38 - Newspaper Research Journal • Vol. 27, No. 4 • Fall 2006 Problems Found in Reporting USA Today Pre-Election Polls by Kelli S. Burns During the 2004 presidential campaign, USA Today demonstrated several lapses in reporting polling data, including reporting the sampling error as an actual difference or change. __________________________________________ The rise in polling firms has naturally led to an increase in the reporting of poll results by media organizations, many of which are co-sponsors of these polls. The publication of pre-election polls is a questionable practice that generates controversy between media critics and the journalists who are report- ing these polls. While journalists should not withhold this information from the public, they should make every effort to provide fair and accurate reporting of the data. One criticism of pre-election polls is that the availability of poll results tends to breed “horse race” journalism, where campaign issues are downplayed in favor tracking which candidate is in the lead. 1 A study of the 1996 presidential election found 70 percent of stories in several print sources contained a discus- sion of the “horse race.” 2 Polls are being used as predictors of the election outcome, rather than aids that help campaign strategists understand how messages are persuading a constant group of people. 3 Others question the influence of pre-election polls on voting behavior. 4 The reporting of polls can affect the momentum of a campaign by creating false impressions about the standing of the candidates, which may in turn affect voters’ thinking. 5 A candidate who trails in the polls generates campaign pessimism, resulting in an inability for the candidate to raise money and recruit volunteers. A trailing candidate also receives more critical news coverage or less coverage in general. 6 Burns is an assistant professor in the School of Mass Communications at the University of South Florida.