By Josh Pasek (University of Michigan)

At the heart of democratic governance is the notion that what government does should be responsive to what people want. To reflect citizens’ desires, political leaders need to have a sense of which policies most people would prefer, what kinds of concerns they have, and ideally why they are making the choices they are making. Public opinion polling is the primary vehicle through which the desires, hopes, and preferences of members of the public trickle up to influence the decisions of social and political leaders.

Institutions of governance in contemporary democracies offer citizens relatively limited opportunity to express their preferences and provide no meaningful mechanism for explaining those preferences to leaders. When voters enter the voting booth, they typically indicate who they think should represent them at various levels of government and sometimes also get to express up-or-down views on ballot initiatives. These choices are not particularly revealing about public desires. Election results don’t tell us who made which choices or what their motivations might have been. Indeed, knowing only who won does not reveal whether voters were expressing a preference for the candidate they chose or against that candidate’s opponent. So, although representatives often begin their terms in office asserting that they have a mandate to lead, the results of elections provide little insight into what, if anything, that mandate is for.

“Public opinion polling is the primary vehicle through which the desires, hopes, and preferences of members of the public trickle up to influence the decisions of social and political leaders.”

Public opinion surveys fill this gap. They contextualize the results of elections, allowing us to discern who made which choices, what they were thinking about, and how the events and messages of a campaign cycle shaped the ways people came to their decisions. Public opinion surveys also provide a window into what the citizenry wants on a more regular basis, letting us understand how preferences are changing and gauge responses to the novel issues and events that emerge during periods between elections. They also reveal what people do and do not know about political issues and thereby how informed their preferences are.

In many cases, the greatest attention to polls comes as they track responses to the events and messages of a campaign, before voters fill out their ballots. Surveys that are primarily used to describe who is likely to turn out to vote and which candidate or candidates they will choose are referred to as trial-heat or horserace polls. These polls can provide important insights into how different parts of an electorate are behaving or how things are changing over time.

Notably, there are a number of valid concerns about the misuse of public polling. News reports sometimes cover only the overall head-to-head results rather than focusing on the unique information that surveys can tell us about elections. More perniciously, when polls are treated as if their primary purpose is to estimate who will win, they have the ability to influence the results of the very decisions they aim to measure. Voters who are convinced that a preferred candidate is hopelessly behind or comfortably ahead may feel less compelled to canvas on behalf of that candidate or even to show up on Election Day. This sort of difference should be expected, because these polls capture the state of a race at a particular point in time and are not adjusted to forecast how attitudes and behaviors are likely to be different on Election Day.

“News reports sometimes cover only the overall head-to-head results rather than focusing on the unique information that surveys can tell us about elections.”

For the last few election cycles, there has been a growing tendency to aggregate polls and other data into election forecasts. Polls are indeed useful in setting expectations about who will win elections, which can be valuable for industries to ready themselves for the policy changes and economic shifts that often come with governance transitions. But as this process often understates the inherent error in polling (especially about future behaviors) and largely omits the contextual information that polling provides, this is probably not polling’s most useful purpose.

But it is important to remember that surveys tell us a whole lot more than who is likely to win. Polls are an incredible opportunity for political and social leaders to ask questions of members of the public and to find out what people think and what they want. Understanding popular desires and how they relate to behaviors both during and between elections is one of the central processes that allows government officials to take actions that represent what the public wants. And when those in power are not responding to the public, polls provide information to the political opposition on what kinds of policy proposals and initiatives might win public support and aid their prospects in future elections. While critiques rightly note that public polls only reflect views on the questions being asked and are frequently under-informed and only partly considered, it is hard to imagine that we would better reflect the pubic by ignoring their views summarily.

Indeed, it is not an exaggeration to say that public opinion polling is the most prevalent approach for contemporary democracies to reflect the desires of the public.

Josh Pasek is Professor of Communication & Media and Political Science, Faculty Associate in the Center for Political Studies, and Core Faculty at the Michigan Institute for Data & AI in Society at the University of Michigan. His research explores how new media and psychological processes each shape political attitudes, public opinion, and political behaviors. Josh also examines issues in the measurement of public opinion including techniques for reducing measurement error and improving population inferences.

Download as PDF