How Far Are We? The Triumphs and Trials of Generative AI in Learning Software Engineering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Choudhuri, Rudrajit, Liu, Dylan, Steinmacher, Igor, Gerosa, Marco, Sarma, Anita
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909128434122752
author Choudhuri, Rudrajit
Liu, Dylan
Steinmacher, Igor
Gerosa, Marco
Sarma, Anita
author_facet Choudhuri, Rudrajit
Liu, Dylan
Steinmacher, Igor
Gerosa, Marco
Sarma, Anita
contents Conversational Generative AI (convo-genAI) is revolutionizing Software Engineering (SE) as engineers and academics embrace this technology in their work. However, there is a gap in understanding the current potential and pitfalls of this technology, specifically in supporting students in SE tasks. In this work, we evaluate through a between-subjects study (N=22) the effectiveness of ChatGPT, a convo-genAI platform, in assisting students in SE tasks. Our study did not find statistical differences in participants' productivity or self-efficacy when using ChatGPT as compared to traditional resources, but we found significantly increased frustration levels. Our study also revealed 5 distinct faults arising from violations of Human-AI interaction guidelines, which led to 7 different (negative) consequences on participants.
format Preprint
id arxiv_https___arxiv_org_abs_2312_11719
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle How Far Are We? The Triumphs and Trials of Generative AI in Learning Software Engineering
Choudhuri, Rudrajit
Liu, Dylan
Steinmacher, Igor
Gerosa, Marco
Sarma, Anita
Software Engineering
Human-Computer Interaction
Conversational Generative AI (convo-genAI) is revolutionizing Software Engineering (SE) as engineers and academics embrace this technology in their work. However, there is a gap in understanding the current potential and pitfalls of this technology, specifically in supporting students in SE tasks. In this work, we evaluate through a between-subjects study (N=22) the effectiveness of ChatGPT, a convo-genAI platform, in assisting students in SE tasks. Our study did not find statistical differences in participants' productivity or self-efficacy when using ChatGPT as compared to traditional resources, but we found significantly increased frustration levels. Our study also revealed 5 distinct faults arising from violations of Human-AI interaction guidelines, which led to 7 different (negative) consequences on participants.
title How Far Are We? The Triumphs and Trials of Generative AI in Learning Software Engineering
topic Software Engineering
Human-Computer Interaction
url https://arxiv.org/abs/2312.11719