A random survey of 205 drivers measured each driver's "anger" score. Higher scores indicated higher propensities for road rage. The average anger scores were calculated for drivers aged 16-25, 26-35, 36-45, 46-55, and 56-65. A one-way analysis of variance was performed to assess whether or not there were differences between anger scores of drivers in the different age categories.If we would reject the global null hypothesis and were interested in all of the possible pairwise comparisons, how many additional tests would we have to perform?