Overview

1. Introduction

In rural Africa, access to clean and safe water can be a challenge. Thousands of water points have been installed over the past decade, but they are of poor quality or are not well maintained, causing them to become non-functional. This can cause waterborne diseases, which can destroy lives and livelihoods. One country in West Africa which faces this issue in Nigeria.

According to a research paper1 which utilises information from the 2015 Nigeria National Water and Sanitation Survey, more than 38 percent of all improved water points in Nigeria are non-functional. Hence, it would be in the interest of the Nigerian government, non-governmental organisations and researchers to discover water points patterns in Nigeria and potentially figure out what is causing the problem and how to solve it.

2. Motivation

This app is designed to help various stakeholders, whether technical or non-technical, to be able to visually analyse the distributions of data points by attributes and identify geospatial insights and patterns statistically and geographical segmentation. For this launch, we are using waterpoints dataset in Nigeria to demo the intuitive functions and non-technical users friendly features.

Through this app, every interested stakeholder, such as government, local community planners, and academic researchers, can effectively and visually communicate the findings on a common ground, reducing potential misinformation gap. This will potentially help government and community planners make informed policy decisions, decide where to allocate their resources optimally to fix or maintain water points and eventually improve the quality of lives of the people living in Nigeria.

3. Data Description

The data is focused on the water points in Nigeria. It will make use of two data sets.

  1. The boundary data set, at the Nigeria level 2 administrative boundaries, contain GIS data for the polygon feature. The data is obtained from https://data.humdata.org/dataset/nigeria-admin-level-0-1-and-2-administrative-boundaries

  2. The water points’ attributes data set is obtained from https://data.waterpointdata.org/dataset/Water-Point-Data-Exchange-Plus-WPdx-/eqje-vguj/data.

Below are some of the key variables included in the data visualisation:

  • Operational status of the water points - Functional, Non-functional and Unknown.

  • Technology deployed in water points - Hand pump, Mechanical pump, Tap stand.

  • Water source - Spring, Well.

  • Pressure score of water points (0-100%) - The ratio of number of people assigned to the particular water point over the theoretical maximum population which can be served based on the technology

  • Crucial score of water point (0-100%) - The ratio of number of likely current users to the total local population within a 1km radius of the particular water point.

  • Usage capacity of water points - Recommended maximum users per water point.

  • Community - Indicates how many percentage of the water points fall within Urban or Rural communities.

4. Approach

Using various R packages such as ggplot2, ggplot2 extensions and tidyverse packages, the team will extract, analyse, and visualise geospatial patterns of functional and non-functional water points and build an interactive R Shiny application with visual analytics techniques.

It will be split into 3 sections, as shown below:

a) Overview

This section shows an overview revealing high-level patterns. It displays locations of functional, non-functional and unknown water points at the district level throughout Nigeria. This distribution visualisation is further supplement by various available parameters for users to choose.

Find out more about this section here.

b) Visual Inferential Analysis

This second section applies visual analytics techniques to the dataset through analytics approaches. This will help users identify relationships or correlations between different features, such as technology used or pressure score with non-functional water points through correlation matrices. It may also allow users to compare statistics across different districts through one-way ANOVA test visualisations.

Find out more about this section here.

c) Geographical Segmentation

This third section helps users to visualise geographical segmentation based on the Hierarchical Clustering algorithm by grouping the water points according to the selected attributes and spatial location, methods to determine the optimal number of clusters and a heatmap for clustering analysis.

Find out more about this section here.

Footnotes

  1. “Andres, Luis; Chellaraj, Gnanaraj; Das Gupta, Basab; Grabinsky, Jonathan; Joseph, George. 2018. Why Are So Many Water Points in Nigeria Non-Functional? : An Empirical Analysis of Contributing Factors. Policy Research Working Paper;No. 8388. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/29568 License: CC BY 3.0 IGO.”↩︎