Accessibility of ATMS for the functionally illiterate through icon-based interfaces

• Published in: Behaviour & information technology Vol. 25; no. 1; pp. 65 - 81
• Main Authors: Thatcher, A., Mahlangu, S., Zimmerman, C.
• Format: Journal Article
• Language: English
• Published: London Taylor & Francis Ltd 01.01.2006; Taylor & Francis
• Subjects: Asynchronous transfer mode; Computers; Exact sciences and technology; Human-computer interaction; Icons; Information and communication sciences; Information retrieval systems. Information and document management system; Information science. Documentation; Interfaces. Software; Networks; Sciences and techniques of general use; User interface; Graphics; Bank system; Illiteracy; Accessibility; User interface; Process improvement; Graphical interface; Man machine relation; Transaction processing; Self service; Usability
• ISSN: 0144-929X; 1362-3001
• DOI: 10.1080/01449290500102128

This study looks at the preliminary development and evaluation of an icon-based ATM interface for use with functionally illiterate bank account holders. In the first part of the study functionally illiterate bank account holders were closely involved in the development of an icon set for the entire ATM withdrawal transaction. This involved target subjects providing descriptive information on possible icons and then evaluating these descriptions using an icon identification test. Although only seven of the 15 icons met the ISO criterion of 67% correct identification, the 'best' icons for each instruction were evaluated using one of three prototype interfaces - icon-only, text-only and text-and-icon interface. Comparisons were made between these three interfaces, and a functionally illiterate group and a comparison literate group. Results suggested that within each group there were few significant differences based on the type of interface, although there were significant differences between the groups. Significant differences in the literate group were primarily due to prior exposure to similar ATM interfaces, whereas significant differences in the illiterate group were primarily due to icon comprehensibility.