Pre-poll surveys: Great reads but pure fiction
New Delhi: How gullible is the average Indian when it comes to poll analysis for which several crores are spent by TV channels in order to boost their ratings? If past statistics are any indication, polling agencies have zero credibility when it comes to forecasting election results. Last week, there was a wide difference between what the C Voter was predicting as a possible result in the three States going for assembly polls, and polling results by C-fore and Times Now. One predicted a Congress victory in the three States while the other just predicted the opposite.
Who does one believe? Neither.
Just look at the last Lok Sabha polls 2014.
BJP’s overwhelming win in the 2014 General Elections was credited to Prime Minister Narendra Modi’s pull and popularity, promises and bombastic style. The Modi wave was cited as the single factor that turned the tides for BJP. No one came to even close to predicting the outcome. To be fair, even the BJP had no inkling they would win so overwhelmingly.
In the previous Lok Sabha elections, 2004 and 2008, almost all pollsters got the numbers wrong. In the 2004 elections, most polls predicted that NDA would retain power but they got it so wrong. In the 2009 elections, most polls correctly predicted that UPA would retain power, but they were way off the mark on seat projections.
In the Assembly elections too that followed, no one was able to gauge the extent of Mamta’s popularity or the clean sweep by the BJP in Uttar Pradesh or the first BJP government in North East by completely demolishing Congress in Assam that had been in power for 15 years straight.
These are the figures for Lok Sabha 2014.
2014 Lok Sabha elections: NDTV – Hansa Research came close but was nowhere near the final tally.
In the Lok Sabha elections, CNN-IBN –Lokniti–CSDS predicted 211-231 seats for NDA, 107-127 for UPA and 205 seats to others. The ABP- Nielsen survey gave NDA 236 seats, UPA 92 seats and 215 seats to other parties. Times Now-India – TV-CVoter predicted 227 seats for NDA, 101 for UPA and 215 seats for other parties.
NDTV- Hansa Research suggested 275 seats for NDA, 111 for UPA and 157 for others.
Most agencies maintained that NDA would emerge victorious but none had predicted that it would gain majority.
In the final tally, NDA won 336 seats, UPA won 60 seats and 147 seats were bagged by other parties.
|Parties||CVoter||CNN-IBN –Lokniti–CSDS||ABP-Nielsen||NDTV- Hansa Research||Outcome|
West Bengal Assembly Elections 2016: NDTV comes very close
The West Bengal Assembly elections of 2016 were highly charged, with a resurging BJP that was still riding on the Modi wave, a battered CPI-M desperate to revive and a boisterous TMC that was certain of its victory. But no one, least of all, the pollsters had any faith in her winning ability.
Mamata Banerjee was the single-most factor capable of pushing TMC through the roof. She had maintained her image of being aggressively pro-poor, pushed education policy, brought in general improvement and development in public works and infrastructure, besides working on rural connectivity. All these factors ensured she had a great people connect.
On the other hand, expectations were also high that the CPI-M, which was in power for 34 years, would put up a good fight, especially when it got into alliance with Congress.
In the actual results, the CPI-M got wiped out.
Here are the predictions against the final tally.
|Congress – Left||120||44||126||103||67|
Assam Assembly Elections 2016: Nielsen comes within earshot’s distance
In Assam elections, BJP did something unprecedented; it came to power in North East despite the region having a phobia for the ‘outsiders’, especially the Hindutva and everything related to it. Congress in North East always had the anti-incumbency factor, but there was no other choice beyond the grand old party, whereas the regional parties did not have enough resources with polarising ideologies.
BJP was smart enough to realize that the dynamics at play in North East were different from other parts of the country. Like in the case of Delhi and Bihar, it realised that Modi – wave may not necessarily work. Hence, it did two things extremely well: one was to choose the right chief ministerial candidate in Sarbananda Sonowal and second was giving importance to local leadership.
It also got lucky with ‘kingmaker’ Himanta Biswa Sarma walking away from Congress over differences with the then CM Tarun Gogoi and unhappy with Rahul Gandhi’s style of working. He described Rahul’s behavior as ‘humiliating’ him. Sarma joining BJP was nothing short of a blessing for the party, gaining an immaculate strategist renowned in the region for making and breaking governments.
The pollsters simply piggybacked on the Congress’ stronghold and fear for the saffron bridge in the region, to not even put BJP in the eye. They expected BJP to gain ground, but only on the back of its regional allies. One thing they grossly miscalculated – Sarma’s acumen and counter strategies.
The results speak for themselves.
|Parties||India TV – Cvoter||ABP-AC Nielsen||AVC||Times Now – India TV – Cvoter||Outcome|
Karnataka Assembly Elections 2018: None gets it right, TV9 – CVoter comes close
Karnataka Assembly Elections of 2018 was full of drama, both before and after. The BJP was just so certain of its win that it practically ruled out any pre-poll alliance but it ultimately failed to get the desired numbers. Congress’s performance was poor. But it was the regional party JD(S) that turned out to be a game-changer.
The Modi-Shah Jodi could not replicate their magic as their election campaign was reportedly not liked by the local masses. Their speeches were found short on facts and their language unethical, which clearly hurt the voters. On the other hand, the Congress had cemented its vote bank and even slowly furthered its vote share. Hence, the final tally stood very different from the projected figures.
But pollsters had absolutely no idea that this would be a tight race. The BJP was given the edge but the Congress – JD(S) alliance sneaked through.
Following are the figures.
|Parties||TV9 – Cvoter||ABP News-CSDS||Times Now-VMR||C-fore||Outcome|
Uttar Pradesh election 2017: India Today-Axis distant close
The Uttar Pradesh elections of 2017 proved to be a game changer for the BJP, establishing it as a party that others needed to take very seriously. It also showed how swiftly the party was gaining ground across the country. But the pollsters were all off the mark by a huge margin.
If anything, the Modi wave was again a huge factor here. Modi’s campaign was very focused, he did not try appeasing ‘all’ but pitched to the Hindu-strong belt, which worked in his favour. He also touched upon issues that others had so far underestimated, like appealing to women. Add to this, his image of someone working towards propelling the country to new heights.
The infighting within SP also helped BJP gain from uncertain voters.
BSP’s splurging of state funds on needless monuments and parks in its previous tenure, without anything on offer for the Dalits, meant Mayawati’s penance would be long drawn. Further, they felt marginalised under the Samajwadi Party rule.
Did anyone really predict this?
Look at the predictions and actual results.
|Parties||India TV-CVoter||India Today-Axis||ABP News-Lokniti-CSDS||The WEEK-Hansa Research||Outcome|
|Congress – SP||138-162||97-104||187-197||178-182||54|
Punjab Assembly Elections 2017: India Today-Axis comes close
Congress was always expected to do well in Punjab. BJP was betting big on the Shiromani Akali Dal, with the election of 2012 being a successful alliance. AAP was expected to be a game-changer. The pollsters were again way off the mark, especially with regard to the AAP.
While the Congress had a strong chief ministerial candidate in Amarinder and the SAD-BJP alliance banked on the tested leadership of Parkash Singh Badal, the AAP refused to project a name.
But none of the pollsters were expecting the Congress to romp home in the manner that they did.
The figures tell their own story.
|Parties||Huff Post-CVoter||India Today-Axis||Lokniti-ABP-CSDS||The Week-Hansa Research||Outcome|
|BJP – SAD||11||18-22||50-58||28-30||18|
|Congress – SP||43||56 – 62||41-49||49-51||77|
|AAP||63||36 – 41||12-18||33-35||20|
Tamil Nadu Assembly Elections 2016: Times Now-India TV-CVoter comes closest
Prior to the 2016 elections, Tamil Nadu had been somewhat similar to Rajasthan in its election-on-election verdict, alternating power between DMK and AIADMK every election. Pollsters were certain that DMK would come to power, replacing AIADMK. But AIADMK retained power, becoming the first party to do so since 1984 in the state.
Late J. Jayalalitha initiated schemes that appealed to the poor, but did not promise total prohibition of alcohol in the state and besides, there were too many candidates in the fray which split the voters. These were some factors. But most importantly, many simply followed the alternating power factor to come into play.
For one, Times now – C Voter came close to an accurate prediction. The rest were off the mark.
|Parties||Times Now-India TV-CVoter||News Nation||Cauvery News||Puthiya Thalaimurai – APT||Outcome|
It is evident from the above data that pre-poll predictions are usually way off the mark. How do they get it right sometimes? Figures seem to indicate that whenever pollsters get it right, it is generally more by fluke. And yet, many of us lap up all polling data each time. For many, poll pundits’ election predictions are served only to reinforce one’s own views on who will win. Partisan readers and political parties brandish predictions to convince others of an imminent victory.
The methodology is akin to crystal ball gazing. And when served with a flourish, it is nothing short of pulp fiction!