In May this year, Datastory launched the service Election Tracker. In this blog post, I wanted to shed some light on how it's being maintained, and how crowdsourced information from Wikipedia and its sister projects can be utilized without sacrificing accuracy.

The idea behind Election Tracker is simple: list all upcoming national elections in the world on a timeline along with some basic context. With this service, we want to surface blank spots in the traditional media's global coverage on politics.

What's not simple is maintaining the data. While some countries announce the date of their next general election years ahead of time, others set it just months before due date. On top of that, you have snap elections, postponed elections, coup d'états, and other complicating events.

Wikipedia to the rescue

It's in situations like this that Wikipedia really shines. Leveraging the power of thousands of contributors, the online encyclopedia maintains articles like 2021 national electoral calendar and List of next general elections.

That's only the start though. In order to have a complete overview of which countries do and do not have information on their next general election, you need to be able to query the dataset. A simple list cannot answer questions like which countries are missing from it, or whether there are any internal inconsistencies.

And its little sister

Enter Wikidata, a sister project of Wikipedia containing structured data, as opposed to text written for human consumption. Our first queries to Wikidata returned a few dozen items relating to upcoming elections. That's a good start but there were still too many gaps. To patch them, Datastory's editors decided to go through all countries and update the information on their next election, and ever since, we've made sure that once a country holds an election, a new date is entered for the next one.

What's beautiful about this setup is that we are not the only ones working on keeping this data up to date. We are working in tandem with the whole Wikidata community. Working together we can fill the gaps and fact check everything before we import it to our own database where we can visualize the information from multiple angles, helping journalists and citizens around the world to learn more about upcoming elections.