Tuesday 29 April 2014

Credibility, integrity of opinion polls at stake!


A sting operation by a Hindi news channel appeared to show opinion polling agencies willing to manipulate the results on behalf of political parties for a price.
The contents of the sting operation conducted by News Express and aired first at a press conference on Feb 25 afternoon and then on the channel. Undercover reporters for the channel approached 13 polling agencies and were turned down by two (AC Nielsen and the Centre for the Study of Developing Societies). The remaining 11 offered to manipulate, to various degrees, their findings to suit the interests of a political party, the channel claimed.
C-Voter appeared to offer to raise the margin of error from 3 to 5 per cent to allow more seats to be shown for one political party or the other, the channel claimed.  The India Today group, which has in the past commissioned polls from C-Voter, announced the same evening that it was suspending all opinion polls by C-Voter for the group and had issued it a show-cause notice.
Another pollster featured in the sting, QRS (Quality Research and Services) appeared to be telling the undercover reporter that he had projected 200 seats for the BJP in Uttar Pradesh under one agency’s name and 200 for the Samajwadi Party under another agency’s name.
The global market research company Ipsos’s Indian head too appeared to offer to use margins of error to favour a party and project “fence-sitting seats” as belonging to the party in the media, even while privately informing the party of the “real” situation.
Mid-stream Marketing Research’s director Sanjay Pandey offered in the clip to manipulate survey data, the channel said.
A Development and Research Services (DRS) representative appeared to offer to delete unfavourable data, but a company representative said the person had since resigned.
Lead Tech Management Consulting’s Vivek Singh Bagri said the undercover reporter had put words into his mouth, and while he had played along at the meeting, he subsequently sent an SMS refusing to do anything unethical.
Five other little-known pollsters could not be contacted for comment as no official representative was available on the phone.
A number of opinion polling agencies approached by undercover reporters agreed to manipulate poll data, a television news channel has claimed, sparking a fresh controversy in a heated election season as senior ministers and political parties called for an investigation.
Clips from the sting operation aired by the channel showed many pollsters agreeing to produce favourable numbers by leveraging the so-called margin of error, a statistical concept meant to indicate the quality of sampling and the accuracy to be expected from survey results. “For a price, the prediction of seats tally can be changed to suit the interests of political parties.
 The agencies have no qualms accepting even black money for this purpose,” the channel, News Express, said in a statement. Representatives from these agencies are seen responding variously to the undercover reporters, who posed as consultants for political parties.
Congress and the Aam Aadmi Party called for an investigation. Union Law Minister Kapil Sibal urged the Election Commission to urgently take up the matter. “This (the expose) is a very serious development. It shows these agencies are trying to manipulate public opinion. It is nothing but paid news, manufactured to serve vested interests. This is an extremely serious offense that warrants a full-fledged inquiry... I also call upon the Election Commission to take all effective steps to ensure such shady operators are not allowed to manipulate the public opinion in the run up to the elections.”
In November last year, the Election Commission had written to the government asking that a law be passed to restrict release of opinion poll results after the notification of elections. Fourteen out of fifteen national parties had agreed to the proposal.
It is no surprise that the BJP is opposed to the proposal, saying such a move would impinge on freedom of expression, because it is that party and its Prime Ministerial candidate Narendra Modi who are the sole beneficiaries of opinion polls of sorts so far in creating an illusion of  a ‘Modi wave’ sweeping through the nation.
 “We got the idea for the sting operation from the Election Commission letter. And then the fact that there seemed to be a new opinion poll almost every day now. Everybody is a pollster and a psephologist it seems,” said Vinod Kapri, editor in chief of News Express.
Congress spokesperson Rashid Alvi said the sting operation confirmed the party’s long-held position that the integrity of opinion polls was suspect. Aam Aadmi Party spokesperson Ashutosh said the tapes should be investigated and those found guilty should be prosecuted. “It is the channels which engage agencies and the agencies which conduct these opinion polls.
There’s a puzzle. Opinion polls, which newspapers and TV channels conduct before elections, sometimes get their predictions some what right, sometimes dead wrong, often in between. Many voters, and indeed politicians, eagerly await the outcomes of surveys to assess which way the wind is blowing, but parties for whom they spell doom often raise the spectre of polls being “motivated”. These parties are shrillest in their call for a ban on opinion polls, while those whose prospects appear rosier beam in the ‘feelgood’ created by such surveys. How does one make sense of the scepticism and the frenzy that seem to go hand in hand?
The loudness of the Congress’s call for a ban is proportional to the scale of opinion stacked against it. Whether it would have sought a ban if surveys had veered even remotely in its favour is anybody’s guess. Similarly, there’s no surety how the BJP, now shown as a frontrunner, would have behaved if the popular mood had seemed against it. The Election Commission, in receipt of petitions from the Congress, is seeking a ban on opinion polls by asking the government to bring in the requisite law. In the recent past, the EC has banned publication of exit poll results till the last vote had been cast.
This thumbnail sketch does not fully describe the richness or complexity of the situation. For one, there is a lot of scepticism about the process of opinion polls among the political class. Even electorally well-placed parties such as the BSP, JD(U) and TMC are against them. Second, neither opinion polls nor the debate about them seem to factor in the smaller parties and independents whose views—and cumulative effect on elections—get systematically overlooked.
So should opinion polls be seen as just another seasonal tamasha in the run-up to the real event, much like the high-decibel TV debates and opinions in print? Or, because pollsters can get their predictions wrong, can it be asked if they are spreading disinformation? Do the polls have the ability to swing votes in favour of a party on D-day? Equally important, is freedom of expression at stake here?
That opinion polls generate a lot of heat in the media is beyond doubt evident in the space and airtime devoted to them. As for their benefits in an age of politician- or corporate-owned media, the question is answered by Bhaskar Rao, director, Centre for Media Studies, credited with having started opinion polls in the country. “A quarter or more of news media is under control of political interests and corporates, raising the question of motives. What a poll survey brings out and projects is context-specific, and its relevance and even accuracy would altogether be different in a different context and at a different time, making the very exercise futile, even misleading. Taking recourse and linking the concern in this regard to press freedom or free speech is an industry perspective meant to divert the core issue,” says Rao.
 On the other hand, the constitutional fathers have guaranteed the freedom of expression. “The freedom of expression includes the right to express and communicate the same,” says Ashok Desai, former attorney general of India. “The restrictions on this freedom are also spelt out in the Constitution and the courts have always taken note of it,” says Desai.
Incidentally at an all-party meeting in 2004, the EC had recommended some restrictions on the publication of results of poll results. As of now, the ban on election survey results 48 hours before the elections is binding on the media. This was done on the apprehension that such surveys influence voting.
However, there is, as of now, no scientific study to prove opinion polls influence voter behaviour. But in the battle of perceptions, opinion polls do come in handy for political parties. 
(Design: Courtesy ‘Outlook’ magazine)
The final sanctified outcome, of course, is known only when the results are out. The surveys, say some, are a subversive threat. The threat lies, they say, in their ability to influence the voter. Hence the need for greater transparency in how they are conducted. Influence, however, is confined to the 4-6 per cent undecideds likely to cast their votes, which, in terms of seats, can swing in favour of any party.
Do opinion polls force people to come out and vote? Take the examples of Mumbai and Delhi, which have the lowest turnout in elections, compared to other cities. They also happen to be places where media and opinion-makers are concentrated in large numbers. “Are opinion polls actually preventing voters (in scientific parlance they are called pre-emptive polls) from coming out to vote following projections of a clear winner?” In other words, does the voter think that, since a particular candidate is winning anyway, why bother to vote? There are no easy answers.
But one thing is for sure: a ban is certainly not the solution. Some call for transparency in the manner in which such polls are conducted and says pollsters should stop short of predicting the actual seats. The way out, as suggested by some pollsters themselves, is mandatory disclosure on the sample selection and its social profile as well as the sponsors of the polls. In the proliferation of media and the questions being raised about media groups’ ownership, the opinions displayed by them would carry greater credibility if the track record of the pollsters is put on display. A greater rigour going into analysis rather than merely predicting the outcomes of political parties will go a long way in informing the voter at large.
But what with the latest exposure, such ideal conditions are long way to go ant till then the credibility and integrity of opinion polls would be at stake and for people at large it is better to keep themselves uninfluenced by any!

No comments:

Post a Comment