Title: Randomization in Designed Experiments
Author: Keith M. Bower
Publication Source: American Society for Quality, Six Sigma Forum, December 2004

Abstract

This article addresses the importance of randomization in designed experiments.

Notes

A key reason for this paper was to hopefully have more practitioners read George Box's referenced paper: "Must We Randomize Our Experiment?" My paper is essentially a trivial mathematical simulation showing how something can go wrong when a non-stationary disturbance is in the "background," and shows how randomization can address such issues. Box's paper provides some excellent advice, and I strongly advise practitioners to read it.




© Smoother Pebble LLC. All rights reserved. Disclaimer