Study uses students' 'traditional' social networks to predict grades
Graduates of Harvard Law School wave gavels as their degrees are conferred during Harvard University commencement exercises, in Cambridge, Mass., Thursday, May 24, 2012. (AP / Steven Senne)
Christina Commisso, CTVNews.ca
Published Friday, December 28, 2012 2:33PM EST
New research out of an Israeli university suggests that an analysis of students’ social networks can predict how well they do in an academic course.
Scientists out of Ben-Gurion University of the Negev looked at interactions between students, such as questions emailed to one another, assignments completed in pairs, course forums and conversations in labs to predict a student’s grade before the final exam.
Researchers also tracked login time and computer usage for students who worked in groups.
“While most papers about social network analysis deal solely with information gathered online, this study draws some of the information from the real world -- social interactions which were conducted off the grid,” said study co-author and PhD student Michael Fire.
Fire told CTVNews.ca on Friday that the study revealed a clear connection between students’ social networks and their final grade.
“In the past many studies demonstrated the existence of similar types of correlations,” Fire said via email from Israel, pointing to a 2008 study that suggested individuals who are surrounded by many happy people more likely to become happy in the future.
“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,” he said.
“Alternatively, social networks in courses offer conditions whereby good students will pair with other good students, and similarly weaker ones will pair with weaker students.”
Fire said the research team successfully predicted final grades to within a mean average error of 9.979.
While there are no immediate plans to use the research in university courses, Fire said the study could be used to determine which students may need more help in a course and which ones should consider a career in the subject area.