled experimental environment, the amount of valid motion analysis data was limited. However, we identified trends in gaze and head movements correlated with overall course evaluations, confirming the potential of these metrics as objective evaluation indicators in online classes. Moving forward, we aim to increase the dataset to develop an automated evaluation system for microteaching and a practice support system.

Keywords:

Eye movement, faculty development, head rotation, microteaching.