WHY THE POLLSTERS GOT IT ALL WRONG

Published June 16, 2024
A man at an appliance store in Ajmer watches the exit polls on June 1: several exit polls predicted the NDA would comfortably return to power, saying that the BJP alone would win more than 300 seats | AFP
A man at an appliance store in Ajmer watches the exit polls on June 1: several exit polls predicted the NDA would comfortably return to power, saying that the BJP alone would win more than 300 seats | AFP

Disturbed by the hugely misleading exit polls that projected a runaway victory for Indian Prime Minister Narendra Modi and his Bharatiya Janata Party (BJP) coalition partners in the National Democratic Alliance (NDA) in the recent general elections?

If so, you may be primed to side with the opinion that American writer E.B. White offered in 1948, saying: "The so-called science of poll-taking is not a science at all but mere necromancy.”

In other words, to White’s mind, psephology (the statistical study of elections and trends in voting) was no better than summoning ghosts and goblins to figure out one’s fate, often spawning tragic consequences, such as the one encountered by Macbeth.

Shakespeare scholar Professor Andrew Cecil Bradley called the witches' prophecy an "equivocation of the fiend." But there was no equivocation in the Indian exit polls. There was no doublespeak or a forked tongue that would insidiously support both sides of an argument. The exit polls, as Indian National Congress (INC) leader Rahul Gandhi led the field in suggesting, were a straightforward trick, a confidence trick, played on the small investors, for the benefit of market punters.

India’s leading pollsters, nearly all of whom had predicted a clear sweep for the BJP, turned out to be so spectacularly off the mark that speculation is on the rise about who benefitted

At its core, psephology is best described as the science of reading the pebbles. The term draws from the Greek word for pebble — psephos. The ancient Greeks used pebbles to vote. Similarly, the word ballot is derived from the mediaeval French word ballotte, meaning a small ball. In its less suspect avatar, psephology attempts to both forecast and explain election results. Exit polls, like the ones seen in India, are offered as a reliable tool in this endeavour to divine the people’s will, before the vote is counted.

Why on earth would anyone want to know the results of any elections, in this case Indian elections, before the results are officially out? Correction. Why would anyone be interested in the likely and uncertain results when the incontrovertible and verifiable verdict would be published soon? Correction. Why would anyone be interested in exit polls about a major election when exit polls are generally wrong?

There’s a possibility for yet one more correction here. Why are exit polls allowed in gambling-prone countries like India, where they are known to be used primarily to wreak havoc at the stock market, targeting the small investors mostly?

People bet on anything for a quick profit. The number plate of the arriving car, odd-even, cards, rain. Indira Gandhi banned public access to monsoon forecasts because Gujarat traders allegedly exploited gullible farmers with their insight — not too different from insider trading — in the days when the farmers relied on instinct, not satellites for weather outlook. At the height of the Gujarat massacres in 2002, there were reports of betting going on about where the next killings would be and on how many victims there would be.

A sadhu shows his ink-marked finger after voting in Ayodhya, Uttar Pradesh: even Ayodhya, which was assigned a special place in Modi’s chauvinist politics, decided to dump him | Reuters
A sadhu shows his ink-marked finger after voting in Ayodhya, Uttar Pradesh: even Ayodhya, which was assigned a special place in Modi’s chauvinist politics, decided to dump him | Reuters

Hansie Cronje and Mohammed Azharuddin, among others, led the field in cricket and also allegedly with betting syndicates. There has been a cultural somersault with the economic opening, as India has embraced its free market regime. Gone are the days when Raj Kapoor movies, Shri 420 in particular, berated the “satta bazaar” [betting market] and where a card sharper could become a villainous investor on the bourse.

In Delhi in the 19th century, Mirza Ghalib was jailed for hosting gambling under the watch of his patron, the last Mughal emperor. In terms of chronology, note the legend of the Mahabharata, where the lure of gambling took a heavy toll on Draupadi, the wife of the five brothers who lost her in a single throw of the dice.

Therefore, it was not an unusual event that the Indian stock markets went up with the exit polls and crashed with the actual results. The actual outcome being the opposite of what was predicted by the so-called pollsters with their so-called scientific tools was not surprising.

Rahul Gandhi was not alone in pointing out the difference between previous episodes, which too had a habit of missing the mark often enough by a wide margin, and the exit polls this time around. It was clear through data that there was suspicious and mysterious stock market activity around the exit polls and election results, by which a group of foreign investors gained, and millions of Indian small investors lost wealth.

The key point in Rahul Gandhi’s accusation was that Prime Minister Modi and the Indian home minister Amit Shah had publicly exhorted people to invest in the stock markets, because the results would lead to a boom. Were Modi and his fellow Gujarati aide culpable in a potential crime? Rahul Gandhi has demanded a parliamentary investigation into the matter, similar to the 1980s’ Bofors guns scam, which had targeted his father.

Exit polls in India, as shown in the data above, greatly overestimated the NDA’s popularity
Exit polls in India, as shown in the data above, greatly overestimated the NDA’s popularity

Let’s turn to a more phlegmatic view of what happened and if it were tantamount to a scam. So this is what happened, as The Hindu explained in its diligent way. Ahead of election results on June 4, several exit polls predicted the return of the NDA to power, with a tally of more than 300 seats for the BJP alone.

One pollster, Today’s Chanakya, predicted 400 seats (plus or minus 15) for the NDA, and another pollster, Axis My India, said the NDA would win an average of 381 seats. All these polls were way off the mark, as the results showed.

The well-regarded CSDS-Lokniti post-poll predicted that the NDA would receive a vote share of 46 percent, while that of the INC-led Indian National Development Inclusive Alliance (INDIA) bloc (excluding the Left, Trinamool Congress in West Bengal and Aam Aadmi Party in Punjab) would be 35 percent. The error margin was 3.08 percent points.

The results showed the NDA bagged 293 seats (43.63 percent vote share) and the INDIA bloc 205 seats (excluding the Trinamool Congress, which won 29 seats) with a vote share of 37 percent. CSDS-Lokniti did not project seats for the alliances but predicted that the NDA would return with a majority. Its vote share figures were within the error margins but were 2.5 points higher for the NDA.

The election results led to the NDA amassing far fewer seats in the Lok Sabha (lower house of the Indian parliament)  than had been predicted by the pollsters
The election results led to the NDA amassing far fewer seats in the Lok Sabha (lower house of the Indian parliament) than had been predicted by the pollsters

Axis My India projected 47 percent for the NDA and 39 percent for the INDIA bloc and the actual results showed that the projections overestimated the NDA vote share beyond the error margins. C-Voter projected a seat tally of 353 to 383 seats for the NDA, with a vote share of 45.3 percent and 38.9 percent for the BJP alone, which was 2.3 points higher than the actual vote share for the party — 36.56 percent. Its figures for the INDIA bloc were also roughly 2.4 points lower than the actual mark. While the vote shares were within error margins nationally, its seat tallies were way off, across several states.

Naturally, when the stock market reopened on June 3, after the weekend, it rose to an all-time high, driven by the exit polls’ prediction of a third successive term for Modi with an untrammelled majority. Evidently, the group of foreign investors who suddenly bought huge amounts of shares on May 31 saw their values rise enormously.

On June 4, when the actual results were being declared, it became clear that every single exit poll was way off the mark, and Modi was struggling to get even a simple majority. The stock market panicked, and we all know what happened.

On October 9, 2021, the then chancellor of Austria was forced to resign in a unique scandal. He had rigged opinion polls in his favour, forced the media to show these surveys, and won his election. Indian pollsters may have done even better.

The writer is Dawn’s correspondent in Delhi. He can be reached at jawednaqvi@gmail.com

Published in Dawn, EOS, June 16th, 2024

Opinion

Editorial

Afghan challenge
Updated 15 Jul, 2024

Afghan challenge

Foreign states must emphasise to the Afghan Taliban diplomatic recognition and trade relations all depend on greater counterterrorism efforts.
‘Complete’ justice
15 Jul, 2024

‘Complete’ justice

NOW that the matter of PTI’s reserved seats stands resolved, there are several equally pressing issues pertaining...
Drug fog
15 Jul, 2024

Drug fog

THE country has an old drug problem. While the menace has raged across divides of class and gender, successive ...
Miles to go
Updated 14 Jul, 2024

Miles to go

Some reforms agreed with the Fund are going to seriously impact economic growth and fresh investments, at least in the short term.
Iddat ruling
14 Jul, 2024

Iddat ruling

IT was a needless, despicable spectacle which only ended up uniting both conservatives and progressives in ...
Cricket shake-up
14 Jul, 2024

Cricket shake-up

SOMEONE had to take the blame and bear the brunt of the fallout from Pakistan’s disastrous showing at the T20 ...