Have any of you data journalists visualized election results in such a way that *includes* non-voters?
For example, 40% voted for x, 35% voted for y, 15% didn’t bother to vote, and 10% never even registered. Yay, x wins! 🙄
The problem seems analogous to trying to count unemployment. You can know people left the workforce but it's harder to get a good idea of why. If people never engage with the system, if anything that seems like it could be a stronger vote against the system than if they had been engaged and stopped for some reason. In the latter case maybe they got engaged because of some compelling candidate. How many people registered just to vote for Obama?
A great chart would have y-axis upper bound be population represented in the district, then stacked ghost bars would be registered voters in parties & stacked solid bars would show actual votes
I think turnout would be 100% if we could only vote up 1 candidate but could vote down as many as we want. Voicing displeasure removes a mandate
@scott @freakazoid @ted the “honest election results” by Welt for German election of 2017 https://www.welt.de/politik/deutschland/article168983338/Das-ehrliche-Wahlergebnis-der-Bundestagswahl.html
@scott I did a little thing for a local mayoral election here in Austria, showing the shares of the winner and runner up based on the eligible voters (came up as turnout was about/below 50 percent): https://www.drawingdata.net/ibk2018/ (from left to right it shows eligible voters, turnout and votes for the leading candidates, the inner circle is the run-off-result)