The 2024 Parliament election results have surprised everyone. It was taken for granted for last few months that BJP is going to cross the halfway mark on its own and the whole discussion was about how high it will go?

The exit polls seem to suggest a bigger victory for BJP/NDA. All the ten polls that I saw on various TV channels gave at least 350 seats to the NDA and some of them even predicted 400 seats. All of them have been proven to be wrong. Where did they go wrong? What does the result mean for the governance of the country in the next five years?

It is clear that there will be a coalition government. With BJP being only 30-35 short of the majority mark, it will be close to 290 along with its pre-election allies; so it is likely that we will again see Narendra Modi as the next Prime Minister.

The government will have to be more careful in taking decisions and will have to take along its allies.

It should be noted that while BJP has kept its vote share, its seats went down by about 60! The major factor for this is that the leading contenders came together in several States and formed an alliance against BJP/NDA. This factor had been named the ‘index of opposition unity’ in the era when Congress was the dominant party. While this ‘index’ varied from State to State, it worked very well in UP and Maharashtra — the two largest States (in terms of seats in Parliament). These are the two States where BJP/NDA suffered the biggest loss. While the INDI alliance was a loose alliance, the fact that SP and Congress fought as allies in UP had a big impact.

Also Congress’ vote share in UP went up significantly. These led to the biggest upset for BJP/NDA. In Maharashtra, NDA suffered a big loss as the erstwhile partner Shiv Sena went out of NDA and aligned with Congress along with NCP. Later Shiv Sena and NCP split, but had limited impact. The next big loss is Rajasthan, where there was a direct contest between BJP and Congress. Perhaps several local factors may have contributed for the loss. In Bengal, the TMC on its own snatched about 6-7 seats from BJP.

Indeed, if BJP and NDA had not made gains in Andhra and Telangana, NDA would have been just above the majority mark.

Exit polls fiasco

What went wrong with exit polls? Since I was not associated with any exit poll this time, I can give my views! Firstly, the exit polls have their limitations as implementing proper sampling when choosing respondents as they exit from the polling booth is very difficult, if not impossible. We had worked on an alternative to exit poll — which we called the day-after-poll — where sampling is done by proper methodology and the chosen respondents are asked questions by going to their residence the day after they have voted. This had yielded very good results over the years.

People at large, including experts in media, politics and other walks of life, perhaps do not understand the importance of randomly choosing the respondents (or the sample) in statistical decision-making. They forget or do not know that the statistical guarantee that the sample proportion (of any observed attribute, including say supporters of a party) is close to the true proportion (in the population under consideration) kicks in only when the sample is chosen randomly! A randomly chosen sample of, say, 30,000 (even in such a large country as India) is better than a sample of size three lakh chosen by some other means. By better I mean that the chances of being widely off the mark are rather low.

The other issue is that perhaps psephologists make projections based on data and mix it with expert opinion in their decision. I can say that if projections were solely based on randomly chosen data, devoid of mixing it with expert opinion, we would not see a scenario where several agencies who have done surveys independently of each other go wrong by a big margin, and all errors are one-sided.

One good outcome of the results seems to be that the attack on EVMs and the Election Commission that we had seen over the last several weeks has now taken a back seat. After all, after making a big gain, the Opposition cannot attack EVMs and the EC, questioning their own gains.

There will be a lot of introspection among all parties about what was right and what was wrong. One would hope that the bickering among the ruling party/alliance and Opposition parties ends or at least comes down soon and the government goes on with decision-making.

The writer is Professor Emeritus, Chennai Mathematical Institute

Related Topics