Content area
Abstract
Model life tables describe typical age and sex patterns of mortality and have been useful to demographers and other social scientists for a variety of purposes, including evaluation of the demographic data normally collected in population censuses and surveys. The objectives of this study are to discuss the modelling of age patterns of mortality, with specific reference to developing countries, and to illustrate the use of such model age patterns for evaluating the quality of recorded census age distributions in the South Asian region. Namely the age distributions from the 1961 and 1974 censuses of Bangladesh, the 1961 and 1971 censuses of India, the 1961 and 1971 censuses of Nepal, and the 1961 and 1972 censuses of Pakistan are considered. The methodology used is the integrated procedure developed by Samuel Preston for estimating demographic parameters. This dissertation takes the procedure one step further by combining the estimated demographic parameters with the age-specific intercensal growth rates to produce a "correct" intercensal age distribution. This correct distribution can be compared to those recorded. The study finds consistent patterns of errors in age recording among these four South Asian countries, which need to be taken into consideration when using demographic data from this region. The study also provides concrete evidence of the value of the integrated method for demographic estimation and the value of model life tables in this process.
