Data visualization is one of the most important skills to learn in contemporary journalism, two prominent journalists agreed.
Bloomberg Terminal helps professionals analyze data and place trades on the electronic trading platform. Credit: Carlos Osorio | Toronto Star | Getty Images
“The best-paying jobs in the media and communications require you to use data and statistics,” says Lee Miller, a senior editor who has spent nearly 30 years at Bloomberg News.
More and more data appear in our breaking news stories, including coronavirus infection cases, U.S. presidential election polls, and the financial records of big companies.
Journalists often use data visualization to tell stories in four circumstances: to show change over time, to compare values, to show connections, and to trace flows, according to the Data Journalism Handbook by the European Journalism Centre.
The fact that journalists are “allowing people to explore data in visual formats is becoming a bigger trend to leverage readers’ experience,” says Pam Tobey, a former graphic artist at The Washington Post and currently the visuals director of Beijing Review.
So far, technology helps journalists produce graphs or charts in various digital formats and animate data on a responsive website, making these numbers more accessible for readers.
Several news agencies even develop their technology services to make use of the data generated in the market. The Bloomberg Terminal enables professionals to monitor and analyze real-time financial market data.
Some people argue that those artificial intelligence technologies could replace man in journalism one day, but “human creativity element is what makes you stand out,” Tobey adds.
To add that value, it is important for journalists to offer context for visual information, she added. In the two maps illustrating the 2016 U.S. presidential election’s voting outcome below, the left graph showed that the Republican candidate landslide dominated sparsely populated rural America, but the one on the right, based on population distribution, displayed more of a tie between the two parties.
Illustrations of the 2016 U.S. presidential election’s voting outcome. Credit: Core77
The final result was that Donald Trump won the Electoral College vote, but Hillary Clinton won the popular vote. This outcome is better represented by the second map with colored dots varying in size proportionally to represent the population in each county.
“The audience demands the numbers and they demand accuracy,” Miller affirms. “Numbers and data can be our friends. We can tell different stories from those numbers.”
It is the job of journalists to avoid misleading readers as they visualize data. Based on inventory turnover, people could say that Inditex SA (Zara’s owner) is a more efficient company than its two biggest competitors.
“But if we look at the stock market, in the last five years, the Fast Retailing [Uniqlo’s owner] is the most successful one,” he explains further.
The stock prices of Inditex SA (Zara’s owner), H&M, and Fast Retailing (Uniqlo’s owner) are represented in the Bloomberg Terminal. Photo by Jamille Tran.
The “human” involvement, however, could make bad use of data and mislead the public. In February, the BBC reported that the viral image describing how many people had left Wuhan before the region was locked down due to COVID-19 was wrong. The picture was tweeted by a group of scientists, initially to accompany their study. However, it actually showed global air-traffic routes and travel for the entirety of 2011, instead of how the virus would spread.
To avoid misunderstanding of a graphic, “run it by colleagues and friends to see if they understand what you are trying to represent,” Tobey suggests. Adding logos, watermarks, and instructions on the picture could help readers to get the right idea.
“If you find a graphic to use, make sure of what it is really showing,” Tobey tells young journalists. “You want to present things as they are, instead of stereotypes or personal opinions.”