एग्‍जिट पोल्‍स के नाम पर टीवी शो तो नहीं बनाने लगे हैं ‘चुनावी विशेषज्ञ’?

The results of Haryana Assembly elections have become clear. BJP has got absolute majority. It has got 48 seats with about 40 percent votes. If we look at exit polls, these results are surprising. In the exit polls, less than 30 seats were estimated for BJP. The Congress, which was given 44 to 65 seats by the exit pollsters, could not even reach 40. Despite getting almost equal number of votes as BJP, Congress got stuck at 37 seats. The special thing was that this time the opinion of almost all the exit polls (I am not saying the results) was that BJP is leaving Haryana and Congress is coming back.

But, the results of EVMs have exposed the layers of exit pollsters. Once again it has been proved that a farce is created in the name of exit poll. Although Congress alleges that the Election Commission has acted under pressure, there is no solid basis for this allegation. The concrete thing is that the exit polls have been proved wrong.

Exit poll means knowing the political preferences of voters immediately after voting and drawing results based on the analysis of that information. Simply put, after voting, the voters should immediately know which party they voted for? Based on their answers, assess which party’s actual results may be in favor.

Exit polls are a scientific process
Exit polls are a scientific process. To complete this process properly, a lot of money, time, resources and networks are required. But, here in every election, surveys are basically conducted for TV channels. And most of the agencies conducting these surveys have much less resources and networks than required. Apart from this problem related to infrastructure, there are many other reasons due to which exit polls prove to be wrong. However, this reason is also related to lack of infrastructure somewhere.

If seen directly, if the question is asked in the right way and to the required number of selected voters and the answers are correct, then it is not difficult to correctly predict the results by analyzing it scientifically. Especially, when agencies that claim to be experts and have years of experience are engaged in this work.

Then why do exit pollsters go wrong again and again? Experts say that the results of exit polls depend on many things. Such as: not taking proper sampling, not asking questions properly, not getting correct answers, giving wrong weightage etc.

Sample Size: Meaning, how many people should be asked the question? The sample size must be in a certain ratio. A solid estimate can be made even by surveying 20,000 people out of a population of 100 crores, but it is important to have proper representation of these people. This means that if people with different electoral moods are not represented, the results will not be correct. For example, if the opinion of a large number of people is taken outside the polling station considered to be the stronghold of one party and the stronghold of the other party is left out, then the results will definitely be wrong.

Sample Selection: Meaning, how should the people to whom questions are asked be selected? This was discussed above. It is good to choose a random sample so that the possibility of the responses being biased towards one side can be reduced. Method of conducting survey: A lot depends on how the questions were asked. It also makes a difference whether the question was asked face to face or through phone etc., how the question was put to him… etc. Whom did you vote for? Many voters do not like to give correct answers to such direct questions.

How accurately has the weightage been given to the sample according to the population? This means that the voters should be included in the sample in proportion to the categories into which they are divided. For example, the ratio of the number of youth, elderly, women etc. and their sample number should not be mismatched.

what is the process
This is how surveys are done. Now let us talk about the process after conducting the survey. Many things have to be kept in mind while analyzing the answers received from the voters included in the sample. For example, Prannoy Roy wrote to Dorab R. In his book ‘Verdict’, written jointly with Supariwala, he has stated that while converting vote share into seat share, there is a need to keep in mind the political mathematics of that particular state or region.

Similarly, Rai has given an example and said that if a party has won the assembly elections of a state and Lok Sabha elections are being held there after a few months, then that party can get the benefit of ‘honeymoon effect’. In such a situation, to calculate the swing of parties there, the assembly elections should be made the basis and not the last Lok Sabha elections.

David W Moore has also said in his book ‘The Opinion Makers’ written in 2008 that exit polls often prove to be wrong because they are not able to remain completely biased. Many times, questions are proved wrong due to not asking them properly and being influenced by the narrative created by the media.

Tags: assembly elections, exit poll