This post is part of a blog series and originally appeared over at my friends’ from Idalab, a Berlin-based specialist on data science and machine learning.
We live in exciting times. Not many generations in history do get the chance to witness a technological revolution as profound as the current one. While it can be challenging to constantly adapt to new devices and services, there’s also the thrill to observe just how the new technologies change the ways we go about things.
One of the fields that is literally turned upside down these days is journalism. Newspapers and magazines have a history of many decades, their first ancestors appeared in early modern times. For a long time, the technology of printing has shaped how they investigated and edited stories. Now, digital publishing allows for completely new approaches to reporting.
One such new method is subsumed under the term “data journalism”, and as with many fancy terms, it leaves many people wondering what this is supposed to mean. At a very basic level, it means that news outlets are increasingly tapping complex statistics as an independent source for new stories.
At first sight, this may not sound like a revolutionary innovation. We’ve had statistics for a long time (the field began to rise as a science in the early 19th century), and we had newspapers for a long time. Why would it need a digital revolution for the newspapers to pick up on statistics?
Data journalist Simon Rogers, who is currently Data editor at Google News Lab in San Francisco, has attributed this change in journalistic methods to the change in the available tools. Statistics had very well been publicly available before the advent of the computer, but the data was buried in books and printed documents – not an accessible format for it to be analysed or visualised any further.
It was only when data sets were published in a machine-readable format and personal computers came with software that allowed to “crunch” the numbers that journalists began to analyse data as a source for their reporting. Simon Rogers, who was a leading figure in establishing this new genre of journalism when he edited the Guardian Data Blog a few years ago, has described how the new tools and the availability of data — from big data leaks to public data deposits such as Data.gov — have sparked activity among investigative journalists:
“It’s hard to know what came first: the data or the demand for it […] In the past, when we all relied on official bodies to tell us what we needed to know, it didn’t matter if the data was aggregated and analysed for us. But now… we want to know the numbers behind the story for ourselves – to see if we’re being told the truth and discover our own stories.”
Over the past years, the field of data journalism has diversified and grown exponentially. Many news outlets are engaging in this field of reporting, with the topics ranging from national statistics and election results to local reporting on issues such as housing, urban noise or service themes such as the clock change piece above.
Munich-based Süddeutsche Zeitung has published a piece which beautifully exemplifies a general mind set that data journalism has helped to bring about: using data to look behind the well-established public narratives. Based on the notion that a national parliament should represent a country’s population, and that it could do that best if all demographic groups are proportionally represented, the interactive piece analyses the German Bundestag according to gender, religions, age, education and rural vs. urban population.
The piece not only shows how data journalism can provide in-depth background information which adds to and enhances the minute-to-minute reporting of current events. It is also a good example of how many data journalists strive for data transparency – in this case by adding a whole “methodology” note at the bottom detailing how the statistical results were obtained.
All images are screenshots from the respective news sites. You can visit these sites by simply clicking on the images. Simon Rogers is quoted from Rendgen/Wiedemann: Information Graphics, p. 60.