Between 23 and 26 May 2019, the current European Parliament was elected. Some days after the election, Zeit Online gathered a dataset which contains the geometries and election results of all 78,818 municipalities that have voted.
All parties that were taking part in the election were previously categorised by their political classification. That enabled us to colorize and compare all regions with the same criteria and create this super detailed election map of Europe.
We have worked on multiple election maps before, for example the results of the federal election 2017 in Berlin and Germany, but due to the massive amount of data, this one was more challenging.
Although we have already developed a map with the same number of geometries, this time we wanted to use more modern technologies and switch from raster-only to a combination of raster and vector tiles. This would have the advantage of showing a mouse-over effect on higher zoom levels and improving the general performance of the map.
In addition to the election map that is colorized by the party family which won the region, we rendered a map for each individual party family to show their strongholds.
This allowed us to use the legend of the map as a navigation and give more insight about the electoral behaviour around Europe.
All the maps that we are showing in this interactive are rendered automatically and from the same dataset. The input dataset is a GeoJSON file which was rendered to vector tiles using Tippecanoe by Mapbox and raster tiles using Node Mapnik. The vector tiles contain the data for each region and their boundaries, whereas the raster tiles are used for colorization.
Both tilesets are composed together with the mapping library mapbox-gl. Through our approach of combining raster and vector tiles, we were able to show a super detailed colorized map and have an interactive tooltip and mouseover effect as well.