Measuring media exposure and the effects of negative campaign ads

Recent controversy over negative television campaign commercials has focused on their effects on voters. Proponents of the demobilization hypothesis claim that negative ads undermine political efficacy and depress voter turnout. Others have suggested a stimulation hypothesis, arguing that such advertising may have an invigorating effect on the electorate. Empirical tests of competing claims demand improved measures of real voters' exposure to real ads in the context of real campaigns. We develop a new approach to estimating exposure outside the lab that combines respondent viewing behavior and the strategic decisions of campaigns. Using this combined measure, we find no evidence that exposure to negative advertising depresses turnout. Instead exposure to negative ads appears to increase the likelihood of voting. We find this effect when we estimate exposure with our new measure, as well as when we use a very different perceptual measure of ad tone.