Pair Programming's Impact on Effort: A Comparative Discussion

Table of Links Abstract and 1. Introduction Abstract and 1. Introduction 2. Experiment Definition 2. Experiment Definition 3. Experiment Design and Conduct 3. Experiment Design and Conduct 3.1 Latin Square Designs 3.1 Latin Square Designs 3.2 Subjects, Tasks and Objects 3.2 Subjects, Tasks and Objects 3.3 Conduct 3.3 Conduct 3.4 Measures 3.4 Measures 4. Data Analysis 4. Data Analysis 4.1 Model Assumptions 4.1 Model Assumptions 4.2 Analysis of Variance (ANOVA) 4.2 Analysis of Variance (ANOVA) 4.3 Treatment Comparisons 4.3 Treatment Comparisons 4.4 Effect Size and Power Analysis 4.4 Effect Size and Power Analysis 5. Experiment Limitations and 5.1 Threats to the Conclusion Validity 5. Experiment Limitations and 5.1 Threats to the Conclusion Validity 5.2 Threats to Internal Validity 5.2 Threats to Internal Validity 5.3 Threats to Construct Validity 5.3 Threats to Construct Validity 5.4 Threats to External Validity 5.4 Threats to External Validity 6. Discussion and 6.1 Duration 6. Discussion and 6.1 Duration 6.2 Effort 6.2 Effort 7. Conclusions and Further Work, and References 7. Conclusions and Further Work, and References 6. Discussion In this section we discuss some results of other experiments, and we contrast them with our results regarding duration and effort. 6.1 Duration The experiment run by Nosek [24] employed 15 practitioners grouped in 5 pairs and 5 solos. Subjects wrote a database script. Results show a decrease of 29% in time duration in favor of pair programming. Williams et al. [28] used 41 students grouped in 14 pairs and 13 solos. During the experiment, subjects completed four assignments. Authors reported that pairs completed the assignments 40 to 50 percentage faster. Nawrocki and Wojciechowski [23] employed 16 student subjects (5 pairs and 6 solos). Subjects wrote four programs. Authors did not find differences between pairs and solos. Lui and Chan [19] used 15 practitioners grouped in 5 pairs and 5 solos. Authors reported 52% decrease in time in favor of pair programming. Müller [22] used 38 students (14 pairs and 13 solos). Students worked on four programming assignments where tasks were decomposed into implementation, quality assurance and the whole task. Author reported that pairs spent 7% more time working on the whole task, however this difference is not significant. Arisholm et al. [1] used 295 practitioners grouped in 98 pairs and 99 solos. Subjects performed several change tasks on two alternative systems with different degrees of complexity. Authors reported 8% decrease in favor of pairs. In contrast, the results reported in this paper infer a significant (at a=0.1) 28% decrease in time (in favor of pairs) and an effect size d=0.65. With respect to duration, our results reinforce those reported in [24]. Authors: (1) Omar S. Gómez, full time professor of Software Engineering at Mathematics Faculty of the Autonomous University of Yucatan (UADY); (2) José L. Batún, full time professor of Statistics at Mathematics Faculty of the Autonomous University of Yucatan (UADY); (3) Raúl A. Aguilar, Faculty of Mathematics, Autonomous University of Yucatan Merida, Yucatan 97119, Mexico. Authors: Authors: (1) Omar S. Gómez, full time professor of Software Engineering at Mathematics Faculty of the Autonomous University of Yucatan (UADY); (2) José L. Batún, full time professor of Statistics at Mathematics Faculty of the Autonomous University of Yucatan (UADY); (3) Raúl A. Aguilar, Faculty of Mathematics, Autonomous University of Yucatan Merida, Yucatan 97119, Mexico. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. available on arxiv available on arxiv