Measuring affective development in AI-integrated classrooms: development and validation of a multidimensional scale
Year: 2026
Type: article
Abstract: As Artificial Intelligence (AI) transforms K-12 education, monitoring student emotional growth is imperative. However, existing affective instruments often measure transient states and overlook the unique interpersonal dynamics of intelligent environments. Departing from traditional Social Emotional Learning (SEL) tools that focus on human-only contexts, this study introduces the Student Affective Development Scale (SADS) to capture the longitudinal co-evolution of students and AI agents. Grounded in Regulation and Interaction theories, the SADS encompasses four dimensions: affective experience, regulation, interaction, and human-technology interaction. Two rounds of empirical testing with a total of 2,083 participants ( N 1 = 752; N 2 = 1, 331) demonstrated that the scale possesses robust psychometric properties. Confirmatory factor analysis (CFA) indicated a superior model fit (χ 2 / df = 2.45, CFI = 0.952, TLI = 0.941, RMSEA = 0.058), and internal consistency was high across all dimensions (Cronbach's α ranging from 0.813 to 0.918). This instrument provides a validated metric for evaluating human-AI collaborative teaching and offers a practical tool for fostering student emotional wellbeing in the digital age.
Source: Frontiers in Education
Authors: Xiangqiong Zhang, Meilan Peng, Xiaotong Li, Fuquan Huang
Institution: South China Normal University
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: gold
Estimated APC paid: $2,353
OpenAlex ID: W7135196824