Assessment of the Validity of Assumption of Zero Net Migration in Estimation of Adult Mortality Using Preston Integrated Approach

Abstract

This study examines the implication(s) of ignoring net migration in estimation of adult mortality from non-stable population using the Preston [1] integrated method and proposes a method of assessing presence of net migration in a study data. In deriving the model for estimation of adult mortality from non-stable population, Preston [1] has assumed that the study population is closed to migration, that is, the net migration is zero or negligible. However, in most developing countries this assumption is not necessarily true. In this study, the method proposed for assessing the need for adjustment for net migration is the ratio (AF) of the observed  proportion of mid-period population reported as aged x years (c(x,t))  when net migration is not zero to the corresponding proportion when net migration is zero (\(\hat{c}\)(x,t)). The Preston [1] approach to estimation of adult mortality is to relate characteristics of the observed population to some life table functions and using this relationship to obtain estimate of the implied level of adult mortality. The results indicate that when net migration is actually zero, the ratio AF is equal to one and different from one when net migration is not zero. Data on age- sex distribution of populations of three selected developing countries were used to illustrate the methods. It has therefore, been recommended that when this ratio is not one an adjustment should be made for net migration.