This talk will take as its starting point the definition of demography as the study of people, rather than populations. It will also emphasise the key role of demography as the multidisciplinary approach to study of peoples' life-courses: the link here is to the common usage of 'demographics' as covering a broad range of socioeconomic and other characteristics of people. Illustrations of the changing nature of demographic research over my life-course will be given and links made back to the broad and innovative approaches of David Glass. The term life-course is used in the original meaning by Glen Elder, as the interweaving of pathways through life and is taken to include both biological and social scientific pathways. Particular emphasis will be given to emerging concerns involving these biosocial interplays, with some illustrations from my recent research. The exciting opportunities provided by new ESRC investments in large-scale prospective studies, especially Understanding Society and the new Birth Cohort Study, will be stressed, as will the opportunities from SHARE and the Generation and Gender Programme.

Development Goals and Mortality Measurement in the Age of Immediate Gratification

Kenneth Hill, Harvard Center for Population and Development Studies

Millennium Development Goal (MDG) 4 sets a target of reducing under-5 mortality by two-thirds from 1990 to 2015, and MDG-5 sets a target of reducing the Maternal Mortality Ratio by three-quarters over the same time frame. Donor agencies in the health field are increasingly looking for hard evidence that interventions work, increasingly linking actual disbursements to measures of performance. The recently-published report of the United Nations Secretary-General's Information and Accountability Commission also calls for annual reporting, though of process indicators rather than mortality measures. Given the weakness of most developing country vital registration systems, can demography satisfy this obsession with rapid measurement and real-time assessment of mortality change? The full birth history included in Demographic and Health Surveys is very expensive, and can significantly detect only a huge annual change in under-5 mortality. The situation for maternal mortality, with far fewer events and definitional issues, is worse still. The paper proposes and illustrates some analytic approaches requiring less expensive data collection procedures, but concludes that effective rapid mortality monitoring will have to wait for the development of adequate vital registration systems.