Simulation‑mediated drone‑integrated STEM education : design, learning processes, and higher‑order thinking development /

Open in iSearch

Student Theses

2026

Yeung, Chung Yiu Richard

Thesis

While integrated STEM education is widely championed for its potential to foster higher-order thinking and authentic problem-solving, a persistent “doing–thinking” gap remains a challenge in many classrooms. This thesis addresses that disconnect by reframing integration as epistemic coordination — the process through which learners connect conceptual knowledge, procedural action, evidence-based judgement, computational and design reasoning, and metacognitive reflection to make and revise warranted decisions in authentic STEM problem spaces. The work demonstrates how this coordination can be intentionally designed and traced longitudinally to open the “black box” of the STEM learning process.Guided by Kolb’s Experiential Learning Theory, a Drone-Integrated Cross-Disciplinary STEM Curriculum was developed as a hybrid learning ecology. This environment blends physical drone enactments with web-based simulations, which were co-designed using Large Language Models to provide responsive feedback and rapid scenario variants. The curriculum is structured as a developmental continuum, moving students from multidisciplinary foundations toward sophisticated transdisciplinary engineering design. Following six stages of design-based research, the study culminated in a stabilized eight-lesson intervention implemented across 14 Hong Kong secondary schools with 227 students.A convergent mixed-methods design examined both educational outcomes and the underlying learning processes. Beyond assessing standard gains in STEM knowledge and computational thinking, the study utilized Epistemic Network Analysis to model the evolution of reasoning over time. By analyzing 4,776 segmented student reflections, the methodology successfully mapped longitudinal coordination trajectories and identified how different subgroups — including gender and age levels — navigated the epistemic demands of the curriculum.The results reveal more than just improved test scores; they demonstrate a systematic reorganization of student discourse. Over the eight lessons, reasoning shifted from basic operational or affective engagement toward a more evaluative, design-integrated coordination. While gender subgroups showed similar overall progress, their internal paths differed: males frequently foregrounded collaborative implementation, while females prioritized evaluation-regulated inquiry. Similarly, younger students remained closer to identity-based links, whereas older students moved toward consistent metacognitive regulation.Ultimately, this thesis contributes a framework for measuring integration as a tangible epistemic achievement. By offering scalable, AI-augmented design principles and simulation resources, the research provides a technological blueprint for the future of AI-enhanced epistemic analytics, allowing educators to foster and visualize the “epistemic ascent” of students within routine school constraints

LG51.H43 Dr 2026eb Yeungcyr

https://educoll.lib.eduhk.hk/records/k4sR53fY