Visualizing World Map Indicators and Data Analysis
Introduction
The World Bank has compiled data on many countries across a span of decades. This data includes information across a number of categories, such as Agricultural and Rural Development, Aid Effectiveness, Climate Change, Economy and Growth, etc. Given the sheer volume of data, there are a number of interesting visualizations possibilities. . These include a world map showing the distribution of data across countries, comparing two countries in a certain metric, looking at the change of a metric across time for a given country, checking the correlation of two indicators for a given year and country, etc. This Shiny app attempts to display some interesting features of some of the indicators.
Data
The World Bank has a database which contains indicator data that spans several decades. It also has an API that allows a user to access information quickly. There are two libraries in R that allow one to connect to the World Bank database: wbstats and WDI. Rather than downloading the data onto a local machine, I decided to use the WDI library to access the data. Using the library and WDI API allows quick access to the information without having to waste space storing it locally.
Insights
Insight 1
The image below was generated by setting the year to 1993 and the indicator to GNI per capita. We can see that countries such as Russia and Brazil fall in the middle of the countries in terms of their GNI per capita.
The image below shows the GNI per capita for the year 2013. We see that Russia and Brazil have made significant progress with respect to their position on GNI per capita. The user can use the world map to check for changes in a country or region, and generate hypotheses to explain the observed changes.
Insight 2
The scatterplot page allows the user to select two indicators and a year, to see whether there may be any kind of relationship between the two indicators. As an example, we can look at the relationship between male life expectancy and tertiary school enrollment. We would expect there to be some positive relationship between the two. The scatterplot below indeed does show that there is a strong positive relationship between these two indicators. Further, the data is somewhat segmented by continent. The African countries score lower on both male life expectancy and tertiary school enrollment, while the European and Asian countries score higher. These observations could serve as the basis for policy decisions by government officials to make strides to close the gap between developing countries and developed ones.
Conclusion
This app allows the user to look at different indicators from different perspectives. The user can compare countries using different indicators and try to find relationships between indicators. In the future, I would like to clean up the plots and data that are available now and expand the app by adding the ability to look at the data across time.
About Author
