Search

There is increasing concern about the lack of accuracy in population projections at national levels. A common problem has been the systematic underestimation of improvements in mortality, especially at older ages, resulting in projections that are too low. In this paper, we present a method that is based on projecting survivorship rather than mortality, which uses the same data but differs technically. In particular, rather than extrapolating trends in mortality, we use trends in life expectancy to establish a robust statistical relation between changes in life expectancy and survivorship using period life tables. We test the approach on data for England and Wales for the population aged 50 and over, and show that it gives more accurate projections than official projections using the same base data. Using the model to project the population aged 50 and over to 2020, our method suggests nearly 0.6 million more people in this age group than official projections. [ABSTRACT FROM PUBLISHER]

Population estimates play an important role in the allocation of resources at many levels of government and commerce but little is known about the accuracy of age-specific population estimates. Such knowledge is crucial, as resource allocation is often targeted at populations of particular age, and decisions need to be based on the reliability of the estimates. This paper presents a multi-level statistical analysis of the accuracy of age-specific population estimates made for British local authorities in 1991. The aim of this work is to identify the factors that influence accuracy, and to investigate how these influences interact. Our analyses show that the following area characteristics are key factors: true population size; intercensal population change; and percentages of unemployed residents, armed forces residents, and students. In addition, we find that the overall type of method used to calculate estimates is important, and that its effect varies both with area characteristics and with age-group. Local census methods are found to be generally superior, but a low-cost apportionment method, if implemented well, may be as effective. [ABSTRACT FROM AUTHOR]