Tracking Test First Pair Programming — An Experiment

Abstract

The authors ran an experiment where a group of professional programmers working in pairs and a control group programming alone implemented a small system from predefined requirements. Most programmers spent between 50% and 60% of time on testing; only the most inexperienced spent less. Programmers reported more problems with refactoring than testing. The rhythm of switching the driver and navigator role is essential for test-first pair programming. The experiment showed that partners switched roles 21 times per day on average. The comparison of the control group of individuals and the group programming in pairs showed that both groups spent almost the same amount of time to complete the tasks. The result of this comparison is by applying a t-test not statistically significant. We believe that more detailed research apart of evaluating test-first programming is needed to compare solo vs. pair programming in the investigated group.