As we noted at the beginning of this book, election polls play a powerful role in contemporary American democracy. They can and do affect the candidates and their campaigns; reporters, editors, and the campaign coverage that their news organizations provide; and the attitudes, candidate preferences, and voting behavior of the public.
We believe that election polls used to their best and fullest potential could have a significant positive impact on the American electorate, in part because of their singular ability to provide an accurate reflection of the attitudes and preferences of the people on a timely basis. Too often, their potential for good is not realized, whereas their potential for harming democracy seems to manifest itself routinely.
Candidates and campaign strategists appear to use polling more often to determine merely how to beat an opponent rather than how best to serve the public. Journalists and their news organizations appear to use polling more often than not to help identify likely winners, whom they then decide will "merit" news coverage, rather than to identify, investigate, and portray the appropriate complexities of issues that can and should affect public policy formulation. Given the typical content and format of reporting based on media polls, the electorate appears to use polling results too often to decide whether or not to vote rather than to further educate themselves