“Thanks to the exit poll, people get alternative data”

Russians will be able to find out the first preliminary results of the presidential elections even before the results are announced by the Central Election Commission – from exit polls on the evening of March 17. Exit polls have long been firmly established in Russian political sociology, but still raise many questions and even criticism. The general director of the All-Russian Center for the Study of Public Opinion (VTsIOM) told Kommersant about why exit polls are needed in the age of new information technologies and how they affect voters’ confidence in elections. Valery Fedorov.

— When will we see the first exit poll data for the 2024 presidential election?

— Immediately after 21:00 Moscow time, when the last polling stations in Kaliningrad close.

— Will there be any novellas?

– For what? The technology has been known since the 1960s. Everything is worked out to the smallest detail. There will be a completely classical methodology, but technology corresponding to the times.

— Will you conduct exit polls in new regions?

– No. For security reasons.

— There are different opinions regarding exit polls. It is believed that they could influence the voting results if published ahead of time. Do you think it is really important to meet these deadlines?

— Yes, the publication influences the voter’s opinion. The question is – how much? At one time, George Gallup himself explored this issue in his book “The Pulse of Democracy” and came to an important conclusion: familiarity with the data of sociologists can change the position of 1-2% of those acquainted. Let’s assume that every third person will meet (that’s a lot). This means that 0.33–0.67% of the total number of voters may change their position. Frankly speaking, not much. Much less than the influence of advertising, debates, campaigning, news… So the influence of our publications on changing the position of voters is vanishingly small.

— Do exit polls remain relevant these days, when voting results are counted faster and faster, and we learn the first results literally a few hours after the polls close?

— Technically, the relevance is decreasing: previously the CEC provided more or less adequate data in the morning of the next day, but now this happens in the evening of voting day.

But, fortunately, people are not robots, and technology by itself does not solve anything. And social relations decide. And if they are built in such a way that the technology can be perceived, they use it; if not, they discard it. Or they find a completely different use for it. Take gunpowder: in peaceful China it was used mainly for fireworks, but warlike Europe made it into a destructive weapon.

So everything is decided by the attitude of society, and the key point here is trust. In the modern world, trust is low and becoming less and less. Political systems are objectively interested in strengthening their credibility. Especially in countries where democracy is in question. This is partly why the exit poll, invented in the United States, began its triumphal march around the world. It reached Russia in April 1993, then VTsIOM conducted the first exit poll on the referendum (we are talking about a referendum on April 25, 1993 on four issues: confidence in President Boris Yeltsin, the policy of the Russian government and the need for early elections of the president and people’s deputies. — “Kommersant”). Thanks to the exit poll, people receive alternative data, and in conditions of low public trust, this helps to shape their attitude towards official data. This is an important moment, and it doesn’t go away. I think he won’t leave, even if the Central Election Commission sums up all the election results a few minutes after the end.

— But these are your figures that almost 70% of Russians trust the election results?

— A considerable part of them trust “conditionally”: they seem to trust, but they seem not. Their position is unstable. In general, up to a quarter of voters are such “conditional” voters. These people need reinforcement, confirmation that everything is honest, orderly and noble. And if it is not there, then “conditional trust” easily turns into distrust. If there are any signs that something is fishy, ​​the majority of those who trust will turn into a majority of those who do not trust.

This already happened in the 2011 State Duma elections. A lot of short and unclear videos suddenly appeared online, showing: look, here at the polling station the chairman of the polling station grabbed the ballots and moved them from one stack to another! There is clearly something wrong here… Our elections were stolen!!! Observers from opposition parties came to the polling stations with smartphones, filmed what seemed suspicious to them, and showed it all on social networks, commenting accordingly. And it fell on well-prepared ground: that very significant part of the “conditionally trusting” voters. And then these people swung towards mistrust, which was followed by mass rallies, protests, a split in society and other sad things.

— So, an army of trained observers and a Central Election Commission, which is ready to declare any violation a provocation, are not enough?

– Remember the legend about Pandora’s box, which one foolish woman once opened – and released all the diseases and troubles that have tormented humanity since then? Apparently the exit poll was among them… Just kidding. The exit poll has become a tradition, and traditions must be preserved. If an exit poll is not carried out, people have questions: why? Something is probably unclean here! This is bad. After all, for any political system to work stably, legality alone is not enough (this is ensured by the Central Election Commission). We also need legitimacy, that is, recognition from the people. And there is no legitimacy without trust in the election results.

— To what extent do exit poll results usually coincide with the official voting results?

— I would pose the question differently: to what extent do the official figures correspond to the exit poll? In general they usually agree, but in particulars there are discrepancies. Sometimes these particulars become important. For example, if the cut-off threshold is set at 3%, and the party received 2.7% of the votes according to official data, and 3.3% according to the exit poll. There is already a political difference! Unfortunately, an exit poll, like any sociological study, has its own errors and cannot determine the result with an accuracy of tenths. But it also cannot be wrong by more than a couple of percent, if the methodology and technology are followed. So there is a limit of accuracy, upon reaching which the exit poll loses its function of “ultimate truth”.

How can this work in practice for us? In the case when the candidates are very crowded, “head to head”, the difference between them can be 1-2 percent – this is within the margin of error. In other words, if the gap is small, we will not be able to authoritatively say based on the exit poll data: this candidate took second place, and that candidate took third place. And in this case, it will be necessary to trust the Central Election Commission more than sociologists. After all, the election commission takes into account every vote, that is, its error – purely theoretically, if all the methodology and counting technology is fully complied with – should be zero.

– But there are people who refuse to tell sociologists how they voted – how is their opinion taken into account?

“It doesn’t affect anything at all.” I will not go into methodological details, but the overall result does not suffer in any way from the fact that a number of citizens refuse to answer. There are always those who refuse, and now more often than, say, 10 years ago. This is a worldwide trend, but it has little impact on the reliability and accuracy of our data.

— Do exit polls provide any additional information after the elections?

— There are two approaches here: pragmatic – when they only ask for whom the respondent voted, and research – when they want to get as much data as possible and ask several dozen questions at once. In America, by the way, they try to ask a lot of questions. Our approach is rather pragmatic, so we manage to collect little additional information. But there are other ways to get it – for example, our daily all-Russian survey “VTsIOM-Sputnik”. We can ask voters all the questions that interest us on March 18.

– And what will you ask about?

— Why did you come to the polls, why did you stay at home? Preferences for voting method: pawn or DEG? How did the “three-day” work out, what day did you go? Do you like or dislike the new elements of the electoral procedure? Motivation: why did they vote for this one and not that one? Do you trust the vote count and official election results? Etc.

— Do you note any features of this campaign? Well, it was really boring – the candidates didn’t even file a single complaint against each other.

— The whole point is that the 2024 campaign is being held as a “referendum” or “plebiscitary”. There is this type of election campaign, which, by the way, is quite common. For example, in the American Congress, up to 90% of deputies are elected according to this procedure; such districts are called “safe seats”. There is no competition there; it is known in advance who will win. In a referendum scenario, everything is basically clear already at the beginning of the campaign: people decide whether they will vote and for whom, even at the start. The share of those who found it difficult to choose is very small. There are few cross-flows and runs between candidates. All the media attention is focused on “micro candidates” who make no difference. And at the same time, everyone (well, except journalists, maybe) is happy! Everyone understands that this is how it should be… But, as they say, you need to live in Russia for a long time. Wait, in just two years there will be elections to the State Duma. Here everything will be much more dynamic and less predictable!

Interview conducted by Anastasia Kornya