SYSTEM DESIGN FOR E-MENTOR SUPPORT SYSTEM WITH PREDICTION OF LEARNING STYLE

As online education expands, the importance of mentoring has been emphasized for quality education. The more meaningful and customized supports mentors provide, the larger their burdens become. This study aims to design a mentor support system to reduce their burden to predict learners’ behaviors and learning styles priori to a course. For the learner type prediction, the authors’ developed e-Learning Self-regulated Learning Scale (eSLS) was employed (Goda, et al., 2010). It was based on the sample items of Wolters, et al. (2003) and consists of 40 Likert-items with 4 Factors, (1)Affective Strategies, (2)Cognitive Strategies, (3)Help-Seeking, and (4)Self-Independency. This system was developed as one server module for Learning Management System (LMS). It allows e-mentors to see would-be dropout learners visually. First, learners answer learning style questionnaire through online questionnaire system in ADLE. This module starts factor analysis function, and calculates learner’s factor score in each factor based on formula. Next, this module divide learners into three ranks; low: -1 >= factor score, middle: -1 < factor score < 1, high: 1 <= factor score. Therefore, this module displays appropriate emoticon to factor score. For example, sad face for low score in each factor is displayed to e-mentors. Thus, this module allows e-mentors to know whom potential dropout learners are intuitively, and helps them to concentrate on mentoring potential dropout learners. We report and discusse the research background and procedures of the instrument development, prediction model equation formulation, and interface design for mentors. The significance of this research is to introduce a new approach to improve online learner supports from the perspective of mentor-supports. As future works, we should be engaged in several task and evaluation as following: (1): Development of the prediction module of learning style, (2) Development of learning style visualization function, (3): Development of mentoring method recommendation function, (4) Evaluation: after development of this module, we evaluate whether e-mentoring module will be helpful for e-mentors and learners. We will compare learning management system embedded this module with plain one from the viewpoints of educational psychology and cognitive science.