Administrative data are collected primarily for non-statistical purposes and adopted for producing statistics, whereas statistical data are collected for generating statistics. The concept of administrative statistics was appreciated when the important micro objects of the society as interrelated and interacting elements of a system was reflected on the macro level by the System of National Accounts in the 1950s and 1960s.

In India, administrative statistics have grown along with the federal structure of the government. Administrative records contain a wide variety of data on demographic and social including health, economic, cultural and environmental subjects. Administrative authority of the data are vested with statutory authority to canvass the requisite information to ensure public compliance and response. In instances where such legislative provisions support data collection, the coverage and completeness of records is usually good. Collected data is relevant and direct, and handled by agencies that have special knowledge of the subject. The set-up where this data is collected has good knowledge of the subject and is well-versed with local language, conditions.

Advantages and challenges: The data through administrative sources eliminates survey errors, removes (or significantly reduces) non-response, and provides more accurate and detailed estimates for sub-populations. Although data collection costs are not incurred, but uses of administrative data will require transforming it to the required statistical outputs and, therefore, involve data management costs.
Confidentiality to the wider community, in using administrative data by the statistics office, is important and it may be ensured that administrative data collecting organisations are complying with the law and information privacy requirements. Administrators responsible for collecting data may not be properly trained on methodology for recording. Administrative data is aggregated at a broader level, and some information content is lost due to individual variation or equity concerns. Private service providers may be less inclined or not required to contribute data to the aggregated dataset, thereby sometimes leaving out a significant population that utilises these services. The concepts and definitions used as required in terms of laws and regulations may not be as required for the statistical purpose and also not conform to international standards.

To use administratively sourced data in statistical system, the statistician needs to:

-Clearly understand the rules and processes that define administrative data;
-These rules and processes will influence the coverage, timing, quality and completeness of the administrative data;
-Understand the differences between the model defining the statistical data requirements and the administrative data that is available; and
– Develop appropriate methodology and processes for the statistics office to transform, model or adapt the administrative data to the statistical model.

Digitisation and the future of administrative statistics: There is enough research on Digital India. One is Iyer et al, who, in their book, focused on the need of Digital India and the role played by e-governance in our country. Two, Jaini & Tere, who accepted that Digital India is the need of our economy and a tool for development. They also concluded that Digital India reduces the paperwork and digitisation increases GDP of the economy.

Through digitisation, India is far more data-rich than it has ever been, with biometric data of its citizens available through Aadhaar. It will soon have detailed records of company-level transactions through GSTN. Other examples are MGNREGA and Jan-Dhan Yojana, which collect and disseminate detailed disaggregated data.

According to Frederick J Gravetter and Larry B Wallnau, statistics “creates order out of chaos” by summarising and simplifying complex human populations. Statistical analysis is a critical component in a needs assessment. The use of this kind of administrative data in the production of statistics can be categorised by purpose as controlling the processing of statistical data and quality evaluation of final products, producing new statistical products either separately or in combination with data from multiple sources, and preparing improved collection frames for sample surveys and censuses. It’s time to take advantage of digitisation to rediscover the administrative database.

By- Rajnesh Jain, Deputy Director General, Ministry of Health and Family Welfare. Views are personal.