The contemporary classroom is shaped by economic efficiency and is frequently exposed to technological interventions from researchers and emerging innovations. It falls short of the potential it bears as a collective learning environment for learners and teachers. Unsurprisingly, classrooms are not perceived as malleable beyond the superficial decoration of their physical form by their users. The trope of the future postdigital classroom offers itself as a conceptual construct to re-imagine under which circumstances we want to learn and teach together in the future. This paper, situated in the domain of postdigital educational research, proposes a guide to conduct a variant of future workshops to generate local collective visions of the future postdigital classroom grounded in the current challenges faced by educational stakeholders. In this sense, the proposed variant of the future workshop aims to empower the participants in shaping their future learning environment, fostering agency. The guide is designed to be continuously imagined and re-imagined by those who use it, acknowledging the widely varying context of educational stakeholders. The authors acknowledge its variable form through space and time as a postdigital artifact of postdigital research, of which this written representation can be but a snapshot; nevertheless, this is a first necessary step to the active construction of new imaginaries in the future postdigital classroom.
Learning analytics researchers often analyze qualitative student data such as coded annotations or interview transcripts to understand learning processes. With the rise of generative AI, fully automated and human–AI workflows have emerged as promising methods for analysis. However, methodological standards to guide such workflows remain limited. In this study, we propose that reasoning traces generated by large language model (LLM) agents, especially within multi-agent systems, constitute a novel and rich form of process data to enhance interpretive practices in qualitative coding. We apply cosine similarity to LLM reasoning traces to systematically detect, quantify, and interpret disagreements among agents, reframing disagreement as a meaningful analytic signal. Analyzing nearly 10,000 instances of agent pairs coding human tutoring dialog segments, we show that LLM agents’ semantic reasoning similarity robustly differentiates consensus from disagreement and correlates with human coding reliability. Qualitative analysis guided by this metric reveals nuanced instructional sub-functions within codes and opportunities for conceptual codebook refinement. By integrating quantitative similarity metrics with qualitative review, our method bears potential to improve and accelerate the process of establishing inter-rater reliability during coding by surfacing interpretive ambiguity, especially when LLMs collaborate with humans. We discuss how reasoning-trace disagreements represent a valuable new class of analytic signals advancing methodological rigor and interpretive depth in educational research.
A large variety of tools for collaborative learning has been developed in the last 20 years. While the impact of these tools on collaboration is acknowledged by the research community, a fine-grained understanding of links between the various functionalities offered by these tools and collaborative processes does not exist yet. In this paper, a meta-analysis was carried out on 49 collaboration studies to examine the relationship between individual functionalities on collaborative processes.