Our results indicate that more frequent and intense social interactions generally imply better score for students engaging in them. We find that these interactions are hosted within a “rich-club”, mediated by persistent interactions among high performing students, which is created during the first weeks of the course.
Hmm, that's a possible explanation.
But it feels pretty insufficient, given the evidence shows children from more affluent backgrounds perform better regardless of other factors when it comes to education. Is it possible they have the causation here backwards? I think sociology has a lot to say on why people tend to network with their own socio-economic class background, and the education factor could be separate from that.
I'm just speculating though, I've not read the paper to form any specific conclusions about whether they've sufficiently controlled for this factor already or not (work night, aint nobody got time fo that).
If it helps, it's about college students, who are, unfortunately, generally already selected for a certain socioeconomic class.
I also haven't read the study with enough depth to see whether they controlled for nontraditional college students, who have other obligations such as jobs, families, or other external factors that limit their opportunities for social interaction with classmates.