Randomized Block and Latin Square 203

Although the blocked 2k factorial design is a common approach to compensate for non-uniformities in the experimental conditions, there are many other variations of the blocked experimental design concept. While the underlying motivation remains the same in all blocked designs, particular restrictions on randomization have resulted in some very clever schemes and treatment combination patterns to overcome the problem of not being able to randomize completely. Picking the proper design almost becomes intuitive and automatic once the basic concept is understood. The analysis and interpretation is where the difficulties arise. In this chapter, we shall look at some of the other types of blocked designs. Chapter 18 will then show how to properly analyze the data from these designs.