An Error and A New Process

Sidhant Koshi

In the edition of the First Draft published on Monday, September 21, 2020, the lede story (“British universities take a hit, Indian universities likely to benefit” by Priyam Sharma) carried two charts (Fig. 1 and Fig. 2). There were errors in both the charts.

We would like to apologize to our readers for the errors. You can read the corrected version of there story here.

The online medium allows us to correct our errors easily. However, at First Draft there is no five-second rule for errors — every time we make an error, we will apologize, issue a corrigendum, and try to implement processes so that these are not repeated.

For the sake of full transparency, we have investigated how the error happened.

These were pointed out to us by Professor Ruchira Sen, an economist and faculty member at Jindal School of Journalism and Communication (JSJC).

“The article says that the Covid-19 crisis has made Indian students anxious about their plans for studying in the UK,” she said. “But in the graph which compares the number of students who applied in 2019 and 2020, you see an increase in the latter year.”

“There is no source for the graph and there is no analysis of the claim that it makes,” she added. “In fact, the graph contradicts the previous statement.”

“There is a hyperlink to a Guardian survey,” said Prof. Sen, who is also the data editor for First Draft. “This is not enough. We should be able to see the graphs from the survey in this article itself.”

“Also, the survey is four months old,” she added. “This, in itself, is problematic.”

The second graph, too, had its set of problems.

“There is a line which states that UK saw a 63% increase in the number of Indian students in 2019,” said Prof. Sen. “But the graph (on the right) has depicted the years as a separate category and 2018 and 2019 are shown to be on a par with each other. The graph on the left does not let readers know what the figures mean as there is no legend.”

She added that the graphs seem to have been made in a hurry — on Microsoft Excel rather than on Datawrapper.

“The first graph doesn’t have a source and the second one has a source but it is not replicable,” she said. “Will I be able to go back to the website and replicate the numbers? This would have been better if it wasn’t a data story.”

Sharma, the author of the article; Bhavya Vemulapalli and Ishita, who edited it; Pujit Tandon, who was responsible for the graphics; and Shaardhool Shreenath, who was the proof reader, were also interviewed by First Draft.

“I wasn’t really happy with the graphs, but I wasn’t sure what to do about it,” Sharma said. “We should definitely communicate our dissatisfaction to the news editors of the edition before we publish.”

Every week, three or four students serve as news editors for the edition of First Draft. They decided if the edition will have a theme, commission the articles, coordinate with the writers and producers, help out with the editing, and finally publish the edition with the help of the website team.

“After I had designed the graph, I let Bhavya and Shaardhool know,” said Tandon. “Bhavya told me that Professor Uttaran Das Gupta, editor of First Draft, would check the graphs in class — but, unfortunately, that did not happen.”

“I did not realize what was wrong when I checked the graphs,” said Vemulapalli. “I am not that good with numbers — but we should have been extra careful. I was the second editor and should have looked more thoroughly.”

“While editing, I contacted the writer and asked him about the data and checked the data’s reliability by myself as well,” said Ishita, who was the first editor. “But I feel after that, I slipped while checking if the data connected to the structure of the story.”

“I corrected the title as there were some mistakes,” said Shaardhool. “But I did not notice the mistakes with the graphs.”

“There were several people involved in the editing process, but we did not manage to catch the error,” said Prof. Das Gupta. “We also missed a very important step in the process — the story and the graphs were not shown to the data editor, who could have helped us avoid this error.”

To ensure this does not happen in the future, every data story First Draft publishes will be checked by Prof. Sen, he added.

Spotted an error? Tell us! Disagree with our views? Let us know! Have a story you want First Draft to cover?

Write to us: editorial_firstdraft@jgu.edu.in