Measuring Software Test Execution in Agile Development

Abstract

Software Testing is the process of executing a program with the intent of finding defects. While it is almost impossible to release software which is completely bug free, the objective of software testing is to achieve high quality within reasonable costs. It is therefore important to explore methods and techniques which can help us achieve better quality by optimizing test techniques while keeping costs under control. One of the challenges in real world software testing projects has been to accurately quantify and track progress of software testing projects. Effective software coverage measurement can ensure better execution of projects with controlled costs and accurate forecasting. This paper evaluates various approaches proposed in current work to measure testability and test coverage. Conventional approaches have included code coverage or test case count as methodologies for test coverage measurement. However, these techniques when implemented in real world projects have shown some shortcomings – either they do not account for scenarios of test coverage which were missed in coding/development or they discount size of a test event considering all events of equal size thereby resulting in skewed test coverage data. In addition, modern agile software development techniques demand ways that are easily adaptable and require minimal documentation or processing to track this data. This therefore leads to the need to evaluate better ways for test coverage measurement – which are more accurate and less error prone. This paper makes an attempt to address this problem by proposing a methodology which can be implemented in most software projects for effective software test coverage measurement and project planning.