Tuesday, 22 April 2014


Opinion Poll in India – A Necessary Evil?




An opinion poll surveys a particular sample and presents an explicit shape to public opinion before an incoming election. One of the pioneering opinion polls happened in history was conducted during the contest for the United States Presidency in 1824 by the Harrisburg Pennsylvanian. However, the methodology adopted for opinion polls must be nation-specific because of a particular nation’s unique form of democracy.
Opinion polls, such as the ones conducted by the joint venture of the Centre for the Study of Developing Societies (CSDS) and CNN-IBN, prior to India’s general election in 2014 brought forward a variety of concerns associated with the system of opinion polling. Firstly, in response to the issue whether extrapolation of opinion of a small fraction of the electorate can be accepted as an authentic exercise, Rajeeva Karandikar, Director of the Chennai Mathematical Institute, claimed to have developed a model which can determine, among more than two candidates, the winner with ninety-nine percent confidence if the gap between the winner and the candidate coming second is at least four percent [1].
Secondly, the conversion of vote share into seats in an opinion poll is a potential pitfall in the prediction of the Indian election because of the possibility of not winning any seat by a hopeful party even after having significant vote share in the opinion poll.
Thirdly, opinion polls cannot estimate bull-whip effect – the influence of undecided voters on the day of actual polling [2]. Although exit polls can capture such effect, it is impractical to choose respondents from a pre-chosen list, and hence sample profiles in exit polls often differ widely from population profiles.
Fourthly, the underestimation of minorities and new political entrants can alter the prediction of opinion polls in totality. For instance, no one predicted Sheila Dixit’s defeat at the hands of Aam Aadmi Party (AAP) in the recently concluded Delhi Assembly election.

The feedback effect of opinion polls on the electorate has a mix of constructive and disastrous consequences. For one, voters can make informed choices because of the deliberation of several electoral manifestos in different channels of mass media. However, such polls can be potential tools of misuse depending on the user. Therefore, the latest demand for ban of opinion polls is not thoroughly justified and such polls may have a scientific basis albeit taken with a pinch of salt.

References
1. “Having their say”, Rajeeva Karandikar, Frontline, Vol. 31, No. 6


Priyadarshika Talukdar
M.Sc Media, P.G 1

No comments:

Post a Comment