ABSTRACT: People’s thought processes have a significant impact on software quality, as software is designed, developed and tested by people. Cognitive biases, which are defined as deviations of human mind from the laws of logic and mathematics, are likely to cause software defects. However, there is little empirical evidence to date to substantiate this assertion. In this research, we focus on a specific cognitive bias type called confirmation bias. Confirmation bias is believed to be one of the factors that lead to increased software defect density. Due to confirmation bias, developers might perform unit tests to make their program work. This results in the propagation of more defects to testing phase and hence probably an increase in software defect density. In this research, we present a metric scheme to explore the impact of developers’ confirmation bias on software defect density. In order to estimate effectiveness of our metric scheme in quantification of confirmation bias within the context of software development, we perform an empirical study which addresses prediction of defective parts of software. In our empirical study, we applied confirmation bias metrics to five datasets obtained from two industrial partners which are from Telecomunications and Enterprize Resource Planning (ERP) domains respectively. Our results provide empirical evidence that people’s thought processes and cognitive aspects deserve further investigation to find out empirical evidence about their effectiveness in software defect prediction as well as their relation to software quality.
Recommended citation: Gül Çalikli, Ayşe Bener. (2013). "Influence of Confirmation Biases of Developers on Software Quality: an Empirical Study. " Software Quality Journal. 21(2): 377-416 (2013).