Student's Use of Generative AI as a Support Tool in an Advanced Web Development Course

TL;DR

This study investigates how Generative AI is used as a learning, creativity, and productivity tool in an advanced web development course for experienced undergraduates. It combines quantitative surveys, reflections, logs, and assignments to analyze usage patterns, prompting practices, and perceptions over 19 weeks, within an unrestricted GenAI environment. Findings show broad GenAI utility across code generation, idea generation, and documentation, with reported gains in productivity and learning, yet persistent concerns about reliability and over-reliance, and a desire for formal prompting training. The work offers practical guidance for educators on designing GenAI-rich upper-level courses, including policy considerations, assessment alignment, and targeted prompting instruction to maximize benefits while mitigating risks.

Abstract

Various studies have studied the impact of Generative AI on Computing Education. However, they have focused on the implications for novice programmers. In this experience report, we analyze the use of GenAI as a support tool for learning, creativity, and productivity in a web development course for undergraduate students with extensive programming experience. We collected diverse data (assignments, reflections, logs, and a survey) and found that students used GenAI on different tasks (code generation, idea generation, etc.) with a reported increase in learning and productivity. However, they are concerned about over-reliance and incorrect solutions and want more training in prompting strategies.

Figures (2)

• Figure 1: Answers to question 'For this course, I regularly used GenAI when working on... (left) and to... (right)'.
• Figure 2: Answers to questions about the perceived effect of using GenAI on productivity (top), learning (middle), and concerns (bottom).