BIO361H5 Lecture Notes - Lecture 9: Restricted Randomization, Complement Factor B, Block Design
Document Summary
Factorial design: when all factors are crossed, when there is data point for every combination of every factor (ex. every level of factor a is crossed with every level of factor b) Nested design: levels of one factor are not the same for all levels of another factor (there can be crossing) One distinction b/w split plot and nested design: No, thus student = blocking variable (random) bc we don"t care about individual influences on y (students rep. variability in population) But if i tell you we used poetry distractor, it tells us something about the level, we know what level is used. So if you repeat a study, you can use the poetry distractor again, but you probably can"t use the same individual again. (answer = yes, effect = fixed: wordlist = fixed. Student = random: observations are usually random. Ex. rat study where we sample their liver.
