A study of code inspection performance and personality traits
Since their introduction, software inspections have proven to be an effective and cost efficient means of identifying defects and improving software quality. However, performance can vary significantly between inspectors. The influence of personal characteristics – such as personality – on inspection performance is not well understood. This thesis used regression analysis to investigate whether or not the Big Five personality traits could be used as predictors of software inspection performance. Undergraduate students completed a personality inventory measuring the Big Five personality traits, as well as a code inspection task, using an online software inspection tool. The personality trait scores were used as predictors in a regression analysis of personality and inspection performance. Results showed that three personality traits – conscientiousness, agreeableness, and extraversion – were statistically significant predictors of inspection performance. The strength of association between inspection performance and each of these traits was modest, indicating that they are not sufficient to predict performance.