The Dos and Don’ts of Data Journalism
Blastland, along with fellow data journalists Mona Chalabi of The Guardian’s Datablog, Dan Knowles of The Economist and Nicola Hughes of The Times, chaired by Conrad Quilty-Harper of Ampp3d and formerly The Telegraph, had been brought together by Grapevine to give aspiring journalists an insight into the industry. The evening was a follow up on the organisation’s first event in April 2013 which brought together the country’s top student newspapers.
Read highlights of the second panel discussion here.
Quilty-Harper started the discussion by asking Blastland how data journalism had changed since he published his book The Tiger That Isn’t in 2008.
“The origin of the data does get better, we have a lot more people watching it for a start…[but] there’s huge amounts of uncertainty in recording numbers, there’s great difficulty in the interpretation numbers that go up and down all the time…there’s a lot of data.”
Hughes said that part of a data person’s job is trying to find out where problems can arise and to be constantly asking questions.
“A data person would be able to see whether the numbers are telling the truth or has an agenda. It’s about really understanding the integrity.”
Knowles said that a lot of his job is looking at data, which has already been told, and debunking it:
“Mostly it’s just how do you get past the headline statistic and digging through spread sheets and finding a trend that nobody’s spotted… You have to self police and make sure that something that looks brilliant and gives you a fantastic statistic isn’t actually a blip.”
“Interrogating the data is an intrinsically journalistic activity. You’re checking verifying, finding out whether it’s true essentially.”
Chabali said that the key to data journalism is going deeper into the story than just the data and interviewing people on the ground:
“They provide us with the backstory of the ‘why’ because so much of what we do is just describing ‘what’.”
“You have to combine it with interviews, it’s not enough to have a spread sheet and go ‘oh this is really interesting’… you have to start with the spread sheet but…then you go visit somewhere and you interview people and then you write the story.”
Oliver Franklin of GQ in the audience asked how has the Internet and social media changed the representation of data?
There are many more ways to tell it Knowles replied:
“The freedom of the internet is that you have an unlimited amount of space. You can have this story told through data visually as well as the text underneath it.”
But with so many ways of representing the data do all journalists need to be able to handle data to some degree?
“What you need is more data journalists to let you deal with the raw ingredients and not wait for the press release or the end result statistics…what they’re [organisations] doing now is they’re releasing the data raw… saying ‘we’ve done what we said we’d do we’re not hiding anything’. The problem previously was access to the data, now it’s too much data.”
Quilty-Harper ended by asking the panel, what are their dos and don’ts for those wanting to get into data journalism?
Do have an instinct to not merely describe – also analyse why.
Don’t be afraid of being able to master different things, you can’t just be a good writer, you can’t just be familiar with spread sheets, you need to know the basics of coding, you need to know several different tools (or know someone who does).
Do not be arrogant – not only checking other people but check yourself.
Learn to use the ONS (Office for National Statistics) website – that will give you an advantage.
Learn how to find statistics quickly.
Learn how to pick a statistic that’s valid and that can debunk or prove a story.
Do not feel you need to be taught something to be able to do it, do not rely on anyone else to teach you – Google it. There are so many free resources.
“Don’t be seduced by the glamour of exciting flashy stuff, remember that you can produce rubbish very easily and seductively with all those techniques. If you do not have the skills of statistical inference to make sure that you are saying something legitimate, all the rest is rubbish. Exciting rubbish.”
Following the success of their events, Grapevine are launching a data-focused site in the coming months. Get in touch with Harry Lambert (@harrylambert1), Max Benwell (@maxbenwellreal) or Rebecca Choong Wilkins at [email protected].
Watch and listen back: