Open data was seen for journalists and editors as a solution to rescue an industry in decline, says research by Jonathan Stoneman. Suddenly, there was an access to all sources of information and the need to hire people with different backgrounds to crunch this data.
Simon Rogers discusses on his book Facts are Sacred that this new wave represents a way to save journalism. Journalists can acquire a new role and act as a bridge between those who have the data and the public that need help to understand it.
Even though this new role can help journalists to inform better the audience, professionals differ in how they define data journalism and people who work in this field. Do we focus on data, journalism, or both?
Paul Bradshaw writes on the Data Journalism Handbook:
Data can be the source of data journalism, or it can be the tool with which the story is told — or it can be both. Like any source, it should be treated with scepticism; and like any tool, we should be conscious of how it can shape and restrict the stories that are created with it.
Following the NODA and Tutki!2016 conference on data and investigative journalism, I asked a few professionals in this field to define data journalism. Look at the word cloud below to see the words they used:
It seems to be an endless tug of war between reporting stories and gaining technical skills such as computer programming, statistics, and graphic design. Most interviewees pointed out the importance of using computational skills to process data and answer the ‘what’ of the story.
Stories go first
Even though investigation techniques and coding skills are important, the backbone of data journalism is explaining complicated subjects to audiences.
There are several techniques that help journalists to find stories and acquiring data, such as Freedom of Information Acts, open data or creating own datasets. Peter Sjöholm uses code to gather, clean and read data before turning it into a news story.
Hendrik Lehmann highlights that a data journalist uses data to find new socially relevant information or to tell a story.
Doing journalism with data
The majority of participants point out that data journalism is journalism based or that uses data.
Jens Finnäs says that it is journalism that uses data to “answer the question that you are investigating.”
Andy Dickinson explains that it is the process of doing journalism with data. Even though it seems a genre of journalism such as print, broadcast, mobile or digital; data journalism is “a mix of practice and ideological positioning; the why and the how.”
Henk van Ess writes that it is journalism based on data that has to be processed with tools “before a relevant story is possible.”
Doing journalism with ‘structured’ data
Other experts narrow the previous definition down saying that data journalism works with ‘structured’ data.
Nicolas Kayser-Bril says that data journalism implies doing journalism with structured data. However,this definition is incomplete if journalism is not defined, something that he argues as impossible.
Julian Ausserhofer also highlights the use of structured information by data journalists as a source in their reporting: “they employ computational and quantitative methods from the social sciences to analyze and explore datasets.”
Eirik Stavelin discusses on his PhD-thesis that data journalists take ‘structured’ data as input to produces stories and fall into one out of two categories. The first one is the “old-school CAR-journalist type”, producing text or broadcast news stories based on the “quick” analysis of datasets. The second category is the “new generation computerised journalist” that gathers data both from the old school and scraping from the web.
Knowing methods to research and analyse data
Research on data journalism in Quebec says that “computational journalism aims to enable reporters to explore increasingly large amounts of structured and unstructured information as they search for stories.”
This approach was mentioned by other professionals. Julius Uusikylä says that data journalism uses modern digital tools “to make sense of information around us by systemizing it, making serious journalism searchable and ‘connecting the dots’ in complicated subjects.” If newsrooms have tight budgets, he says that data journalism could also help automating work or know better how and what to cover.
Florian Stalph says that journalists in this field analyse data combining “socio-scientific methodology with journalistic investigation techniques and journalistic questions.”
Mattias Östmar points out that journalists should be comfortable using quantitative, computational research methods as well as qualitative.
These definitions show that journalists should have data-related skills to identify uncovered trends, as a research by Katherine Fink and C.W. Anderson remarks.
Does data journalism need data visualisations?
Some of this definitions include visualising the data as a form of presenting the story. Julian Ausserhofer says that these pieces are usually web-based and have a visualisation component, which can be explored interactively by its readers.
Eirik Stavelin says that data journalists seek to visualise the information on the web or mobile. Therefore, they mainly look for API data sources that enable good formats for visualisations and spaces that allow the audience to interact with the story.
Katarina Lind uses data to make news more unique, correct and interesting. However, she also thinks that this field needs to make data easier to understand and amaze the user with a nice design when possible.
Working as a team
Nicolas Kayser-Bril says that no person can have the best programming, statistics and graphics design skills to deal with structured data.
This lack of computational skills or access to information drives data journalists to seek internal or external collaboration, says a research by Eddy Borges-Rey. Some connections are done through online platforms such as NICAR, the University of Missouri and the IRE.
Some data journalists admit in a research by Katherine Fink and C.W. Anderson that they have “many hats in the newsroom”. Therefore, new job titles are arising such as programmer-journalists, news apps developer, editorial programmer, and hacker-journalist, explains Eirik Stavelin.
Open -and processed- data
Relying only on open data portals might jeopardise journalism because they contain nothing but “friendly” data’. In Quebec, a study sample revealed that only 2% of data journalism projects were based on self-built databases between 2011 and 2013.
Jonathan Stoneman discusses on Does Open Data Need Journalism? that open data is just one of the means that journalists can use for their pieces. They can also send FOI requests or create their own sources. Therefore, it is important to also know if journalists need open data.
Nicolas Kayser-Bril argued at the Noda conference that data journalists should collect more independently data and be aware of the government purposes when presenting datasets:
Being able to access bogus data is pointless. What is needed to make sense of the world around us is better data, free from government interference. To be free to contextualize news, free to interpret events and free to think independently, we need independently collected data.
How many of the following data journalism books have you already read?
I asked the participants to recommend a book for people who were starting their first steps in data journalism. There are two options to test your knowledge in this field:
1.Using a Telegram bot to see who recommended what in less than five minutes!
2.Complete and check this Google form while having a look at the list:
Do you have more examples? Let me know in the comments or on Twitter at @mcrosasb