Over the years, the demand for statistics has increased multi-fold, while the process of data collection and analysis has become very complex.
Recently, Ministry of Statistics and Programme Implementation (MOSPI) proposed the development of a new “National Policy on Official Statistics”.
This is significant as it comes in the midst of declining trust of the masses in official statistics due to the proliferation of post-truths (fake news).
What are the proposed institutional reforms in statistics?
As trust in official statistics is vital for democracy, public trust needs to be won through competent and honest institutional mechanisms.
In this context, the proposed policy envisions addressing statistical aspects like complexity, ethics, quality, documentation and data retention.
“Sample surveys” forms the bedrock of Indian statistical system and the instructions for surveying are dealt in a detailed and explicit manner.
Specialisation - Increasing need for statistics in diverse areas is pressuring our statisticians to seek assistance of scholars from multiple disciplines.
Notably, National Sample Survey (NSS) switched over to a new set of codes in 2009 for data collection involving industrial labourers.
These change made it difficult to differentiate between farmers and farm managers and shopkeepers and sales managers via occupational codes alone.
This resulted in the exclusion of a large chunk of the Indian workforce in the unorganised sector from our statistical analysis.
The problem arose mainly due to our statisticians merely aping models followed in industrialised societies, without contextualising it for India.
This fumble could’ve easily been avoided had there been a sociologist with knowledge on ‘occupational studies’ on board.
Reliability - How surveys are designed and questions are developed has evolved into a science that in itself needs specialists.
Notably, RBI has adopted an inflation-targeting approach that relies on data on inflation expectations of individuals (which is highly unreliable).
Such an absurd approach to data is what is driving our policy, which calls for a serious rethink on our data collection modules.
Honesty - Data collection is increasingly being done by contractors and for-profit organisations, which has raised honesty related concerns.
Supervising them and ensuring their honesty remains challenging despite the commissioning of some technology based monitoring mythologies.
This aspect has reiterated the fact that there is no substitute for empathy and personal truthfulness, which needs to be culturally ingrained in organisations.
Decentralisation - Currently, statistical systems are largely centralised at the ministerial levels in the central and state governments.
Instead of this, creating a data ecosystem that harnesses the energy of diverse institutions innovatively should be considered.
Statistics needs to get internalised into the fabric of our institutions even at the lowest level and the higher-ups should merely collate the data points.
