John E. Newhagen University of Maryland
Byron Reeves Stanford University
If negative advertising does not work, its increasing use across the American political landscape would be difficult to explain (see Guskind & Hagstrom, 1988; Nugent, 1987). During the 1988 Presidential campaign, for instance, viewers saw 30-second spots charging that Michael Dukakis was soft on criminals, allowed pollution to go unchecked, and was weak on national defense. There were images of criminals walking to freedom through penitentiary gates, disgusting pools of industrial pollution, and Dukakis smiling from the turret of an Army tank as it drove in circles around an open field. Late in the campaign, Dukakis countered with his own negative attacks on George Bush, and at one point tried to make the negative tone of the Bush campaign an issue itself.
Yet a review of research results casts doubt on the effectiveness of negative political advertising. Negative commercials can boomerang on the sponsor ( Garramone, 1985; Shapiro & Rieger, 1989) and can produce unstable attitudes ( Krugman, 1981). And correlations between attitude shift and memory are weak at best (see Petty, Cacioppo, & Kasmer, 1988; Rothschild, 1974).
Much of the work in negative political advertising is based on persuasion models using attitude as the independent variable and behavior change as the dependent variable (see Fishbein & Ajzen, 1981, for a description of attitude-behavior persuasion models; and Berry, 1983,