Quantitative and qualitative methods are used to analyze media content about polled support for the invasion of Iraq before it took place. A tendency for journalists to emphasize data suggesting majority support rather than opposition is identified, and an associated perception of a
pro-war majority is seen in letters to the editor. Analysis contrasts conceptual and methodological imprecision attending discussions of public opinion with numerical precision in survey data, arguing that a scientific reading of poll results should not be equated with public opinion. A perceived
pro-war majority aligned with a pro-war political climate is explained in terms of a distinction between scientific, instrumental, and symbolic readings of polls, the mediated nature of public opinion, and symbolic power.