On the Relation between Code Elements and Accessibility Issues in Android Apps

  • Henrique Neves da Silva UFPR
  • Andre Takeshi Endo UTFPR
  • Marcelo Medeiros Eler USP
  • Silvia Regina Vergilio UFPR
  • Vinicius H. S. Durelli UFSJ

Resumo


Mobile apps have gone mainstream and become part of our daily lives. Currently, many efforts have been made to make apps more accessible to people with disabilities. However, little is still known on how to implement more accessible apps. In the Android API, there are (code) elements that may be employed to (in)directly improve the app's accessibility. This paper aims to investigate the prevalence of accessibility code elements and their relation to potential accessibility issues. First, we identified code elements of the native Android API that may be related to accessibility features, and mapped them to principles and success criteria of the Web Content Accessibility Guidelines (WCAG) 2.1. Using a sample of 111 open source mobile apps available in Google Play, we conducted a characterization study to examine the prevalence of accessibility code elements. We also analyzed how these code elements are related to issues detected by the static analyzer Android Lint and the accessibility testing tool MATE. Our results indicate that code elements are not widely used; the ones directly related to accessibility are present in only a few apps. Additionally, our results would seem to suggest that apps that adopt accessibility code elements, tend to have less accessibility issues. By analyzing our results from the standpoint of the WCAG principles, we conclude that there is room for improvement in terms of how both the Android API and automated testing tools deal with accessibility-related issues.
Palavras-chave: Mobile apps, Accessibility
Publicado
19/10/2020
SILVA, Henrique Neves da; ENDO, Andre Takeshi; ELER, Marcelo Medeiros; VERGILIO, Silvia Regina; DURELLI, Vinicius H. S.. On the Relation between Code Elements and Accessibility Issues in Android Apps. In: SIMPÓSIO BRASILEIRO DE TESTES DE SOFTWARE SISTEMÁTICO E AUTOMATIZADO (SAST), 5. , 2020, Natal/RN. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2020 . p. 40–49.