Opinions, Opinion Polls, and Democracy

The Election Commission’s move of seeking political parties’ opinions on banning pre-election opinion polls has sparked off a heated debate. On the one hand, the Congress has called for an outright ban on such polls, while the BJP, which in 2004 had favoured such a ban, has now decried the Congress’ call for a ban as an attack on press freedom and freedom of expression. Most media organisations too have dubbed any restrictions on opinion polls as an attack on freedom of expression.

The positions of Congress and BJP at different times on opinion polls have been opportunist, coloured by whether the opinion polls tend to favour them or not in the forthcoming months. How can we go beyond the terms of the debate fixed by opportunistic ruling class parties and corporate-funded media, and arrive at a reasonable and democratic opinion on opinion polls?

One argument being made in favour of unrestricted opinion polls is that opinions by laypersons, editorials and media commentators, pollsters and politicians float around in any case. If one calls for restrictions on opinion polls today, it might go to the extent of restricting the free expression of opinions tomorrow. A related argument is that scientific opinion polls are preferable to lay rumours and ‘expert comments’ that do not enjoy the backing of data or scientific analysis.

What these arguments ignore is that opinion polls are not simply mirrors reflecting opinions. They are tools that seek to shape, influence, and organise political opinions. An editorial in a newspaper, or a comment by a TV anchor or ‘expert’ is simply that – an opinion of a particular media institution or individual, with no claims to reflecting popular will. But an opinion poll lays claim to representing people’s opinion, and that is what sets it apart, qualitatively, from mere ‘opinions’. An editorial is a self-avowedly subjective opinion – whereas an opinion poll claims to be an objective, even scientific, representation of, or guide to, popular opinion. In any elections, laypersons or media outfits may express opinions projecting poll outcomes and may attempt to persuade others to accept these opinions. But these subjective attempts at persuasion are likely to be viewed by voters very differently from the ‘objective’ polls presented authoritatively by media houses and various other bodies.

Any opinion poll claiming to objectively predict the outcome of elections (as opposed to simply expressing a subjective opinion about the outcome) do, then, have a potentially distorting effect on voting behaviour. In the backdrop of widely prevalent ‘paid news,’ the possibility of opinion polls being used to misinform and manipulate voters is immense, given the fact that it is corporate-funded media houses which generally conduct these surveys and polls. There is simply no denying that such surveys conducted and propagated by corporate-funded media houses have a far greater potential to distort voting behaviour than mere opinions of media-persons, laypersons or political figures. The voter is well equipped and experienced to assess the predictions of the local ‘expert’ at the tea stall or street corner or even the poll ‘pundit’ on the TV news. An ‘opinion poll’ that projects itself as the opinion of the people, claiming to be gathered through scientific methods, is far more capable of insidious influence and far more difficult for the layperson to challenge.

Currently, most opinion polls in India fail the test of transparency. Opinion polls that fail to disclose their sampling frame, size and social profile of the sample, techniques used, wording and sequence of questions asked, details of organisations conducting and sponsoring the poll, and fail to make their raw data available for scrutiny and challenge, nevertheless propagate their findings as ‘objective.’ However, even if the highest standards of transparency are met, the fact remains that the ordinary voter lacks the methodological tools to assess sample size, profile, etc to test the objectivity of various opinion polls. The ordinary voter will therefore be under pressure to accept the findings of such polls without really subjecting them to analysis.

No doubt, voters have proven opinion polls wrong time and again – exposing the lack of objectivity and even the possible political bias of those who conducted the opinion poll. But this cannot be an argument to allow opinion polls to continue unregulated.

Calls for a blanket ban on opinion polls, as well as calls for unregulated opinion polls, must be rejected. With the announcement of the election schedule, the model Code of Conduct comes into force, which imposes several restrictions (such as bans on the announcement of any new schemes by the ruling party) aimed at keeping out undue distortions of voting behaviour. During this period when the Code of Conduct is in force, media houses, parties, and other bodies should be restrained from publishing opinion polls or claiming to base their analysis on opinion polls. Any opinion polls conducted and published before this period must be subjected by the Election Commission to reasonable restrictions including mandatory disclosures to ensure transparency, and availability of raw data for scrutiny.

Liberation Archive