ABSTRACT
Research into the intended and unintended consequences of negative campaigns has spawned a variety of approaches and a variety of contradictory findings. This paper suggests that some of the contradictions are likely due to methodological issues. First, we examine the use of elite assessment as an improved means of operationalizing the tone of political campaigns. Second, several hypotheses about the mediating effects of local electoral culture are tested. That is, we first derive a measure of negativity from the perceptions of newspaper editors and reporters who covered the 1998 congressional campaigns in the USA, and then merge this information with the 1998 NES data to measure the effects on respondents in different electoral subcultures. We find that the unintended effects of negative campaigns are more pronounced in U.S. states with little or no history of partisan competition. We conclude that elite assessment should be considered as a method for measuring a race's “negativity” and that context is a key component in sorting out the “unintended effects” mystery.