The databases for population counts have increased manifold in recent times with the advent of large scale surveys and also innovative and improved ways of organising administrative data to serve this purpose.
The challenge that these databases pose is external validation through a process of triangulation that is a hallmark of any estimate emerging out of such an exercise. With the pandemic, the need for population data beyond age and sex has been recognised towards identifying multiple vulnerabilities and provisioning according to specific needs. The state has to also assess the impact of the various containment efforts across population sub-groups and differences in mitigation strategies. To effectively perform these tasks, we need population data with characteristics to identify and provide those rendered vulnerable. In this context, postponement of the conduct of census 2021, a source of vital information needed for planning and provisioning, is of serious concern.
The Indian Census represents a marvel of statistical and administrative capability that has remained uninterrupted for more than a century. It may be a decadal event, but it provides the backbone for all other data systems, either by way of a sampling frame, or providing denominators through projections for estimating number with specific characteristics. Its significance is not realised adequately because its role is invisible even to researchers who work on population data.
Population databases can perhaps have no alternative to complete enumeration like the census. The various largescale surveys being planned cannot be based on a population frame based on the last census i.e. more than a decade old. The other alternative has been the projected population counts made available by the Expert Committee on population projections.
However, these efforts at projection involve assumptions which can at best be approximate and the projections are limited to count devoid of any characteristics beyond age, sex and residence. When many other large scale enumeration operations by both government and other private agencies are afoot, there is a need to awaken the public systems to the urgency regarding the conduct of the census.
We need to be aware of the consequences of neglecting this phenomenal resource, it will lead to mis-counting many other things that depend on the census count for denominators and external validation of population counts.. The conduct of the census digitally may provide quicker and possibly more accurate estimates. However, generating this information with a reasonable delay of two years will pose its own challenge to maintain figures for routine periodic comparisons in every sphere.
At this juncture, the common challenge is to have a population denominator that is either getting assumed or approximated. When these assumptions and approximations are not clear, the resultant is varying counts leading to more confusion and unnecessary controversy.
In the time of pandemic, population information with its vital characteristics is invaluable to make any assessment of interventions as well as provisioning for various population sub-groups. In this dire situation, efforts to gauge the population count and characteristics with alternative parameters of population dynamic available to us are needed. There is the problem of reliability of these parameters generated and their comparability across geographies.
The positive side of recent developments has been the appreciation of a range of administrative information data bases that have been overlooked in the past. This can be considered an opportunity in the current crisis not only to make use of them but also verify their robustness.
One wonders as to whether this helplessness has emerged out of the neglect for the scientific and technical expertise in indirect estimation that was decimated systematically over time by the scientific community as we moved towards the progress of obtaining direct estimates to replace them. The tools and techniques along with scientific thinking need to be re-purposed to generate and evaluate population parameters in the evolving scenario of incomplete information and their rough approximations, which are, at times, a far cry from reality.
The responsibility lies squarely with the demographic community to reinvigorate the scientific rigor of their methods to offer the required approximations with scientific reasoning. They also need to evaluate and contribute to corrections in estimations that emerge out of incomplete and/or inaccurate denominator bases. Although one may argue against the need for indirect measures given the availability of direct estimates obtained in surveys, there is every need to evaluate their robustness and consistency not merely in terms of their statistical features like confidence intervals and standard errors but also their consistency with other parameters.
The message therefore is to urge authorities to conduct the census operation at the earliest and put to rest the nascent idea that largescale surveys can be a substitute to a complete enumeration. Population data without the census is like diffused day light, without the direct light of sun in the sky. In delaying it further, we may miss out on documenting the silent transition that is underway in relation to reproduction, marriage, mobility, family formation and survivorship at large as all our assumptions emerge from the past, not from the bright light of the present.
(Uday S Mishra is Professor, International Institute for Population Sciences Mumbai. Mala Ramanathan is professor at Achutha Menon Centre for Health Science Studies, Sree Chitra Tirunal Institute for Medical Sciences and Technology (SCTIMST), Thiruvananthapuram.)
