Hello. I'm Doug Montgomery. I'd like to welcome you to our introduction to experimental design basics course, and I'm going to tell you a little bit about this course. It is, as the title suggests, an introduction to the subject of experimental design. So you might be asking yourself, what's an experiment and why design an experiment? Well, humans have been doing experiments, I think since the dawn of time. I suspect that the first human to build the fire actually did some experimentation to figure out how to do this, and there have been lots of famous experimenters through the years. Thomas Edison, for example, is famous for having run thousands of experiments. The Wright brothers developed the notion of powered flight and the ability to control an airplane and flight entirely by experimental methods. They actually built a wind tunnel in their bicycle shop in Dayton, Ohio, and used it to master the principles of aerodynamics required to control an aircraft and flight. Particularly the notion of role control, they developed a technique called wing warping to do that, which later on became generalized into the notion of ailerons, which give that lateral control to an aircraft. But most of the experiments that we're talking about here, are not scientifically designed experiments, they're not statistically designed experiments. The field of experimental design, as we're going to talk about it, originates in agricultural science. In England, between about 1915 and about 1920, in that era, a British statistician and geneticist named RA Fisher was hired by the British government to go to their agricultural experiment station at Rothamsted, to try to improve the efficiency with which the egg scientists there were running agricultural field trials. Now, the scientists at Rothamsted had been running field trial experiments trying to learn how to grow crops better since the 1850s, but with very mixed results, and Fisher rapidly took over the planning of these experiments and developed the concepts and ideas that today form the basics of the field of statistical experimental design. He developed three concepts. Randomization, running the tests in an experiment in random order. Replication, repeating at least some runs in the experiments so that you have an idea of experimental error. Blocking, a technique to control nuisance variation during the experiment. He also developed the basic idea of a factorial design, which we will talk about in this course. He also developed the idea of the analysis of variance, one of the standard techniques today for analyzing data from design experiments. So we're going to take a look at the origins of the field. We're going to talk about the terminology in the field, and then we're going to talk about our two of the principal analysis methods that we'll use throughout this course, and that's the two-sample t-test, which is used when you want to compare two levels of a factor, and then the analysis of variance, which is a more general technique developed by Fisher, which can be applied and extended to virtually any type of designed experiment. We'll also get an introduction to Fisher's blocking concept. Blocking is a technique to control variability from nuisance sources in the experiment, and typically experiments have nuisance variables. It can be variation due to different batches of material, variation due to the different operators or different people, or just variation over time. Some way to control that to reduce the amount of background noise in your data is really important, and that's what blocking is all about. We'll see how that works in this class. The other thing that we'll talk about that's very important in here is a procedure, a step by step process for successfully designing experiments. All work is done in processes and you need a well-defined process to help carry out a successful design experiment, and we will go through a seven-step process that if you follow this rigorously, you will get very, very good results. With that, that's an introduction to the course, and I hope you enjoy this.
