By Sunil Garodia
First publised on 2021-03-25 08:55:37
Only yesterday, this correspondent had written about the results of the ABP News-CNX opinion poll about the forthcoming elections in West Bengal. That poll had predicted a hung assembly with massive gains for the BJP. Later in the evening, the results of the latest Times Now-C-Voter opinion polls were published. This poll predicted that the TMC would return to power in the state as the gains made by the BJP will not be enough to force a change of guard. The C-Voter poll also predicted a higher vote share for the TMC than predicted by the CNX poll.
What is the reason behind different opinion polls conducted virtually at the same time throwing up different results? There could be two main reasons: sample size and choice of constituencies (or even the weightage given to different parameters). For example, a poll with a lower sample size is likely to come up with an erroneous forecast. Similarly, a poll conducted mainly in constituencies which are considered to be strongholds of one or other party is likely to be biased. Hence, it is very difficult to compare two polls that do not have exact or almost similar parameters.
Yet, when numerous polls are being conducted simultaneously, analysts have to give their opinion. The Times Now-C-Voter poll has by and large returned results that were published in polls conducted a couple of months earlier. In fact, it has showed that there has been renewed support for the TMC in these couple of months. So have the people of West Bengal begun to sympathize with Mamata Banerjee after the 'attack' on her in Nandigram? Has Mamata successfully played the 'victim' card? Or is it that they have begun to believe that "outsiders" (as Mamata calls the BJP leaders who come from other states) should not be given a chance? The C-Voter poll also shows that not many people believe that the Left-Congress-ISF alliance will be able to wean Muslim votes away from Mamata Banerjee.
The difference in seat projections in the above two polls are mainly due to the difference in projected vote share. While both the polls give almost the same vote share to the BJP at around 37.5 percent, CNX gives TMC just 40 percent while C-Voter gives it 42.1 percent. This 2 percent extra results in more than 25 extra seats for the TMC in the C-Voter poll. While the CNX poll predicts that the Left-Congress-ISF alliance will get 16.2 percent vote share (thereby also predicting that some Muslim voters will desert the TMC), the C-Voter poll gives the alliance only 13 percent vote share. This is why the end results in seat projections are so vastly different.
While psephologists are having a field day now by juggling with sample size and field reports, it is upon the people of the state to decide their next government by voting in eight phases from March 27th. It remains to be seen whether the people of Bengal vote for "asol poriborton" or stay with "banglar meye".