Social Learning Frameworks for Analyzing Collaboration with Marginalized Learners

Collaborative learning has potential to serve as a platform for fostering social connection, particularly in non-traditional contexts. Recent years have seen an increase in (long overdue) interest in supporting communities within such contexts through research. This has been often approached through ethnographic methods such as naturalistic observations, which are suitable for smaller, marginalized populations, However, the analysis of data produced by such methods often lack standardization, which limits generalizability of results and makes comparison across populations and learning contexts challenging. In this paper, we argue how greater grounding of data analysis in collaborative learning theories can provide standards for more meaningful comparison across contexts. We review Vygostky's social learning theories, shared social regulation of learning, and the trialogical approach. We discuss how anchoring inductive and deductive approaches in social frameworks may yield standardization metrics for unstructured, qualitative data from studies of social learning. We base this in our ongoing research on collaborative language learning between immigrant grandparents and grandchildren.