Researchers at the Ben-Gurion University of the Negev’s (BGU) Social Networks Security Research Group have developed a method to predict how well or badly a student will perform in an academic course by looking at social relations both online and offline. It can look a bit strange that they extrapolated some of the offline interactions through their online interactions, but there is some smart thinking. Btw, don’t use this research to argue that you better team up a good student with a lesser one.
From their press-release:
The researchers analyzed data from a BGU course that included assignments submitted online and Web site logs (containing 10,759 entries) to construct social networks of explicit and implicit cooperation among the students. The implicit connections are used to model all the social interactions that happened “offline” among the students: e-mails with questions, conversations in the lab while preparing the assignments and even course forums.
“These connections were very important, as we sought to model the social interactions within the student body,” co-author and Ph.D. student Michael Fire explains.
In addition to analyzing the online submissions of the students who had to work in pairs or in groups, they also tracked login time and computer usage. For instance, if two students submitted their assignments from the same computer, it was a likely indication that the two had worked together to complete the assignment. If two students submitted assignments from different computers, but one right after the other on more than one occasion, the authors gave a value to that data, as well.
“One explanation for what we discovered is that your friends influence your grade in the course, so, if you pick your friends well, then you will get a higher grade,” Fire says. “Alternatively, social networks in courses offer conditions whereby good students will pair with other good students and similarly weaker ones will pair with other weaker students.” he continued.
Abstract of the research paper, freely available here:
In this paper, we propose a novel method for the prediction of a person’s success in an academic course. By extracting log data from the course’s website and using network analysis, we were able to model and visualize the social interactions among the students in a course. For our analysis, we extracted a variety of features by using both graph theory and social networks analysis. Finally, we successfully used several regression and machine learning techniques to predict the success of student in a course. An interesting fact uncovered by this research is that the proposed model has a shown a highcorrelation between the grade of a student and that of his “best” friend.





