Estimation of infant and childhood mortality in Kenya when mortality conditions are declining: evidence from 1969 and 1979 censuses.

Abstract

This study is concerned with investigating the accuracy of Brass estimates under conditions where mortality schedules are known or suspected to have been declining in the recent past. To achieve this task, three techniques suited for a changing mortality condition have been applied: namely, Additive Hypothetical Synthetic procedure; Kraly-Norris procedure and Palloni's technique. These techniques are applied at National, Provincial and District levels. The investigation is divided into five chapters. Chapter I deals majorly with historical development of the problem we set out to investigate. Chapter II deals with the intercostal mortality estimation between the period(1969-L979). Additive synthetic technique, a procedure suited for fluctuating mortality schedule is used. Chapter 1 deal Kraly-Norris simulation process for a declining mortality conditions. Chapter IV deals with yet another technique suited for a declining condition of mortality:Palloni's model. Chapter V deals with a summary of the findings, conclusions and recommendations. The mortality level estimated from Indirect procedure employing Additive hypothetical synthetic procedure behaves exactly the way they would be expected excatly under declining conditions of mortality. The violation of the static assumption of mortality brings inconsistency and unreliability in the estimates. Errors emanating from such a violation showed that they (the errors) ascended strictly monotonically by childhood age. Kraly-Norris procedure and Palloni's both yielded results which are superior to those obtained form unadjusted values ( i.e Using constant assumption of mortality schedules).