Network analysis of university courses

Abstract

Crucial courses have a high impact on students progress at universities and ultimately on graduation rates. Detecting such courses should therefore be a major focus of decision makers at universities. Based on complex network analysis and graph theory, this paper proposes a new framework to not only detect such courses, but also quantify their cruciality. The experimental results conducted using data from the University of New Mexico (UNM) show that the distribution of course cruciality follows a power law distribution. The results also show that the ten most crucial courses at UNM are all in mathematics. Applications of the proposed framework are extended to study the complexity of curricula within colleges, which leads to a consideration of the creation of optimal curricula. Optimal curricula along with the earned letter grades of the courses are further exploited to analyze the student progress. This work is important as it presents a robust framework to ensure the ease of flow of students through curricula with the goal of improving a university's graduation rate.