The announcement that the government has decided to merge the National Sample Survey Office (NSSO) into and under the Central Statistics Office (CSO) has caused both surprise and concern. What exactly the ‘merger’ means remains unclear. Recent attempts to question the veracity of National Sample Survey (NSS) data and the way the issue has been handled have given rise to apprehensions within academia, State governments and the media about the prospect of radical changes in the present system for deciding substantive issues of scope, design, scrutiny and validation of the surveys.
The present system
Under the present system, every year various departments of government send a list of subjects that they would like to be investigated by the NSSO. The requests are sent to the National Statistical Commission (NSC), which has respected economists, subject matter specialists and statisticians from government, including the head of the CSO and senior officials of the NSSO responsible for technical aspects of design and conduct of field work, as well as representatives of State governments. Subject matter specialists in particular fields are also brought in. The proposals are discussed at length keeping in view the budget allocations, availability of trained field staff and supervisors. In doing so, the conduct of periodic surveys on important issues is also considered. (It should be noted that budget allocations, and personnel of the NSSO have always been under the Department of Statistics.)
After providing for periodic repeat surveys (at quinquennial or decennial intervals) of some important aspects (notably consumer expenditure, employment, social consumption, land holdings, rural savings and investments), the subjects to be covered in a particular year and the scope of the inquiry are decided.
The tasks of sampling design, the scope and content of information to be collected, design of schedules and protocols of field work are left to be decided by special working groups. These groups are chaired by experts from academia, and senior officials of the CSO and the NSSO, State government representatives as well as select non-official experts. These working groups are in continuous session from the inception of each round through all the subsequent steps. Discussions of concepts, questionnaire design, field work schedules and supervision are continuous, detailed and highly professional. Once the field work is over, the groups decide the detailed tabulation programme, and the tables to be prepared for publication. The tabulated results are discussed in detail by the NSC and are published after its approval.
After considerable hesitation and prodding, the government decided some years back to put all tabulations and the primary data on open access, especially to academic and other interested users. This decision has stimulated and facilitated the use of these data for intensive analyses by numerous researchers. They have been used extensively for monitoring of trends and critical assessment of several important aspects of the economy and society, such as poverty and inequality, consumption patterns, employment, household savings and investment, and health-seeking behaviour. They have spawned intense as well as creative controversies over survey methodology, quality of data, and interpretation of structure and trends. These have played an important role in shaping policy and in improving the surveys.
The NSSO surveys command wide respect among academics, State governments and non-governmental organisations as the most reliable and comparable basis for discussions in the public, policy and even political arenas. This is based on their well-earned reputation for professionalism, independence and integrity. Widespread apprehensions that the proposed absorption of NSSO into the CSO could compromise the surveys by subjecting their review and publication to government approval must therefore be allayed promptly in an unqualified manner. The existing institutional arrangement in which the NSC, as a professional body independent of government, has not only functioned smoothly but also commands confidence and respect both within the country and abroad must be maintained. Any attempt or even a suggestion that its substantive work, publication and free dissemination of data are subject to the department’s approval will hugely dent the credibility of the Indian statistical system.
Scope for improvement
Urging this forcefully does not in any way suggest that the present institutional arrangements are flawless or that the NSSO is perfect. On the contrary, it is widely recognised that there is scope for improvement in the functioning of the institution and the way data are collected. These problems are well known: the NSSO doesn’t have adequate budgetary allocations; there is an acute shortage of trained field staff; the scale of surveys is un-manageably large mainly because the users demand a degree of detail in content and regional disaggregation of estimates. The NSC is fully conscious of these difficulties. The solutions call for action by the institutions responsible for gathering data by investing in continuing research on improving sampling design, field survey methods and validation of data. Correcting these deficiencies is entirely in the domain of government.
But there are also serious difficulties inherent in trying to get reliable and complete information through the interview method. Most respondents do not maintain any records or accounts of their transactions. Since most respondents rely on recall, it is unrealistic to expect them to provide reliable information on the scope and detail sought by questionnaires. Memory lapses and respondent fatigue lead to high incidence of non-response, indifferent response and biased response. These problems are particularly serious among the more affluent and better-educated sections of respondents. Increasing the role of CSO officials in running the NSSO will not solve these problems, but they can help by providing funds for specialised research on survey design and methodology. The necessity and importance of such research calls for far greater attention and resources than they receive at present.
A. Vaidyanathan, an economist, is a former member of the Planning Commission and RBI board
