The loss of percentage, the increase in popularity or the castling of candidates – we are talking about all this, focusing on sociological research, the number of which keeps increasing on the eve of the election. Opinion found out whether we should always believe these results?
How representative are the results of social surveys?
Volodymyr Volia, the political and international affairs expert and the master of public administration, assured Opinion: the surveys conducted by sociological companies with a long-standing reputation, as a rule, are always representative.
“Sometimes there are cases when several well-known companies receive similar results, and the results of another individual company differ for some reason from the general ‘trend’. In such a situation, it is necessary to specifically find out the cause of the discrepancy. Is it a consequence of errors in the development of the tools, or the consequence of violations of the rules during the field phase of the survey? We cannot rule out the deliberate manipulation as well.”
But the political scientist and the Deputy Director of the Ukrainian Institute of Analysis and Management of Policy Kyrylo Molchanov believes that even reputable sociological companies are prone to false research.
“In general, sociological research, especially during the election period, turned in Ukraine into a profitable business for the owners of these services. Sociological companies spring up like mushrooms after the rain, although everyone knows that the authoritative structures can be counted on the fingers of one hand (Razumkov Centre, KIIS, Sociological group RATING, Oleksandr Yaremenko Ukrainian Institute for Social Research), although even they sometimes don’t disdain to do pseudo-surveys.”
Dmytro Sinchenko, the head of the public organization Association of Political Sciences, is convinced that the ratings compiled on the results of the surveys, can be considered representative and real, but it is important to remember: any positions may change very quickly.
“The representativeness of a sociological survey should be indicated in the information about this survey. It depends on the sampling. Typically, in case of the national election, the survey is (or should be) representative for the adult population from all over Ukraine, that is, for the voters. They should be perceived as the real positions of the candidates at the time of the survey. However, it is important to understand that these results may change at any time, and we will know the final results only after the day of voting.”
Oleksii Buriachenko, the political expert, urges voters not to accept the results of social surveys at face value: we should always pay attention to the sociological error, the questions, and the sampling criteria.
“Speaking about the representativeness and the results of sociological research (SR) it is necessary to pay special attention to the sampling criteria, its coverage and of course the quality of the formulated questions. Thus, SR methods differ by the coverage of respondents by regions, the total number of respondents, the age gradation of respondents, the percentage of rural and urban population and many other parameters. It is incorrect to accept the results of SR at face value. It is always important to study the sampling criteria, the questions asked, and to estimate the declared sociological error.”
How is it possible to manipulate with the help of social studies?
According to Dmytro Sinchenko, everything depends on the goal. The most common method of manipulation, according to the expert, is to improve the results of the candidate who is the company’s client, and, accordingly, reduce the percentage of competitors.
“However, the client is not always interested in such distortions, because it may turn out as a plus or a minus for him during the vote. Why? When a candidate artificially overestimates his rating and underestimates the rating of a competitor, he tries to provoke voters to feel more confident in him, to believe in the possibility of his victory, and to cast their doubts aside. However, overestimation of the rating may mobilize the competitor’s voters to come to the polling stations.”
In his turn, the political scientist Vladyslav Serdiuk notes that manipulation may occur not only at the stage of results formation but also directly in the process of collecting information.
“There is a lot of manipulation in the process of carrying out the sociological researches. The most common of them is the creation of a non-representative sampling. The attention should also be paid to how the question was formulated during the collection of information since it is the wording that may often affect the respondent’s response. Moreover, the researcher himself may indirectly influence the response by his behavior during communication.”
Volodymyr Volia believes that manipulations occur at the stage of interpretation of the research results, mainly, he talks about the incorrect submission of quantitative data.
“As a rule, the results of surveys are presented in the form of quantitative indicators, tables, charts and small comments to them. At this stage, manipulations are excluded if they have not been carried out before. Typically, manipulations occur when political teams, support groups etc. interpret the survey results. We are talking about various forms of incorrect presentation and interpretation of quantitative data.”
Kyrylo Molchanov told about the main goals of rating manipulations. According to him, they differ from the positions and chances of candidates. So, the leaders and the average candidates’ goals will be very different.
“The manipulation by the ratings has three global goals. The first is for outsiders. Their spin doctors claim that this way they have a chance and in fact, continue to campaign mastering the pre-election budgets. The second is for average candidates. Thus, they can negotiate to create blocks with other candidates or withdraw in favor of the favorite in exchange for some preferences. The third is for the leaders of the race. This is an opportunity to assure the public that they are either in the first place or at least will make it into the second round, to mobilize the undecided or hesitant to vote or not.”
But Oleksii Buriachenko explained that in this context manipulation is rather an ambiguous and not very correct concept. The specialist is convinced that those companies that are regularly engaged in research, care about their reputation too much to publish results that are not true.
“It should also be noted that sociological studies also differ in terms of publicity. That is, some are closed to the public and indeed reflect the state of affairs, while others, public ones, might carry certain loads associated with the “technical task” of the client.
In the format of the electoral race, the candidates’ campaign offices are the main customers of social studies. Even the bare eye can see how many different studies appear daily in the media space. It happens when they are in demand.
Goals (technical tasks), which are placed by campaign offices for those who conduct sociological research are quite diverse and depend, for the most part, on the desired result of the election race for a candidate.”
How are social surveys useful for voters and politicians?
According to Volodymyr Volia, the results of the surveys are mainly of value to experts and consultants in the process of forecasting and planning.
“First of all, we are talking about the information that is necessary for making a personal decision in political matters. For some people, the results of the survey can be an argument for ‘mobilization’, for someone those quantitative indicators can be an argument for refusing to participate in the election. It all depends on the subjective perception of political life by each individual, as well as on the interpretation and use of data by political teams. Of course, some things also depend on the mass media.”
The Political scientist Vladyslav Serdiuk assured Opinion that the results of social surveys have a considerable, though not always positive, impact on the eve of the election.
“First, they may force voters to vote not for their own, but for the “strongest” candidate. Secondly, the voter may refuse to support anyone because of the low ratings of his candidate. That is, sociology may demoralize the electorate. Third, there is the notion of ‘self-fulfilling prophecy’, when sociology affects reality in such a way that the forecast is correct.”
However, the Deputy Director of the Centre for Information Issues of Territories of the NAS of Ukraine Roman Solomoniuk believes that sociology has no special meaning to the voter, while the trends of opinion change are important f politicians.
“There are certain trends. For example, an idea or a politician that is popular among the intellectuals, with time become popular among other layers, although at the beginning it might have had a low rating. Economic ideas that are popular among officials, quickly spread among other layers, because officials are perceived as an authority. Workers tend to change their opinion quickly, so if the high rating of a politician is based on the support of ‘workers’, it may quickly change. Most sociologists do not consider all these details. Therefore, as a rule, there is a ‘cross-section relevant for today’, which may change in a few days.”
What should we pay attention to in the “fresh” surveys, and how to protect ourselves from manipulations?
According to Volodymyr Volia, there is a number of indicators that require the attention of the voter who decided to get acquainted with the new ratings.
“When new survey results appear, you should pay attention to the name of the company that conducted it. Time and territory of the survey are important. Pay attention to whether there are “abnormal differences” from those data which are published by other companies at the same time. For example, if there were some high-profile political scandals after the survey was finished, then some of the survey data may be outdated.”
But Maryna Bahrova, a member of the board of the international union Institute of National Policy, advises paying attention to who finances this or that sociological service.
“If the organization that finances the relevant sociological service is behind a particular candidate, it cannot but indicate that the surveys are conducted in favor of the respective candidate. This is the mystery of why different sociological services have different data of social surveys, in which different candidates are at the lead.
Practice shows that sociological surveys are conducted, as a rule, in those sociological groups which are most loyal to this or that candidate. Basically, domestic social surveys are based on a shallow sampling of respondents, for example, from 1000 representatives of the relevant social group, an average of 5 to 7 people are surveyed. All this is the reason for the discrepancy between the results of social surveys and the results of the election.”
Roman Solomoniuk advises paying attention to the trends and dynamics: when the growth or decline took place. This, according to the expert, will help to understand the situation in which the country was, and what changes may happen next.
“Although the best method proven by practice is operational sociology. This method was used by the campaign office of Donald Trump in the last election. There was a case when the accusations of “nationalism” towards Trump were considered by his opponents as a minus and were actively spread. But Trump’s campaign office tracked that people perceive “the nationalist” in a good way, and therefore did not provide obstacles, but rather play along with the dissemination of information by the behavior of the candidate.”
Dmytro Sinchenko believes that the only option to protect ourselves from fake ratings is.. self-ordering surveys and monitoring the compliance with the methodology. According to the expert, nothing else will help.
“It is possible to reveal rough manipulations and imitations; however, it is practically impossible to reveal high-quality forgery. The results of sociological surveys cannot be blindly believed. Even truthful ones, because public opinion may change at any time. The survey does not show what the result of the election will be. It can only show trends.”
By Dmytro Zhuravel