Incomplete block design
Problem description
Design
The experiment has 9 genotypes, 2 treatments, and 3 blocks. Each block only has a limited number of experimental units and thus a block was limited to have 3 genotypes under both treatments and 2 checks/controls. Typically the checks/controls are replicated while some of the other genotype-treatment combinations may not be replicated.
SAS is having difficulty fitting the model even when constructing a new “treatment” variable that combines genotype-treatment into a single variable.
Advice
SAS should be able to fit this model perhaps the new treatment variable is miscoded?
blog comments powered by Disqus