A Visual Data Mining Approach to Understanding Students Using Computer-Based Learning Technology

Educators are increasingly using online computer-based training and assessment software— especially with large classes or in distance education settings. This technology is often criticized, however, for hampering personalized interaction with students. This paper introduces a unique approach for analyzing student characteristics influencing their adoption and use of computer-based educational technology so that instructors can better meet student learning needs. Using visual, selforganizing mapping, our data mining approach clustered students based on input data from thirty-six survey questions posed to over 400 students with experience using computer based training and assessment. The data mining technique provided clear descriptions of four different student clusters. Based on the unique characteristics of the four clusters, instructors could optimize classroom resources as well as provide individualized support once specific students are matched to their respective cluster group. In this manner, continual computer-based assessments of students can be used to maximize computer-based learning and evaluation.