Augmented Reality Simulations for Conceptual Chemistry Education Assessment

Thesis Abstract

Augmented reality (AR) has the potential to transform conceptual chemistry education by enabling learners to interact with molecular structures, reaction mechanisms, and thermodynamic processes in context-rich, manipulable visualizations. Despite advances in AR-enabled learning tools, there is limited empirical evidence on how AR simulations specifically influence conceptual understanding and assessment practices in chemistry education. This study aims to evaluate the effectiveness of AR simulations in enhancing students’ conceptual grasp of core chemistry topics and the accuracy and efficiency of assessment within AR-enabled environments. The specific objectives are (1) to determine the impact of AR simulations on conceptual understanding of atomic structure, bonding, and reaction energetics compared with traditional simulations and hands-on activities; (2) to examine the validity and reliability of AR-based diagnostic and formative assessments; (3) to investigate students’ cognitive load, motivation, and affective responses when engaging with AR simulations; (4) to analyze teachers’ perceived feasibility and instructional integration of AR assessment tools; and (5) to develop a conceptual model linking AR engagement, assessment validity, and learning outcomes grounded in constructivist and cognitive load theories. The study adopts a mixed-methods, quasi-experimental design conducted across three public secondary schools and two tertiary institutions in a metropolitan region over an academic year. A total of 420 students in high school chemistry and first-year university chemistry courses will be stratified and randomly assigned to either an AR-assisted instruction group (n = 140), an active control group using conventional simulations (n = 140), or a traditional instruction group (n = 140). The AR intervention comprises a suite of tablet-based AR simulations that visualize submicroscopic phenomena (e.g., orbital hybridization, electron-pair repulsion, reaction pathways) over real-world lab contexts, integrated with formative AR-based diagnostic assessments and real-time feedback dashboards. Data collection instruments include (i) standardized conceptual chemistry assessments administered as pre-, post-, and delayed post-tests; (ii) AR-specific assessment instruments assessing validity, reliability, and interpretive accuracy; (iii) cognitive load measurements using the NASA-TLX adapted for AR tasks; (iv) motivation and engagement scales (Intrinsic Motivation Inventory); (v) semi-structured interviews with students and teachers; and (vi) classroom observations using a validated rubrics protocol. Validity and reliability of instruments will be established through pilot testing, Cronbach’s alpha analysis, test-retest reliability, and factor analysis. Data analysis will proceed in multiple strands. Quantitative data will be analyzed using ANCOVA to compare post-test gains across groups while controlling for pre-test scores, and multilevel modeling to account for nesting within classrooms. Regression analysis will examine the relationship between AR engagement metrics (time-on-task, interaction frequency) and learning gains. Diagnostic and formative AR assessments will be evaluated for construct validity using item response theory and differential item functioning analyses. Qualitative data from interviews and observations will be analyzed using thematic analysis, triangulated with quantitative results to illuminate mechanisms of learning, cognitive load, and instructional feasibility. A structural equation model will be tested to evaluate the hypothesized pathways from AR engagement through cognitive load and affective factors to learning outcomes, informed by constructivist and cognitive load theories. Expected findings include (a) statistically significant gains in conceptual understanding for the AR group relative to both control groups, with larger effects for topics involving submicroscopic representations; (b) improved validity and reliability of AR-based assessments, demonstrated by higher item discrimination indices and consistent diagnostic feedback; (c) manageable cognitive load with high intrinsic motivation and engagement reported by students, mediated by intuitive AR interfaces and immediate feedback; (d) positive teacher perceptions regarding feasibility, integration into existing curricula, and perceived enhancements in formative assessment; and (e) a validated conceptual model linking AR engagement to learning outcomes through cognitive load and affective mediators. The study contributes to knowledge by providing rigorous empirical evidence on the efficacy of AR simulations for conceptual chemistry education and by outlining a validated framework for AR-based assessment design and implementation. Practical implications include guidelines for developers and educators on selecting AR features that optimize learning and assessment validity, as well as implications for policy and teacher professional development in technology-enhanced chemistry pedagogy. Recommendations emphasize scalable AR interventions, context-sensitive assessment design, and longitudinal studies to assess long-term retention and transfer of conceptual understanding.