Data journalism has transformed how journalists share stories and is particularly effective in reinventing election coverage.

During last week’s INMA South Asia Webinar, The role of data journalism and visualisation in election reporting, Gurman Bhatia, founder of India’s Revisual Labs, shared insights into how journalists can leverage data journalism to cover elections in India.

Madhavi Sekhri, head of INMA’s South Asia Division, explained the current Indian elections provide media companies with the ideal opportunity to explore data journalism: “It is reshaping the way news is gathered and presented. The primary role of data journalism is to provide context for new stories, helping readers understand the significance and implications of events or issues — or elections, in this case.”

The challenge is not in finding the data — there’s plenty of that available — but rather in assembling the data cohesively and presenting it in a format that the audiences find credible, engaging, and appealing, Sekhri said. Bhatia, whose company specialises in merging data, design, and storytelling, assured attendees, “There’s nothing overwhelming or scary about data. Data is beautiful and today is just about  that.”

Why use data?

In the ‘90s, a new idea emerged in journalism: computer-assisted reporting. Today, Bhatia noted, all reporting is computer-assisted, but that doesn’t mean it is only about numbers and statistics.

Quoting from a book published in 1999, she said: “It remains one of the best ways available to check tips, generate new sources, and help ensure that the moving human stories we tell reflect the larger truth…  it’s just one more layer to add to your story.”

When data is paired with people, it creates a stronger, more trustworthy story, she said. And while the technology used to collect data is new, the marriage of elections and data is not. To prove that point, Bhatia shared an election map published in The New York Times in 1896.

Using data to tell stories checks two important boxes for readers: exploring and explaining.

“It can help you explain concepts and trends that might otherwise be hard to explain,” she said, adding it can also lead to additional stories readers find relevant and allow them to take a deeper dive.

Discovering different data sets

Journalists have numerous ways to approach election data. Bhatia outlined several sources that can be used, with the main one being the Election Commission of India (ECI), which offers access to both current and older elections. And there are many options for stories:

“Within the ECI, they have a dedicated site for the 2024 election where if you were to try and find voter turnout by state or by constituency, that’s something you could do.”

Local election sites are good resources for information more relevant to individual states, such as showing the location of each polling station.

Another popular data source is the Association for Democratic Reforms, or ADR, which has a WhatsApp channel and provides excellent information on the candidates, including criminal cases, education, number of women candidates, and more.

“So if I were to pick a constituency, it’ll give me information on candidates from that seat, and that’s the most popular use case of mine,” Bhatia said, adding it also allows users to go deeper: “If you went to that person’s profile you’ll also get information on how much gold did they declare, which cars did they declare, and stuff like that.”

Finding new angles for election coverage

One way to think about data around elections is to consider what researchers are writing about it. Bhatia suggested doing searches on programmes such as Google Scholar and Microsoft Academic to see what has been said about certain issues: “When researchers do these sort of projections and analysis, that can be one way of incorporating it in the work that you do.”

Covering the polls and explaining the methodology behind that coverage provides a wealth of story opportunities, as can looking at some unexpected aspects of India’s elections.

“If you can think of things to talk about within the process of elections, and then there are some things that are unique to India — those become interesting as well,” she said.

For example, in 2017 she wrote a story for a state election where she looked at candidates with the same name. “So essentially your rival party might pay a person with the same name to stand as an independent to include voters so that the vote gets split. Fun things like that happen, so those are quirks you can look for.”

How to improve data-driven storytelling

To make stories “pop,” Bhatia offered three tips:

1. Make the data playful and engaging. This means finding ways of letting the reader insert themselves into the story and understand how they might be affected by an issue. That’s more effective than “just shouting numbers at them.”

2. Make data beautiful. She shared an example from The Washington Post, which created a mini-golf game that helped explain gerrymandering. “They just made a game out of it and that’s a way to make it both playful and beautiful.” 

3. Provide context. “Make data meaningful with context —that’s the primary duty we have as journalists, right?” She noted numbers often mean little to readers on their own, but telling readers what has changed — or why — can go a long way toward providing perspective. “Bringing [data] into context is something that will add layers to your data and your story,” she said.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .