DIGITAL LIBRARY
COVID-19 LOCKDOWN, ONLINE LEARNING AND STUDENTS’ PERFORMANCE
1 University of Plovdiv "Paisii Hilendarski" (BULGARIA)
2 Khulna University (BANGLADESH)
3 Al-Balqa Applied University (JORDAN)
4 Apex Professional University (INDIA)
About this paper:
Appears in: ICERI2020 Proceedings
Publication year: 2020
Pages: 3500-3506
ISBN: 978-84-09-24232-0
ISSN: 2340-1095
doi: 10.21125/iceri.2020.0784
Conference name: 13th annual International Conference of Education, Research and Innovation
Dates: 9-10 November, 2020
Location: Online Conference
Abstract:
The COVID-19 Pandemic is the most serious ongoing pandemic around the world. Every sector, as well as the education sector, is being hampered by this pandemic. The COVID-19 forces many universities all over the world to adopt online learning, remote work, and other activities to reduce the spread of the virus. This situation has led to an increased interest in the use of learning management systems and tools for virtual meeting. As a result of the COVID-19 pandemic and the increased amount of the data generated by students, many higher education managers and faculty staff started to use more actively Learning Analytics tools to inform their decision-making and planning for future academic terms. The paper presents an attempt to use learning analytics techniques to improve student success in a given course as well as to find out the prevalence and cause of the learning condition of the student during the lockdown and before lockdown. It presents the organization of the training in LMS Moodle and Zoom of 51 students teaching "Object-oriented programming and design". During each week teachers have access to learning analytics tool that helps them to track students’ activity, their progress and achievements and to make decisions about the timely support of students with unsatisfied results. To confirm the effectiveness of online learning, and the Learning Analytics tools, students’ final grades are compared with their grades from the previous year (by subject from the same field of study - "Programming"). The dataset consists of 51 students’ records, divided into two sets: one contains the academic outcome of the students before the COVID-19 pandemic, and another set belongs to the pandemic time when the classes were held in online mode only. For the comparison of the results are employed three most frequently used classifiers in the educational data mining domain - Bayesian Network, Random Forest and Multilayer perceptron classifiers. The statistical analysis showed that the overall result of students during COVID-19 lockdown is better than their overall results before lockdown. The obtained results prove the effectiveness of the conducted online learning and the applicability of the Learning Analytics techniques to increase students’ success.
Keywords:
COVID-19, Online Learning, Student Performance, Learning Analytics.