how partisan attachments affect candidate assessments (part 1)
by nick
Few factors are of greater importance for our national elections than the lasting attachment of tens of millions of Americans to one of the parties….Most Americans have this sense of attachment with one party or the other. And for the individual who does, the strength and direction of party identification are facts of central importance in accounting for attitude and behavior.
–The American Voter (Campbell et. al 1960, 121)
While it may seem absurd today, what with a 24 hour news cycle dedicated to bringing you the news in your flavor of reality, the idea that partisanship is the primary driver of voter attitudes and behavior hasn’t always engendered widespread acceptance.
In fact, in a highly critical response to a ground-breaking 1960′s book titled The American Voter, well-known political scientist V.O. Key rejected the proposition that psychological partisan loyalties were responsible for voter behavior. Famously arguing that “voters are not fools,” Key believed that electoral decisions were instead a function of a highly-rational, “reward-punishment” calculus which drew upon careful retrospective judgments about past economic success (Key 1966, 7). What good was a theoretical framework of voting behavior based on partisan loyalties that dated back to the Civil War era and that was, furthermore, rapidly evolving in a post-Depression world that spawned the Johnson landslide in 1964 and the South’s abandonment of the Democratic party in favor of Republicans (Bartels 2008b, 11)?
Suspicions about the salience of partisanship persisted for another 30 years, with varying explanations building on Key’s original analyses. In a textbook written in the mid-1990′s, two authors asserted that perhaps major parties didn’t resonate with voters as they had in the past, citing evidence which suggested that strong party identifiers had decreased. Some researchers suggested that a general partisan decline was the result of a drift towards candidate-centric politics (beginning with the Johnson campaign of ’64), while others thought that voters were less likely to connect their circumstances to their choice of political affiliation (just plain bizarre).
In a very real sense, however, party affiliation did suffer a marked decline beginning in the early 1960’s and continuing through the mid 1970’s. Because the National Election Studies survey (the most well-known, academic-quality series of election polling in American politics) not only tracks respondents’ party association, but the strength of that preference as well, we are able to examine nuanced party affiliation trends that extend back to 1952. Figure 1 illustrates that strong Republican identifiers began to decline after the 1960 presidential election, while strong Democrat identifiers began to show weakening signs of support after 1964. Although Figure 2 suggests that the number of weak Republican partisans has remained fairly consistent over the course of the last sixty years, weak Democratic support has dropped farily significantly since the 1964 elections. As a consequence of this decline, as illustrated in Figure 3, a great many of these partisan “defectors” apparently became Independents.
According to Fiorina (2002), the era of dealignment pictured above was the direct result of the organizational decline of the parties, facilitated by the implementation of new federal regulations which ensured that public sector jobs were largely removed from party control and that the payoff system for public office was abolished. When combined with the civil rights upheaval, the war in Vietnam, and the further realignment of the traditionally southern Democratic bedrock, the theoretical framework behind The American Voter’s partisan voter seemed to deteriorate at an alarming rate—so much so that even its authors began to hedge on their original analysis of the salience of partisanship on voter behavior (Miller et. al 1976; Bartels 2008b). In place of party loyalties, it was suggested that perhaps issues of public policy, a transition to decidedly candidate-centric elections, or even an increase in the consistency of mass political attitudes had become the new prominent influences on voter behavior (Converse, 1964; Key 1966; Nie, Verba and Petrocik 1976).
There seem to be two plausible responses to these substantial criticisms. First, the idea that voters had developed more consistent attitudes about issues and candidates was shown by Brunk (1978) and Sullivan et al. (1978) to be the result of a structural change in the way public opinion surveys asked and formatted certain questions. As Bartels writes in a detailed history of the study of partisan behavior, the increased issue-coherence found by critics “was probably an artifact of the change in question wording, rather than reflecting a fundamental transformation in the structure of political attitudes (2008b, 14).”
Second, while there was some decline in the level of partisan loyalties during this period, the stability of individual voter partisanship appears to be quite consistent over time when we examine votes cast by respondent party identification. Figures 4 and 5 demonstrate that over the past sixty years, strong Democrats voted for Democratic candidates at an average rate above 80 percent, while strong Republicans voted for Republican candidates well over 90 percent of the time. Weak partisans showed similar average support for their party’s candidates, with weak Democrats voting more than 70 percent of the time for Democratic candidates and weak Republicans voting 85 percent of the time for Republican candidates. Thus, even though the number of partisans may have decreased, partisan adherence to strict own-party voting patterns certainly did not.
If partisans’ voting habits showed little change over time, then what about those of the growing ranks of Independent voters? Were they as open-minded as their new, party-neutral affiliations insinuated? A closer examination of the behavior of this voting bloc suggests they were not. Because the National Election Study survey not only measures the strength of partisan affiliation, but the party-leaning tendencies of Independents as well, we can actually observe how Independent-leaners voted. The groups of Republican and Democratic leaners—who made up, on average, two-thirds of all Independents—were actually more likely to vote in a manner reflective of their partisan predilections than weak party identifiers: Independent Democratic leaners voted for Democrats an average of 77 percent of the time, while Republican leaners voted for Republicans more than 86 percent of the time. Thus, even though the numbers of strong and weak partisans slightly decreased over the course of time, when the behavior of Independent leaners is accounted for, partisan voting hardly seems to have declined. This echoes Keith’s (1992) findings in The Myth of the Independent that any level of partisan identity has an extremely high likelihood to be reflected in vote choice.
———
Bartels, Larry M. 2008b. “The Study of Electoral Behavior.” Jan E. Leighley, ed. The Oxford Handbook of American Elections and Political Behavior. Web manuscript: http://www.princeton.edu/~bartels/electoralbehavior.pdf
Brunk, Gregory G. 1978. “The 1964 Attitude Consistency Leap Reconsidered,” Political Methodology. 5: 347-35
Converse, Philip. 1964. “The Nature of Belief Systems in Mass Publics.”
Key, V.O. 1966. The Responsible Electorate. Belknap Press.
Fiorina, Morris P. 2002. “Parties and Partisanship: A 40-Year Retrospective.” Political Behavior. 24: 93-115.
Miller, Arthur H., Warren E. Miller, Alden S. Raine, and Thad A. Brown. 1976. “A Majority Party in Disarray.” American Political Science Review 70: 753-778
Nie, Norman H., Sidney Verba, and John R. Petrocik. 1976. The Changing American Voter. Harvard University Press
Sullivan, John L., James E. Piereson, George E. Marcus, and Stanley Feldman. 1979. “The More Things Change, the More They Stay the Same: The Stability of Mass Belief Systems,” American Journal of Political Science. 23: 176-186.
