Quarter truth

“Did you know that only 3% of the villagers who are supposed to be helped by the rural employment guarantee scheme actually get anything!” informed a colleague.

That was too bad to be true. So I looked it up on the Web.

The Comptroller & Auditor General of India has come out of a draft report on the National Rural Employment Guarantee Scheme (NREGS). And all hell has broken lose.

A news organisation reporte that '97% of households in India did not receive the 100 days of employment under NREGA that was promised across India.'

A pro-NREGS columnist said that it 'is well known even from the official figures which show an all-India average of 33 days of work provided to 25.5 million households.'

An pro-market commentator wrote that 'only 3.2% of the registered households could avail of 100 days guaranteed work. The average employment under NREGP was just 18 days.'

The bits and pieces made no sense. So I asked a statistician friend what sort of distribution would have 3.2% scoring 100 yet an average of 18. (Admittedly, this is not what our right-wing friend said; though I suspect he implied something thereabouts.)

My friend replied that it's idiotic to even start imagining when we have is two data. Do we know how many got 99 days' work? 90 days? 5 days? What is the standard deviation? Are we referring to the same thing throughout, or does the first figure pertain to household and the second to persons?

He is right, of course.

My question is: 'Why are we clogging the airwaves and drenching newsprint with quarter truths? Don't commentators feel it their duty to connect the dots for the voting public? If not, what should the public do? Kill Jews?'

By the way, the report is up on the Net though I cannot vouch for its authenticity. Apparently, it's not supposed to be there. Also, I'm no wiser after scanning the relevant portions.