Pollsters got it wrong in T.N.

But forecasts for other States were more or less on target.

May 20, 2016 02:45 am | Updated September 12, 2016 07:18 pm IST - New Delhi:

Barring one exit poll prediction, pollsters failed to see an AIADMK win in Tamil Nadu as the verdict in four States and one union territory became apparent on Thursday, once again raising questions about how pollsters get their numbers wrong, repeatedly.

Pradeep Gupta, chairman of Axis My Media, explained how his team of surveyors missed Jayalalithaa’s victory. “While a day before the elections, we were sure that women were voting for Ms. Jayalalithaa. On the day of the polling on May 16, for some reason my surveyors ended up mostly speaking to male voters who said they had voted for the DMK.”

Mr. Gupta also said that on polling day, pollsters do not have the luxury of going to the electorates’ homes. “But we have learnt a lesson this time,” he said.

His prediction was 89-101 for the AIADMK.

‘Examine the factor’

While that could be a serious admission to make raising questions about the demographic sample, others like Today’s Chanakya who too got the numbers wrong in the State, said “they have to examine the factors leading to the error.”

A spokesperson in the agency said, “We need to examine whether the AIADMK has won more seats with a narrow margin or that the DMK lost with higher margin.”

The spokesperson also said that while in closely fought elections, the wisdom is, the ruling party is almost favourably placed, it is not always right.

To drive his point home, Today’s Chanakya pointed to the Punjab elections of 2012 where there was a marginal difference between the Akali Dal alliance and the Congress’s votes but the difference was more than 25 seats when it came to actual seats. Their prediction: 90+/-9.

The pollster who got the State right, C-Voter’s Yashwant Deshmukh, said it was quite possible that the others were doing their exit-poll predictions for the first time in the State.

“I had got it right in 2001, 2006 and 2011 and it is unfair to compare C-Voter with those who got it wrong this time as the others were first-timers to the State,” he said. C-Voter’s prediction: 139.

The forecasts for other States were more or less on target from landmark gains in Assam for the BJP alliance to the Trinamool Congress victory in West Bengal. Says political scientist/politician Yogendra Yadav, “We shouldn’t read too much into the success or failure of the exit polls.

These polls are useful not for the predictions they make but for the wealth of archival information about how people voted. And that is important for a student of politics.”

Mr. Yadav said the reasons why pollsters got their predictions wrong were either because the sampling was not representative or there is an overrepresentation of one demographic. Another reason cited by Mr. Gupta for the mistake in Tamil Nadu was that the women came to the polling booths unaccompanied by men.

0 / 0
Sign in to unlock member-only benefits!
  • Access 10 free stories every month
  • Save stories to read later
  • Access to comment on every story
  • Sign-up/manage your newsletter subscriptions with a single click
  • Get notified by email for early access to discounts & offers on our products
Sign in

Comments

Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide by our community guidelines for posting your comments.

We have migrated to a new commenting platform. If you are already a registered user of The Hindu and logged in, you may continue to engage with our articles. If you do not have an account please register and login to post comments. Users can access their older comments by logging into their accounts on Vuukle.