1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	New metric for analysis and productivity measurements of pair and solo programming
Authors:	Marwan N. Moftah Solla (P50140)
Supervisor:	oosajee Patel, Prof. Dr.
Keywords:	Software engineering Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia
Issue Date:	28-Mar-2011
Description:	Software engineering has become a scientific discipline that is concerned with methods, principles, approaches, measurements and tools for development of all kinds of software systems, from the early stages of problem definition, requirements capture, system specification through to testing and maintaining the system throughout the life of its use. Currently various methods such as agile methods has gained popularity within the industry and academic research, and pair programming is one of the key software development practices of extreme programming which is one of the most popular agile methods. Measuring productivity of pair programming is a challenging exercise since it has unbounded rules and complicated set of parameters. Using existing solo programming metrics for measuring performance of pair programming are inadequate. They are not to be suitable because of the advances in software engineering practices and new ways of designing and developing programs and systems. The goal of this research is to generate new software metric consisting a set of formulae which can be used to measure performance productivity of pair programming as well as solo programming in a much more objective and comprehensive manner and with the proviso to provide equitable comparison and determine the effectiveness of pair programming over solo programming. To prove the effectiveness and reliability of the metric in a real working environment to achieve this goal, an controlled experiment was performed using first year Computer Science students at Faculty of Information Science and Technology and (FTSM), Universiti Kebangsaan Malaysia , the National University of Malaysia. Two classes contain 30 students were participated in this experiment and it was run twice for the two classes during two weeks. Each week they had to solve a programming task in the computer laboratory supervised by the course's teaching assistants. After each experiment run a questionnaire had to be completed by the students. The result of the study indicates that pair programming was faster than solo programming in solving problems. The majority of pair programmers managed to implement nearly all the requirements of the task, and all solo programmers were slower and didn't manage to complete the task. A significant result of some new importance shows that students who have never tried pair programming before actually enjoy pair programming and they would like to practice it again. A whopping 87% agreed to this fact. They felt more confident with their solutions in the parts of the development that they worked as pairs and felt more satisfied with the outcome. The results illustrate that new developers are more productive when they work as pair programming than solo programming. Through these experiments the new metric is proved to be more reliable and effective in providing productivity measurements. The new metric is an original contribution from this research., Certification of Master's/Doctoral Thesis” is not available,Master Computer Science
Pages:	126
Call Number:	QA76.758.S6665 2011 tesis
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.