Data-driven Bayesian network modelling to explore the relationships between SDG 6 and the 2030 Agenda.

The Sustainable Development Goals (SDGs) are presented as integrated and indivisible. Therefore, for monitoring purposes, conventional indicator-based frameworks need to be combined with approaches that capture and describe the links and interdependencies between the Goals and their targets. In this study, we propose a data-driven Bayesian network (BN) approach to identify and interpret SDGs interlinkages. We focus our analysis on the interlinkages of SDG 6, related to water and sanitation, across the whole 2030 Agenda, using SDG global available data corresponding to 179 countries, 16 goals, 28 targets and 44 indicators. To analyze and validate the BN results, we first demonstrate the robustness of the BN approach in identifying indicator relationships (i.e. consistent results throughout different country sample sizes). Second, we show the coherency of the results by comparing them with an exhaustive study developed by UN-Water. As an added value, our data-driven approach provides further interlinkages, which are contrasted against the existing literature. We conclude that the approach adopted is useful to accommodate a thorough analysis and interpretation of the complexities and interdependencies of the SDGs.