2025 ASEE Annual Conference & Exposition

This abstract is for a full methods paper. Artificial intelligence (AI) has recently emerged as a powerful tool used to conduct sophisticated analyses on different types of data. In this study, we use generative AI through an embedding similarity technique in the context of graduate Engineering Education (EngE). As more students graduate with terminal degrees in EngE, there is an increasing demand for relevant academic job opportunities. While EngE as a field continues to prioritize both research and teaching, less is known about what skills EngE graduate programs prioritize to prepare students for their futures and how this preparation compares to the expectations of the academic job market.

The purpose of this study is to demonstrate the utility of sentence embeddings to illuminate similarities and differences between EngE degree-granting programs and the EngE academic job market. To address our purpose, we undertook an exploratory qualitative study. We identified 12 engineering education doctoral programs and 95 job postings. We used qualitative coding, sentence embeddings, and cosine similarity to establish that EngE doctoral program outcomes are predominantly research and teaching focused, academic job postings are teeming with doctoral entry-level positions, and program outcome and job posting alignment varies significantly by program. Our findings suggest that the embedding similarity technique is appropriate and feasible to use in EngE research.