Content area
Abstract
This study explores the relationship between political knowledge, representation, and political equality in opinion surveys. Contrary to much of the literature on collective opinion, I find that the low levels and uneven social distribution of political knowledge in the mass public often cause opinion surveys to misrepresent the mix of voices and interests in a society. Correcting for information asymmetries reveals that many collective policy preferences would look quite different if all citizens were equally well informed about politics.
Although it is commonly thought that statistical aggregation and information shortcuts allow the mass public to compensate for its inattentiveness to politics, I show that the promises of "collective rationality" and cognitive heuristics often fail to materialize. Aggregation turns out to have limited information-pooling qualities, and few survey questions on policy issues include the kinds of heuristic cues most helpful to people lacking factual knowledge of politics.
Because knowledgeable respondents are better able to form opinions consistent with their political predispositions, and because they tend to give opinions more frequently than other people, the demographic characteristics of relatively informed people--who tend to be more affluent, older, white, and male than the ill-informed--can cause collective preferences to reflect disproportionately the preferences of some groups more than others.
To measure the bias introduced by information effects, I present a method for simulating "fully informed" collective preferences using actual survey data. I find that the mass public, as seen through the lens of opinion surveys, appears more progressive on some issues and more conservative on others than it might be if all citizens were equally well informed about politics. To the extent that opinion polls influence democratic politics, this finding suggests that information effects can impair the responsiveness of governments to their citizens.
In this light, I argue that the results of opinion surveys never should be equated casually with the "will of the people." Instead, I emphasize the importance of identifying quality in measures of surveyed opinion, and suggest that survey results are of higher quality when information effects are minimal.
