Public opinion polls have become an important part of everyday life for contemporary humans a long time ago. Thousands of people answer endless questions of curious researchers every day and express their vision on almost everything that happens in their city, country and around the world. The outcomes of such polls are actively used not only by scientists, who seek to identify the general patterns of the development of society and write new articles and books about these patterns. The data of these surveys are of great interest to politicians and entrepreneurs, whose effectiveness is impossible without a clear understanding of the needs, moods, and desires of society. Whether we like it or not, public opinion polls are increasingly becoming the basis for making important management decisions that directly affect the lives of each of us. However, can we fully trust the results of such polls? Do their results really reflect the real moods and desires of society? Can they be used to predict the behavior of large communities of people under different circumstances? Although I have been organizing public opinion polls for more than 15 years, I must admit that, unfortunately, I have great doubts that all these questions can be answered with a confident yes. In this paper, I would like to talk about the reasons why it is necessary to be very careful with the outcomes of public opinion polls.
Probably we can say that public opinion polls became very popular primarily because of the technical simplicity of their initial idea. Basic logic of any public opinion poll assumes that you can very easily understand the desires and intentions of society: the only thing you need is to circulate a short (or a not so short) questionnaire among people and ask them directly what is on their mind. The deceptive simplicity of this logic is very attractive and explains the impressive popularity of this research instrument. But unfortunately, this logic is not perfect. It is based on a whole series of controversial assumptions, none of which can be considered an obvious fact. It seems obvious that a simple survey of people can help to understand the desires and intentions of society only if at least three conditions are met simultaneously. The first condition is that people must have clear desires and intentions and feel able to characterize these desires and intentions by answering simple questions. The second condition is that researchers must formulate appropriate questions which will allow people to clarify their desires and intentions as accurately as possible. The third condition is that people need to answer the questions of the researcher sincerely and honestly. Violation of even one of these conditions destroys the fragile logic that underlies the use of public opinion polls. And, unfortunately, each of them may not always be observed.
Let's focus on the first of these conditions, which assumes that people must possess clear moods and desires and feel able to express them during the research of public opinion. The problem is that typical topics of many public opinion polls go far beyond the borders of the limited range of issues that concern the average person in his or her everyday life. Daily activity of most people is focused around solving their personal and everyday problems, which may be significant for them and mean nothing to others. Most of the people do not have any reasons to constantly think about international politics, human rights, justice, and other specific topics that may be of interest to researchers. When they participate in a public opinion poll related to these topics, they are forced to discuss topics they may never have thought about. Of course, people are able to talk about topics of which they have a limited understanding: they can choose answers randomly or rely on fragmentary information received from the Internet or from the media. But the value of such answers is very doubtable. Hypothetically, these answers may reflect not the characteristic features of the moods and desires of society, but just a random set of opinions mentioned by a limited number of people during answering the questions they did not understand. And, unfortunately, we have to admit that this problem inspires serious doubts about the existence of “moods” and “desires” of society that can be clarified by implementation of a public opinion poll.
There are also some complications with the second condition, that assumes that researcher must formulate appropriate questions for participants of the research and provide them an opportunity to express their opinion correctly. It seems obvious that the same question can be asked to a person in different forms. You can use different words to refer to the same phenomenon (for example, the same problems in the economy can be called a "deceleration" or a "crisis"). You can offer them different options of answer to the questions (for example, you can ask them to choose their favorite movie from the list you provided, or you can leave them the opportunity to name their own option). These tricks are known to any person. But in the context of public opinion polls, these details play a crucial role and directly affect the outcome of the research. I will give a simple example. Let's imagine that we have the results of a public opinion poll that show that 73% of the inhabitants of a certain country agree with a ban on visiting public places without medical masks. This result of this survey could be obtained in different ways. People could simply be asked: “Do you agree that you should not enter public places without a medical mask?” with only two options of answer: “Yes” and “No”. The question could have been asked in another way: “Do you think it is better to completely close public places for visiting or allow people to enter them only in a medical mask?”. The question could also be asked in an open form without any offered options of answer: “Which measures of prevention of viral infections you would prefer to see in your country?”. Potentially answers to all of these questions could show that 73% of people supported the ban on visiting public places without medical masks. But we need to admit that it is possible to make extremely different conclusions about the society, depending on the answers to which particular question showed such a result. Therefore, it is important to understand not only the nominal results of public opinion polls, but also the peculiarities of the wording of the questions that were used by the researcher. Unfortunately, this is important detail is often ignored by careless researchers.
Finally, there are certain difficulties with the third condition, which assumes that people must answer questions sincerely and honestly. Unfortunately, there are quite a few circumstances which can make people provide you the answers contradicting their real thoughts and behavior. First of all, people tend to hide their true opinion if they believe that their view differs from the generally accepted one or may offend someone (that is why many people find it difficult to sincerely answer questions about religion, bad habits, sex, and other sensitive topics). Distortions in people's answers often appear also because they can feel a temptation to choose the answers that can show them attractive, cultural, and competence in fashionable topics. Finally, it is important to understand that people can simply change their minds, and such changes can happen quickly. And if today you heard a certain answer to your questions from someone, you cannot be sure that tomorrow he or she will answer your questions in the same way. I do not want to say, that all of the participants of public opinion polls are liars and everything they say is not credible. But I suppose it is important to consider that the psychology of any person is much more complicated than the elementary model of thinking on which public opinion polls are based. In fact, there are many reasons why people’s answers to the researcher's questions may differ from their actual moods and desires.
Thus, we can say that all of the basic conditions necessary for basic logic of public opinion polls do not fully work. People do not always have a clear opinion on topics that interest the researcher. The answers that people provide in public opinion polls depend on the unpredictable quality of questions.
So, does it all mean that public opinion polls are a poor analytical tool that should be abandoned? Despite all the critical implications of the text that I wrote earlier, I believe that it is still not worth posing the question in this way. Although public opinion polls have significant vulnerabilities, they still remain a fairly promising tool for analyzing society. However, such surveys have their limitations, and it is important to interpret their outcomes with caution. Public opinion polls cannot be regarded as a tool that allows you to understand exactly what society wants in a given situation. In order to correctly understand the results of public opinion polls, it is important to approach them critically and think about exactly which people took part in the study, what kind of questions were asked to them and what is the likelihood that the answers received from them were sincere. Such reasonable criticism will allow you to work with data from public opinion polls and understand them correctly.
I hope that in this text I have been able to show the main reasons for the vulnerability of public opinion polls and show those nuances that are important to pay attention to for the correct interpretation of their results.