# Reconceptualizing learning engagement: evidence for a context-sensitive structure in STEM education

## TL;DR

This study shows that student engagement in STEM education is better understood when linked to specific learning activities rather than general categories like behavior or cognition.

## Contribution

The paper introduces a novel, context-sensitive model of engagement based on specific academic activities in STEM education.

## Key findings

- A context-based model of engagement fits student data better than traditional behavioral/cognitive models.
- The model generalizes across different STEM disciplines and educational settings.
- This approach challenges existing theories and suggests new ways to design educational interventions.

## Abstract

Understanding student engagement as a psychological construct remains a persistent challenge in educational psychology, particularly in higher education STEM contexts. Traditional models distinguish engagement into behavioral, cognitive, and affective dimensions, yet often overlook how the structure of engagement may be shaped by contextualized learning activities. This study introduces and tests a novel, activity space-based model of engagement, hypothesizing that behavioral and cognitive engagement are organized by the specific academic environments in which they occur (e.g., lectures, exams, projects, recitations). Applying new engagement survey instruments that were iteratively developed to be contextually meaningful, we first present an exploratory factor analysis applied to 1,176 students from two different courses and institutions. Then we present a confirmatory factor analysis applied to 772 students in a third course. We find that a model organized by activity contexts—rather than by behavioral and cognitive distinctions—better fits the data and generalizes across STEM disciplines. The findings challenge conventional engagement theory and support a reconceptualization of engagement as a partially context-sensitive construct. This theoretical shift has implications for psychological models of learning and for the design of more precise, equitable interventions that address varied patterns of engagement within and across STEM domains.