The advent of configurable software has shifted the implementation of software solutions from total reliance on software developers to towards increased participation of end-users. End-users are now able to create software solutions without the need for writing code but through configuration and customisation. Despite the increasing use of configurable software challenges on designing the software platform architecture, process of testing and usability exists in configurable software. The research aimed at evaluating how available interface elements influence usability in DHIS2. Empirical data was collected by studying the design of custom data collection forms for routine health data collection with two groups of users. 80% and 90% were recorded as completion rates of the designed task and overall efficiency of 86.23% and 89.94% was achieved between the two groups. Lack of relevant editing features, increased distance between related objects, lack of conformity to Keep It Simple, Stupid (KISS) and minimalistic design principle were found to be the major challenges affecting the usability.
A Rule-Learning Approach for Detecting Faults in Highly Configurable Software Systems from Uniform Random Samples