Data Journalism: Meyerism and Vox's new data team.
A few weeks back I wrote a post about Data Journalism and how it was defined (on Wikipedia at least). So I was interested to read Vox’s take on it when they announced, last Wednesday, that they were creating a Vox data team:
Interestingly for me,Vox co-founder Melissa Bell, sees their kind of data journalism as a direct descendant of Philip Meyer’s Precision Journalism work on the Detroit Riots (1967).
Forty-eight years after Detroit, precision journalism has given rise to data journalism, which has become a much-touted new media trend.
So Vox’s ‘data journalism’ is 21st Century Precision journalism.
Philip Meyer has become something of an adopted parent to data journalism. The work was not just groundbreaking, more importantly in my view, it was disruptive. It was disruptive to the status-quo of accountability – the assumptions made of those about the rioters. It was also disruptive to journalism. Meyers first iteration of Precision Journalism was directly challenging a prevailing form of literary journalism that many saw as undermining truth and trust. It put science before journalistic belief. In doing that it was also part of a bigger disruption of sociology – a new wave. It’s no surprise then that, like a patron saint, he is invoked by any new data journalism project looking to define the data journalism they do.And Meyer is a very useful starting point.
It doesn’t matter what hue of data journalism you might be, Meyer fits. For many , Meyer is CAR through and through. But if you don’t like the hypothesis driven, 20th Century trappings of CAR, well, Data Driven journalism has all the same tech but with a nicely positivist, scientific approach. A reading of Meyer that is just as likely to keep those exploring the boundaries of computational and algorithmic journalism happy.
But as much as Meyer offers an agreed (and agreeable) starting point for those looking to unpick the “much-touted new media trend” that is data journalism, for me it’s the fundamental philosophical approach that Meyer disrupts (and suggests in that disruption), that is more useful as a tool to think about data journalism and what it means.
For me, in trying to get a flavor of what’s driving (those involved in) the data journalism conversation, it often comes down to this – which comes first. The data or the question?
A proponent of CAR informed data journalism would tell you that you start with the question: ‘I know that there are dodgy MP’s there, I need the data to tell me how dodgy’. It’s all about sampling. Your DDJ fan would tell you that by analyzing and linking data we would ‘discover’ that there were dodgy MP’s. It’s about having all the data.
In a Q& A in the comments (nice idea) Bell gives Vox’s perspective on the which comes first question:
It is definitely both. You can start with an idea and seek out data to help answer the questions, or you can start with a data set and surface stories from the changes discovered within that set. Either way, it’s always about being constrained only by your imagination!
So, very much story driven. If we have the data we’ll do something with it.
Editorial Products Engineering Director Ryan Mark steps in to answer a question about the amount of raw data and covers similar ground:
It’s difficult to give a direct answer… it depends on the topic, what data we can get a hold of, and whether that data can help us bring clarity to the thing we’re trying to explain.
Digging for data takes time and doesn’t always yield fruit. Raw data usually comes in drastically different formats and structure and takes work to process and understand.
I think we’ll be collecting as much raw data and we can handle. We’ll have to focus in on the stuff that we think can add the most to our reporting
Both of those answers speak more to the ‘longitudinal’ issues of data journalism than any definition. How will resources and editorial line impact on the way you use data? How long can you stick to a Data Driven approach when resources and editorial line don’t let you gather and develop databases of raw data? For what it’s worth, I think Bell’s comments about the structure of team tell us more about where Vox are going, alluding to a more visual, editorially responsive mode.
I’m excited to see what Vox come up with. As much as anything else, because what they come up with will excite others – they will be saying we want data journalism like VOX. As much as Meyer might be the motivation, Vox and their ilk are now the dominant blueprint.
For me, Vox’s position underlines the importance of Meyer as a reference point; common ground on which to start the conversation and not much else. We can’t say that Vox would be any more or less Meyersian in its data journalism. At best it means* I don’t know where you are going but I do know where you are coming from. *
In helping me understand what data journalism is for Vox, that’s as much as I can ask for.