TECHNIQUE OF ONLINE-LEARNING DATA ANALYTICS AS A TOOL FOR IMPROVING QUALITY OF THE MASSIVE OPEN ONLINE COURSES
Ural Federal University (RUSSIAN FEDERATION)
About this paper:
Conference name: 11th annual International Conference of Education, Research and Innovation
Dates: 12-14 November, 2018
Location: Seville, Spain
Abstract:In a digital economy age transition to management of economic processes using big data creates new opportunities in all spheres of social life. Of course, crucial changes take place in education too. These changes include digitalization of educational content, wide use of information technologies for data storage and transfer, along with new approaches to teaching methods, personalization of learning, and adoption of learning according to the demands and capabilities of the students achieved by using artificial intelligence. Massive open online courses (MOOCs) became a response to the society demanding new education, and gained widespread acceptance in recent decades. MOOCs provide new opportunities for building individual learning paths and permanent education.
Besides the evident benefits including openness of educational content, flexibility, and affordability of education both in terms of required funds, and lack of limitations due to student location, basic level and time required to master a course, MOOC provides another benefit in form of a digital learning path captured within the electronic learning system. That allows tracking these paths, revealing correlations among student activities within the course and learning results, and studying student behavioral patterns. Learning analytics also provide objective metrics of MOOC quality, and act as course improvement tool.
The study presents a method for processing learning analytical data, and covers analysis of efficiency metrics for learning that included use of MOOCs by Ural Federal University that were located on the National Open Education platform. The study was aimed at revealing possible disruptions of course cohesion, errors in material presentation logics, methodological errors, and irrelevant testing tools. Learning efficiency metrics included dynamic conversion of the course users, student involvement in learning process, and a degree of course individualization with regard to success of different student categories. Developed database containing depersonalized data about students passing the courses did allow learning student academic performance in dynamics, and designing recommendations for course authors regarding improvements in online learning methods, and support of education process implemented using online courses.
Keywords: Online learning, massive open online courses, higher education, learning analytics.