Democracy is meant to be a reflection of the people. The principles of popular sovereignty and majority rule obligate a government to pay attention to what its citizens want. By the same token, they obligate citizens to convey what they want to government. After all, the people cannot have a say in how they are ruled, nor the majority’s preferred course of action be ascertained, unless the public somehow collectively expresses its political views.
Tourists' images are distorted and reflected back at them by Cloud Gate, a sculpture in Chicago's Millennium Park.
But how effective is government based on public opinion anyway? Are the people as a whole capable of sending clear, accurate signals to policymakers about what they want? If so, should policymakers listen? We all like to get what we want, but we also recognize that sometimes what we want isn’t what we need, and that other times we don’t even know what we want.
Before deciding whether or how much government should rely on public opinion, we must answer two more fundamental questions. First, does the public actually have opinions about political issues that could, in principle, be used to chart a course for a democracy? Second, is there a mechanism by which those opinions can be extracted and compiled in such a way as to give government officials an accurate sense of what the public wants them to do?
The answer to both of these questions is a qualified yes. People do have political views, though these views are often less stable than is commonly assumed. And these views are discoverable through sophisticated polling techniques, though even the best polls are limited in their ability to accurately sum up how the populace thinks and feels about political issues. Behind every presidential approval rating or issue poll is a thorny thicket of human psychology and statistical methodology that is both challenging and fascinating to untangle.
Attitudes & Opinions
An attitude is an orientation or predisposition toward some object. This object could be a person (such as the president), a group (such as the Republican Party), an idea (such as the death penalty), or something else. An attitude must have an object: approval of the president, warmth toward the Republican Party, and opposition to the death penalty are all attitudes, but approval, warmth, and opposition by themselves are not.
Attitudes are highly influential on political behavior. It is no surprise, therefore, that many people — politicians, journalists, political scientists, average citizens — want to know what attitudes the public holds. But attitudes as they exist in people’s minds are too intricate to be measured directly. The most advanced technology available for brain imaging cannot tell us exactly how someone thinks or feels about tax cuts or drone warfare, much less how strong or durable those thoughts or feelings are.
Although attitudes themselves cannot be directly measured, the expression of those attitudes can be. An opinion is a measurable manifestation of an attitude. Opinions can be spoken, written, or signaled in some other way that expresses an underlying attitude. Thus, the term public opinion refers not to the collective attitudes of a populace but rather the collective expressions of those attitudes.
At best, an opinion is only an imperfect representation of an attitude. Consider the classic public opinion survey question of whether someone approves or disapproves of the job the president is doing. You can probably think of at least a dozen factors that influence your approval or disapproval of the president’s job performance, but pollsters cannot discern those factors from your one-word answer of “approve” or “disapprove,” nor can they distinguish your answer from someone else’s which is the as yours same but for completely different reasons. Like a two-dimensional photograph of a three-dimensional object, no opinion can fully express the attitude on which it is based.
Opinion Dynamics
Opinions — political or otherwise — shift frequently. When asked the same or similar questions at different times, we often change our answers whether those times are months, days, hours, or even just minutes apart. Attitude change is one possible explanation for opinion change, but opinions can also vary even if the attitudes on which they are based remain constant.
The instability of opinions is rooted in human psychology. Generally, we do not store preformed opinions in our minds, ready to deploy one if someone asks. Instead, we generate opinions on the fly when they are required of us, based on pieces of information stored in our memory. This is why you can encounter totally novel decisions — whether to try an unusual food, what to tell a friend about his or her new outfit, how impressive an artistic or athletic performance was — and make snap judgments with no preparation. You won’t always be satisfied with these snap judgments, but your ability to make them at all instead of being paralyzed with indecision demonstrates that opinion formation often happens in the moment.
When people generate opinions on demand, they generally do not consider all the relevant information in their minds before reaching a conclusion. You would likely think about only a handful of reasons to approve or disapprove of the president before stating your opinion, even though you might be able to list a hundred reasons if given enough time. Your attitude toward the president may consist of many factors, but only a few will influence the opinion you render. The salience of the information in your head — how likely each piece of information is to come to mind — determines which factors you consider when forming an opinion. The more salient a factor, the more likely you are to consider it.
The salience of information in a person’s mind changes over time due to priming. Events and experiences can “prime” certain information, making it temporarily easier to recall and therefore more influential in opinion formation. This is why Americans are likely to report feeling more patriotic if they are asked in July than if they are asked in April. In July, Americans are primed by Independence Day celebrations to think about the benefits of their citizenship, whereas in April they are primed by the deadline for paying their federal income tax to think about its costs. Primes don’t necessarily change people’s attitudes, but they do change which parts of those attitudes come to mind most readily, for at least long enough to affect which opinions are expressed.
Although primes can be surprisingly effective at changing opinions almost instantaneously, it’s important to not overstate their impact. Primes tend to be most influential right when a person is first exposed to them, after which their potency declines rapidly. One reason for this phenomenon is that people are exposed to so many primes that the effect of any one prime is quickly superseded by multiple other primes, unless the initial prime is particularly powerful or repeatedly reinforced.
How opinions are solicited can influence the salience of factors in a person’s mind, thereby influencing the opinions they report. Framing is a special category of priming which stems from how questions or issues are presented. For example, supporters of abortion often refer to “reproductive rights” when describing the issue, whereas opponents of abortion often mention “the right to life.” These expressions are used specifically to frame the issue of abortion by emphasizing different factors (the rights of the mother versus the rights of the child) with the goal of making those factors more salient. Most public opinion researchers try to avoid biased framing when designing survey questions. Nevertheless, unintentional framing effects often produce noticeable effects on opinions, even when they have no lasting impact on underlying attitudes.
Primes and frames do not impact all people equally. The same prime or frame might impact one person strongly, a second weakly, and a third not at all. It is impossible to know based on a single survey response how much of a role a particular prime or frame played in shaping that response. Still, primes and frames can influence enough people to cause substantial shifts in public opinion overall. Understanding how and in what contexts opinions are solicited are asked is therefore crucial for interpreting public opinion data.
Sources of Public Opinion
Political attitudes differ because people experience politics in different ways. Government enables a diverse society to solve collective action problems, but those problems (and their solutions) impact certain people more than others. Even defining what counts as a problem is tricky: a status quo that some find intolerable may be ideal for others. It is therefore no surprise that people disagree with one another about politics.
People’s differential experiences with politics are often linked in some way to their identities. An identity is a stable, defining characteristic of a person that affects many aspects of their life. A person’s identities might include their sex, age, race, ethnicity, education level, income level, religion, sexual orientation, occupation, and party, among others. (Note that identities are distinct from character and personality traits, such as “hardworking” or “shy.”)
Not every personal characteristic is a strong identity for every person who shares it. Two people may both classify themselves as Catholic, but the one who regularly attends Mass, strives to abide by Catholic teachings, and is actively involved in parish activities probably identifies more strongly as a Catholic than the one who grew up Catholic but is no longer practicing. Similarly, two people may both consider themselves Chinese-American, but that identity is probably more central to the one who speaks Chinese, practices Chinese cultural customs, and maintains a social network of many other Chinese-Americans than to the one who does not. A person can be technically classified as a member of a group without necessarily identifying with that group.
Public opinion data suggest that identities strongly influence political attitudes and opinions. Large gaps persist between the political opinions of different demographic groups, indicating that these identities shape the ways people react to politics. For example, in the United States, on average, blacks tend to be significantly more supportive than whites of the Democratic Party. This does not mean that all blacks support the Democratic Party, nor does it mean that any specific person supports the Democratic Party because he or she is black. Nevertheless, the trend is sizable and stable enough to suggest that there is some important relationship between race and party support in the United States.
Not all influences on public opinion are as stable as identities. Local, state, regional, national, and international events can have temporary or lasting effects on how people think and feel about politics. The president’s job approval (as recorded in Figure 7.1 below) tends to sharply increase at the start of a war but gradually decrease as costs and casualties mount. Presidents also tend to get credit or blame for economic booms and recessions that happen while they are in office, regardless of whether they deserve it. Natural disasters, terrorist attacks, and political scandals can have both short-term and long-term impacts on what issues citizens think are most important, what policies they support, whether and how much they trust their government, and many other elements of public opinion.
Figure 7.1: Presidential approval, 1945-2021 (Source: The American Presidency Project)
Events can also interact with identities to produce differential effects on public opinion. The September 11 attacks on the World Trade Center and the Pentagon shocked and horrified many Americans, but New Yorkers and Washingtonians experienced them most directly — and American Muslims encountered heightened levels of fear, suspicion, and hatred from non-Muslims in the aftermath of the attacks. The COVID-19 pandemic affected just about everyone around the world, but its effect differed depending on whether a person was a healthcare worker, whether he or she had school-aged children, etc. In these and other cases, our identities shape how we experience events, which can in turn influence how those events impact our political perspectives.
The many different factors that influence our opinions make it impossible to credibly claim that any one person’s opinion is caused by any one factor, or to determine precisely how much of that opinion is attributable to a specific identity or event. Even asking people why they believe, feel, or think a certain way may not produce an accurate answer, as we often don’t know or don’t want to admit the true reasons behind our opinions. Still, the public opinion trends we do observe related to identities and events are robust enough for us to make claims about public opinion overall, even if the individual opinions which make it up remain difficult or impossible to fully explain.
Measuring Public Opinion
Public opinion is most commonly measured by polling. A poll or survey is a process of soliciting opinions from people about a particular topic. Polls purport to speak about a group’s views, but they almost never include every member of the group, as this would be impractical or impossible unless the group is very small. Rather, pollsters ask questions of a sample (a small subset) of a population (the group whose opinions they want to know and understand).
Outside of polling, the term population usually refers to every person living in a given territory. The population from which a sample is drawn for a poll, however, can be (and usually is) much narrower. A typical election poll might draw a sample of 1000 or so respondents from a population defined as all likely voters or all registered voters. People who can’t or probably won’t vote aren’t part of the population, because the poll’s purpose is to predict the result of the election. Including kindergartners (or anyone else unlikely to vote) in the population from which the sample is drawn wouldn’t help the poll fulfill its purpose.
Although polls typically use samples which are less than 1% of the population they want to learn about, they can still capture public opinion quite accurately thanks to statistics. If 50% of Americans approve of the president’s job performance and a pollster interviews a random sample of 100 Americans, it is likely that the number of respondents in the sample who approve will be close to 50. The pollster could be unlucky and draw a sample in which only ten people approve, but such instances are rare if the sampling is done properly. These likelihoods and unlikelihoods can be calculated mathematically, so we can look at the opinions expressed by a well-drawn sample and be reasonably confident that they closely resemble the opinions which would be expressed by the entire population.
The trustworthiness of a poll depends in large part on its representativeness. A representative sample is one that resembles the population from which it was drawn. If the population is 50% women, for example, a representative sample would also be 50% women (or very close to that). If the sample were 80% women instead, we might wonder whether it could accurately reflect the population’s views, given how severely it overrepresents women and underrepresents men. The same concerns apply to other potentially relevant demographic characteristics, including age, race and ethnicity, education, income, ideology, and partisanship.
Although public opinion polls have been conducted in the United States since the 1820s, scientific polling based on representative samples only became common in the middle of the 20th century (which is why the presidential approval chart on the previous page begins with Harry S. Truman and not George Washington). Prior to this period, many polls relied on “convenience samples” which overrepresented the types of people who were easiest for pollsters to reach, often to the detriment of their accuracy. Modern polling techniques achieve a high degree of representativeness by using census and election data to draw samples that closely match the population in terms of demographics and “weighting” their results to correct for over- or underrepresentation.
Even a well-executed poll based on a representative sample will rarely match the population’s opinions exactly. To acknowledge this, responsible pollsters report a margin of error alongside their poll results. This margin of error is calculated statistically and describes a range within which a pollster is reasonably sure — usually 95% sure, to be precise — the true value of public opinion is contained. If a poll indicates that 48% of Americans approve of the president’s job performance with a ±3% margin of error, that means the pollster is 95% certain that the president’s true approval rating is somewhere between 45% and 51%. (This still leaves a 5% chance that the president’s true approval rating is less than 45% or greater than 51%.) The larger the sample, the smaller the margin of error (as shown in Figure 7.2 below).
Figure 7.2: Margin of error by sample size (Note: Margins shown are based on evenly split public opinion. When public opinion is lopsided, margins of error are smaller.)
Problems in Public Opinion Measurement
Polling is a science, but an inexact one. Often two polls purport to describe the same population over the same time period but contradict one another in such a way that both cannot be correct. There are several potential sources of error which can bias the polling process and therefore the accuracy of poll results.
Coverage error occurs when not everyone in the population is equally likely to be sampled. Some of the earliest (unscientific) polls in America were conducted in taverns and on trains, resulting in samples that were easily accessible but poor matches for the overall public. Until relatively recently, modern telephone pollsters often did not call cell phones, which led to them undersampling young people and non-homeowners (who tended to not own landline phones). These biases can change the composition of the sample in numerous ways, making it an inaccurate representation of the population.
Nonresponse error occurs when not everyone who was sampled responds to a poll. Over time, nonresponse rates in polling have increased: today, a telephone pollster who calls ten people can expect nine of them (on average) to hang up without completing the poll. If this pattern of nonresponse is nonrandom, it can make the poll less accurate, even if the sample itself is representative. The young, the employed, and the politically disengaged are less likely to respond to an invitation to take a public opinion poll about politics than the old, the unemployed, and the politically engaged, all of whom tend to have more time for and interest in sharing their political views.
A jubliant Harry S. Truman holds up a newspaper erroneously proclaiming his defeat in the 1948 presidential election, based on early polls that failed to capture Truman's eleventh-hour comeback.
Measurement error occurs when the questions asked on a poll fail to adequately capture the opinions being sought. This failure can occur because the questions are worded in a confusing or biased manner, because the answer choices given to respondents do not cover all plausible options, or because the respondents themselves are unable or unwilling to answer the questions accurately. In the United States, for example, self-reported voter turnout is typically much higher than actual voter turnout, because nonvoters have a tendency to lie to pollsters and claim that they performed their civic duty.
A poll about marijuana legalization could easily produce each of these types of error. The poll could frame marijuana use in unflattering terms, thereby artificially depressing support. It could also force respondents to choose only “support” or “oppose,” without including an option for people who believe marijuana should be legal for medical but not recreational purposes. Finally, respondents might lie when asked whether they smoke marijuana to avoid the social stigma attached to marijuana use. In each case, public opinion would be distorted by the error.
No poll is completely without error, and even an error-free poll would at best be a measurement of the way things are at a particular moment in time. Attitudes and especially opinions change with new events and circumstances, and there is no guarantee that a poll conducted today will still accurately reflect public opinion a year, month, week, or day later. If an election poll taken in October does not match the results of the election held in November, that does not mean the October poll was wrong, only that it was no longer accurate by the time the votes were cast.
Vox Populi, Vox Dei?
Machines take inputs and convert them into outputs. The American political machine takes public opinion as its input and converts it into government policy. Americans tend to like this arrangement, at least in theory. To some, it seems like the most effective method of governing a country in a way that will please the populace. To others, it is a moral obligation, and it would still be the the right thing to do even if it weren’t particularly effective.
Americans’ faith in democracy can be summed up by the Latin phrase vox populi, vox Dei: “The voice of the people is the voice of God.” We cite polling data on the popularity or unpopularity of politicians, parties, and policies with an air of finality, as though they were judgments issued from some almighty being. Whenever politicians pursue unpopular agendas over the public’s objections, there is a sense that our democracy is in some way being diminished.
Of course, the will of the people is not almighty in America. Many crucial government decisions are made by political and legal experts, insulated from public influence. Often this is done for the public’s own benefit. After all, public opinion has many shortcomings. Our opinions are inconstant, changing with the times and not always for good reason. We are uninformed or underinformed on many important issues, and it shows: many government policies that were popular to begin with turned out unsuccessful or even disastrous. The principle of “garbage in, garbage out” warns us not to be surprised when the machine that is America produces shoddy outputs from poor-quality inputs.
Thus, vox populi differs from vox Dei in at least two respects: the public is neither all-powerful nor all-knowing. As an indirect democracy, America is structured to channel the uninformed and fickle force that is public opinion into effective government action. It attempts to balance letting the people have a say in how they are ruled with trusting the judgment and expertise of government officials to occasionally go against the public’s wishes when circumstances call for it. How well it achieves this goal is — like so many things in American politics — a matter of opinion.
