Playing poll games
13 Oct 2013 9:32 PM GMT
Television debates on Indian news channels seldom get to the heart of an issue. One aspect of these debates revolves around opinion polls that news channels use to predict election results. Conducting opinion polls and debating them almost a year prior to the general elections in 2014 seems far-fetched at best. The country is politically fragmented, with a large number of regional parties playing a significant role in determining how large chunks of our populace vote. Besides, there are crucial sociological elements of caste and religion that will make their presence felt. In this era of coalition politics, significant doubts remain as to whether we would witness an alternative to the UPA or NDA in the ‘Third Front’.
Immediate doubts arise as to which parties would join the UPA or the NDA, prior to or after the announcement of the 2014 general election results. Most significantly however, remains the choice of candidates that each party will decide upon. The only rationale behind predicting election results via opinion polls, a year from now seems to be how the country would have voted, if polls were held at the time these opinion polls were conducted.
In opinion polls, the standard practice is to divide ?the country ?into regions and conduct surveys in randomly chosen constituencies in each region. At this particular juncture, these results are used to determinethe possible distribution of vote-share in each region. These numbers are then then put through possibly complicated formulas, which will then be converted into the number of seats.
In a Times Now-C Voter-India TV National Poll Projection, held at the end of July this year, predicted that no single pre-poll formation looked to be in a position to form the government. NDA emerged as the biggest alliance with 156 seats. BJP was seen to have become the single largest party with 131 seats. The study gave the UPA only 136 seats, with Congress winning 119 seats. The projections say, ‘This is the lowest cumulative total seats that the two big national parties Congress and BJP have ever won.’
The poll categorically stated that the a third or fourth front led by the SP or the AIADMK, which could emerge as the single largest non-Congress, non-BJP party, could achieve the magic number of 272 with the support of one of the larger national parties. All these predictions give you a hazy picture at best, as none of the above formations have been determined. For the 2009 general elections, the C Voter-Times Now poll had predicted (in March, 2009): UPA 201 (Congress 146), NDA 195 (BJP 138), and Others 147. The eventual results read: UPA 262 (Congress 206), NDA 159 (BJP 116).
Polls have begun trying to determine which figure would make the better Prime Minister. According to the Times Now survey, ‘To the key question, who should to be the Prime Minister of India, an overwhelming 37.7 percent of those surveyed opted for Narendra Modi. Rahul Gandhi came second with 17.6 per cent; Manmohan Singh got 6.2 per cent and Sonia Gandhi gets 3.9 per cent.’ In a Westminster-style structure, that our polity follows, such polls rarely signify anything.
For example, if I’d want Rahul Gandhi as my Prime Minister, and his party or their allies fails to put up a candidate in my constituency, my desire for Rahul Gandhi as PM achieve virtually nothing. For example, if the Congress were to put up a crook as its candidate in my constituency, it is unlikely I will vote for him even if I strongly desire that Rahul Gandhi becomes PM. The same equation will hold for Narendra Modi too, despite the cult of personality. Essentially, preference of one’s prime ministerial candidate doesn’t play a significant part in the overall analysis regarding whomthe voter finally votes for.
Moreover, the biggest problem that comes with analysing opinion polls is that there is no real disclosure on the methodology adopted by them. With the sole example of CNN-IBN-The Hindu-CSDS poll, which did up a detailed note on their methodology, none of these polls showed complete disclosure.
More significantly, however, on the question of opinion polls, is what Rajeeva Karandikar, director of the Chennai Mathematical Institute and one of India’s top election statisticians, said. In a well-articulated article, he said that the churn (change in voting intentions) that happens in the last few days before the elections is much more than that in the preceding weeks. Therefore, the ability of an opinion poll to accurately predict results, even weeks before the elections are conducted, is weak at best. This begs the question, as to why Headlines Today would spend a one hour segment on prime time discussing possible election results in 2014. Surely, they’ve got more urgent issues to address.
The Election Commission has sought to ban opinion polls, once the election polls begin. The rationale used by the commission falls in line with the passing of an amendment of Representation of People Bill in 2009, which banned exit polls till after completion of polling. Essentially, there is a belief that opinion polls confuse the voter and thereby affect the ‘sanctity’ of the electoral process. The ban on opinion polls has been supported by all political parties across the board. Looking at the figures mentioned in Times Now-C Voter poll, it is hard to see, how much of effect opinion polls have on the overall electoral outcome.
Immediate doubts arise as to which parties would join the UPA or the NDA, prior to or after the announcement of the 2014 general election results. Most significantly however, remains the choice of candidates that each party will decide upon. The only rationale behind predicting election results via opinion polls, a year from now seems to be how the country would have voted, if polls were held at the time these opinion polls were conducted.
In opinion polls, the standard practice is to divide ?the country ?into regions and conduct surveys in randomly chosen constituencies in each region. At this particular juncture, these results are used to determinethe possible distribution of vote-share in each region. These numbers are then then put through possibly complicated formulas, which will then be converted into the number of seats.
In a Times Now-C Voter-India TV National Poll Projection, held at the end of July this year, predicted that no single pre-poll formation looked to be in a position to form the government. NDA emerged as the biggest alliance with 156 seats. BJP was seen to have become the single largest party with 131 seats. The study gave the UPA only 136 seats, with Congress winning 119 seats. The projections say, ‘This is the lowest cumulative total seats that the two big national parties Congress and BJP have ever won.’
The poll categorically stated that the a third or fourth front led by the SP or the AIADMK, which could emerge as the single largest non-Congress, non-BJP party, could achieve the magic number of 272 with the support of one of the larger national parties. All these predictions give you a hazy picture at best, as none of the above formations have been determined. For the 2009 general elections, the C Voter-Times Now poll had predicted (in March, 2009): UPA 201 (Congress 146), NDA 195 (BJP 138), and Others 147. The eventual results read: UPA 262 (Congress 206), NDA 159 (BJP 116).
Polls have begun trying to determine which figure would make the better Prime Minister. According to the Times Now survey, ‘To the key question, who should to be the Prime Minister of India, an overwhelming 37.7 percent of those surveyed opted for Narendra Modi. Rahul Gandhi came second with 17.6 per cent; Manmohan Singh got 6.2 per cent and Sonia Gandhi gets 3.9 per cent.’ In a Westminster-style structure, that our polity follows, such polls rarely signify anything.
For example, if I’d want Rahul Gandhi as my Prime Minister, and his party or their allies fails to put up a candidate in my constituency, my desire for Rahul Gandhi as PM achieve virtually nothing. For example, if the Congress were to put up a crook as its candidate in my constituency, it is unlikely I will vote for him even if I strongly desire that Rahul Gandhi becomes PM. The same equation will hold for Narendra Modi too, despite the cult of personality. Essentially, preference of one’s prime ministerial candidate doesn’t play a significant part in the overall analysis regarding whomthe voter finally votes for.
Moreover, the biggest problem that comes with analysing opinion polls is that there is no real disclosure on the methodology adopted by them. With the sole example of CNN-IBN-The Hindu-CSDS poll, which did up a detailed note on their methodology, none of these polls showed complete disclosure.
More significantly, however, on the question of opinion polls, is what Rajeeva Karandikar, director of the Chennai Mathematical Institute and one of India’s top election statisticians, said. In a well-articulated article, he said that the churn (change in voting intentions) that happens in the last few days before the elections is much more than that in the preceding weeks. Therefore, the ability of an opinion poll to accurately predict results, even weeks before the elections are conducted, is weak at best. This begs the question, as to why Headlines Today would spend a one hour segment on prime time discussing possible election results in 2014. Surely, they’ve got more urgent issues to address.
The Election Commission has sought to ban opinion polls, once the election polls begin. The rationale used by the commission falls in line with the passing of an amendment of Representation of People Bill in 2009, which banned exit polls till after completion of polling. Essentially, there is a belief that opinion polls confuse the voter and thereby affect the ‘sanctity’ of the electoral process. The ban on opinion polls has been supported by all political parties across the board. Looking at the figures mentioned in Times Now-C Voter poll, it is hard to see, how much of effect opinion polls have on the overall electoral outcome.
Next Story
