2026 ASEE Annual Conference & Exposition

Describing Data Structures through Culture: Developing Resources for Computing Students with Large Language Models

Presented at DSAI-Session 9: LLMs for Content Creation, Research Support, and Student Writing

Data structures are used to format, store, and access data and are a critical foundation of computing education and programming pedagogy. However, students unfamiliar with them may struggle to conceptualize these computational building blocks more abstractly. In recognition that the same descriptions may not resonate with all populations of students, we sought to explore how large language models (LLMs) can aid in contextual content creation of data structure explanations. Given that learners may come from a range of backgrounds, in this work, we focused on tailoring content to connect with a particular community, Hispanic or Latine students in the United States (U.S.). As a first step, we developed six prompts, each asking about six data structures, using them as input across six different LLMs. We applied discourse analysis to compare the (n = 216) outputs, utilizing established measures of culture to quantify patterns. The responses generated often centered on cuisine, followed by familial relationships, and then events. The majority of the outputs were in English, with some Spanish phrases applied, although the proportion of Spanish could increase in prompts specifying role play scenarios. Yet, given the broad framing, many of the examples followed suit. As such, the second phase applied the prompt identified as having the highest cultural relevance in the first phase to further scope to the population of a specific region, nationalities predominant in the southeastern region of the U.S. In particular, we explored possible explanations of data structures based on specifying 10 nationalities as part of the input across the six platforms. Additional analyses of the (n = 360) responses yielded further insight into more tailored examples to resonate with different populations with a high population density in this region. Again, culinary examples were the most prevalent; however, there was more specificity of dishes referenced. The outputs often provided step-by-step instructions, and these were frequently either entirely or mostly in English. In this paper, we share our approach, detailing the implications and potential ethical concerns. We also provide guidelines around how LLMs may aid in creating educational materials that could resonate with different groups.

Authors

Kevin Lemus University of Maryland Baltimore County [biography]

Kevin Lemus is a graduate assistant researcher and Master’s student at the University of Maryland, Baltimore County (UMBC). Kevin earned his Bachelor of Science in Information Systems along with a Certificate in User Experience, Web, and Mobile Development from UMBC in 2024. He is currently advancing his expertise as a candidate for a Master of Science in Human-Centered Computing, with an expected completion date of December 2026.
Stephanie Jill Lunn http://orcid.org/0000-0003-3840-1822 Florida International University [biography]

Stephanie Lunn is an Assistant Professor in the Multidisciplinary Engineering and Computing Education, Systems and Management (ESM) department and the STEM Transformation Institute at Florida International University (FIU). She also has a secondary appointment in the Knight Foundation School of Computing and Information Sciences (KFSCIS). Previously, Dr. Lunn earned her doctorate in computer science from the KFSCIS at FIU, with a focus on computing education. She also holds B.S. and M.S. degrees in computer science from FIU and B.S. and M.S. degrees in neuroscience from the University of Miami. In addition, she served as a postdoctoral fellow in the Wallace H. Coulter Department of Biomedical Engineering at the Georgia Institute of Technology, with a focus on engineering education. Her research interests span the fields of computing and engineering education, human-computer interaction, data science, and machine learning.
Edward Dillon University of Maryland Baltimore County [biography]

Edward Dillon is an Associate Professor in the Department of Information System at the University of Maryland, Baltimore County (UMBC). Edward received his B.A. in Computer and Informational Science from the University of Mississippi in 2007. He would go on to obtain his Masters and Ph.D. in Computer Science from the University of Alabama in 2009 and 2012, respectively. Prior to his arrival to UMBC in August 2024, Dr. Dillon served as a Computer Science Instructor at Jackson State University (2012-2013), and a Postdoctoral Researcher at Clemson University (2013-2014) and the University of Florida (2014-2016), and Assistant/Associate Professor in the Department of Computer Science at Morgan State University (2017-2024). His research focuses on human-centered computing with emphasis on computing education, broadening participation in computing, and tech workforce prep.
Dr. Krystal L Williams University of Wisconsin-Madison | Education Policy & Equity Research Collective (Ed_PERC) [biography]

Krystal L. Williams, Ph.D. is an Associate Professor of Higher Education in the University of Wisconsin Department of Educational Leadership and Policy Analysis. She directs the Education Policy and Equity Research Collective (Ed_PERC), including the Computing Sciences Education Research Team (cSERT) where she and her team explore issues regarding race and public policy with an emphasis on Historically Black Colleges and Universities, and broadening participation in STEM. She attended the University of Michigan where she completed her doctoral studies in the Center for the Study of Higher and Postsecondary Education, and Clark Atlanta University where she earned a BS and MS in mathematics, graduating as valedictorian.
Christian Ruiz

Download paper (450 KB)

Are you a researcher? Would you like to cite this paper? Visit the ASEE document repository at peer.asee.org for more tools and easy citations.

» Download paper

« View session