Kolkata:
Six of Pradeep Gupta’s surveyors spent 24 days in jail in West Bengal. Their offence was asking voters about their polling preferences which the police charges alleged could lead to names being struck off the electoral rolls. The incident is a window into why West Bengal is “very different and difficult compared to the rest of India”, as Gupta, Chairman and Managing Director of Axis My India puts it in a conversation with NDTV Editor-in-Chief Rahul Kanwal.
Gupta, with a track record of 74 correct calls out of 81 polls since 2013, said what his teams found on the ground amounted to a quiet admission that Bengal may be the one state where even a competent pollster is flying partly blind.
“When we talked to 10 people, only two to three were willing to reveal whom they voted for,” Gupta said. “Those who haven’t spoken, which party has the more share or the less share in those particular 60 per cent, that is problematic,” he added, referring to the percentage share of voters unwilling to reveal their preferences in Bengal.
In most of the country, the share of voters unwilling to reveal their preferences is around 10 per cent, Gupta said. In Gujarat, it can reach 20 per cent; in the north-east, roughly 15 to 20 per cent.
But in Bengal, after this round of field work, the figure has crossed 60 per cent which is more than double what it was in the 2016 assembly election, when 20 to 30 per cent of respondents stayed silent, Gupta said.
“Fear plus SIR fear factor” is driving this silence, he said. While fear has been historically woven into Bengal’s political culture across decades, the newer factor is the Special Intensive Revision (SIR) of electoral rolls that resulted in the deletion of a significant chunk of names ahead of the election. SIR has added a specific anxiety to the existing one, he said.
“People think they don’t know who we are, what we are trying to ask. And if they say X party or Y party, their name may strike off from the voter roll. So this is another fear,” Gupta said.
It was this same fear that voting preferences could cause electoral roll deletions that formed the basis of the police action against Gupta’s surveyors.
On the massive 93-per cent turnout in the first phase, Gupta said that if the voter rolls were trimmed by roughly 10 to 11 per cent through deletions, the denominator shrank. If the same number of people who voted in 2016 when turnout was around 82 to 83 per cent of a larger base came out again, that same absolute number now yields a significantly higher percentage against the reduced base.
“If you divide 84 who voted by 90, what you see is total voter turnout in excess of 92 per cent,” he said.