Examining Image-Based Button Labeling for Accessibility in Android Apps through Large-Scale Analysis

Abstract

of their abilities or use of assistive technologies. However, prior work has shown there still exist important accessibility barriers within apps [20,21,37,40,42,43]. Awareness of the need to create more accessible apps is increasing. Google and Apple are the primary organizations that facilitate mobile technology and the app marketplace, through the Android and iOS platforms. Both have released developer and design guidelines for accessibility [16,30], provide accessibility services as part of their platforms [2,15], have app development libraries that include built-in compatibility with assistive technologies, and have released accessibility testing scanners [17,28] and suites [10]. Some companies creating popular apps have also made statements and taken actions to create more accessible apps [19,45]. Such approaches have included creating internal guidelines [45], having specialized accessibility teams [19], actively prioritizing accessibility [19,45], and working with people with disabilities during app development and testing [45]. Despite these accessibility-focused efforts, studies of relatively small groups of apps have found they still include significant accessibility barriers [20,21,37,42,43]. This suggests a continuing need for accessibility improvements, however, the field lacks a detailed understanding of the state of mobile app accessibility at a large-scale, "population" level. Many design patterns in mobile apps are image and icon focused and use image-based buttons for main functionalities. One key component of accessibility for screen reader users is labeling image-based buttons. This need is parallel to the need for alt-text for images on the web. However, there is no large-scale understanding of the prevalence of unlabeled image-based buttons, how effective tools are at promoting labeling, nor the potential causes of failure to label. Ross et al. [42] present an epidemiology-based framework that suggests large-scale analyses can help answer some of these questions. The framework emphasizes that apps do not exist in isolation. It suggests that, in addition to testing individual apps, additional benefits and insights can be gained by exploring app accessibility at the population-level, situated within the richer ecosystem of influential factors. Such analyses can give unique insights into the state of app accessibility and opportunities for improvements. An epidemiologically-inspired analysis can also establish a baseline against which to measure the evolution of app accessibility over time.

Portuguese and Spanish DMOs’ Accessibility Apps and Websites

“…The review found no author or journal with extensive publication in the field of the accessible addressing app mobile and websites, despite the increasing number of articles published in high-impact journals within the tourism field that have taken an interest in the topic. Therefore, other studies on websites [58] and mobile applications in other sectors have been taken into account [59][60][61][62][63][64]. No similar studies have been found directly related to Spain and Portugal.…”

Screen Recognition: Creating Accessibility Metadata for Mobile Applications from Pixels

“…These technologies depend heavily on the availability of accessibility metadata provided by developers to expose the underlying semantics of apps [2,37]. [65][66][67], and our analysis in this paper (Section 3.3), show that developers routinely fail to include this information, and many mobile apps remain inaccessible.…”

Semantic Web Accessibility Testing via Hierarchical Visual Analysis

“…within the accessibility research community rather than software engineering. This research area involves studying certain categories of websites (e.g., airline websites [31], [32], education portals [33], other categories [34], [35], [36], [37]) or certain platforms (e.g., Android [38], [39], [40]), and then focusing on manually observing how non-sighted users would use those apps or websites in order to identify any patterns or trends in accessibility, with the purpose of publishing improved accessibility guidelines. Another line of work focuses on researching software development best practices and how do they impact the accessibility of the end product.…”