Evaluation of Learning Efficiency of Massive Open Online Courses Learners

Abstract

This study selected 175 massive open online courses (MOOC) learners  of the School of Marxism in a university in Henan Province as respondents. A hierarchical clustering analysis was carried out using SPSS22.0, and the learning styles of learners were classified by k-means clustering. The learning efficiency of learners was estimated by data envelopment analysis (DEA), and the differences of learning styles in learning inputs and learning outputs were analyzed through a variance test. Results demonstrated that according to hierarchical clustering analysis, the learning behavioral indicators of MOOC learners could be divided into four classes. According to the results of k-means clustering, learning styles could be divided into four types, namely, high-input-high-output, high-input-low-output, low-input-high-output, and low-input-low-output. Clustering results could explain significant differences in learning inputs well, thus showing significance (P<0.05). A total of 125 respondents were non-DEA effective, accounting for 71.43%. Moreover, 114 respondents had fixed or increasing returns to scale, accounting for 65.14%. The conclusions of this research are of important significance to analyze the progress effectiveness of students, increase the scientificity and rationality of teaching evaluation theory, train teaching managers to control the teaching effects, and make scientific evaluations of the learning efficiency of university students.