Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Towards User-Focused Cross-Domain Testing: Disentangling Accessibility, Usability, and Fairness

Matheus de Morais Leça, Ronnie de Souza Santos

TL;DR

The paper investigates how fairness testing relates to usability and accessibility in data-driven, AI-powered software. Using a tertiary study of 12 systematic reviews from 2015–2024, it synthesizes definitions, characteristics, and interplay among the three testing domains. It finds that boundaries are blurred, with strong links between fairness and both accessibility and usability, but notes a lack of established methods for routine implementation of fairness testing. The authors advocate a holistic, cross-domain testing approach and call for methodological work to integrate fairness into standard software development practices, aiming to ensure fair, inclusive, and ethically sound user experiences.

Abstract

Fairness testing is increasingly recognized as fundamental in software engineering, especially in the domain of data-driven systems powered by artificial intelligence. However, its practical integration into software development may pose challenges, given its overlapping boundaries with usability and accessibility testing. In this tertiary study, we explore these complexities using insights from 12 systematic reviews published in the past decade, shedding light on the nuanced interactions among fairness, usability, and accessibility testing and how they intersect within contemporary software development practices.

Towards User-Focused Cross-Domain Testing: Disentangling Accessibility, Usability, and Fairness

TL;DR

The paper investigates how fairness testing relates to usability and accessibility in data-driven, AI-powered software. Using a tertiary study of 12 systematic reviews from 2015–2024, it synthesizes definitions, characteristics, and interplay among the three testing domains. It finds that boundaries are blurred, with strong links between fairness and both accessibility and usability, but notes a lack of established methods for routine implementation of fairness testing. The authors advocate a holistic, cross-domain testing approach and call for methodological work to integrate fairness into standard software development practices, aiming to ensure fair, inclusive, and ethically sound user experiences.

Abstract

Fairness testing is increasingly recognized as fundamental in software engineering, especially in the domain of data-driven systems powered by artificial intelligence. However, its practical integration into software development may pose challenges, given its overlapping boundaries with usability and accessibility testing. In this tertiary study, we explore these complexities using insights from 12 systematic reviews published in the past decade, shedding light on the nuanced interactions among fairness, usability, and accessibility testing and how they intersect within contemporary software development practices.
Paper Structure (13 sections, 1 figure, 2 tables)

This paper contains 13 sections, 1 figure, 2 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Method
  4. Search Strategy
  5. Study Selection
  6. Data analysis
  7. Findings
  8. Usability, Accessibility and Fairness Definitions
  9. Usability, Accessibility, and Fairness Testing Interplay
  10. Discussions
  11. A Holistic Software Testing Perspective on Usability, Accessibility, and Fairness
  12. Limitations
  13. Conclusions

Figures (1)

  • Figure 1: Holistic Testing Perspective on Usability, Accessibility, and Fairness