Analysing inter-relationships among water, governance, human development variables in developing countries

Abstract

Abstract. The "Integrated Water Resources Management" principle was formally laid down at the International Conference on Water and Sustainable development in Dublin 1992.

Bayesian networks modelling in support to cross-cutting analysis of water supply and sanitation in developing countries

“…A better understanding of which and how variables impact WSS will help to orient efforts and set priorities for intervention. To do so, the analyses performed previously on the WatSan4Dev database (Dondeynaz et al, 2012) are used as a basis for the modelling phase proposed in this paper. WatSan4Dev was developed by the Joint Research Centre (JRC) of the European Commission (EC) and contains 42 indicators (called variables) representing environmental, socio-economic, governance and financial aid flows data in developing countries.…”

“…WatSan4Dev was developed by the Joint Research Centre (JRC) of the European Commission (EC) and contains 42 indicators (called variables) representing environmental, socio-economic, governance and financial aid flows data in developing countries. As explained in Dondeynaz et al (2012), a subset of 25 variables is used for analysing WSS behaviour because of the correlations observed through multivariate analyses of the WatSan4Dev database. The description of the data and multivariate analyses performed both on variables and observations (countries) are detailed in Dondeynaz et al (2012).…”

“…As explained in Dondeynaz et al (2012), a subset of 25 variables is used for analysing WSS behaviour because of the correlations observed through multivariate analyses of the WatSan4Dev database. The description of the data and multivariate analyses performed both on variables and observations (countries) are detailed in Dondeynaz et al (2012). In line with this previous analytical work, this paper proposes various probabilistic models based on Bayesian network methods.…”

“…Despite the efforts made towards the Millennium Development Goals targets during the last decade, improved access to water supply or basic sanitation still remains unavailable for millions of people across the world. This paper proposes a set of models that use 25 key variables and country profiles from the WatSan4Dev data set involving water supply and sanitation (Dondeynaz et al, 2012). This paper suggests the use of Bayesian network modelling methods because they are more easily adapted to deal with non-normal distributions, and integrate a qualitative approach for data analysis.…”

“…Modelling is run with the five groups of countries as defined in Dondeynaz et al (2012). Models for profile 4 (essential external support) and profile 5 (primary material consumption) are specifically detailed and analysed in this paper.…”