Empirical Analysis of Pair Programming Using Bloom's Taxonomy and Programmer Rankers Algorithm to Improve the Software Metrics in Agile Development

Abstract

Collaborative programming is a co-operative effort of 2 teams to n-teams to share knowledge, synergize and produce better code. Pairing, Swarming and Mobbing are the standard of agile technology which are adapted by many organizations. In this paper, software development is carried out using Pair Programming(PP) in a medium sized organization developing mobile applications using android is presented. The first method uses the Programmers Competency Matrix (PCM) based on Blooms Taxonomy to assess the skill of the programmers. Since, the pairs are chosen randomly in the PCM method, a novel algorithm is proposed to pair the programmers by Programmer Ranking Algorithm (PRA). The two proposed methods are evaluated in an organization and the results are validated. The results prove that PP definitely improves the software development process than when it is developed by individual programmers. The PRA methodology outperforms the PCM because the PRA chooses the pair wisely using the programming skills of the programmer.