William G. Hunter Award 2005: Douglas M. Hawkins

The recipient of the 2005 William G. Hunter Award is Douglas M. Hawkins. The Statistics Division of the American Society for Quality (ASQ) established the Hunter Award in 1987 in memory of the Division’s founding chair to promote, encourage and acknowledge outstanding accomplishments during a career in the broad field of applied statistics. The attributes that characterize Bill Hunter’s career—consultant, educator for practitioners, communicator, and integrator of statistical thinking into other disciplines—are used to help decide the recipient. The chair for last year’s Hunter Award Selection Committee was Joseph Voelkel.

Dr. Hawkins, Professor of Statistics, University of Minnesota, has made an impact in statistics that is both broad and deep. A short list includes; CUSUM, regression diagnostics, outlier detection, multivariate methods, recursive partitioning, and numerous contributions in chemometrics.

I never met Bill Hunter. In accepting the Hunter award last year, Tony Greenfield noted that he was the first recipient who had never met Bill Hunter. My nomination this year continues an inexorable but sad trend away from the circle of Bill’s personal acquaintance, and into the wider community who know his writings and his achievements, but who know of his philosophy and guiding principles only at second hand.

Tony had another advantage over me; Bill Hunter knew and cited his work. I don’t know that I can make the same claim. I had half a dozen or so papers in Technometrics during Bill’s time as Associate Editor: perhaps he saw my work, though I am not aware of his having cited it approvingly, as Tony was.

I will tentatively claim my own Hunter awardee first – for the person who started out furthest from Bill’s base in the Upper Midwest. I spent most of my first 40 years in South Africa, with a large chunk of my work in mining, mineral processing and refining, and things generally related to chemistry. The statistical profession there was small (the national society had some 300 members at all levels), but had the challenge of providing infrastructure for major industries such as precious metals miners and processors, as well as the more conventional commerce, industry and government. It was hard to provide competent statistical advice for the variety of problems seen without the large support network of smart colleagues that we here take for granted and call on for ideas. This meant that print resources had an importance to us well beyond that to most workers in the US.

Bill's numerous papers provided specific ideas about methods to use, but his contribution to the whole print fabric may have been of greater value to us far-away folk. Hands-on journals such as Technometrics were a boon. In his acceptance speech, another Bill (Meeker) noted Bill (Hunter)'s astounding record of service on the Technometrics editorial board. I wonder whether Bill knew what a help this was to workers so far outside his geographic ambit. Though if not, probably his stay in Nigeria showed him that aspect of his contribution to society.

Statistics for Experimenters was a godsend when it appeared in the late 70s. It provided an authoritative source to use, not only in designing industrial experiments, but as a readable resource for engineers and scientists wanting to know more about how they could advance their own experimental programs. We used it extensively for research into such problems as optimization of platinum refining and organics synthesis.

Those who knew him stress Bill’s qualities as someone who most enjoyed working on real problems from the real world. I did not have the benefit of this experience firsthand, but was informed by a couple of others with similar predilections. My advisor, John Kerrich, loved making a statistical contribution to problems in a wide variety of subject matter areas. Another role model, Herbert Sichel, was a perfect example for how one could take a problem in, for example, gold mine evaluation, and in parallel solve the specific problem and develop a general statistical methodology. He also shone at recognizing the commonality of statistical formulations across completely different subjectmatter problems.

A chance remark of Herbert's showed me that SPC was more than rote plotting of points on Xbar charts, and led me to carry out research motivated by problems of chemical assay bias I saw in the mining industry and write the results up for general purpose use – much as Bill did.

I arrived in the USA about the time Bill died. I had hosted George Box on his visit to Johannesburg some years previously, and made a return trip to the CQPI in Madison once I had settled in. This was too late to meet Bill, but the inscription on Christopher Wren’s crypt Si monumentum requiris, circumspice seemed appropriate. Looking around, the smart, energetic, interesting people in the Center, Bill’s monument: the structure he and his colleagues had worked hard to bring to fruition, was clear to see.

Apart from the tangible monument though, we are left with Bill’s philosophy of the place of statistics in the world. As I understand, we share the beliefs that we are here to make a difference; that we can do so by generously adding our wisdom to that of others in and outside our profession; and that we must impart this wisdom to our students, our statistical colleagues, and our subject-matter collaborators. Enthusiasm is infectious. And as Yogi Berra might have said, it is amazing what you can see just by watching provided you do so in a thoughtful well-designed way.

The ranks of previous Hunter awardees include many of the people I most respect in the area of usable statistics. It is a signal honor to me to be joining their ranks, and a pleasure to be seeing some of them here today. I am grateful to the Society and the committee for their selection. Along with the awardees, I have greatly enjoyed my interactions with George Box and Stu Hunter, two of the SfE triumvirate. So it is particular poignant that I never met the third member, Bill Hunter.