Exit polls add to the suspense

Counting of votes in Assam, Kerala, Tamil Nadu, West Bengal and the Union Territory of Puducherry is scheduled for May 2. (Representative image)

Exit poll numbers have been declared for four states and one union territory where elections were held in March and April. While there is consensus among pollsters on Assam, Kerala, Tamil Nadu and Puducherry, it could go down to the wire in West Bengal.

The Communist Party of India (Marxist)-led Left Democratic Front (LDF) is predicted to break the near four decade trend in Kerala and retain power. In Tamil Nadu, the Dravida Munnetra Kazhagam (DMK)-led alliance is projected to sweep the polls and the state could revert to one-sided elections as witnessed in the past.

The Bharatiya Janata Party-led National Democratic Alliance (NDA) is predicted to win Puducherry. It is also expected to retain Assam despite anti-Citizenship Amendment Act protests and a formidable alliance led by the Congress.

The verdict is split in West Bengal. Agencies like C-Voter predict a Trinamool Congress (TMC) victory, while others like Jan Ki Baat feel the BJP could dethrone the TMC. Agencies like Axis My India and CNX predict a very close contest.

Exit Polls Vs Opinion Polls

The increase in the number of channels and the advent of the social media has increased peoples interest in elections. There is a big market for exit polls since the results are not announced on the same day. Earlier, there were only three to four names, nowadays there are 8-10 exit polls for each election.

Exit polls are normally a better barometer of results than opinion polls. They may not be able to predict accurately the seats tally and vote share, but they do suggest the direction of the trend. However, increasingly exit polls are failing to predict election outcomes, more so in close contests. Bihar being the latest and prime example, and now West Bengal.

COVID-19 has complicated matters for pollsters hampering efficient sampling and data collection. Some agencies have opted for telephonic discussions, while others a mix of telephonic and on-the-ground personal interviews. In the first exit poll after the pandemic, in Bihar, all pollsters got it wrong.

Big Ranges Create Confusion

While there is consensus, big ranges have been predicted by pollsters, thereby creating confusion. The low and high end of range of predictions are in some cases bizarre. For example, the LDF is expected to win anywhere between 71 to 120 seats in Kerala. The Congress-led United Progressive Alliance (UPA) is expected to win anywhere between 40 and 66 seats in Assam. Come May 2, so wide are the projections, it is sure one of the pollsters will get it right.

In the battle of battles in West Bengal, some pollsters are essentially playing it safe by taking the middle ground. As per agencies, the TMC could win anywhere between 64 and 185 seats, and the BJP 104 to 192 seats. This suggests that the contest is very close. Another factor could be that because of the violence witnessed during elections, many voters are apprehensive of divulging their preferences.

In a fiercely fought election amidst a polarised environment, the margin of victory in many seats (64 seats as per Axis) is expected to be very low (<2%), normally 2-3 percent is the margin of error in any poll. This further complicates it for agencies.

Challenges Faced

Exit polls are based on the assumption that voters speak the truth. However, some may deliberately lie, others too scared to tell the truth, especially poor, downtrodden, lower caste. A lot of reliance is placed on historical trends, which may not be always correct, more so with the increase in the number of booths due to COVID-19.

The women voter is becoming an important demography. However, no survey in India can achieve, 50 percent female respondents. Thus, extrapolations lead to errors.

Another factor that affects exit poll predictions is the quality of work done by agencies. These days there are many agencies which are ready to do exit polls for nominal costs. Thus, cutting corners while collecting data and implementing approved methodologies impacts the outcome.

COVID-19 restrictions and the fear of violence, especially in West Bengal, could have led to the temptation among some surveyors to sample convenient locations/booths, such as urban booths, thus distorting the picture.

To sum up, exit polls quench our curiosity and anxiety between the last day of polling and counting day. They also have become a source of entertainment. In the end, the brouhaha yields nothing and we need to wait for the results day.