The Sleeper Effect and Negative Political Advertising.

The sleeper effect phenomenon is examined to explain how the impact of negative political advertising persists--and even increases--over time. Using a true experimental design with attack and defensive message treatments, adult subjects were asked to report their vote and its certainty immediately

after message exposure and in a delayed telephone callback. Results indicate that a defensive advertisement following the attack is initially effective; however, over the next few weeks, the impact of the attack ad increases substantially. Similarly, an initial perception that the assailant has low credibility has only a temporary suppressive impact on the effectiveness of the attack ad. Our documentation of both an order-driven and a credibility-driven sleeper effect poses strategic challenges for those candidates who are attacked. Further, these results offer theoretical insights into the determinants and the pervasiveness of the sleeper effect in political campaigns.

In the 1980's and the late 1970's, when negative advertising strategies were still regarded as somewhat "maverick," the volume and style of negativity in campaigns became a mainstay of election coverage (Advertising Age 1998; Television/Radio Age 1988; Will 1986). While such advertising is consistently disliked (Ansolabehere and Iyengar 1995; Hill 1989) and thought to alienate large numbers of potential voters (Freedland 1994; Rothenberg 1990; Teepen 1995), its effectiveness is endorsed by its continuing and increasing use at every level of campaigning (Jamieson 1992; Newsweek 1996; Tinkham and Weaver Lariscy 1997).

The prevalence and increasing legitimacy of negative attack ads (West 1992) raises the question of how to respond to and discount the influence of an attack particularly among candidates who will almost inevitably be attacked. In the 1980's a popular strategy was to "take the high road" if attacked, and not dignify the accusation with a response. Michael Dukakis' lack of response to the "Willie Horton" ad in 1988 is cited by political analysts and commentators (Jamieson and Campbell 1992) as well as by Dukakis himself (Washington Journal, C-Span, 1997) as a contributing factor to the landslide Bush victory.

Similarly, the use of positive messages unrelated to the attack (as attempts to counter negative ads) does not offset the specific message of the attacker. While these ads may bolster the evaluation of the attacked candidate, they are less powerful than the attack ad. Labeled the "negativity bias," results from numerous information processing and judgmental decision-making studies indicate that individuals weigh negative information more heavily than positive information (Kellerman 1984; 1989; Lau 1982; 1985) and are more likely to remember it (Ito, Larsen and Smith 1998; Lang 1991; Lang, Dhillon and Dong 1995).