Why did you get into data journalism?

In researching my book chapter (UPDATE: now published) I asked a group of journalists who worked with data what led them to do so. Here are their answers:

Jonathon Richards, The Times:

The flood of information online presents an amazing opportunity for journalists, but also a challenge: how on earth does one keep up with; make sense of it? You could go about it in the traditional way, fossicking in individual sites, but much of the journalistic value in this outpouring, it seems, comes in aggregation: in processing large amounts of data, distilling them, and exploring them for patterns. To do that – unless you’re superhuman, or have a small army of volunteers – you need the help of a computer.

I ‘got into’ data journalism because I find this mix exciting. It appeals to the traditional journalistic instinct, but also calls for a new skill which, once harnessed, dramatically expands the realm of ‘stories I could possibly investigate…’

Mary Hamilton, Eastern Daily Press:

I started coding out of necessity, not out of desire. In my day-to-day work for local newspapers I came across stories that couldn’t be told any other way. Excel spreadsheets full of data that I knew was relevant to readers if I could break it down or aggregate it up. Lists of locations that meant nothing on the page without a map. Timelines of events and stacks of documents. The logical response for me was to try to develop the skills to parse data to get to the stories it can tell, and to present it in interactive, interesting and – crucially – relevant ways. I see data journalism as an important skill in my storytelling toolkit – not the only option, but an increasingly important way to open up information to readers and users.

Charles Arthur, The Guardian:

When I was really young, I read a book about computers which made the point – rather effectively – that if you found yourself doing the same process again and again, you should hand it over to a computer. That became a rule for me: never do some task more than once if you can possibly get a computer to do it.

Obviously, to implement that you have to do a bit of programming. It turns out all programming languages are much the same – they vary in their grammar, but they’re all about making the computer do stuff. And it’s often the same stuff (at least in my ambit) – fetch a web page, mash up two sets of data, filter out some rubbish and find the information you want.

I got into data journalism because I also did statistics – and that taught me that people are notoriously bad at understanding data. Visualisation and simplification and exposition are key to helping people understand.

So data journalism is a compound of all those things: determination to make the computer do the slog, confidence that I can program it to, and the desire to tell the story that the data is holding and hiding.

I don’t think there was any particular point where I suddenly said “ooh, this is data journalism” – it’s more that the process of thinking “oh, big dataset, stuff it into an ad-hoc MySQL database, left join against that other database I’ve got, see what comes out” goes from being a huge experiment to your natural reaction.

It’s not just data though – I use programming to slough off the repetitive tasks of the day, such as collecting links, or resizing pictures, or getting the picture URL and photographer and licence from a Flickr page and stuffing it into a blogpost.

Data journalism is actually only half the story. The other half is that journalists should be **actively unwilling** to do repetitive tasks if it’s machine-like (say, removing line breaks from a piece of copy, or changing a link format).

Time spent doing those sorts of tasks is time lost to journalism and given up to being a machine. Let the damn machines do it. Humans have better things to do.

Stijn Debrouwere, Belgian information designer:

I used to love reading the daily newspaper, but lately I can’t seem to be bothered anymore. I’m part of that generation of people news execs fear so much: those that simply don’t care about what newspapers and news magazines have to offer. I enjoy being an information designer because it gives me a chance to help reinvent the way we engage and inform communities through news and analysis, both offline and online. Technology doesn’t solve everything, but it sure can help. My professional goal is simply this: make myself love news and newspapers again, and thereby hopefully getting others to love it too.

6 thoughts on “Why did you get into data journalism?

  1. Matt

    I must confess I’m a little unsure of the exact definition of data journalism, but I’d like to share some of my experiences. My personal understanding of this term is obtaining a decent headline through harvesting large quantities of data and following up leads thrown up by that data.

    My experience doesn’t really encompass representing that data in a flash-based online applications or spreadsheets (though these techniques could be brilliant for representing the sort of info I occasionally get my mitts on – just wish I had the chance), but in using the information as the basis for TV/radio reports and the occasional online article.

    There is an awful lot of data out there, often in gov.uk, nhs.uk or police.uk domains, but I’m not convinced that the really interesting or newsworthy stuff is so readily available – public bodies tend to hide embarrassing info or only give it up with great reluctance. Often this info is obtained it via third parties. I’m happy to be corrected on this if anyone can show me where to find scoops by speculatively trawling for stuff which is already out there.

    I have spent a fair few hours at my kitchen table going through reams of paperwork obtained via FoI, trying to find stories or nuggets of information which have prompted further inquiries and further FoI requests, which have in turn led to stories.

    Examples of the sort of info which can be a goldmine for journalists but aren’t neccessarily available online include HSE RIDDOR notifications, Serious Untoward Incident (SUI) reports from NHS Trusts, IR1 notifications from Ambulance Trusts and information entered onto Strategic Executive Information Systems operated by NHS Trusts. These are the things which will reveal medical accidents, staff politics, data breaches, lab accidents, complaints – you name it – but needs to be dug out, prised out of them. This stuff just isn’t searchable on the internet.

  2. Jerrod

    I am currently reading Ken Doctor’s book Newsonomics, and it talks about several of the same ideas that you are researching. With the indefinite future of journalism looming, all signs are pointing to technology and computer-based outsourcing. I have also closely follow Romenesko’s aggregated media posts, and I have seen the same trends emerging.

  3. Pingback: L’informazione vera “è” pallosa. Ovvero il giornalismo dei dati » Scene Digitali - Blog - Repubblica.it

  4. Pingback: Let’s explode the myth that data journalism is ‘resource intensive’ | Online Journalism Blog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.