Numbers for the nation

At the launch of her book Whole Numbers and Half Truths, journalist and author Rukmini S spoke of the importance and flaws of data collection in India
(L-R) Rukmini S in conversation with Palanivel Thiaga Rajan and M Gunasekaran | Debadatta mallick
(L-R) Rukmini S in conversation with Palanivel Thiaga Rajan and M Gunasekaran | Debadatta mallick

CHENNAI: Countries around the world — even those that followed a more liberal definition of a COVID-19 death — accept that they were likely missing some deaths from Covid. To fill in this gap, they turned to estimates of excess mortality as a proxy for confirmed and suspected Covid deaths. India last released all cause mortality data in 2019. Yet, I knew that updated data on death on all causes — from the smallest village to the biggest city — were being uploaded on government servers every day.

In June 2021, I began to access secret official government portals that held civil registration system data for every state. I found that in May 2021, at the peak of India’s second wave, mortality from all causes shot up to nearly five times the usual average in Madhya Pradesh, for instance. Behind these statistics is the story of what undercounting means. 

For every systematic reality in India (and otherwise) is data that explains the nuances of the matter, points to interconnected concepts and reasons, and disproves or proves the general consensus. With her book Whole Numbers and Half Truths — from which the above excerpt is taken — journalist and author Rukmini S looks to answer 10 questions about how things work in India and how the data answers it or misses to do so.

Written over the summer of 2021, the book features diverse discourse, backed by statistics and humanised with real stories, and interrogates facets of Indian politics and culture. “The writing process was truly organic I think, in part, because of my training as a journalist — we are accustomed to weaving these anecdotes into our reporting, as also integrating data into reporting. Some of these were stories picked up over years of reporting. For others, I used contacts to seek the sort of person I was looking for. Some of these were open-ended conversations, and I went along with where they led me,” the author told CE on the sidelines of a book launch at Madras Music Academy. 

Lessons at the launch

As a field journalist-turned-data journalist, both roles have given Rukmini an understanding of the Indian system. “The reason the data is very important is because we live in bubbles and we think the world around us is the only one. So, one of the reasons I love to share information about India’s income and consumption distribution is that all of us think we’re middle-class, but you are most certainly in the top one per cent of the country and this is something you would not know if you relied only on what you have seen around you,” she began the discussion with journalist M Gunasekaran, the moderator of her subsequent conversation with Palanivel Thiaga Rajan, Minister for Finance and Human Resources Management, Government of Tamil Nadu.

But data alone cannot take you very far, she added, pointing to her chapter on police statistics. “If I was not a field journalist who had spent nights in the police station and seen what goes into writing an FIR, I, too, would have a lot of belief in NCRB data. But having seen what goes on before an FIR is filed, I find it impossible to take those numbers at face value,” she noted. So, how does one trust the nowadays questionable government data on which the book heavily relies, posed Gunasekaran. “A lot of younger people ask me this since they are skeptical about the official data. You need to understand the statistics well and go in with skepticism but not blanket suspicion. For example, we now know that there is now a problem with household surveys because we know from talking to people that there haven’t been problems all this while.

There has been an oversight in mechanism and accountability; we’ve had surveys conducted by international standards since the ‘50s. When we know something has changed, that’s how we know there is ground for suspicion,” observed Rukmini. But there are also ways to check the quality of data if you wish to, according to Palanivel. “The data is systematically biased; it is not randomly biased. The count of deaths and unemployment will be undercounted. There is inherent bias depending on the statistic. You must start with that perspective. The next is to check for internal consistencies. For instance, what is this year’s data compared to the last? Or, how does the Covid data reported in the medical data compare to the number of bodies cremated? So, have ways of both, understanding the skew and checking how great it is. Of course, data is imprecise at best; we are a large country and most of it (data collection) is by sampling. But there are ways to check which is more likely to be accurate and which is rigged,” he mentioned. 

Understanding data

The collection and analysis of data is vital for a nation. When it comes to policy, this is crucial information to formulate one that is beneficial for the parties intended. But gathering accurate data is difficult, according to Palanivel, and it depends on the purpose for which it is collected and used for. However, data can only get you so far. It is how you interpret, decipher and compare it with past data that is important for the government and people, he added, quoting an example.

“In a review conducted with government departments about expenditure and revenue every six months, one department had initially projected an increase of 20 per cent. But the results were only 13 per cent and the department claimed that this was because some places had 30-50 days of lockdown so this is a good result. However, we were on lockdown for four months last year. The 20 per cent projection is already poor — it should be 30-33 per cent. The goal, itself, was wrong,”  he said. 

The same ability required from politicians is also necessary to a citizen. How would one understand policies and the need for them if we couldn’t decipher any data. But how can one ensure that this data is accessible and moreover, impactful to the people of the country without diluting it? The key is to simply explain the data in detail, agreed both Palanivel and Rukmini.

“The way you make numbers accessible is, in my opinion, simply explaining what it means. Many people have not had that exposure to numbers; we teach Maths poorly in schools. So, I try to give as much nuance as possible. In addition, I try to draw on stories,” said Rukmini. But whose story gets told? “One person’s experience is not representative of everyone. I think the pursuit of objectivity and lack of bias is almost futile but what I do aim to achieve is fairness. So, I hope the anecdote I am putting out is fair to the broader point the data is making. Everyone loves stories; I take great pride in well-told stories through numbers.”

The other way to reach people, as Palanivel has had to do in politics, is by taking a step back. “Whenever I spoke as the opposition, I used to start from the fundamental principle because many of the MLAs have not really had a reason to come across this concept — why is a deficit bad or what is the interest to revenue ratio and how does it affect etc. These concepts are not inherently obvious or familiar to them, so if you just give data without explanation, it is harder to penetrate,” shared Palanivel. 

With so many interesting points already made, there was little left to add . The event may have come to a close, but the conversation is open to more deliberation and discussion.

In the Book

Rukmini S looks to answer 10 questions about how things work in India and how the data answers it or misses to do so. The book features diverse discourse, backed by statistics and humanised with real stories, and interrogates facets of Indian politics and culture.

Related Stories

No stories found.
The New Indian Express
www.newindianexpress.com