The Scottish demographer and statistician William Brass is best known for his imaginative, elegant, and innovative contributions to methods for demographic estimation. Although always insistent on methodological rigor, Brass was, as he used to say, "a practical man" for whom the ultimate justification of a method was that it helped answer an important question and solve a problem.
After graduating from Edinburgh University in 1947 with a degree in mathematics and natural philosophy, Brass joined the East African Statistical Department as a statistician. While working there he developed a lifelong interest in sub-Saharan Africa and the problems of demographic estimation based on deficient or defective data. In 1955 he joined Aberdeen University to work on the application of mathematical models to medical statistics but maintained his interest in demographic estimation.
A year's leave of absence at Princeton University's Office of Population Research in 1961–1962 to work on the demography of tropical Africa proved to be a turning point in his career. During that period Brass developed several of his signature demographic estimation methods, notably methods for estimating child mortality from women's reports of their children ever born and their children who had died, for evaluating reported fertility rates by comparison with measures of lifetime fertility, and for the use of the logit transformation to fit life tables to fragmentary data. Those methods were applied systematically to the survey data from Africa then available.
In 1965 Brass moved to the London School of Hygiene and Tropical Medicine, first as a reader and then as a professor of the new field of medical demography. He continued to develop new methods as data availability and quality improved, including methods for estimating adult mortality from data on the survival of parents, the evaluation of data on deaths by age through comparisons with age distributions of the living, the relational Gompertz fertility model, methods to evaluate birth history data, and ways to use truncated data on parity progression for women of reproductive age to study fertility change. He applied those methods, among others, in an authoritative analysis of fertility trends in Kenya that was published as part of a series of reports on the population dynamics of sub-Saharan Africa produced by the U.S. National Academy of Sciences.
Many of the methods originally developed by Brass have been developed further by others. A full bibliography of Brass's writings is included in the commemorative volume Brass Tacks (2001).
selected works by william brass.
Brass, William. 1975. Methods for Estimating Fertility and Mortality from Limited and Defective Data. Chapel Hill: University of North Carolina Press.
——. 1981. "The Use of the Gompertz Relational Model to Estimate Fertility." International Population Conference, Manila, 3: 345–362.
Brass, William, John Blacker, and Basia Zaba, eds. 2001. Brass Tacks: Essays in Medical Demography: A Tribute to the Memory of Professor William Brass. London: Athlone Press.
Brass, William, and Ansley J. Coale. 1968. "Methods of Analysis and Estimation." In The Demography of Tropical Africa, W. Brass, A. J. Coale, P. Demeny, D. F. Heisel, F. Lorimer, A. Romaniuk, and E. van de Walle. Princeton NJ: Princeton University Press.
Brass, William, and Kenneth Hill. 1973. "Estimating Mortality from Orphanhood." International Population Conference, Liège, 3: 111–123.
Brass, William, and Carole Jolly, eds. 1993. The Population Dynamics of Kenya. Washington, D.C.: National Academy Press.
Brass, William, and Fatima Juarez. 1983. "Censored Cohort Parity Progression Ratios from Birth Histories." Asian and Pacific Census Forum 10:5–13.
Brass, William, and Hoda Rashad. 1992. "Evaluation of the Reliability of Data in Maternity Historoes." In The Analysis of Maternity Histories, ed. Allan Hill and William Brass. Liège, Belgium: Ordina Editions.
Kenneth H. Hill