AI-Assisted Analysis of Learning Outcomes from a Faculty Development Program on Green Technologies

Abstract

This study evaluates a faculty development program (FDP) on green technologies using a mixed-methods approach combining traditional statistical analysis with basic artificial intelligence tools for text analysis. Main purpose of this paper is to assess knowledge improvement in sustainability domains and efficiently analyze qualitative feedback from engineering faculty participants. A total of approximately 200 faculty members completed Pre-Faculty Development Program (FDP) and Post-FDP surveys across multiple institutions all over India. Quantitative data included nine knowledge items on sustainability topics measured using 5-point Likert scales. Qualitative responses covering learning experiences, implementation intentions, and program suggestions were analyzed using sentiment analysis (TextBlob in Python) and keyword frequency methods. Paired t-tests were employed for statistical comparison. Statistically significant knowledge gains were observed across all measured domains (p < 0.001), with average improvements ranging from 0.4 to 0.6 points. Sentiment analysis revealed predominantly positive or neutral feedback (approximately 72%), indicating strong program reception. Keyword analysis identified recurring themes including "curriculum integration," "renewable energy applications," "institutional audits," and "practical implementation strategies." These patterns aligned with program objectives and validated quantitative findings. Simple AI-assisted text analysis tools can efficiently complement traditional evaluation methods, enabling faster processing of large volumes of qualitative feedback while maintaining interpretive validity. This approach offers educational institutions a practical, resource-efficient framework for program evaluation without requiring specialized data science expertise or expensive software infrastructure.