Evaluating Machine Learning Techniques for Qualitative Institutional Data

Session Abstract: This research presents useful approaches to enhance data-driven decision-making in higher education, particularly by leveraging advanced machine learning (ML) techniques to analyze large and complex structured and unstructured qualitative datasets from higher education climate surveys. Our study applies various ML techniques to process these data more efficiently and effectively than traditional methods. We will evaluate the performance of each technique using established ML metrics (accuracy, precision, recall, F1-score) to determine the relevance and applicability of each method in institutional research. By comparing the outcomes of these techniques, this research aims to demonstrate how ML can streamline the analysis of qualitative data, providing higher education professionals with actionable insights to inform decision-making, improve institutional practices, and enhance campus climate.

Keywords: machine learning, qualitative data analysis, survey methodology, faculty