Open data
Conversations with Data: #5
Do you want to receive Conversations with Data? Subscribe
Hello again! Welcome to the 5th edition of Conversations with Data.
Each conversation, we’ll be featuring exclusive data journalism advice, sourced directly from you or an expert in the field.
Last time, we asked Journalism 360 about immersive storytelling. Today, we’re talking open data.
What you said
The term ‘open data’ might imply that it’s readily available, but it was clear from your submissions that accessing relevant data isn’t without its challenges.
Fabrizio Scrollini shared some examples from Ojo Publico, a Peruvian news agency, which created a website to monitor the country’s health service delivery. Although they were able to access data through freedom of information (FOI) laws, it took a lot of effort to gather the data through paper trails, before getting lucky with information hidden in a few pdfs. (Eick!)
When reporting on child disappearances Pinar Dağ also faced issues getting data through FOI requests. Her tip: "it is always important to create an alternative open data source".
But, how to do this?
Before you start, said Jonathan Stoneman, ask "who would have this data?", "where would they put it?", and "what might the dataset be called?".
Eva Constantaras also provided suggestions: "use global databases, find publications from local universities, check out the research being produced by local think tanks. If government data is unreliable, find alternative data sources to get the most accurate data available, just like you would when interviewing different sources."
In short, she said, "start with the data you have, not the data you would like to have".
Becky Band Jain, from the Centre for Humanitarian Data, added: "it’s helpful to first have a hypothesis of what you’re looking for in the data, i.e. the story you want the data to tell. For journalists working with the datasets available on the Humanitarian Data Exchange (HDX), we’ve tried to make it easy for anyone to search for topical datasets or find datasets specific to each country".
And then there’s the trouble with data quality. The European Journalism Centre’s own Eric Karstens knows about this all too well from his work on the YourDataStories project. He explained that "hardly any dataset is entirely error-free", but "even the tiniest inaccuracy may undermine trust in open data wholesale".
With this in mind, "don't assume the data is correct, and make follow up calls," said Paul Bradshaw.
"When we worked on the unsolved crime data a couple police forces said that the data was wrong, and resubmitted it. When The Bureau Local worked on a story on councils dipping into reserves, they found a bunch of them were 'inaccurate and unchecked'. Here's another example on ambulance waiting times."
But, he warned, "if someone says it's wrong, don't take their word for it. One of the forces who said the data was wrong couldn't really back that up."
Engaging with other journalists and the open data community can also help you overcome these challenges.
Nikesh Balami (CEO, Open Knowledge Nepal) and Giuseppe Sollazzo (founder, Open Data Camp) both recommend collaboration as a means to overcome common difficulties working with, and accessing data. Some good places to network and find potential collaborations include chapters of the Open Knowledge Foundation and open government networks.
You should never be afraid to reach out! Mollie Hanley, of OpenCorporates, sees it as the responsibility of open data projects to connect with wider audiences and share skills. This, in turn, can help them measure impact.
While we’re talking about collaboration, a lot of you mentioned the need for journalists to open up data from their own projects.
From Anastasia Valeeva, co-founder of School of Data Kyrgyzstan: "I consider it crucially important to publish our own datasets under the open data license together with the story. Not only does it make your story bulletproof, it also shows that you hold onto the principles you require from the others".
Gianfranco Cecconi went further to say that "journalists should be concerned when they are not using open data". When journalism is based on non-open data, "the reader won’t be enabled to 'check the calculations' for themselves and perhaps get to a different conclusion".
In this way, Natalia Carfi, Deputy Director at Open Data Charter, pointed out that open data can help in the fight against fake news.
Our next conversation
I don’t know about you, but we’ve been eagerly following the World Cup through all the brilliant data journalism that it’s inspired.
With the final coming up on Sunday, we’ll be putting together a special early edition of Conversations with Data next week.
Until next time,
Madolyn from the EJC Data team
Time to have your say
Sign up for our Conversations with Data newsletter
Join 10.000 data journalism enthusiasts and receive a bi-weekly newsletter or access our newsletter archive here.
Almost there...
If you experience any other problems, feel free to contact us at [email protected]