Analysing the model of online learning in mathematics and science using a two-step least squares method

• Published in: AIP conference proceedings Vol. 2500; no. 1
• Main Authors: Surianshah, Sarimah, Hassan, Suriani
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Melville American Institute of Physics 08.02.2023
• Subjects: Bias; Costs; Distance learning; Education; Expenditures; Incentives; Least squares method; Mathematical analysis; Mathematics; Students
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0109920

This study analyses the model of online learning. Specifically, the model examines the effects of student education expenditures on student perceptions on online learning of mathematics and science subjects. We expect that student education expenditures are higher upon the introduction of a new method of learning which is online learning during the Covid-19 pandemic. Even though many incentives have been implemented, unobserved factors such as availability of transportation to commute and distances to acquire reliable internet connection might affect student's expenses and performance. Hence, in this study, we employ the two-step least squares method to address potential endogeneity bias in the baseline model. The findings show some evidence of bias with statistically significant results of the endogeneity test and strong instrument results of the overidentifying restriction test. Male students in particular significantly tend to have high education expenses and less likely to prefer online learning compared to face-to-face learning of mathematics and science. It suggests for the government and policymakers to intervene strategically by allocating digital endowment such as network externalities at appropriate platforms to reduce students’ expenses on education. It also perhaps may help to reduce students gender inequality in education.