Implementation of Shiny-R in active learning strategies for teaching statistics in engineering education

Briones-Núñez F.I.; Vasquez J.; Sarango J.; Garcia V.; Rodas F.; Aguirre F. · Frontiers in Education (2026) · DOI: 10.3389/feduc.2026.1734627

Resumen

The digitalization of engineering education demands pedagogical strategies that go beyond passive instruction. While R is a standard tool for data science, its steep learning curve often creates a “syntax barrier” that hinders conceptual understanding, leading to cognitive overload for novice students. This study evaluates ShinyStats-ESPE, an interactive web-based environment designed to bridge the gap between statistical theory and computational practice. Grounded in Cognitive Load Theory, our approach isolates conceptual exploration from coding mechanics using a “Predict-Observe-Explain” pedagogical cycle. We employed a quasi-experimental design ((Formula presented)) to compare learning outcomes between a control group ((Formula presented)) using traditional script-based methods and an experimental group ((Formula presented)) utilizing our interactive modules. The results, analyzed via Welch’s t-test, indicate a significant improvement in the experimental group ((Formula presented), (Formula presented)), with pass rates doubling from 37% to 76%. These findings suggest that temporarily bypassing syntax through interactive GUIs allows students to scaffold deeper statistical reasoning, offering a scalable model for engineering curricula.

Instituciones: Universidad Nacional de Tumbes