Initially the Unit of Biometry, Applied Statistics was an offshoot of the then Department of Agricultural Botany, and a natural destination after an agricultural research station. In addition to teaching its own students on statistics degrees and joint degrees involving statistics, the Department had responsibility for teaching service courses and providing statistical advice to staff and postgraduate students in other departments. Roger was very much involved in all these activities, in particular contributing greatly to the advisory work at first largely within the Agriculture Faculty but later very much more widely across the University and with external clientele. Video recordings of some advisory sessions were used in teaching students how to interact with those coming for advice. He was also a pioneer in developing computer games to teach the design and analysis of experiments. Thousands of students, colleagues and others had good reason to appreciate Roger’s kindly, while piercingly intelligent, support and guidance to their research. Over the years, generations of later additions to the Applied Statistics staff and postgraduate students were attracted to the University of Reading in particular because of the unique quality of the advisory involvement, and over decades all of them owed great debts to Roger as they learned the reality of making statistics meaningful and useful, frequently depending on his wisdom and generosity with his time. Roger developed a system for measuring workloads that was used to allocate work within the Department and between the Department and the Statistical Services Centre that provided training, and consultancy and research services for clients outside the University. The M.Sc. in Biometry was started in 1966 and its graduates have risen to senior positions in the U.K. and overseas.
Roger’s overseas work included standard visits to established groups of statisticians in many countries. During his period as International Secretary of the IBS he encouraged the establishment of new Regions and Groups of the Society. Often as a result of contacts with M.Sc. in Biometry and Ph.D. graduates he visited and assisted the work of international agricultural research institutes in several countries including Brazil, India, Kenya, Mexico and Nigeria. Exchange visits to the Universities of Ege in Turkey and Colombo in Sri Lanka strengthened the formal links between these Universities and the Department at Reading. In 1985, Roger Mead and Robert Curnow gave a course on statistical methods to plant breeders in Nanjing, with their wives Nina and June assisting in the practical classes! They also visited a number of agricultural research institutes and universities in China.
Possibly the most widely-known of any of Roger’s research is his joint paper with John Nelder in the Computer Journal some 50 years ago, modestly titled “A Simplex Method for Function Minimization”. This paper has some 20,000 citations and, known world-wide as the Nelder-Mead algorithm, not only remains a foundation to current research but is also widely applied in a diverse range of computer software. Other areas where Roger made important contributions included the modelling and analysis of data from plant competition, intercropping and agro-forestry experiments; with Robert Smith, an early attempt to introduce stochastic elements into the study of population dynamics, and numerous wide-ranging publications on the design of experiments. Roger presented many paper at IBS meetings and four read papers to the Royal Statistical Society- on intercropping research (1981 with Janet Riley); the modelling of spatial variation in field crop experimentation (1986 with Alan Brewer);the non-orthogonal design of experiments (1990), and the statistical analysis of public lending right loans (1991 with Anne Hasted, J. Sumsion and Sharon Wheeler) . Roger was a Joint Editor of the Journal of the Royal Statistical Society Series C from 1983 to 1986. Roger chaired the Research and Ethics Committee of the local Health Authority for several years.
The Department’s consulting and teaching work in agriculture provided the background to the highly-successful 1983 textbook with Robert Curnow, with second (1993) and third editions (2003) with Anne Hasted as additional author (Mead, R., R. N. Curnow, and A. M. Hasted (2003). Statistical Methods in Agriculture and Experimental Biology. Boca Raton, Fla: Chapman & Hall/CRC).
His magnum opus (Mead, R. (1988). The Design of Experiments: Statistical Principles for Practical Application. Cambridge, UK: Cambridge University Press.) provided a unique approach to his favoured field, imbued with his incisive insights, abounding intuition and massive scholarship. The most recent book (Statistical Principles for the Design of Experiments) was co-authored (2012) by one of Roger’s former Ph.D. students, Professor Steven Gilmour, and his son Andrew Mead, now Head of Applied Statistics at Rothamsted Research.
Amongst the many other facets of a very rich life was his long established Madrigal Group that performed at his Memorial Service. Roger was very involved with his wife Nina in the work of the United Reformed Church and with the local Labour Party.
The memories of his colleagues and many friends are of a man of great integrity, calmness and kindness to all. This was combined with quite exceptional skills and intuition in interpreting data and translating into practical advice the individual and particular needs of experimenters and modellers. His life and his writings will continue to influence those who were fortunate to work with him or know him personally or through his publications.
Roger’s family were very important to him and supported him through the many years of his last illness. Roger is survived by his wife Nina, sons Simon and Andrew, and daughter Rachel, and his seven grandchildren.
Submitted by Robert Curnow and Colleagues