Logistic Regression Approach in Classifying the Effectiveness of Online Education

Abstract

The developments and changes that have accompanied the Covid 19 pandemic have affected the educational world and all sectors. Educational institutions around the world have implemented emergency and online educational practises to ensure continuity of education as opposed to the planned distance education activities that were implemented for continuity of education. Due to the Covid 19 pandemic, face-to-face classes have been held in universities across the world for about a year in many disciplines through various platforms. In this process, determining the effectiveness of distance education practises in universities for students is critical for programmes to achieve their goals. This study aims to highlight the variables and effects that influence university students' decisions regarding the efficiency of online instruction. To this end, 821 university students were surveyed. Their willingness and attachment to online education, socioeconomic level, and gender were tested using logit regression analysis to build a model that predicts university students' decision about the efficiency of online education. Age, gender, high school graduation, willingness to Online Education, and attachment to Online Education are among the variables in the logit regression model that significantly predict university students' decision about whether they consider online education to be efficient or not. When analysing the result of classifying students whether they consider online education efficient or not using the logit regression model, 291 of the 409 students in the group who consider education efficient were classified correctly and 118 of them were classified inaccurately, with the rate of correct classification being 71.1%.