Towards Designing Mobile Banking User Interfaces for Novice Users

Abstract

A lot has gone into research aimed at establishing design guidelines to guide developers in developing mobile applications usable by low literate and novice users due to their technological inclination. However, despite quite a number of valuable usability constraints unveiled, corresponding recommendations made and a list of synthesized design guidelines established across various research works, those guidelines are still incomplete and not standardized. In a similar context, the World Wide Web Consortium (W3C), an international standards organization for the World Wide Web, has addressed a similar problem for web based applications. They developed a set of standard guidelines called Web Content Accessibility Guidelines (WCAG 2.0) that specifies how web content can be made more accessible to all types of user on the World Wide Web. The existence of such standard for mobile phones application would greatly impact mobile application development for such users. However none of such standards as WCAG exists for mobile applications development. It has been established that more effort is needed towards uncovering more low literate and novice user centered usability constrains with corresponding recommendations. Thus there exists the need in this research area towards achieving a set of standard guidelines for the development of mobile applications. This study aim to achieve four (4) objectives: i. to explore for low literate or novice user centered usability constraints in mobile applications and their associated design recommendations from existing literatures. ii. to build a mobile banking prototype based on design recommendations from existing literatures. iii. to perform empirical test on some selected banking applications against the developed prototype. iv. to evaluate (comparative analysis) of objective (iii.) aimed at exposing more novice user centered usability constraints using System Usability Scale (SUS) tool. Levels of significance were tested via a two sampled t-test for mean.