As political ratings are made and when they should not be trusted – explains sociologist Anton Hrushetsky

Photo: pexels.com

After Volodymyr Zelenskyy and Donald Trump exchanged sharp statements (Trump said that Zelensky had only 4% of trust in Ukraine, and Zelensky replied that we would like a rating not only Trump but also to other world leaders), the Ukrainians were interested in ratings. This is against the backdrop of the scandal in the oval office, according to the new rating from KIIS, the rating of Zelensky rose to 68%. And in the event of a truce can be prepared for elections and candidates will be those ratings to measure.

The ratings are based on the results of social survey, on the same grounds are determined by folk sentiment and generalized public opinion. As sociologists derive it as the respondents are looking for and what can be manipulations – Briefly about told Deputy Director of the Kiev International Institute of Sociology Anton Hrushetsky.

From corruption to reforms and possible concessions – Antone, the polls are conducted on different topics. Who and how do they determine?

Anton Hrushetsky. Photo: facebook.com/anton.grushetskyi

– There are several reputable private research companies in Ukraine that can conduct surveys on both their own initiative and custom. For example, our KIIS conducts at least four of our own surveys during the year when we form the topics. We also conduct dozens of surveys to order customers. For the most part, these are universities, scientific institutions in Europe, international organizations that are interested in certain aspects of life in Ukraine: from the fight against corruption, socio-political sentiment to health care reform or energy in energy.

Now some universities in the West are interested in readiness for certain concessions for the end of the war and what combinations Ukrainians consider more perceived for negotiations. At the request of one of the universities, we conducted such a survey in 2022: how to survive the experience of hostilities and whether it increases readiness for concessions. It was a highly specialized study with its specificity. Because one situation when people survived the shelling of them, the other – if they lived in the frontline.

There are certain indicators that we “measure” society every year or almost every year are geopolitical moods, attitudes towards Russia, readiness for concessions, trust in the president or various institutions in Ukraine. And there are point polls that arise situational. For example, the attitude to suspension funding from USAID. It happened, it was important to ask – we asked. It turned out that the most respondents – 47% – believe that there will be some negative impact, but at the same time recognize that it is a certain justice for those who have passed money or engaged in unnecessary affairs.

– If the topic goes from the customer, does he formulate a question?

– The customer cannot formulate questions to respondents himself. And if it formulates, then a company that values ​​its renome will do a methodological test: whether the questions are correctly asked, whether they push the respondent to a certain answer, whether they contain manipulations.

– And what can be manipulations?

– For example, the order of questions. Suppose the main topic of the study is corruption. People are asked many questions about corruption, and then asked: who will you vote for? In the ranking, this will worsen the situation of the current authorities. Or, first, a question is raised with a slope on what a bad opposition or grentoids-activists have-so it is possible to beat a few percent support of the opposition.

Important not the volume but the correct sampling – How many people are involved in polls? Millions of Ukrainians learn from the ratings what they think about someone, although they have never been asked about it.

– This is the most difficult moment of perception of the results of polls by the public. There is an option to conduct a continuous survey, but to be interested in the opinion of 30 million Ukrainians is simply unrealistic. Therefore, many decades ago, scientists have decided that the experience and achievement of statistics should be used. And she showed, if you choose the right people, even small samples can produce representative results. You can interview a thousand or two thousand citizens and have more accurate results than 10,000. It is not in volume, but how they were selected. Is it really random whether men and women or residents of different regions are represented.

The illustrative history was in 1936 during the US election campaign. This is a point that has recorded the success of selective surveys. The popular Riderz Digest magazine interviewed several million people – by phone numbers, they sent questions to owners of cars. And there was a company that interviewed one thousand Americans. So the forecasts of this thousand were more correct, because they were professionally built a sample.

Even if we conduct 50 surveys a year that will cover two thousand people, the probability that you will get to the sample is very small.

– Is there a quality restriction on professions? It was once said that journalists should not participate in ratings. How about the military?

– Citizens of Ukraine are interviewed. If the military gets into the sample and agree, we interview them. In total, it is 1.5-2 percent of respondents. There are less of them, because people do combat missions, they are not talking. Similarly, journalists are citizens of Ukraine, they have the right to express their opinions.

Restrictions can only be administered in narrow -target polls. Basically, they are marketing research, where customers are afraid that they will give their competitors, and therefore put some restrictions on the profession. Journalists can be excluded from the respondents for reasons to prevent the leakage of information. Although the likelihood that they will be sampled is very low.

There are no professional restrictions in socio-political polls.

Results of a survey on the suspension of USAID funding. Infographic KIIS

At the crossroads in the Gorishny floodplains – And is there a age -old qualification?

– A typical sample is 18 years and older. In online -Pompations can be the age of “18 -60 years” because older people use the Internet less. In socio-political polls, the upper limit is not established.

Among the youth under 18, social orientation surveys can be conducted. For example, what profession is considered prestigious school graduates. Sociologists do not work with children under 15 years of age, because there are still difficulties with personality formation.

– Let’s say, it is a rating of confidence in the courts. The judges themselves can also speak?

– Yes, the judge can also express his opinion as a citizen of Ukraine. But again, the appearance of a judge is unlikely, because the judges for the population are not so much. Although this idea was to conduct a survey among judges and those who participated in lawsuits, but the courts did not support it.

– In what way is the sample?

– Now there are two most popular methods. The first is telephone surveys by accidentally generated mobile operators. That is, we do not know whom we call – a man or a woman whose inhabitant of the region.

The second option is personal interviews. The questionnaire’s work is selected here. For example, there are 10 points in Kiev, there are 5 points and 3 villages in the Poltava region. Interviewers go, go to households. Or street version. The address is selected, for example, in the Gorishni Plavni, there are interviewers. Anyone who passes through this point is offered to take part in the survey.

– And do we really get a picture of public opinion?

– If the sample is properly implemented, then yes. You can interview 50 thousand subscribers of telegrams like Trukha and get distorted results, although it will be 50,000. And interview 500 – 600 people on the street and see the real picture.

External migrants are excluded – Millions of Ukrainians expelled the war from home. Is their opinion taken into account in the ratings?

– internally displaced persons, of course, fall into the sample. In telephone polls, they make 10-12%. Unfortunately external migrants are excluded from the sample, and this is one of the biggest problems. Separate surveys can be conducted with them, but it is rare because it is very difficult organizationally and financially. Another problem with refugees abroad is that some of them have become more than more society than ours.

– According to the results of the poll, an error of several percent is always indicated. What does this mean?

– The error is due to the fact that we interview not the whole population of Ukraine, but only part of it. That is, the output figure is averaged – for orientation. If we output the result that Vladimir Zelensky trusts 57%and the error is a plus or minus 4%, it means that the level of trust of the President is in the range of 53%to 61%.

This is how the rating of trust in Vladimir Zelensky as of February 2025. Infographic KIIS

Everything except personal data – Today we have many fraudsters. People can be afraid to answer questions to a stranger on the phone when it comes to a survey.

– It is very easy to distinguish a sociologist from a fraudster. Sociologists do not ask or record any personal information. Unless it will be recruiting in focus groups for further research, but such proposals are rarity.

I repeat, sociologists are not interested in passport data, address, and even more so bank card number. They will be interested in your age, article, nature of employment. They may be asked about earnings if the survey is touched. Everything but personal information.

– You spoke about possible manipulations. Can call, say, political scammers?

– It is not excluded. These will ask, for example, the attitude towards the “bloody dictator regime”. Or about the attitude towards “our courageous president, who has done so much for the state.” That is, when the questions are not neutral, but push for the fact that you have to choose some political side.

– After the election campaigns, people noted that the results of the polls on one or another candidate did not coincide with the results of the vote.

– There is some confusion here, and you need to distinguish a survey on the level of trust and electoral ratings when specifically ask who you will vote for. These are different things. Because you can trust any policy or public figure and not vote for him. And you can not trust, but vote for some pragmatic reasons.

Source: kp.ua