Substance Beats Style: Why Beginning Students Fail to Code with LLMs
This addresses a problem for programming education by identifying key barriers to using LLMs effectively for beginners, with incremental insights into prompt design.
The paper investigates why beginners struggle to prompt LLMs for code generation, finding that a lack of technical vocabulary is only correlated with failure, while the information content of prompts predicts success, and students often get stuck making trivial edits.
Although LLMs are increasing the productivity of professional programmers, existing work shows that beginners struggle to prompt LLMs to solve text-to-code tasks. Why is this the case? This paper explores two competing hypotheses about the cause of student-LLM miscommunication: (1) students simply lack the technical vocabulary needed to write good prompts, and (2) students do not understand the extent of information that LLMs need to solve code generation tasks. We study (1) with a causal intervention experiment on technical vocabulary and (2) by analyzing graphs that abstract how students edit prompts and the different failures that they encounter. We find that substance beats style: a poor grasp of technical vocabulary is merely correlated with prompt failure; that the information content of prompts predicts success; that students get stuck making trivial edits; and more. Our findings have implications for the use of LLMs in programming education, and for efforts to make computing more accessible with LLMs.