Statistician discusses key elements of experimentation
3:26 p.m., April 17, 2012--The individual may be primary in the pursuit of knowledge but the statistician is integral in the design of an experiment, J. Stuart Hunter told attendees at the second annual W. L. Gore Lecture Series in Management Science presentation held Tuesday, April 10, in Clayton Hall on the University of Delaware’s Laird campus.
The series, which recognizes the key role probability, statistics and experimental design have played in the success of W.L. Gore and Associates, Inc., was hosted by the Alfred Lerner College of Business and Economics and sponsored by an endowment from the Gore family.
Sports, race, culture
Hunter, a statistician and professor emeritus of the School of Engineering and Applied Science at Princeton University, discussed the importance of experimental design within marketing, health care, education, biometry and engineering applications.
“We see through the glass darkly and learn through experience,” said Hunter, who opened the lecture by recognizing the valuable role Wilbert Lee Gore played in highlighting the importance of statistics in experimentation.
“Bill Gore was a remarkable guy in his appreciation of the importance of statistics, which was rare in the 1950s,” said Hunter. “I owe him for a lot of the experimental design work I did within chemical engineering.”
Hunter went on to explain that statisticians bring great understanding to the pursuit of knowledge through experimentation because they value aspects of an experiment that others forget.
“A statistician’s first sentence is ‘observations are signals plus noise,’” said Hunter.
According to Hunter, an experiment that does not measure noise as well as the signal is ineffective because without noise there is no signal and without deciphering between noise and the signal data is compromised.
“Data are not information,” said Hunter. “Information lies within the data.”
From classical to factorial to computer-aided designs, Hunter explained the importance of a well-designed experiment and its ability to produce far more valuable and accurate information.
Hunter closed with a call to action for professionals and students to channel the mantra of the statistician and understand the role of noise and to measure it in each experiment.
Hunter is the founding editor of Technometrics and was president of the American Statistical Association in 1992. He became an honorary member of the American Society for Quality in 1999 and was elected to the National Academy of Engineering in 2005.
He has received the Shewhart Medal (1970), the S.S. Wilks Medal: U.S. Army (1987), the Deming Medal (1986), and the Founders Award of the American Statistical Association (1994).
In 2006, he received an honorary doctor of science degree from his alma mater, North Carolina State University, and from the Pennsylvania State University in 2008.
Hunter has published extensively and is co-author with I. Guttman and S.S. Wilks of Introductory Engineering Statistics, and with G.E.P. Box and W. G. Hunter of Statistics for Experimenters. He remains active as a consultant and lecturer.
Article by Arianne Minch
Photos by Ambre Alexander