CS Education Research

Published at SIGCSE TS 2026 as "Bridging the Code Gap: Understanding Pedagogical Patterns in CS1 Programming Courses" with Naomi Chioma Udenze and Dr. Ihudiya Finda Ogbonnaya-Ogburu. → ACM Digital Library

The question

Every intro CS course teaches loops. But what are the loops about? Sorting arrays of integers, or scheduling a bus route, or modelling something a student might plausibly care about? We looked at the examples and exercises used in CS1 courses at top universities and asked what they're implicitly telling students computing is for.

Method

We built a classification framework of eight thematic categories through an open coding process, then applied computational text analysis (LIWC-22) and large language models to characterize the linguistic and contextual nature of the materials at scale.

What we found

An overwhelming prevalence of abstract, mathematical framing. Not wrong — but narrow, and plausibly a factor in who decides early that computing isn't for them. We ended with data-driven recommendations for incorporating more culturally relevant and real-world applications.

Why this sits next to my finance work

It looks like a different field. It isn't. Both are empirical: you form a hypothesis about a system you can't fully observe, design a measurement that could actually falsify it, and stay honest about what the data does and doesn't support. Open coding taught me more about constructing careful evaluations than any ML course did — including the habit that shows up all over the execution research, of building controls specifically to try to kill your own result.

This work came out of the REACH Lab at Virginia Tech, where I also mentored six undergraduate research assistants.