By  SIMON ROGERS

Published on 18, December, 2017

Data journalism has come a long way since I wrote this piece identifying the key traits of the field. And it has never been more important. At a time when the very notions of truth and lies and facts are under attack and disputed every day, data journalism can provide a guide to the mysteries of the daily news cycle.

But right now, what does it look like?

1. Still the new punk

Every year, a group of data journalism students contact me to ask if I still think data journalism is the new punk.

The answer is yes.

There is no other area of journalism where the barriers for entry are so low. There are tools to make maps, clean up data and make charts. Datawrapper, OpenRefine, Carto… there’s a tool for every purpose, all of which are free or low-cost. Yet, for an established field, data journalism is still remarkably unsupported.

And in many newsrooms, data journalists are still in a company of one, often without support, career structure and yet produce work in innovative ways every day. Being scrappy and thinking quickly are part of how data journalism *has* to operate in those environments.

So, truly, anyone can do it. But that doesn’t mean everyone can. A story badly told, is still a story badly told. But you don’t have to be part of wealthy huge newsroom to be a great data journalist.

2. Not about being a coder

That openness means you don’t have to be a developer to make great data journalism. Of course, it can help — the best work is the result of collaboration between those with coding skills and those with story telling skills.

But there are ways around this: Flourish, built by Kiln (which the Google News Lab is making open to newsrooms), allows you to re-use visualizations in a way that wasn’t possible for non-coders. It allows data editors and designers to share their visuals with non-coders.

The great thing about data journalism is that, although some level of coding is eventually inevitable (whether you think you’re doing it or not) you can do as much or as little as you want. It’s a result of necessity — if you want to get something done, you have to do it yourself. One example is John Burn-Murdoch at the FT — a great data journalist who has also taught himself to be a great visual journalist too through coding and designing.

3. Working with big data from big places

In the past, data told us the size of the economy, or the demographics of the nation.

But now we have new sources of data that tell us something more: the state of the global brain — the opinions of millions of people is now something we can see in real time. My team works on am tool that provides one set of insights: Google Trends. We’ve only had real time data for the last two years and the field of search data analysis is comparatively new, which means we’re still exploring new ways of understanding that data. The Pew Research Center have recently conducted this detailed exploration of using search trends to tell the story of Flint, Michigan’s water crisis.

There is never less data. More data breeds a greater need for data journalism to help interpret it.

You can read more about the data that I work with here.

4. The 2016 US election did not kill data journalism

On November 9, 2016, it became fashionable to declare data journalism dead and buried.

I don’t agree. If anything, the election and its aftermath have made data journalism even more vital.

Firstly, predictive reporting and data journalism are not always the same thing. Just because they both use numbers, doesn’t make them identical. But elections are like black holes consuming all the data resources in any organization covering them. And in 2016, that effort often centered on trying to predict the results in advance.

Despite all that attention, there was a lot of other data journalism out there which used the numbers to tell stories in other ways. Projects like Electionland (more below) or the work of data savvy reporters like Philip Bump at the Washington Post showed that there was more to election based data journalism than trying to predict the results.

5. Still about being open

The best data journalism is still the most open and transparent, and there are now many examples that tell that story. La Nacion in Argentina is a model of open data journalism and this year won the prize for its approach to opening up public datasets in a country with no FOI laws and a long history of limiting media access to government information.

Excesses Unpunished, by Convoca in Peru, opened up public data to help its users understand the country’s mining industry better. India Spend, created its #breathe project using sensors to measure air pollution.

Often it’s about revealing that data in environments which are not conducive to transparency. Take KRIK’s online work. The small team created a comprehensive online database of assets of Serbian politicians, launched to help Serbian citizens to better understand the people running their country.

6. It’s about collaboration

All journalists are competitive, however data journalists are particularly willing to work together. This has the advantage for smaller organisations and teams of combining resources.

In fact, there are very few fields of journalism as collegiate as data journalism, maybe because so many of us work in isolation day by day.

Electionland, which our team was involved in, was the biggest single day data-driven reporting exercise in history. Over 1,000 reporters from hundreds of organizations investigating voting related issues on election day 2016. If there had been voting fraud, this exercise — which compiled tip-offs, search data and reports with real time verification — would have found it.

The follow-up to Electionland is Documenting Hate. This collaborative project (supported by the News Lab) is an attempt to bring data to an area where there is none: hate crime reporting. Data journalism has to do one important thing to prove its worth: it has to matter. This is where Documenting Hatecomes in. The project, which includes a number of different news organisations and journalists, is designed to change that by collecting, and verifying, reports of hate crimes.

If you have a lot of journalists working out there in small teams or solo, often developing complementary skills, then bringing them together makes sense. It’s a way to share those skills and make difficult stories stronger.

7. Memory is important

Data journalism didn’t start in 2009. Nor even the 1960s and 1970s with the advent of ‘computer assisted reporting’, beyond even the Detroit riots work of Phil Meyer to the work of Horace Greely the journalist who caused Lincoln so many problems in his early political career. Or Florence Nightingale. Or the first ever issue of the Guardian. It’s as old as the weather report, the stock ticker or the baseball scores.

Obviously, things have changed (not least the threatened nature of the news ‘paper’ itself). The world of social media, digital editions and videos is very different to the world where data journalism was born. But those lessons still carry weight today.

Yet, that work is under threat through code rot and a lack of appreciation of the past. Until there is a data journalism archive, that responsibility falls to us to make sure that we protect the collective knowledge of a field that desperately needs it.

8. Design still matters (in different formats)

While it’s true that a lack of technical skills is no barrier to producing great data journalism, working with designers can always make anything better. Not least because it has become a lot more challenging to do so. At least half of all news readers consume it on mobile, for instance, according to research from the Reuters Institute.

There’s also a perceptible shift away from producing big interactive graphics — with some in the industry questioning whether they should exist at all.

One approach is to take visual techniques that were formally seen as over simple, and give them a new sophistication. Take the work of Lena Groeger in creating more and more sophisticated animated gifs. In effect rendering them as visual tools that can tell complex data stories. It’s worth checking out her tutorial, which shows how this is a field anyone can get into.

9. Human (and fun)

The capacity of data journalism to make a story desperately depressing and sad is, of course, endless. And sometimes necessary. The facts make a story what it is.

But making that data human, that can make it even more powerful. And there are people doing something about it. Mona Chalabi has built up a following for charts which convey often complex stories in ways that make you laugh and was recognised for it.

Giorgia Lupi and her partners at Accurat do the same too. We worked together on WorldPotus before the 2016 US election, which showed how the world was searching for the two candidates and the big issues. The visual is deliberately playful and enticing — something that could otherwise be dull and worthy was transformed into a game.

What ties these two examples together is that they are not patronizing or inappropriate. They work by telling the story in the best way possible, which is the mission of all data journalism. I wrote in this piece about the importance of empathy in data journalism.

Traditionally, a lot of data journalism and data visualisation has not been big on emphasising humanity. Former New York Times developer Jacob Harris wrote a prescient piece in 2015 calling for more empathy in data journalism. He cited ProPublica’s Scott Klein in talking about the importance of the ‘near’ view, in addition to the ‘far’ view that data journalism often takes. In other words, data journalism can often seem abstract and removed from your personal life. Bringing it closer, the ‘near’ view, is what makes it real and concrete. It’s the difference between flying over a place and remarking how the cars look like toys and being there on the road in the back seat of a taxi. One is remote and removed, the other is happening to you.

That empathy is a vital part of what we do.

10. All about telling stories

If data journalism is about anything, It’s about telling the story in the best way possible. This is what I wrote back in 2011 and I would argue it is still valid today. Sometimes that will be a visualisation or a map, sometimes it’s a news story.

Sometimes, just publishing the number is enough.

If data journalism is about anything, it’s the flexibility to search for new ways of storytelling.

Above all, it’s just journalism.

This article is being republished from Data Journalism Awards website with due permission. 

 

 

Adulteration Aged care AIDS ambulances and emergency care birth deliveries blood banks Budget Cancer Cancer cases in India Cancer in South Korea cataract cervical cancer China Cholera disease in India Communicable diseases Comparison of India in Government Health Expenditure Dental Colleges in India Dentist Diabetes Diabetes in Asia Diabetes in India Disease in eastern states of India Diseases Doctors Domestic General Government Health Expenditure Drinking water Expenditure Fertility rate Food adulteration Food testing labs Funds GGHE-D per Capita in India Global Data Government Medical Colleges H1N1 Health Expense Report Health Infrastructure Health Survey Healthcare Education Hepatitis HIV Hospitals Immunisation Immunization coverage Infrastructure Inoculation Insurance Kala Azar disease Life expectancy Maternal health maternal mortality Medical education Medical Research Mental Health Data Milk Adulteration National Organ and Tissue Transplant Organization National Organ Transplant Programme non-communicable diseases NOTTO Organ Donation Organ Transplant Palliative care Pharma PMSSY Medical Colleges Polio Pradhan Mantri Swasthya Suraksha Yojana Public Health Rabies Rabies deaths Registered Dental doctors Registered Doctors with State Dental Council Road Accidents RTI Data Sanitation Sex Ratio Smart Cities Snake bite State Health Expenditure Report in India Suicide rate in Asia Suicide rate in India Technology Tuberculosis Tuberculosis disease in asia Tuberculosis in India Typhoid Vaccination Vector-borne Diseases water borne diseases