Online edition of India's National Newspaper
Saturday, Apr 24, 2004

About Us
Contact Us
News: Front Page | National | Tamil Nadu | Andhra Pradesh | Karnataka | Kerala | New Delhi | Other States | International | Opinion | Business | Sport | Miscellaneous |
Classifieds | Employment | Obituary |

Opinion - Leader Page Articles Printer Friendly Page   Send this Article to a Friend

Not good practice

By V.K. Ramachandran

Does a ban on announcing the results of exit polls before the entire electorate has voted interfere with the freedom of information of the electorate?

IN RECENT years, public opinion polls have acquired an important place in political discussion, and the first round of Lok Sabha elections in 2004 has rekindled the debate on the place and role of opinion polls in the electoral process. The publication of the results of exit polls has been defended on grounds that have to do with larger considerations of freedom; their early publication has also been criticised, particularly by political personalities, as an unacceptable means of electoral intervention.

In the first place, a difference must be made between predictive opinion surveys in general and exit polls. Predictive opinion surveys whose results are announced before the actual voting begins must be considered a legitimate part of the political process, potentially useful sources of information for voters and political parties. We must also differentiate between the legitimacy of exit polls in general and exit polls whose results are announced after a part of the electorate has voted but another part has not, that is, those that, by their very nature, make information available to one section of the electorate that is not available to another.

The question at hand, then, is not "Does a ban on opinion polls and exit polls constitute an interference with the freedom of information of the voter," but a more focussed one: "Does a ban — legal or self-imposed — on announcing the results of exit polls before the entire electorate has voted interfere with the freedom of information of the electorate?"

In order to be able to conduct free, fair, impartial and peaceful elections, the Election Commission has to hold them over a series of days. Geography, the dispersal of habitational settlements, socio-political circumstances, and the availability of personnel to conduct elections and of police and paramilitary forces to protect voters in the electoral process are among the factors that make it impossible for the Commission to conduct elections on a single day all over the country. However, while in practice, circumstances dictate that elections to a single House be held on different days in different parts for the country, in principle, it must be deemed that such elections are a single event, at least to the extent that that principle is necessary to prevent discrimination or unequal access to information by sections of the electorate that cast their votes on different days.

When the results of an exit poll are announced before one or more sections of the electorate have voted, those sections have access to information that voters who have already cast their vote did not. If elections are held on six days, for instance, then the voters who vote on the last day have access to five instalments of important information that voters on the first day did not have. Thus one important argument against publishing the results of exit polls before the last polling booth closes is that such publication interferes with the right of every voter to equal information, and thus distorts the electoral process.

Political parties recognise that the early publication of exit polls interferes with the priority of the principle of equity in access to information. The instinct of a political activist appears, unsurprisingly, more reliable here than the considered opinion of those who demand what amounts to unequal "freedom of information."

Two arguments are commonly raised against the foregoing. The first is that exit polls can hardly have a distorting effect, since only a minuscule section of the electorate acts on them anyway. In reply, one could say that dubious indeed is a measure whose adherents declare simultaneously both its great contribution to the democratic process and its irrelevance to that process. In any case, the question here is not of the reach of published exit polls but whether in fact they interfere with the right to equal information.

The second argument is that the publication of exit polls before all booths close does not affect actual electoral outcomes. This is because, it is claimed, that although the results of exit polls cause some subsequent voters to vote for those who lead the exit polls, and others to vote for those who lag behind in the exit polls, the join-the-bandwagon effect of exit polls and the give-the-underdog-a-hand effect generally cancel each other out. There are two clear arguments against this. First, anyone with even a nodding acquaintance with statistical method would recognise that such a zero-sum proposition can be established inductively only through specialised and extensive statistical testing. Secondly, and more importantly, it does not matter which effect prevails. The very fact that a person chooses either to join a bandwagon or help an underdog after reading the results of an exit poll means that he or she is acting on information that was unavailable to other voters, that is, to those who voted before the exit poll. (It is another matter altogether that the international literature suggests that another effect of the early release of exit polls may be voter apathy: "Why should I vote when the outcome is already decided?").

The fact that the counting of votes begins only after the entire electorate has voted is not only an administrative convenience but is recognition in principle of this right to equal information. Most people would agree that votes should be counted together and that results should not be announced after each phase of voting. Such an agreement is founded on agreement with respect to the right of every voter to equal access to information relevant to his or her decision.

It wasn't always like that. In the general election of 1952, for instance, counting votes (and announcing results) did not wait until voting was complete in all constituencies. And in that election, the early announcement of results had manifest effects on voting behaviour on subsequent polling days.

I have notes from interviews that I conducted in the early 1980s on voting behaviour in constituencies in the agro-ecological zone in Tamil Nadu known as the Cumbum Valley. Two Assembly constituencies fell within the geographical area of the Cumbum Valley in 1952, Cumbum and Uthamapalayam. P.T. Rajan, who went on to be the sole Justice Party representative in the Madras Legislature that year, won the election in Cumbum that year. P.T. Rajan belonged to a major landlord family of the area, one that occupied an important position in the socio-economic hierarchy of the district. The family seat was then at Uthamapalayam, about 110 km south-west of Madurai.

P.T. Rajan stood in two constituencies in 1952. He was defeated roundly in Madurai North, while he won in Cumbum. A voter in the 1952 election described PTR's victory to me in this way: "News of PTR's defeat in Madurai came to us before polling day in Cumbum, and voters in his Cumbum Valley constituency were urged to reverse the blow that the voters of Madurai had dealt to the prestige of this big man of our area."

The early announcement of actual results — and, by extension, accurate exit polling — can certainly affect outcomes, and it is not surprising that the Election Commission no longer follows the 1952 practice of announcing results from some constituencies before voting booths have even opened in others.

These principles have been recognised in practice and in law in different parts of the world. In the 1980s in the United States, the media generally followed a self-imposed code of not announcing the results of East Coast exit polls until booths had closed on the West Coast. In the United Kingdom, the Representation of the People Act 2000 specifically prohibited the publication of exit polls before all voters had finished voting. In France, the findings of public opinion polls may not be published during the week before a ballot (although this ban was circumvented as early as 1997 by means of publication on Internet sites).

In the month before the General Election of 2001 in the United Kingdom, the Financial Times reported that "Internet news sites, television broadcasters and evening newspapers have been warned that a change in the law since the last election means that exit polls must not be reported before the polling booths close at 10 pm on June 7" (Financial Times, May 17, 2001). It reported that the "BBC said it keeps its exit poll `under lock and key' until the polls close `despite getting a phone call from every journalist on the planet.' " It was anticipated that the new law in Britain would help avoid situations such as Florida 2000, perhaps the most egregious recent example of the electronic media erring on how and when to announce the results of exit polls.

Publishing the results of exit polls while elections are still in progress is neither illegal nor unconstitutional. It is, however, not in the best public interest. There are times and situations when the public interest is best served by voluntary, cooperative restraint; this is one of them.

Printer friendly page  
Send this article to Friends by E-Mail


News: Front Page | National | Tamil Nadu | Andhra Pradesh | Karnataka | Kerala | New Delhi | Other States | International | Opinion | Business | Sport | Miscellaneous |
Classifieds | Employment | Obituary | Updates: Breaking News |

News Update

The Hindu Group: Home | About Us | Copyright | Archives | Contacts | Subscription
Group Sites: The Hindu | Business Line | The Sportstar | Frontline | The Hindu eBooks | Home |

Copyright 2004, The Hindu. Republication or redissemination of the contents of this screen are expressly prohibited without the written consent of The Hindu