EUVIMEDDescoperă EUVIMED
Crossref

Risk or Chance? Large Language Models and Reproducibility in HCI Research

Thomas Kosch, Sebastian Feger

Interactions · 2024-11 · 31 (6) · 44-49

DOI: 10.1145/3695765 ↗

At a glance

Journal article in Interactions, published online on 2024-10-29. Written by Thomas Kosch and 1 co-authors. Cited 21 times so far (Crossref); 9 references. Funded by: German Research Foundation.

RezumatDate bibliograficeCiteRelated studies

Abstract (original language)

Reproducibility is a major concern across scientific fields. Human-Computer Interaction (HCI), in particular, is subject to diverse reproducibility challenges due to the wide range of research methodologies employed. In this article, we explore how the increasing adoption of Large Language Models (LLMs) across all user experience (UX) design and research activities impacts reproducibility in HCI. In particular, we review upcoming reproducibility challenges through the lenses of analogies from past to future (mis)practices like p-hacking and prompt-hacking, general bias, support in data analysis, documentation and education requirements, and possible pressure on the community. We discuss the risks and chances for each of these lenses with the expectation that a more comprehensive discussion will help shape best practices and contribute to valid and reproducible practices around using LLMs in HCI research.

Funding

  • German Research Foundation

Cite

APA

Kosch, T., & Feger, S. (2024). Risk or Chance? Large Language Models and Reproducibility in HCI Research. Interactions, 31(6), 44-49. https://doi.org/10.1145/3695765

Vancouver

Kosch T, Feger S. Risk or Chance? Large Language Models and Reproducibility in HCI Research. Interactions. 2024;31(6):44-49. doi:10.1145/3695765

BibTeX
@article{Kosch2024,
  author = {Kosch, Thomas and Feger, Sebastian},
  title = {Risk or Chance? Large Language Models and Reproducibility in HCI Research},
  journal = {Interactions},
  year = {2024},
  volume = {31},
  number = {6},
  pages = {44--49},
  doi = {10.1145/3695765}
}
Start for free →