High school students can influence their casual friends to earn higher or lower grades to match, according to a study done in Binghamton, NY. Scientific American reports that the study, published in PLOS One, is the first to compare social networks with grades over the course of a year.
The study asked 11th graders to identify which students were their best friends, friends, acquaintances, relatives, or people they just didn’t know at all. Based on this self-reporting, researchers constructed graphics to show the social ties in Maine-Endwell High School. Since the class had only about 160 students, and most of them participated, there was a high overlap of acquaintances and friends. Nearly all of the students considered each other known at that level. After separating the social networks into the three levels of friends, best friends, and acquaintances, researchers identified which students had best friends or friends with grade averages that were higher or lower than their own.
Starting with grade point averages in January, 2011, researchers compared averages the following January. They found that students whose reported “friends” had higher grades had mostly brought their own grades up by one point of GPA. The “best friends” links did not have the same effect. Researchers speculated that this is because “best friends” were already very similar to each other. In high school, “friends” is a link that can reach across some mild differences, including academic achievement.
The results indicate that social groupings affect academic results more than previously thought:
… the positive correlation between the neighbors’ average GPA and the student’s academic progress indicates that social contagion of academic success may be taking place in the students’ social network, similar to those reported on obesity, emotions and other cognitive or behavioral traits.
And the presence of such an easily-measured variable — the academic achievement of a student’s friends — suggest to the researchers that it could be used as a cheap and early diagnosis:
… our research suggests the possibility of a quick test to predict a student’s academic progress that does not require large-scale surveys or complicated social network analysis. The key information used is a student’s self-reported friends’ average GPA relative to the student’s own GPA. This information can be easily collected from a single individual student. One could ask a student who are his/her friends in the class, and test if those self-reported friends’ average GPA is higher or lower than the student’s.