Jump to content

2014 Lok Sabha Elections thread |Exit poll results


ganeshran

Recommended Posts

lets wait till results are out then we will see how much they get this time. I am quite sure it will be lesser.
It's not as much about the Muslim vote - BJP will never get a substantial percentage of that till it dissociates itself from the Sangh Parivar. It is more about the moderate Hindu vote. Moderate Hindus are much more likely to vote for Vajpayee than Modi.
Link to comment
Share on other sites

I've been seeing these surveys for the last 4-5 months and every single one of them has been giving very few seats to BSP - come May 16th all these poll pundits would be in for a big surprise regarding BSP's seats IMO.
Problem with these surveys is that there is very little impact factored in on the local candidate, which still plays a very big role in how people vote. Either these surveys are held before the announcement of candidates or do not sample enough constituencies to measure the impact. Unlike, United States there is a critical impact of the local candidate. This always tends to overestimate the number of seats to national parties because in the absence of the information on the local candidate people are much more likely to say Congress or BJP, if they are indifferent to the regional parties. A big part of this vote sways to the regional parties once local candidates are announced by the time of actual voting.
Link to comment
Share on other sites

It's not as much about the Muslim vote - BJP will never get a substantial percentage of that till it dissociates itself from the Sangh Parivar. It is more about the moderate Hindu vote. Moderate Hindus are much more likely to vote for Vajpayee than Modi.
Actually in some seats BJP had been getting Muslim votes but won't get this time. Ex - in Lucknow the BJP had been getting Shia votes since AtalJi started contesting but this time they will not vote for BJP. Agreed. In this election I think some of the moderate Hindus won't vote because of lack of options - they can't see Modi as a viable option and Cong is simply not an option at all.
Problem with these surveys is that there is very little impact factored in on the local candidate, which still plays a very big role in how people vote. Either these surveys are held before the announcement of candidates or do not sample enough constituencies to measure the impact. Unlike, United States there is a critical impact of the local candidate. This always tends to overestimate the number of seats to national parties because in the absence of the information on the local candidate people are much more likely to say Congress or BJP, if they are indifferent to the regional parties. A big part of this vote sways to the regional parties once local candidates are announced by the time of actual voting.
In case of BSP the problem for the News Channels/Survey agencies is that they deliberately try to conduct their campaign under the radar and their voters also stay silent till election day. I would think that it is very difficult to conduct a good opinion poll/election survey in a country like India because the issues that matter can vary drastically from one constituency to another so generalizations/extrapolation wouldn't work well. Also, people can say something now but on election day various factors (be it something like what Shahi Imam said today or alcohol or money or a hate speech like Varun Gandhi's last time etc) can influence their choice.
Link to comment
Share on other sites

Problem with these surveys is that there is very little impact factored in on the local candidate, which still plays a very big role in how people vote. Either these surveys are held before the announcement of candidates or do not sample enough constituencies to measure the impact. Unlike, United States there is a critical impact of the local candidate. This always tends to overestimate the number of seats to national parties because in the absence of the information on the local candidate people are much more likely to say Congress or BJP, if they are indifferent to the regional parties. A big part of this vote sways to the regional parties once local candidates are announced by the time of actual voting.
The problem is much more fundamental - these surveys don't perform random sampling to the 'T'. In addition, they don't take into account basic flaws like priming etc. Also, they don't take into account the "First past the post" system. More often, they calculate vote share and translate it to seats which doesn't really work. Hence, they always miss the point of an opinion poll - predict the "trend" and NOT predict "seat percentages etc.".
Link to comment
Share on other sites

Most of the surveys are saying it repetitively that in multi corner contest, there are more chances of errors as Seat predictions could go totally wrong. But in two party situation where 3rd option is extremely weak, they have been quite correct. After all it is just an opinion poll. It can never predict it accurately. But at least NDA is going to be upward of 200. How much maximum it can get is what we will have to wait until May 16th.

Link to comment
Share on other sites

2014 Lok Sabha Elections thread

Actually in some seats BJP had been getting Muslim votes but won't get this time. Ex - in Lucknow the BJP had been getting Shia votes since AtalJi started contesting but this time they will not vote for BJP.
Yeah, Vajpayee got a lot of the Muslim vote in Lucknow. However, in 2004 a lot of Muslims voted tactically against the BJP overall because of what happened in 2002 and the fact that Vajpayee was unable to replace Modi under RSS pressure. It's interesting that if Vajpayee had replaced Modi in 2002, he not only would have won in 2004 but we would not have had a lunatic PM in 2014. Indian polity as a whole would have gravitated more towards the center, and we would have also been spared the policy paralysis of the last 5 years.
True. Among polls, I usually rate NDTV's as pretty high due to Roy, but I can't get my head around BSP getting only 7 seats and SP losing only 2% votes. Yeah, though I feel towards the end it is local issues which make more of an impact. For example, the Shahi Imam had endorsed Vajpayee in 2004 but it hardly translated into votes whereas Varun Gandhi's speech, which had more of a local impact really dented the BJP in UP in 2009. That's where India is a psephological nightmare because at times issues don't even carry over from one constituency to the next.
Link to comment
Share on other sites

The problem is much more fundamental - these surveys don't perform random sampling to the 'T'. In addition' date=' they don't take into account basic flaws like priming etc. Also, they don't take into account the "First past the post" system. More often, they calculate vote share and translate it to seats which doesn't really work. Hence, they always miss the point of an opinion poll - predict the "trend" and NOT predict "seat percentages etc.".[/quote'] True. I was highlighting one aspect of their flaws, but there are many more. Though on the vote to seat conversion, you can get in the approximate range based on historical figures. That is one of the reasons you see exit polls to being much more accurate than opinion polls.
Link to comment
Share on other sites

Most of the surveys are saying it repeatatively that in multi corner contest, there are more chances of errors as Seat predictions could go totally wrong. But in two party situation where 3rd option is extremely weak, they have been quite correct. After all it is just an opinion poll. It can never predict it accurately. But at least NDA is going to be upward of 200. How much maximum it can get is what we will have to wait until May 16th.
This. A point which every Indian voter needs to understand. How a multi corner contest benefits the non performing incumbent by splitting votes. In a two party situation, a non performing incumbent is sure to lose
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...