In a country where political leaders hop parties, parties switch alliances and voter choices are volatile, opinion polls offered mathematicians a platform to demonstrate the power of statistics, Rajeeva L. Karandikar, Director, Chennai Mathematical Institute said on Thursday.
Addressing a lecture-discussion on ‘The science behind opinion polls’ hosted by The Hindu Centre for Politics and Public Policy, Prof. Karandikar said while the opinion poll was neither “the truth, the whole truth and nothing but the truth” nor wholly about “lies, damned lies and statistics”, sampling, if properly done, had the power of determining the winner.
This held good even if the sample size was as small as 4,000 irrespective of whether a constituency had one lakh voters or 20 lakh voters, he said.
“This is something which is counterintuitive but it is not a matter of faith but of calculation…the point is that it is only the percentages and not the absolute numbers that determine the accuracy,” Prof. Karandikar said.
He pointed out that even in the case of the recent sting expose on some opinion poll agencies willing to manipulate poll projections, the “problem was not with the statistics but its usage”.
However, opinion polls, “which are as much about the power of statistics as their limitations”, can be off the mark too. In a neck-and-neck contest, no survey can tell you who the winner will be, but a psephologist is on safer ground when the differential in vote shares between winner and loser is about 4 per cent. Prof. Karandikar recalled once telling BBC’s psephologist Clive Payne that in India the percentage of safe seats (to predict from a psephological perspective) was only 10 per cent or less and that the majority of the seats in any general election was up for grabs. He pointed to a 1998 survey that showed that 30 per cent of voters had changed preference between the time they were first contacted before the poll and followed up after the election day.
Later, answering a question, he said while he would not rule out opinion poll results influencing undecided voters to go with the winner, Prof. Karandikar still felt that, law permitting, it was better to “do a honest and effective job” with engaging in opinion polls.
P. Ramajayam, assistant professor, Bharathidasan University, Tiruchi, and state coordinator, CSDS Lok Niti, said unlike market research-led opinion surveys Lok Niti field staff discussed with respondents a 20-page questionnaire that elicited virtually every aspect of their social, economic and political life.
Setting the context of the topic, N. Ram, Chairman, Kasturi & Sons Ltd., said the role and credibility of opinion polls was under discussion as some political parties wanted a ban on them, the Election Commission seemed “allergic” to exit polls and opinion polls, and more recently, after a sting operation had raised questions about the integrity of some opinion polls.