Modi Govt Has Withheld Data, But ‘Real’ Issue Is of Collection

“We need much more advocacy to first make sure that the data is collected at all,” writes data journalist Rukmini S.

Image used for representational purposes.

(Photo: Shruti Mathur / The Quint)

On the first day of an extraordinary session of Parliament, the first since the pandemic hit, the NDA government appeared to confirm the characterisation by its critics of it being an opaque government that hides data.

In response to at least four questions asked in the Lok Sabha on the number of migrants workers who had died in road accidents while returning home on foot during the lockdown, the ministries of Road Transport and Highways, and Home Affairs, responded that no such data had been maintained.

The memes and editorial cartoons were swift and scathing: “NDA = No Data Available”, they read.

What We Must Do To Get ‘Better Data’

That the Prime Minister Narendra Modi-led BJP government has been opaque, has denied citizens vital data, and has stonewalled those who seek accountability, is undeniable. India’s annual crime statistics suffered their greatest delay in thirty years when the crime data for 2017 arrived only in October 2019. Also in 2019, the government interfered with and politicised to an unprecedented degree, the functioning of the National Statistical Commission over the release of rounds of the National Sample Survey, the most important source of data for most socio-economic indicators in India. This had shown the government’s record on creating employment and income growth in poor light.

Most recently, the Modi government attempted to claim that the PM CARES fund, set up by him, is not a public authority under the Right To Information Act (RTI), despite the direct involvement of the Prime Minister’s Office (PMO) in its functioning.

But it would be a disservice to the cause of open data to assume that all of our problems are on account of malice alone. Too much of the problem is in fact on account of ossified data collections and classification systems.

To be able to get better data, we need to first put in place better and more nimble systems of data collection.

The Logistical Problem Of Data Collection

Take the question that started it all – on the deaths of migrant workers in road accidents. To be able to answer the MPs’ questions, the Ministry of Home Affairs would need to ask every state, which would then ask every district in the country to comb through FIRs from road accidents during the lockdown period to see if any mentioned that the victim was a returning migrant worker.

Since the profession of the victim or the reason for them to be walking on the road is not part of the standard pro forma that the National Crime Records Bureau (NCRB) uses for its annual ‘Accidents and Suicides in India’ report, or by the roads ministry for its annual report, this would require every police station in the country to depute someone to manually read the FIRs.

The inadequacy of the NCRB in its current form – to answer anything new – pre-dates the current crisis.

When the delayed crime statistics did emerge in October 2019, there was widespread criticism in the media that the 2017 statistics had not looked into ‘hate crimes’ or religious hate-driven lynchings, as the agency’s then director had promised in media interviews. The truth is, it was never going to happen – unless a distinct category exists in the Indian Penal Code (IPC) or under special laws, there is little the NCRB can do but make improbable promises.

On the other hand, when the railway ministry was asked a similar question in the Rajya Sabha about migrant deaths on trains, they were able to respond on 23 September in detail, and included the number of deaths and the broad categories of causes.

This is because the question fell within a pre-existing system of data reporting – all deaths on trains are reported by the state police, and the only passenger trains running at the time were those transporting migrants.

(Graphics: Shruti Mathur / The Quint)

Issues With COVID Data – Or Lack Thereof

Similarly, there was dismay at the government not maintaining separate data on the number of doctors and other healthcare workers who had died of COVID, and rightfully so. This is, however, only one of the countless problems with the way in which the government currently collects and releases COVID data.

The official ministry website only carries cumulative counts updated once a day, and every journalist and academic the world over relies on a crowd-sourced website run by public-spirited volunteers: covid19india.org. This when every small Latin American country is able to produce updated, easily downloadable historical data by region.

Far from having any insight into the professions of the people dying of COVID or how they contracted it – because most states do not have the administrative capacity to collect this data – we rely on press conferences for graphs of even the gender and age-wise break-up of India’s COVID deaths, rather than it being publicly available in updated form.

By the end of April, when India had conducted just one million tests, it had no details on the travel or contact history of over half of those coming in for testing.

(Graphics: Shruti Mathur / The Quint)

More Advocacy Needed To Ensure Data Is Collected In The First Place

Rigid legacy data silos should not absolve the government of its responsibility of collecting and sharing better data.

When the government chooses to, it can be ‘creative’ in data collection; the 2017 crime data did, for example, introduce the imaginative category of ‘crimes by anti-nationals’, by clubbing together three sub-sections involving insurgency and terrorism.

In this year’s migrant crisis, the government was able to marshal data on returning migrants – even though it did not fit into any pre-existing survey or system of classification – by doing something as simple as writing to states to ask them to collect this data. Given that most returning migrants faced some nature of health quarantining, collecting this data became more actionable.

However, broad brush-stroke criticism of the government as being one that is refusing to part with data misses the point: that we need much more advocacy to first make sure that that data is collected at all.

Pushing for more administrative data to be made and kept public, for more useful and relevant questions to be asked in the National Sample Surveys and the National Family Health Surveys, and for the frequency and quality to be improved, must all form part of a peace-time push for better data that will make sure that we aren’t scrambling in times of crisis for data that we never collected.

Small Windows Of Accountability

The ‘NDA’ incidents also place a spotlight on this issue of accountability. Just before the session of Parliament was to begin, the government announced that there would be no Question Hour, and only ‘unstarred questions’ – to which answers are submitted in writing only, and no follow-up questions are possible – would be permitted. Even in this curtailed form, ‘unstarred questions’ were able to elicit responses from the government on migrant deaths – even if it was to admit that they did not have the data.

Another unstarred question got the government to admit to the decline in immunisations that earlier reporting on the National Health Mission’s administrative data had exposed, but that the government had not responded to until then. The first admission by the government that NCRB data was delayed came in response to comments in the Rajya Sabha in 2019 by Leader of the Opposition, Ghulam Nabi Azad.

Politicians should be accountable every day, and not only three times a year when Parliament meets.

But at the very least, we’re going to need to admit that our MPs are doing a good job of trying to get us the data we need, and push to make sure that these small windows of accountability do not close.

(Rukmini S is an independent data journalist based in Chennai. This is an opinion piece and the views expressed are the author’s own. The Quint neither endorses nor is responsible for them.)