Making sense of metadata

The journalistic process gets more complicated, and demands a higher level of engagement and scrutiny when reporting metadata like the Indian Census figures. The historical context when reporting on Census data helps understand the social transformations these numbers tend to capture.

CHENNAI, 16/10/2014: A.S. Panneerselvan, The Hindu Readers' Editor. Photo: V.V.Krishnan

Did  The Hindu  get its story on the population by religion, based on the census figures for 2011 released on August 25, 2015, wrong? Why is its report different from the reports in other publications? Is there a bias in its reporting? What is the need for talking about the numbers in terms of religion? These were some of the questions we received over the past few days.

The central idea here is the difference between a journalist and a stenographer. While the stenographer reproduces a text, a reporter provides the context in which the text plays out its social, economic and political roles. Even while filing a report after covering a routine press conference, a reporter needs to be clear what the lead is; what the implications are; who the beneficiaries or losers in the development are, is there something that is left unsaid that is more important and, finally, what are the elements that contribute to public interest. It is never a mechanical reproduction of every word that has been uttered. The journalistic process gets more complicated, and demands a higher level of engagement and scrutiny when reporting metadata like the Indian Census figures.

Crucial context

It may be useful to share the process adopted by 

The Hindu  team while reporting the population numbers on the basis of religion. First, it was the Editor’s decision to accord the premium space for this story. Second, recognising its import, she took a decision to have the necessary follow-up reports and an editorial, “Questions of timing and context”, on the subject. In 

The Hindu’s editorial judgment, the Census was primarily a demographic document and it decided to look at the findings through a demographic lens. It was a valuable source of over 60 years of historical data. It became imperative to provide a context, framework and milieu to the new findings to put it a proper historical perspective. The historical context helps one understand the social transformations these numbers tend to capture. A systematic disaggregation enables us to understand the full implications of the individual strands — social, economic and regional among many others — that intertwine to form the complex metadata.

We must not forget that the 2011 data was released late in the evening of August 25, 2015 in the form of Excel sheets on the Census of India website without a press conference or an explanatory note. There was a Home Ministry press release on the Press Information Bureau website, giving a bare bones description of the new data with no historical context. Most newspapers faithfully reproduced this press release, choosing not to do any calculations. This was unlike the release of the 2001 census data on religion in 2004. The then Census Commissioner, J.K. Banthia released the report and presented a copy to the then Minorities Commission Chairperson, Tarlochan Singh at a press conference in New Delhi.  Noted demographer Ashish Bose was also invited as a panellist to the press conference and helped to put the numbers in context. Mr. Banthia presented the findings, and took questions from reporters, clarifying some of the key points of contention, chief among them being that Jammu and Kashmir and Assam had not been counted in 1991, thereby artificially depressing the figures.

Numbers decoded

In the case of 2011, 

The Hindu  managed to do the number crunching because it had in-house expertise in data journalism. The team led by the National Data Editor could spot some of the striking features.  First, the population of India’s poorest communities has grown at a faster rate than its richer communities every decade since Independence. In Hindu/ Muslim terms, the Muslim population has always grown faster than the Hindu population. As a result, while the growth in Muslim population in absolute terms was a point worth making with respect to the 2011 data (as the newspaper did), it hardly seemed to be the most revealing point.

What seemed more important, then, was the change. A long-term trend observed in India by demographers was that while the growth rates of all populations are declining, it was slowing faster for Muslims than for Hindus. The new data was an empirical validation of the demographic phenomenon about segments or categories where the fertility rate was higher earlier are now reducing at a faster rate. Several demographers explained this factor to this newspaper.

It is beyond my remit to answer why no other newspaper thought it fit to seek a demographer’s expertise to look at the latest Census figures. This newspaper went the extra mile to provide the historical data.  Since the Census chose not to make any historical data available,  The Hindu’s National Data Editor referred to data she had compiled from past Censuses and from estimates by demographers at the International Institute of Population Sciences, who filled in the gaps for years when some States were not surveyed in the Census. In some States, the growth in Muslim population is not along expected lines, a notable example being Assam and she pointed this out. 

It was the government that released data — that has a potential to polarise a society — into the public domain without the necessary explanations or interpretations. It was  The Hindu  that placed these figures in context and blunted the power of the polarising agents.

  readerseditor@thehindu.co.in