A firm tested the durability of four different types of tires by having 4 Cadillacs each randomly equipped with one of the types of tires. In similar fashion, 4 Fords were used, each randomly equipped with one of the tire types. In a like manner, 4 Volkswagens were each randomly outfitted with one of the four tire brands. Then each of the twelve cars ran over a test track until the tires showed signs of serious wear. At that point the average mileage of the three cars using one type of tire was compared to the average mileage of the three cars using the second type of tires, and so on. This design is:

In the scenario described above, the subjects which are the different car types are grouped into fixed units of 4 called blocks with each unit then assigned randomly to a particular treatment condition, which are the 4 different types of tire. In the end, the mean mileage for each block is taken and compared to that of the other blocks. This type of research design improves the reliability of the result obtained as it also eliminates the occurrence of systematic error in the course of our experiment.