Evidence Based Approach to Discover Effective Non-Functional Testing Technique(s) in Typical Software Domains
Priyanka Mathur, Swati V. Chande · International Journal of Advanced Studies · 2017
An eminent quality software needs to be developed before it is to be brought in use, for this it has to be verified, validated and thoroughly tested. Software Testing has always intrigued the developers as software is prone to defects. To ensure reliability of software, testing needs to be carried out at all stages of development. A major problem faced by testers is that they should have a thorough knowledge of the area or domain to which the software belongs. Different types of software need to be tested using different test methods and more often a combination of these methods, depending upon the domain in which the software is developed. Glenford (2011) suggests that since the organizations spend 50% of their resources on testing so it is advisable that for testing to be error prone a thorough understanding of the system is required. Emin (1999) says that there are several testing techniques available that can be applied to specific domains of software however, the test requirements of all domains vary and so test techniques needs to be identified for typical software domains. Latest applications programming platforms and development tools create highly functional applications, but the quality component in these applications is lacking research interest. Quality component which is checked through Non-Functional Testing is overlooked or superficially tested since it’s considered as time consuming and sometimes novice testers are unable to determine the correct test technique says Eldh (2011). This paper is an attempt to identify, the most preferred and effective non-functional test [NFT] methods in three software domains namely web application, mobile application and e-commerce applications. Using an Evidence Based Approach, evidences were gathered for the commonly used non-functional testing techniques. The method used for gathering evidences is Systematic Literature Review (SLR), which is amongst the most preferred evidence based technique, suggested by Kitchenham (2007) and Keele (2007). The systematic review is based on a comprehensive set of articles obtained after a multi-stage selection process that have been published in the time span of 2010 to 2016. The results of the review show that in all the three domains a variety of NFT are used including Performance Testing, Load Testing, Stress Testing, Compatibility Testing, Usability Testing, Accessibility Testing and Security Testing. The gathered evidences reflected that all though all these testing techniques are used but Usability Testing is one of the most preferred NFT in all the three domains.