Teachings of Prof. S. Ambirajan: From statistics to truth

Excerpts from “From statistics to truth” By S. Ambirajan

• “Indians are usually credited with possessing extraordinary capacity for dealing with numbers, and rightly so because from the early days of our civilisation, when the unknown Indian discovered zero, to our own times, which produced the genius of Ramanujan, we have had countless mathematicians and statisticians. In spite of all this, our official statistics have nothing much to celebrate about. It has caused much harm and not a little confusion. At least one of the reasons for the downfall of national economic planning in our country is the faulty data base with which our planners and bureaucrats had to work. In recent times, much confusion has arisen as a result of our inability to statistically demonstrate the impact of the reform programme in many areas. Quite contrary conclusions have been squeezed out to suit the particular preferences of the analysts. The unreliable statistics may prove a boon to academic economists/ statisticians to make prosperous careers out of correcting and interpreting the available data but official statistics are far too serious a business.
• This was brought home when a semi-literate process worker told me, in reply to the usual defence of increased bus fares by comparing them to statistics from other States, ``Sir, the politicians can prove anything with numbers. Can we believe them?''
• ``You academics take these figures at their face value. It is all fudged as always and I am trying to figure it out''. Perhaps this is nothing unusual as a judge admonished a young civil servant during the hey day of the British Raj when the latter referred to some official statistics: ``When you are a bit older, you will not quote Indian statistics with that assurance. The Governments are very keen on amassing statistics - they collect them, add them, raise them to the nth power, take the cube root and prepare wonderful diagrams. But what you must never forget is that every one of these figures comes in the first instance from the village chowkydar who just puts down what he damn pleases.''
• The fact that the Government has the power over all these stages of data gathering, gives it plenty of opportunities to use statistics to its own advantage. An extreme example of this was provided by the Director of the Chinese State Statistical Bureau, Chia Chi-Yun, during the Cultural Revolution: ``If statistical material does not express a clear political idea but merely reflects real conditions, then obviously it will be used by the enemy... Our statistical reports must reflect... the Party's general line... Statistical work must be something which when the party is using it does not cause embarrassment... Victory is nine fingers, defeat is only the tenth... This tenth is also part of the reality, but.. The question is from what standpoint this one finger is presented and to whom it is presented.'' even if our Governments during the last half a century have not been so blatantly partisan, they have used data to suit their convenience.
• One very clear case of partisanship is withholding of publication due to what the Government considers `sensitive' or making access to it impossible for individual researchers. Apart from the problem of inadequate professional competence and individual idiosyncracies of statistical personnel from top to bottom, there is the question of political bias in the collection of information. Thus the decision not to collect statistics relating to caste during the population censuses since 1951 was essentially political.
• All this means that the Indian statistical system is seriously flawed.”