Mapping the Interaction between Development Aid and Stunting in Nigeria. Bosco, C.; Tejedor-Garavito, N.; de Rigo, D.; Pezzulo, C.; Bengtsson, L.; Tatem, A. J.; and Bird, T. J. In 28th IUSSP International Population Conference.
Mapping the Interaction between Development Aid and Stunting in Nigeria [link]Paper  abstract   bibtex   
For meeting sustainable development goals (SDGs) an improved understanding of geographic differences in health status, wealth and access to resources is crucial. The equitable and effcient allocation of international aid relies on knowing where funds are needed most. For instance, aid for poverty alleviation or financial access improvement requires knowledge of where the poor are. Unfortunately, detailed, reliable and timely information on the spatial distribution and characteristics of intended aid recipients in many low income countries are rarely available. This lack of information also hinders assessments of the impacts of aid; when presented at national scales, development and health indicators conceal important inequities, with the rural poor often least well represented. High-resolution data on key social and health indicators are therefore fundamental for targeting limited resources, especially where development funding has recently come under increased pressure. [\n] In this study, we show how modern statistical approaches can be used to maps for the distribution of indicators with a level of detail that can support geographically stratified decision-making. Using predictive modelling techniques, the rates of stunting in children under the age of five from Demographic and Health Surveys (DHS) geolocated cluster data were exploited to predict high-resolution maps (2008-2013) in Nigeria. An array of different modelling techniques was applied to produce prediction maps. These included Bayesian geostatistical models and machine learning techniques. An ensemble model was also exploited for aggregating the different modelling results. [\n] By combining these maps with information on the disbursement of aid for stunting alleviation in Nigeria (AidData database - http://aiddata.org/ ), we quantified both the distribution of aid relative to population characteristics related to stunting, and how aid disbursement interacts with changes in this index. In spite of the lack of exhaustive information related to aid disbursement, the results here demonstrate the potential of this approach.
@inproceedings{boscoMappingInteractionDevelopment2017,
  title = {Mapping the Interaction between Development Aid and Stunting in {{Nigeria}}},
  booktitle = {28th {{IUSSP International Population Conference}}},
  author = {Bosco, Claudio and Tejedor-Garavito, Natalia and de Rigo, Daniele and Pezzulo, Carla and Bengtsson, Linus and Tatem, Andrew J. and Bird, Tomas J.},
  date = {2017},
  url = {http://www.webcitation.org/75lE10duC},
  abstract = {For meeting sustainable development goals (SDGs) an improved understanding of geographic differences in health status, wealth and access to resources is crucial. The equitable and effcient allocation of international aid relies on knowing where funds are needed most. For instance, aid for poverty alleviation or financial access improvement requires knowledge of where the poor are. Unfortunately, detailed, reliable and timely information on the spatial distribution and characteristics of intended aid recipients in many low income countries are rarely available. This lack of information also hinders assessments of the impacts of aid; when presented at national scales, development and health indicators conceal important inequities, with the rural poor often least well represented. High-resolution data on key social and health indicators are therefore fundamental for targeting limited resources, especially where development funding has recently come under increased pressure.

[\textbackslash n] In this study, we show how modern statistical approaches can be used to maps for the distribution of indicators with a level of detail that can support geographically stratified decision-making. Using predictive modelling techniques, the rates of stunting in children under the age of five from Demographic and Health Surveys (DHS) geolocated cluster data were exploited to predict high-resolution maps (2008-2013) in Nigeria. An array of different modelling techniques was applied to produce prediction maps. These included Bayesian geostatistical models and machine learning techniques. An ensemble model was also exploited for aggregating the different modelling results.

[\textbackslash n] By combining these maps with information on the disbursement of aid for stunting alleviation in Nigeria (AidData database - http://aiddata.org/ ), we quantified both the distribution of aid relative to population characteristics related to stunting, and how aid disbursement interacts with changes in this index. In spite of the lack of exhaustive information related to aid disbursement, the results here demonstrate the potential of this approach.},
  keywords = {*imported-from-citeulike-INRMM,~INRMM-MiD:c-14573315,artificial-neural-networks,bayesian,demographic-indicators,ensemble,geospatial,geospatial-semantic-array-programming,human-health,integration-techniques,machine-learning,nigeria,semantic-array-programming,semantics,stunting,sustainable-development},
  options = {useprefix=true},
  venue = {Cape Town, South Africa}
}
Downloads: 0