“It’s Weird That it Knows What I Want”: Usability and Interactions with Copilot for Novice Programmers
Code (set theory)
DOI:
10.1145/3617367
Publication Date:
2023-08-23T10:51:54Z
AUTHORS (9)
ABSTRACT
Recent developments in deep learning have resulted code-generation models that produce source code from natural language and code-based prompts with high accuracy. This is likely to profound effects the classroom, where novices can now use free tools automatically suggest solutions programming exercises assignments. However, little currently known about how interact these practice. We present first study observes students at introductory level using one such auto-generating tool, Github Copilot, on a typical (CS1) assignment. Through observations interviews we explore student perceptions of benefits pitfalls this technology for learning, new observed interaction patterns, discuss cognitive metacognitive difficulties faced by students. consider design implications findings, specifically terms like Copilot better support scaffold novice experience.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (109)
CITATIONS (81)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....