Student Networks and Achievement: Methodological Lessons from the Literature and a Proposal for a Networks’ Dynamic Analysis Process

Each research domain investigating the links between student networks and academic achievement found inconsistent results between studies. This paper discusses the main methodological issues (e.g., the identification of relevant peers, the confounding of the selection and socialization effects …) encountered when addressing the question of how student networks influence academic performance, and which might lead to inconsistent research outcomes. This discussion will be followed by the introduction of a network analysis process, which mitigates the identified methodological challenges by combining different methodologies, and which includes a social network analysis approach, together with a longitudinal design. The networks’ dynamic analysis process proposed is expected to lead to more consistent research outcomes, which might in turn lead to student grouping policies that facilitate academic achievement. Finally, this methodology might be generalized to predict any behaviors occurring in any network