So you finally have that spreadsheet or database that you’ve long sought. How do you then turn globs of data into a story that people will want to read?
Your newsroom may have recently seen the value of data journalism and are ready to incorporate or integrate datasets into storytelling. At first, the task may seem a bit daunting, but there are various online articles to get you started. One of them is the online Data Journalism Handbook that can be freely copied and is licensed under the Creative Commons Attribution ShareAlike license.
Below we share the process of acquiring the data you want, to interviewing and fact-checking the information to writing the final product as shared by Rox Nixon — a New York Times Homeland Security Correspondent in the United States. He shared these tips at the most recent African Investigative Journalism Conference in South Africa.
Nixon shared these tips through a breakdown on how he carried out the investigations into the border patrol bribery’s that took at least US$15 million bribes in the last ten years. First, he explained that before he began his investigation he had to have what he calls “a data state of mind” and this means that always assuming that the data you want for a story is out there. You just have to figure out how to get it.
He asserts that in investigating a data story it is important to identify which data you need to collect for the story to come together, after collecting the data it is also important that it is cleaned, sorted out and fact-checked before it is used as part of a story.
“Journalists should use data because it allows them to see the big picture of a story. It expands the story from the competing ‘he said/she said’.
“It allows you to find stories that you might otherwise miss, it shifts the focus from looking at one bad fruit to the entire barrel, and it puts the reporter in control, rather than sources,” Nixon said.
Here are seven tips for journalists wanting to venture into data investigative journalism
1. Data journalism 101
- Start with a story in mind (A story idea will lead you to the data you need. The story drives the project, not the data).
- Ask a basic question (A question will help you figure out why the story is important, for example in South Africa if you want to know how much money was stolen from VBS Mutual Bank, then you can start collecting data to answer this question).
2. Determine who has the data you want
- Will the government agencies, non-governmental organisations, websites have it?
- Can you get the data commercially?
- Once you have the data, do you need to build the database from scratch yourself?
3. Acquiring the data from those who have it, ask these questions to figure out the processes of where and how to get the data
- Can you get it for free?
- Will you have to file an open records request?
- Can you get the data from a source?
- Can you get the records to build a database yourself?
4. What tools do I need?
Use spreadsheets for instance, Excel and Google Sheets, to clean up the data and put it in a readable format. Use embedded relational database/database management system such as SQlite. (SQlite is a free platform, but it is not easy to use so you might have to read up on how to use it or watch Youtube SQLite3 Tutorial videos that may provide steps on using it for beginners.)
Also to elaborate and paint a picture with the data you have, consider data visualisation tools like Tableau. Take a look at this review of tools that can be used to visually enhance your stories.
5. Now that you have the data, what’s next? Ask questions about the data.
What’s in the data, who collected it? Over what period of time was it collect? What’s missing from the data? Is it complete enough to tell the story that you want to tell? How much cleaning will you need to do to use the data?
6. Before you write the story you have to humanise the data
Who are the people to best illustrate the information in the data? Are they in the data? Do you need to find people to interview to illustrate the data?
7. Crafting the story, think of a narrative, what or who to lead with? How much space will the data take in the article?
Consider the best lead that would best suit your story. Will it be a hard news lead or will it be an anecdotal lead?
Find good anecdotes and good quotes to add to the story, and maybe consider whether there is a “donkey” you can hang the story on?
“Remember to not load the story with data, the data is not the story, it aids the story,” said Nixon.