Project description:

Since the early stages of Germany’s vaccination campaign, the vaccination monitor tracks the progress both nationally as well as in the different federal states, by age groups, vaccines and in international comparison. A modular concept allows us to react to new developments in the pandemic, newly available data or new questions by adjusting or adding visualizations. For example, in fall 2021, substantial parts of the page were revised to account for the new stage of booster vaccinations.

Impact reached:

As our corona coverage is based on the concept of a web of applications, bundled in a compact form in the main coronavirus monitor dashboard and going into different applications and specific pages from there, it is hard to draw clear lines between individual projects. However, the vaccination monitor was and is a crucial element of the interactives team’s continuous effort and aim to provide all relevant information related to the pandemic and its developments. The vaccination monitor alone has been visited over 15 million times in 2021, and the coronavirus monitor, which includes a stand-alone embed widget of the vaccination monitor, continues to be the most successful article in the company’s history.

Techniques/technologies used:

Data is regularly scraped from different sources with automated node.js scripts and written into csv files. The frontend uses these files and is based on a state-of-the-art frontend technology stack using React with next.js, Mapbox for maps, emotion for CSS-in-js, and d3 for data visualization. As most of our users visited the page on their smartphones, we paid extra attention to a good user experience on small devices with low bandwidth. Additionally, we split parts of the story into smaller pieces that could be embedded on other sites using iframes.

What was the hardest part of this project?

As a page designed with a monitoring function, the project poses a number of technical and conceptual challenges. On a technical end, displaying new up-to-date figures every day means that we automated data-collection processes with various scrapers, often with data sources not especially friendly to scraping (or barely machine-readable in general). On top of that, the issuing authorities frequently change formats, locations or content of the tables they publish without prior notice, so the vaccination monitor also requires continuous and often very quick maintenance or fixes.

On a conceptual level, changes in policy, updated recommendations from health authorities as well as progress in what was originally meant to be a linear process of vaccinating most of the population meant constant reshuffling of priorities, old visualizations becoming obsolete and/or that the need for new visualizations arose, especially as more data was made public or more granular data became available. Major events that made us rethink and rework large parts of the monitor include the moment vaccinations were available to everyone rather than to those belonging to a priority group, changes in the communicated national goal (% of the population vaccinated) as more infectious variants arose, the moment booster vaccinations became necessary and kids could get vaccinated as well.

The quality and validity of data was also a constant question accompanying us, as the methodology behind recording data was not always consistent across time and space (e.g. sometimes vaccinated people were assigned to multiple priority groups, other places just one) and in other cases, the was a discrepancy between reality and the official plans or recommendations that made interpreting data difficult (e.g. officially only people above 12 should be vaccinated, but there already were frequent reports about younger kids receiving shots as well).

What can others learn from this project?

As with all our Covid-19-coverage, this project taught us about creativity in data-driven journalism: even when the overall topic is the same for a long time, it is still both possible and crucial to develop innovative formats that are tailored to users’ specific needs at a specific moment in time – and these change constantly. Modular set-ups provide a lot of benefits with this: they allow creating go-to places for users which can be updated, adapted module by module, expanded and so on, and therefore stay an optimum source of information and stable source of information while adapting to dynamic developments in the pandemic.

An important approach in constantly finding new interesting angles to public data was to match it with other information in order to generate new insights. For example, in the early phase of Covid vaccinations in Germany, the official data included indicators on the professions or age groups of those vaccinated. Rather than just visualizing the published data (how many percent of those vaccinated work in healthcare), we researched data on the different federal state’s approximate numbers of healthcare workers, pensioners, pensioners living in care facilities a.s.o. so we could give an idea about which percentage of these groups had received vaccinations.

We also aim to continuously question standart visualizations and adapt or find new forms of depicting information. With the vaccination monitor, we for example created an abstract queuing line showing the different groups of prioritized people (when vaccinations weren’t available to everyone yet) and depict the numbers of vaccinations in a calendar-based tile plot and bar-chart, making it more accessible and relatable than abstract standard graphs.

Project links:

interaktiv.morgenpost.de/corona-impfungen-deutschland-bundeslaender-weltweit/

interaktiv.morgenpost.de/corona-virus-karte-infektionen-deutschland-weltweit/