Department of Civil and Environmental Engineering (DECA), Engineering Sciences and Global Development (EScGD), Barcelona School of Civil Engineering, Universitat Politècnica de Catalunya, Barcelona, Spain.
Department of Civil and Environmental Engineering (DECA), Engineering Sciences and Global Development (EScGD), Barcelona School of Civil Engineering, Universitat Politècnica de Catalunya, Barcelona, Spain.
Sci Total Environ. 2019 Dec 15;696:133830. doi: 10.1016/j.scitotenv.2019.133830. Epub 2019 Aug 13.
Nationally representative household surveys are the main source of data for tracking drinking water, sanitation and hygiene (WASH) coverage. However, all survey point estimates have a certain degree of error that must be considered when interpreting survey results for policy and decision making. In this article, we develop an approach to characterize and quantify uncertainty around WASH estimates. We apply it to four countries - Bolivia, Gambia, Morocco and India - representing different regions, number of data points available and types of trajectories, in order to illustrate the importance of communicating uncertainty for temporal estimates, as well as taking into account both the compositional nature and non-linearity of JMP data. The approach is found to be versatile and particularly useful in the WASH sector, where the dissemination and analysis of standard errors lag behind. While it only considers the uncertainty arising from sampling, the proposed approach can help improve the interpretation of WASH data when evaluating trends in coverage and informing decision making.
全国代表性住户调查是跟踪饮用水、环境卫生和个人卫生（WASH）覆盖情况的数据主要来源。但是，所有调查点估计都存在一定程度的误差，在为政策和决策目的解释调查结果时必须考虑到这些误差。本文介绍了一种用于描述和量化 WASH 估计值不确定性的方法。我们将其应用于玻利维亚、冈比亚、摩洛哥和印度四个国家，这些国家在地域、可用数据点数量和轨迹类型方面各不相同，以说明为时间估计值传达不确定性以及同时考虑 JMP 数据的组成性质和非线性的重要性。该方法被证明是多功能的，在 WASH 部门尤其有用，该部门在标准误差的传播和分析方面落后。虽然该方法仅考虑了抽样产生的不确定性，但当评估覆盖范围的趋势和为决策提供信息时，所提出的方法可以帮助改善 WASH 数据的解释。
