Measuring Inclusion in Interaction: Inclusion Analytics for Human-AI Collaborative Learning

TL;DR

This paper introduces inclusion analytics, a discourse-based framework that dynamically measures inclusion in human-AI collaborative learning by analyzing interaction patterns, participation, and relational dynamics.

Contribution

It presents a novel, scalable method for real-time, interaction-level assessment of inclusion in collaborative learning environments, focusing on participation, affect, and epistemic equity.

Findings

Inclusion analytics reveals participation disparities and relational dynamics.

The framework can be applied to both simulated and real human-AI interactions.

It offers a process-oriented approach to measuring inclusion beyond traditional methods.

Abstract

Inclusion, equity, and access are widely valued in AI and education, yet are often assessed through coarse sample descriptors or post-hoc self-reports that miss how inclusion is shaped moment by moment in collaborative problem solving (CPS). In this proof-of-concept paper, we introduce inclusion analytics, a discourse-based framework for examining inclusion as a dynamic, interactional process in CPS. We conceptualize inclusion along three complementary dimensions -- participation equity, affective climate, and epistemic equity -- and demonstrate how these constructs can be made analytically visible using scalable, interaction-level measures. Using both simulated conversations and empirical data from human-AI teaming experiments, we illustrate how inclusion analytics can surface patterns of participation, relational dynamics, and idea uptake that remain invisible to aggregate or post-hoc evaluations. This work represents an initial step toward process-oriented approaches to measuring inclusion in human-AI collaborative learning environments.