The Dutch king closes part of his estate to the public for a number of months every year. There is increasing resistance to this, both among the public and in national politics, because the king receives a subsidy for the maintenance of the area. The closed area is actively used by walkers and cyclists for the rest of the year. To find out if the public was concerned about the lockdown, we analyzed data from 300 athletes on Strava.com and showed that their use had already fallen by 97% one week after the closure.
The article has been read 15,000 times, making it one of the most read articles on the day of publication. In addition, the methodology used is explained in detail in a separate publication so that other journalists are able to make a similar analysis.
This manual was also promoted by Jerry Vermanen (a prize-winning Dutch data journalist) in his newsletter with 800+ subscribers, including many Dutch journalists.
In this way, the impact goes beyond just the readers of the original article.
Web scraping was used to collect data on the number of sports activities in the area. To begin with, 300 athletes in the area of the subject have been identified on Strava.com. 1,400 activities were downloaded from these athletes. The data processing was done with R. Several techniques were then used in QGIS to determine how many athletes entered the restricted area. Flouris.studio was used for the visualization. By using a photo slider, the public could easily see for themselves the impact of the king’s decision.
What was the hardest part of this project?
The hardest part was getting the data to back up the story. Data on how much the area is used is not available. That’s why the idea arose to use the athletes’ data on Strava.com as a proxy. However, this data cannot simply be downloaded or scrapped. We had to develop our own approach for this. In the end it came down to creating a number of different profiles on strava.com and then looking for and following athletes from the area. That gave us the opportunity to then download and analyze their activities. This is described in more detail in the manual published as a teaching aid for other journalists.
What can others learn from this project?
How data from social networks like Strava.com can also be a source to substantiate a story. How web scraping can help to collect a lot of information from a website. How open source geographic software like QGIS can do analysis based on location data.