Over the past three decades, political scientists have been developing general models of presidential approval ratings, seeking to determine the structure of aggregate approval. This endeavor has culminated in the broad claim that "peace, prosperity, and probity" drive the public's approval. The unprecedented events of the Clinton Presidency, especially his high approval during and after impeachment, present a strong challenge to this model. However, the existing model explains Clinton's approval remarkably well, suggesting that the public punished and rewarded him for the state of the economy, major political events, and his integrity. Passing this strong test constitutes considerable support for the existing model of approval.
Political science aims to develop general models of political phenomena that can make sense of and account for a wide variety of conditions and outcomes. The simple fact that time marches on, presenting new conditions, outcomes, and events, demands that models of politics be general enough to account for these innovations and idiosyncrasies. New and surprising outcomes often provide difficult tests for models and scholars need to determine whether existing models can explain apparent anomalies or if they need to be adjusted or replaced. Over the past three decades, political scientists have been developing general models of presidential approval ratings, seeking to understand and explain their structure and dynamics. This endeavor has culminated in the general claim that "peace, prosperity, and probity" drive the public's approval (Ostrom and Smith 1992: 128). The unprecedented events of the Clinton presidency present a strong challenge to this model and many interpretations of these events suggest that this model should be examined closely and may need revision. This article argues that the existing model handles this apparent anomaly remarkably well, explaining Clinton's approval ratings just as well as it explained presidents' approval in the previous 40 years.
The oddities of Clinton's approval have been well documented: while approval models suggest that the public punishes presidents for scandals, he remained popular after several scandals and his approval ratings actually increased during the Lewinsky investigation and impeachment proceedings. These events raise the possibility that the public considered the scandal irrelevant to Clinton's performance in office, separating its opinion of his personal integrity from its opinion of his job performance. This view has become quite prevalent, both within popular discourse about the affair and among many political scientists' initial assessments of Clinton's popularity (e.g., Campbell and Rockman 2000; Edwards 2000; Jacobson 1999; Miroff 2000). If true, the existing model does not account for Clinton's approval ratings and must be changed.
This article examines whether explaining Clinton's approval ratings requires a different model by comparing the structure of his approval ratings to the structure of previous presidents' approval ratings. The study has four sections. The first reviews the basic model of approval and presents the Clinton presidency's challenge to the model. The second tests the existing model against Clinton's approval ratings. The third takes a closer look at the model's explanation of Clinton's approval, highlighting the effects of the economy and the Lewinsky scandal. The fourth draws some conclusions about the existing model and the nature of the connection between presidents and the public.
PRESIDENTIAL APPROVAL AND PRESIDENT CLINTON
Presidential approval ratings are the most frequently measured and best known political "fact" in the current American political environment (Ragsdale 1998), the "Dow Jones Index for Politics" (Brehm 1993: 6). Many consider these ratings a form of political capital and argue that the ratings affect various aspects of presidential strategies and political outcomes, including the president's legislative agenda, strategy, and success (e. …