AI Assisted Virtual Laboratories for Engineering without Borders: Quasi Experimental Evidence from Sustainable Agricultural Irrigation Education in Higher Education Contexts

Abstract

Unequal access to physical laboratories in engineering education continues to limit equity, experimental repeatability, and students’ engagement with real-world sustainability challenges, particularly in resource-constrained contexts. This study presents a quasi-experimental evaluation of AI assisted virtual laboratories with intelligent feedback as a pedagogical strategy to support Engineering without Borders principles in higher education. The intervention was implemented in a Physical Experimentation and Statistical Thinking course through an authentic challenge focused on designing a sustainable agricultural irrigation system using a Mariotte Bottle. A mixed-methods quasi-experimental design was conducted with 93 undergraduate engineering students, divided into a control group (n = 47) using traditional physical experimentation and an experimental group (n = 46) combining physical practice with AI assisted virtual laboratories. Data were collected using pre- and post-tests, validated Likert-type surveys (α > 0.85), standardized rubrics, and qualitative reflections. Statistical analyses included independent-samples t-tests, one-way ANOVA, correlation analysis, effect sizes, and power analysis (> 0.80). Results revealed significantly higher conceptual learning gains, motivation, technological acceptance, and problem-solving consistency in the experimental group. The normalized learning gain reached g = 0.67 compared to g = 0.34 in the control group (p < 0.05). Qualitative findings showed deeper integration of the Sustainable Development Goals (SDGs), particularly SDGs 2, 4, 6, 9, 12, and 13. The findings indicate that AI assisted virtual laboratories can complement physical experimentation, offering an ethically grounded instructional approach with potential for application in similar higher education contexts.