The Mainstream Outsider: News Media Portrayals of Presidential Candidate Mitt Romney and His Mormonism, 2006-2008

Abstract

This study examines how news media framed former Massachusetts Gov. Mitt Romney and his Mormonism during his unsuccessful quest for the 2008 Republican presidential nomination. The study's central finding is that, in the aggregate, news accounts framed Mormonism as outside the American religious and cultural mainstream. This framing emerged as part of campaign's "horse-race" coverage, which focused on who was ahead in the nomination race, who was behind and why. That coverage naturally highlighted aspects of Mormonism that caused Romney electoral problems. Journalists zeroed in on the church's history of polygamy, on whether the church is a Christian faith and on current church beliefs that may appear outside the mainstream. Basic beliefs that Mormons share with other American faiths, such as helping the poor, were mentioned, but less frequently. Romney himself was framed as a generally mainstream candidate whose central problem was his faith. This dissertation also describes how news media relied heavily on an analogy between Romney's struggle to overcome his "Mormon problem" and presidential candidate John F. Kennedy's struggle to overcome anti-Catholic sentiments in 1960. Implications of these conclusions are discussed for candidates of other minority religions and further research is suggested. The study proposes a "horse-race influence model" that highlights a candidate's weaknesses, providing voters with reasons to vote against a candidate, which is reflected in the next set of horse-race coverage polls. Horse-race coverage, therefore, may create a feedback loop that increasingly harms a candidate's chances. Quantitative findings are based on a content analysis of 205 news articles that appeared in eight prominent American news outlets between January 2006 and Romney's withdrawal from the race in February 2008. Articles in the sample mentioned Mormonism at least four times and Romney at least once. The content analysis obtained a mean intercoder reliability of .84.