At our local PedR meeting this week, Harriet Jones from UEA discussed Transition to university - getting the most from students. For me, the most stimulating part of this meeting was the discussion around Harriet's view that in the face of the difficult transition from secondary to higher education, our expectations of students are not high enough, and that students of all abilities respond best to challenge rather than dumbing down of the curriculum and over assessment.
In light of this inspiring discussion of a seemingly intractable problem, a paper just published in PLOS One is interesting (Spread of Academic Success in a High School Social Network. (2013) PLoS ONE 8(2): e55944. doi:10.1371/journal.pone.0055944). U.S. high school students whose friends’ average grade point average (GPA) is greater than their own have a tendency to increase their own GPA. Previous studies have shown that a student’s social network can influence obesity, emotional state and other cognitive traits and behavior. These authors found that students whose friends were performing better academically were more likely to improve their own scores over time. The opposite effect was also seen: when their friends GPAs were lower, a given student’s GPA was more likely to decrease as well. The authors also found that the strongest link between a student’s GPA change and that of their peers was likely to be with those they had ranked as friends, rather than best friends or acquaintances.
This suggests that streaming students by ability is a very bad idea, although just as friendship groups tend to form along social and ethnic lines, academic ability (and engagement) is also a determinant of social link formation in my experience. Educators therefore seemingly have a responsibility to mix things up, and where we have any influence, random allocation to groups seems to be about the best way we can achieve this.