1. so what should a representative and acceptably sized sample look like? Isn't all sampling - even if Modi himself went to every sampled house and checked their storage shelves himself - not going to capture every relevant detail? After all, the map is not the territory. Do you discount all sample surveys?

2. Unless Modi can prove that people regularly lie while self-reporting things, this claim is dubious. Exit polls after elections rely on self-reported voting patters and seem to be reasonably accurate. I am going to discount this claim by the GoI.

3. How would a hunger index capture and score "Efforts to combat hunger". There are no prizes for effort. Either the family being sampled is hungry or not hungry. They can't subsist on "effort". Unreasonable to demand that the index track something like effort which, incidentally is self-reported by the government because there is not third party verifying their work.

4. 2/3 weightage to child nourishment - the world's median age is 30.6 while India's is 28.1 so not thaaat much younger. If the world were one country, they would be the 120th oldest country while India is the 143rd oldest country. In other words, India is not as young as you would believe. There are 50+ other countries which are younger but don't seem to be complaining about the methodology. Finally, not sure what relevance the relative median age has to child hunger and its weighting. Can you please elaborate? Are you saying that since India is young, a larger share of its population is hungry and hence its hunger index should be ignored? I genuinely don't understand what point the GoI is making here.