Personally speaking, I do not have much confidence in either the pre-poll surveys or the exit polls during an election in India. As I have written earlier, the Indian voter holds his voting card very, very close to his chest. Added to it is the problem of the sampling size and frequency. Yet another uncounted factor is the lack of homogeneous population or concerns along the length and breadth of the nation.
The current crop of exit polls by half a dozen media houses through various information research companies (that is the latest fancy name for survey guys), have not in any way forced me to change my opinion.
In 2009 General Elections for the Lok Sabha, the exit poll forecasts by the (same, in many cases) media houses, predicted a neck and neck race between the BJP led NDA and the Congress led UPA. By and large the estimates were for a difference of 15 to 20 seats between the two alliances, with UPA in the lead.
We all know what happened. When tabulated the actuals were, on an average, 25 seats short for the BJP’s NDA and 65 seats more for the INC led UPA. That is a mean difference of 45 seats. Keep that figure in mind.
Now let us take a look at the actual projects by the various media house-survey companies, for both India as a whole and also my home state of Tamil Nadu.
First off, I am going to discount the Today’s Chanakya exit poll predictions. There are 2 reasons for it. One of course the off the base projections for India. Two is the tally projected for BJP (7) and DMK (only 5) in Tamil Nadu. For the life of me, I cannot imagine the BJP alliance scoring more than the DMK alliance, in Tamil Nadu. OTOH, Chanakya were the only one who predicted the AAP upsurge during the Delhi Assembly elections. Still, I simply think those projections are far fetched.
A point to note is that all these surveys claim a margin of error within 5%. If we accept it as true then the mean spread between these forecasts should be within that 5%. Is it not?
But take a look at the projections for all India -
It would mean the Indian voter has just flipped the Dosa around!
That she has flipped the current Congress tally of about 206 seats (29% vote share) with that of the current BJP tally of around 116 (19%). Would that be funny or not?
Update: The actual results were Congress 44 seats (19%), and BJP 282 (31%).
It is similar when we look at the Tamil Nadu figures. While I do not have much of a feel for the rest of India, I can confidently write that there is no way the BJP alliance in Tamil Nadu would score 4 to 7 seats in a Lok Sabha election. I have said it before - the BJP party itself will not win a single seat in this election. If at all there is is a win, it would be from Vaiko of MDMK, or Anbumani Ramadoss of PMK - though I doubt even those seats.
Image for Pinterest:
The current crop of exit polls by half a dozen media houses through various information research companies (that is the latest fancy name for survey guys), have not in any way forced me to change my opinion.
In 2009 General Elections for the Lok Sabha, the exit poll forecasts by the (same, in many cases) media houses, predicted a neck and neck race between the BJP led NDA and the Congress led UPA. By and large the estimates were for a difference of 15 to 20 seats between the two alliances, with UPA in the lead.
We all know what happened. When tabulated the actuals were, on an average, 25 seats short for the BJP’s NDA and 65 seats more for the INC led UPA. That is a mean difference of 45 seats. Keep that figure in mind.
Now let us take a look at the actual projects by the various media house-survey companies, for both India as a whole and also my home state of Tamil Nadu.
First off, I am going to discount the Today’s Chanakya exit poll predictions. There are 2 reasons for it. One of course the off the base projections for India. Two is the tally projected for BJP (7) and DMK (only 5) in Tamil Nadu. For the life of me, I cannot imagine the BJP alliance scoring more than the DMK alliance, in Tamil Nadu. OTOH, Chanakya were the only one who predicted the AAP upsurge during the Delhi Assembly elections. Still, I simply think those projections are far fetched.
A point to note is that all these surveys claim a margin of error within 5%. If we accept it as true then the mean spread between these forecasts should be within that 5%. Is it not?
But take a look at the projections for all India -
- the NDA tally is between 257 to 289 - which is well above 10% variation;
- the UPA tally is between 97 and 115 - again, well above 10% variation.
It would mean the Indian voter has just flipped the Dosa around!
That she has flipped the current Congress tally of about 206 seats (29% vote share) with that of the current BJP tally of around 116 (19%). Would that be funny or not?
Update: The actual results were Congress 44 seats (19%), and BJP 282 (31%).
It is similar when we look at the Tamil Nadu figures. While I do not have much of a feel for the rest of India, I can confidently write that there is no way the BJP alliance in Tamil Nadu would score 4 to 7 seats in a Lok Sabha election. I have said it before - the BJP party itself will not win a single seat in this election. If at all there is is a win, it would be from Vaiko of MDMK, or Anbumani Ramadoss of PMK - though I doubt even those seats.
Image for Pinterest:
No comments:
Post a Comment