Page 90 - IJPS-10-4

International Journal of
Population Studies Migration and child mortality estimation

the estimation of rural-urban child mortality disparities. migrated from rural to urban areas in Kenya between the
The results presented in Table 4 reveal that child mortality late 1980s and early 2010s led to an overestimation of
rates in the urban region were overestimated, indicating child mortality rates in urban regions. This overestimation
that the child mortality disparities between rural and urban occurred because, during this period, child survival in
regions in Kenya were larger than previously estimated. rural regions was significantly lower than in urban areas.
Second, the findings have implications for analyzing child The period is also characterized by a persistently high
mortality trends in rural and urban regions. For example, rate of rural-to-urban migration, leading to a situation
some studies in Kenya (Kenya National Bureau of Statistics where children who died in rural regions were reported
(KNBS), & International Classification Function (ICF), by migrant women as if they had died in urban regions.
Macro, 2015; Kimani-Murage et al., 2014) suggest that Our study highlights the necessity for the re-estimation
urban advantage in child survival disappeared in favor of child mortality trends in both rural and urban areas,
of rural regions between the late 2000s and early 2010s. considering the impact of migration. This adjustment
However, these studies did not account for the impact would provide a better understanding of the effectiveness
of rural-urban migration, which could distort the child of various interventions aimed at increasing child
mortality estimates and, consequently, the direction and survival. It would also facilitate a meaningful comparison
extent of child mortality disparities between the two of child mortality between rural and urban regions. In
regions. Although child survival in rural Kenya has made addition, future studies could explore migration within
notable improvements, one must then question whether specific determinants of child mortality, such as changes
the documented timing of convergence in child mortality in economic status.
between rural and urban regions in these studies is real or
merely an artifact of estimation procedures. Evidence from Acknowledgments
this study indicates that the latter is true. This finding calls The authors would like to acknowledge the ICF Macro for
for a re-estimation of trends in child mortality rates in the providing the datasets used for this research.
two regions for better policy direction.
The major strength of this study lies in empirically Funding
establishing how migration altered child mortality estimates None.
in the urban region of Kenya from a statistical point of view,
a point that has not been addressed previously. By pooling Conflict of interest
data from several surveys, we obtained sufficient estimates, The authors declare that they have no competing interests.
which enabled us to carry out meaningful statistical
comparisons. However, the study had some limitations that Author contributions
could have affected the accuracy of rural and urban child
mortality estimates. First, the study exclusively focused on Conceptualization: Alfred Kathare, Titus Agwanda
recent migration between rural and urban areas without Formal analysis: Alfred M. Kathare
accounting for circular migration, a common occurrence Investigation: All authors
among short-distance migrants. Second, it was assumed Methodology: Alfred M. Kathare
that women often stay with their children in the same place Writing – original draft: Alfred M. Kathare
of residence. However, sometimes, women in urban areas Writing – review & editing: Titus O. Agwanda, Kimani
may send their children to live with their relatives in rural Murungaru
areas to reduce the cost of living. These children are included All the authors have read this manuscript and approved the
when computing child mortality in the region where the version for publication.
mother resides, which can lead to errors in estimating child Ethics approval and consent to participate
mortality due to a reduction in the proportion of deceased
children in that region. In addition, data quality issues The KDHS data used in this study is from the MEASURE
such as underreporting of deceased children can affect the DHS repository. The data analysis in this study was,
mortality estimates. However, evidence from KDHS reports therefore, performed under the original consent provided
indicates that errors such as omissions of births and deaths under ICR Macro.
are not severe (KNBS, 2015).
Consent for publication
5. Conclusion This research article is published under the original ethical

This study statistically affirmed that the inclusion of approvals and consents of participation provided during
mortality information from children born to women who the original data collection by ICR Macro.

Volume 10 Issue 4 (2024) 84 https://doi.org/10.36922/ijps.1837