Journalists love a good story. (Who doesn’t?) And in the world of journalism today — a world wrought with data wars and “fake news” crisis after crisis – we as statisticians have a unique opportunity to shine.
So why aren’t we?
I’ve been working with journalists for over a decade now. I’ve seen them quickly and accurately grasp statistical concepts. They hold on to them; remember them, and use them in a moment. They pass them on effectively.
There is a real need for statistical literacy in the journalism sector. More than a need, there is a hunger. The concept of data journalism isn’t new, but the audience for data stories is growing — and statisticians are not the ones leading the way.
Listening and Telling Stories
I’m a P.Stat who loves the data journalism community. There is much to learn from the ways journalists think about and use statistics and much that we as statisticians can contribute as well. It all comes down to listening to and telling stories.
We need to make a real effort to craft the communication of statistical concepts in ways journalists can use and understand. More specifically, we must let go of the need to be viewed as specialized experts. We must be willing to take the magic out of statistics, step out from behind the curtain, and work in partnership with the people telling the stories.
In short, we need to spend less time pointing out what data journalists are doing wrong and more time highlighting what they’re doing right.
A Week at NICAR
I spent a week in early March at the NICAR — short for The National Institute for Computer-Assisted Reporting (the name is a bit of a throwback) conference. The event is an annual gathering of data journalists sponsored by the IRE — Investigative Reporters and Editors — and it showed me some of the fascinating ways journalists are developing to access and understand data.
At one panel, someone made an interesting comment about how the status of communication in journalism vs. statistics is having a profound impact on direction. Effective data visualization, this person suggested, garners high status in the journalism world, while remaining a rather minimally valued skill in the land of statistics.
That fact is just one of the reasons journalists are leading the charge towards bringing effective data visualization to non-academic audiences. We in the statistical community have been slow to pick up on the value of interactivity — but it’s not too late. There is still a real opportunity for statisticians to get involved. We can join in by supporting, lending expertise, and even educating ourselves on how to tell better stories with data.
Data Journalism Success Stories
The data journalism sector is a great place to work. In my years collaborating with journalists, I’ve been able to work with some fascinating data and some clever, curious people. I’ve had a lot of fun and success helping journalists find data and use it correctly.
The work is basically applied statistics; helping journalists locate and access data, showing them how to clean and reformat it, highlighting the various strengths and weaknesses of data and analytic choices, and explaining what questions can (or cannot) be answered in different statistical ways.
Journalists are educating the world by making the data they uncover engaging and accessible to the general public. I’d like to share a few examples of data journalism pieces that have really captured my attention:
When journalists at the Atlanta Journal-Constitution learned that two-thirds of doctors disciplined for sexual misconduct in Georgia continued to practice, they used machine learning and Bayesian methods to uncover how the medical community allows thousands of repeat sex offenders to practice medicine.
Reveal News examined satellite imagery and government data to determine how and why the deadly blazes in Lake County, California, spread so rapidly, and how future firefighting methods will be adapted because of those fires.
Fusion took a deep dive into the FBI’s arrest data to determine where American law enforcement is failing the communities it is supposed to serve. They’ve mapped out arrest rates by county and can even tell you what your chances are of being arrested, based on your address and ethnicity.
The Washington Post’s Wonkblog team collected their own data on UberX wait times in various parts of Washington, D.C., to determine if riders in predominantly white neighbourhoods experienced significantly short wait times than those in other areas.
Be a Part of the Data Revolution
We as statistics professionals have so much to contribute to the world of data journalism. There is a hunger for data stories, and we are uniquely qualified to tell them. It’s time to step down off the pedestal of professional pride and work in partnership with the talented storytellers and journalists who are bringing data to the masses. It’s time for us to be part of the revolution.
Are you a data journalist or nonprofit looking for a data partner? The team at Datassist is happy to work with journalists, nonprofits, and social sector organizations to maximize the value and impact of your data. Get in touch with us today.