From Interviews to Equations: A Multi-Phase System Dynamics Model of Engineering Student Engagement
Abstract
This study presents a systematic approach for converting qualitative data into quantitative parameters within a system dynamics (SD) framework, focusing on modeling engineering student engagement. Although SD typically relies on numerical inputs, important "soft" factors such as motivation, confidence, and a sense of belonging have often been neglected due to the challenge of measurement. Semi-structured interviews were conducted with mechanical engineering students in a Learning Studio environment, capturing stories about hands-on coursework, peer support, and personal growth. Using inductive thematic analysis, frequent mentions of relevant factors were coded and converted into weighted parameters for a Vensim model. The resulting structure includes interconnected submodels illustrating how community cohesion influences motivation, which then affects learning outcomes and career goals. Simulation results show exponential growth in motivation, confidence, and sense of belonging over a sixteen-week period, alongside declines in negative factors like dissatisfaction. Introducing a logistic limit on belonging confirmed that, after social needs plateau, other positive aspects continue to improve. A scenario with a one-week delay in feedback loops reduced the rate of change but maintained the model's overall behavior. These findings align with educational theory, suggesting that community-driven interventions can enhance student engagement. This approach highlights the importance of capturing intangible, evolving student experiences in SD models. While additional validation with larger samples is necessary, the framework shows how incorporating qualitative insights into simulations can yield more actionable findings for educators and researchers.