The winners of ASQ’s prestigious Brumbaugh Award for 2009 are Bradley Jones, PhD, Principal Research Fellow in the JMP Division of SAS, and Christopher J. Nachtsheim, PhD, Carlson School of Management at the University of Minnesota.
Renowned statistician Cuthbert Daniel said that all industrial experiments are split-plot experiments. That’s the reason why we as experimenters need to understand how to create our experiments that reflect the reality of how they have been executed. For years, scientists have fielded their experiments as split-plot experiments and have erroneously analyzed them as if they had been run in completely randomized fashion.
Today, with the tools available to us in JMP, we are easily able to set up and analyze our design of experiments correctly. In fact, after reading the article by Bradley and Chris, you might conclude that you should intentionally implement split-plot designs since they can be much more attractive with respect to cost, efficiency and validity of the experiment. I highly recommend that anyone involved in conducting experiments check out this article, a PDF file of which is available on the JMP Web site. You too may realize that split-plot designs are the correct approach for your future experimentation.