Estimation of Entropy in Kenya
Abstract
The study estimates entropy values in Kenya, which have been lacking. Risks in
death are also estimated. The study uses the 1969 -79, 1979 -89 and 1989 -99
censuses data on expectation of'tife and survivorship.
The derivation of entropy and methods of estimations of entropy and the risks of
death are presented in chapter three and the results of entropy and the applications
ff the entropy are presented in chapter four.
The results show fluctuation of the entropy values for both males and females with
males' values relatively higher than that of females' for the three inter-censal
periods. The males' entropy levels for 1969-79 was 0:37, 1979-89 it was 0.302 and
in 1989-99 it was 0.344. The corresponding entropy values for females were 0.309,
0.242 and 0.294. All these entropy values suggest considerable high risks of death
at lower ages. The increasing values of entropy in 1989-99 from 1979-89 implies
more sift of risks of death to younger ages
The results on change ~ risk of death show significance changes in mortality
associated will small changes in expectation of life. For males the reduction of risks
of death of 35.5 percent in 1979-89 resulted in 10.8 percent increase in expectation
of life. The increase Q.f..,Cisks of death of 25.6 percent in 1989-99 resulted in 8.8
percent reduction in expectation of life. For females, reduction of risks of death by
40.7 percent in 1979-89 resulted to 9.9 percent increase in expectation of life while
increase of risks of death of 28.4 in 1989-99 resulted to 8.3 percent reduction in
expectation of life. The results show also, that, females responded to mortality
changes greater than males during 1979-89 and 1989-99 inter-censal periods.
Chapter five presents summary conclusions and recommendations. Sift of more
deaths towards younger ages in 1989-99 has reduced the expectation of life. We
have recommended for programmes that will increase longevity of life. Further the
results showed exceptionally high increase in risks of death for ages 15 - 34 for
(males and 10 - 29 for females in 1989-99. We have therefore recommended for
youth programmes to deal with age-specific risks of deaths.
We have recommended for investigation of the effect of HIV/AIDS on Kenyan
mortality curve. We have further recommended for invest.i.g" a.tion on the effect of reemerging
diseases on the changes of risks of death in Kenya. We have also
recommended for the investigation of regional disparities in mortality curve.
Citation
A thesis submitted to the population studies and research institute as partial fulfillment of the requirement for the award of master of arts (population studies), university of NairobiPublisher
Institute of population Studies and research