Tuesday, 24 September 2013
Why opinion poll surveys are not to be trusted!
Indian voters have always proved that mood in the opinion polls and the actual results of the election are often contradictory in nature. It is the basic law of nature, “the actual can never be staged, the way exactly it is going to happen.” Being said that, most scientifically conducted opinion polls might throw some light on what is going to happen, but it won’t disclose no matter what, what will actually happen?
*Why opinion surveys are no accurate?
Advantages of survey method of opinon collection:
• Relatively easy to administer
• Can be developed in less time
• Cost-effective
• Capable of collecting data from a large number of respondents etc..
Disadvantages:
• Respondents may not feel encouraged to provide accurate, honest answers
• Not Ideal for Controversial Issues
Finally the whole idea of opinion surveys or polls, is generalization of the results of a few crores[may be 2-5 crore ppl on 100 crore+ ppl]. In a country like India, generalization has least effects.
Now let us consider some facts and figures:
The following data is about the 2009 general elections:
Predictions and Polls for the 2009 Lok Sabha Elections in India
1. According to the second Star-Nielsen poll, the Congress will win 155 seats (203 with UPA allies), while the BJP will win 147 seats (191 with NDA allies) (via TOI).
2. Arun Nehru at the Deccan Chronicle predicts that the Congress will win 157 seats (193 with UPA allies), compared to the 132 seats for the BJP (177 with NDA allies).
3. The Times of India predicts that the Congress will win 154 seats (198 with UPA allies), while the BJP will win 135 seats (176 with NDA allies).
4. According to The Week, the Congress will win 144 seats (198 with UPA allies), while the BJP will have to settle at 140 seats (186 with NDA allies).
5. Shreekant Sambrani at Business Standard predicts that BJP will emerge as the biggest party with 137 seats (184 seats with NDA allies), ahead of the Congress, which will win 119 seats (176 seats with UPA allies, including LJP/ RJD).
6. India TV also predicts that the BJP will emerge as the largest single party with 144 seats (187 seats with NDA allies), ahead of the Congress, which will win 133 seats (178 seats with UPA allies).
7. According to BJP's own survey, conducted by G V L Narasimha Rao, the BJP will win 160 seats (217 with NDA allies) compared to 135 seats for the Congress (180 with UPA allies) (via TOI).
Why so much diversity among various polls? Because, opinion polls are conducted on a small group of people. And then these results are generalized over a larger population. So disparities are expected to occur.
*Also nowadays online opinion polling is a trend. But can one really expect rural people to go to cyber centres in their villages or travel to the nearest city to a cyber cafĂ© and take part in online polling. That’s pretty much bullshit. As most of Indian population comes rural[The Rural population (% of total population) in India was last reported at 69.90 in 2010, according to a World Bank report published in 2012], online based opinion polls showing 80% ppl voting modi lead bjp [http://election.intoday.in/] and many others are totally biased.
Now lets view the facts of actual 2009 General Election Results:
Results of General Elections 2009 (543/543)
Alliance Seats Won
NDA 160
UPA 262
Third Front 79
Fourth Front 28
Others 14
So, when we compare the actual and opinion survey facts, we can see how inaccurate, opinion surveys can be. But this do not mean opinion surveys be abandoned totally. But, in fact, social scientists must try and discover more avenues to obtain accurate opinion results.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment