Efficient 2k Factorials for Blocks of Size 2 with Microarray Applications

Summary: [This abstract is based on the author's abstract.]When the experimental goal is to estimate both main effects and two-factor interactions in two-level designs run in blocks of two sizes, it is necessary to combine replicates of the experiment that use different blocking schemes. Designs are identified that enable estimation with the fewest number of replications. In addition, a construction for general k is given that gives more economical designs than those previously published. Issues particular to the design of two-color microwave experiments are discussed.