Partisan mobilization, cognitive mobilization and the changing American electorate (original) (raw)

One can arguably claim that party identification is the most important concept in modern electoral behavior research. 1 The early analyses of the Michigan election studies demonstrated how partisanship was a core element in political identities and behavior (Campbell et al. 1960, 1966). Since then, party identification is routinely a predictor in a wide array of analyses, ranging from voting to participation to predicting issue positions. Indeed, the developers of the concept stressed its functional importance: The present analysis of party identification is based on the assumption that the ... parties serve as standard-setting groups for a significant proportion of the people in this country. In other words, it is assumed that many people associate themselves psychologically with one or the other of the parties, and that this identification has predictable relationships with their perceptions, evaluations, and actions. (Campbell, Gurin and Miller 1954: 90) But just as the authors of The American Voter were demonstrating the analytic and predictive power of partisan identification as a concept, these ties began to weaken. At first, researchers asked if this was a temporary response to the political controversies of the late 1960s and early 1970s (Converse 1976; Abramson 1979). But now, a quarter century later, partisan attachments remain weaker than during the "stable state" period of The American Voter studies. Indeed, partisanship reached a new low point in the 2000 American National Election Study (ANES); only 41 percent of respondents claimed to be independents in 2000, compared to barely 25 percent in the 1950s. These trends of weakening partisanship are well documented in American electoral research, but their meaning and interpretation remain widely debated. Some analysts discount the significance of these trends, or even the reality of partisan dealignment (Green, Palmquist, and Schickler 2002; Bartels 2000; Keith et al. 1992). Others discuss weakening partisanship as due to the depoliticization of the electoral politics (Dimock 1998; Milner 2002), consistent with Putnam's thesis of Americans' decreasing social engagement (Putnam 2000). This research tests an alternative explanation that begins with the functional theory of partisanship that underlies The American Voter model, and then asks how the socioeconomic transformation of American society during the later half of the 20 th century may have altered this logic. Using the long data series from the American National Election Studies (ANES), we