In order to find out whether reviewers have an impact on review delay and outcome, for each reviewer we calculated the
number of previously reviewed patches and then discretized them according to their reviewing efforts using quartiles.
Applying statistical tests we determined that the difference for response time for A and B groups of reviewers (i.e., the less active ones) is statistically significant when compared to C or D groups (i.e., the more active ones). Since the distribution of delays is very skewed, we report both the median and mean values for reviewers’ timeliness (see Table IV). The results show that the choice of reviewers plays an important role on reviewing time. More active reviewers provide faster responses (with median being 57 minutes and mean being 496 minutes) compared to the individuals who performed fewer code review (the median for time is 84 minutes and 621 minutes for the mean).