All exit poll predictions went awry this time. Pradeep Gupta, Director and CMD of Axis My India, a market survey and intelligence company, found himself on the receiving end of snide remarks from Congress spokesperson Pawan Khera on a TV channel. Khera insinuated ulterior motives behind Gupta’s predictions, linking them to stock market speculations, as the Sensex experienced volatility after election results came in.
Market sentiments built on false dreams of exit polls caved in at the onset of actual results. Gupta is a top psephologist, and his company boasts of 94 per cent accuracy in exit polls. It seems this was a case of six per cent on the wrong side. However, Gupta was not an exception. Predictions of around 10 pollsters, who had tied with different media houses, went horribly wrong.
It would be too much of an accusation that all of them were manipulated to suit the narratives of a particular political party. However, nothing is too much in today’s polarised and raucous political climate. This is not, however, the first time exit polls have gone haywire. They have a non-uniform track record over the last 25 years they have been held in India. They have sometimes gone right, even dead right, and wrong on other occasions. We have repeatedly tried to comfort ourselves into believing that their accuracy has improved over the years, as is the case with weather predictions. That attitude is always a hint of a bigger debacle.
Interestingly, it was DD (then called Doordarshan), that first legitimised exit polls in India. At that time, it was trendy in liberal democracies like the UK, USA, Israel etc. On February 28, 1998, a minute after polling closed for the 12th Lok Sabha at 5 pm, DD aired exit polls conducted by GVL Narasimha Rao for a prominent newspaper and Pritish Nandy Communications. These polls, based on a sample size of 26,000 voters across 1,450 polling booths in 120 constituencies, continued late into the night. Sample sizes have drastically increased since then, reaching an average of around five lakh voters today.
However, the perils of exit polls were identified by Pritish Nandy even at that time. In an article on March 5, 1998, he argued that exit polls, unlike opinion polls, were a ‘dangerous and scary business’. “You can be proved wrong within days — in fact, hours”, said Nandy, “and there are enough people prepared to insinuate motives to what can quite easily be described as systems failure, to quote Dr Manmohan Singh’s famous definition of securities scam.”
A quarter-century later, Nandy’s predictions ring strikingly true. A pollster, risking their reputation to make predictions, faces the potential for swift public scrutiny and accusations of ulterior motives, even if flawed findings stem from systemic issues rather than deliberate manipulation. An exit pollster is actually risking too much by sticking his neck out to call the numbers. It would be inane to think that he would allow his reputation to be destroyed for thirty pieces of silver from the stock market (even if someone is actually offering him a good price for his soul). Nor is he in a position to sway the electorate’s choice, since exit polls are disseminated after voting concludes.
Exit polls have proliferated alongside the rapid expansion of electronic media over the past two and a half decades. But do they hold any real significance, given that election results are usually declared within a few days? What value would they have once the actual results are declared? Their only purpose seems to be providing seasonal employment to some psephologists, and evening-time entertainment to the rest of the nation.
The Representation of the People Act, 1951 under its Section 126A merely prohibits conducting and disseminating exit poll results from the commencement of polls until the completion of all phases. Furthermore, the Election Commission of India’s proposal titled ‘Ban on Exit Polls and Opinion Polls’ (ECI, Proposed Electoral Reforms, P.57-60), does not actually advocate for a complete ban on exit polls. Instead, it merely prescribes additional regulations. Perhaps, the pollsters could get the better of such a ban by invoking their constitutional right to freedom of expression.
However, exit polls require strong disclaimers, clearly stating that, much like tarot card readings in the media, they are for entertainment purposes only and lack legal or scientific sanctity. This columnist can say from his own experience that tarot card readings can be unbelievably accurate on certain occasions. Whether such accuracy can be extrapolated to all situations is debatable. However, such strong disclaimers might be counterproductive, reducing exit polls to mere reality shows. A psephologist would loathe to be compared with a tarot card reader, insisting his methods are more “scientific,” while tarot readings rely solely on intuition.
People chase exit poll numbers in search of the truth. However, the question arises, why do we need exit polls, when the actual results are close at hand?
Could we not do away with exit polls? They appear to be an electoral avatar of the tradition of gambling on Diwali night. We could win, or we could lose, yet we play.
The writer is author of the book ‘The Microphone Men: How Orators Created a Modern India’ (2019) and an independent researcher based in New Delhi. Views expressed in the above piece are personal and solely those of the author. They do not necessarily reflect News18’s views.