Data Strategy and Institutional Research Manager, Institutional Research and Analysis Carnegie Mellon University Pittsburgh, Pennsylvania, United States
Session Abstract: This session tackles a common institutional research challenge: turning hundreds or even thousands of lines of open text data into clear, actionable insights. I will share a successful, repeatable method developed at Carnegie Mellon University's Tepper School of Business that makes qualitative analysis faster and more accessible. By leveraging generative AI tools, this approach allows IR professionals to collect and analyze more narrative data from surveys and focus groups than ever before, while maintaining analytical rigor and trustworthiness. Attendees will learn how to create a custom AI assistant, guide its analysis with a set of instructions, and generate reports tailored for various stakeholders, ultimately amplifying student voices to drive more effective decision-making. This includes a highly effective method for building a "truth directive" to add to qualitative analysis prompts. Together, these methods effectively eliminate hallucinations, maintaining stakeholder trust.
Keywords: Generative AI, Qualitative research, Data analytics, Prompt engineering
