UC Berkeley

Despite the long-standing and prominent place of partisan identification (PID) in many aggregate- and individual-level models of voting behavior and political cognition in the United States, several basic features of this attachment remain poorly understood and significant controversies persist. The research presented here seeks to build upon the recent conceptualization of PID as a social identity (Green, Palmquist & Schickler, 2002; Greene, 1999, 2000, 2004; Huddy, Mason & Aaroe, 2010) in order to increase our knowledge of the ways in which it may function as such and to expand our understanding of partisan intensity and PID's biasing effects. This conceptualization is one that has been put to surprisingly limited use in political science. This work draws upon new data I have generated during my time as a doctoral student, making use of survey experimental paradigms and a new implicit measure. I call upon foundational and cutting-edge concepts and methods from social psychology in addressing several active research programs in political behavior. The first essay presents the most direct evidence to date regarding the presence of an affective identity component of PID (the way in which Campbell, Converse, Miller & Stokes (1960) conceived of the attachment), which sheds light on partisan intensity and measurement of it. Using data from a survey fielded among subjects in the Project Implicit research pool, it introduces a novel measure of implicit PID that directly measures the identity component as it is defined in balanced identity theory (Greenwald, Banaji, Rudman, Farnham, Nosek & Mellott, 2002), and compares it to standard PID measures. Among other things, the findings offer some confirmation that the traditional two-item, seven-point PID measure largely captures respondent identity levels. This is arguably the strongest evidence to date that the measure does, for the most part, what it was designed to do. I also find that Republican partisans, in the current political environment, are significantly stronger partisan identifiers than their Democratic counterparts. The second essay brings new data from embedded survey experiments to bear, assessing, in the case of political party, the presence of the kind of group-based bias often associated with social identities. The manipulation and measure are designed to avoid the confounders present in prior studies that have allowed some to question the biasing effect of PID. Consistent evidence suggestive of group-based bias emerges. These findings establish a new benchmark in this research program by demonstrating, at a micro level, the extent to which partisans are susceptible to a set of standard mechanisms for rationalization, information dismissal and motivated processing. Beyond adding evidence to the debate regarding perceptual bias, though, this paradigm allows for more nuanced analysis of the nature of that bias and heterogeneity in its expression. The final essay uses the notion of "rooting interest" to link this perceptual bias with a social identity model of PID. A manipulation was used to vary the relative salience of an individual's personal and collective self-concepts (Ambady, Paik, Steele, Owen-Smith & Mitchell, 2004), the interplay between which is at the heart of psychological conceptualizations of identity. The effects of this manipulation on the level of bias observed suggest that the strength of rooting interest may vary somewhat, but that the nature of the variation depends upon the political saturation of the context and differs between Republicans and Democrats in the current political environment. The results 1) demonstrate that manipulation of self-concept salience and variations in background politicization can alter the magnitude of bias; 2) provide evidence that this bias is pronounced even in less politicized contexts and when the personal self-concept is made more salient; and 3) suggest that bias is asymmetric across the two parties, with Republicans showing a higher baseline level, but some propensity to have their bias level manipulated downward, and Democrats starting at a lower point, but with the potential to be manipulated upward. Taken together, these new data (from both the experimental and measurement work) demonstrate two important points that were not as apparent in prior studies using other methods. To begin with, it appears that partisans of various intensities (strong Democrat versus strong Republican, for instance) should not be thought of or analyzed as mirror images of each other. Furthermore, it appears that a meaningful "Identity Gap" may exist between Republicans and Democrats in the current political moment. These emergent findings suggest future areas of inquiry, ways in which we might reexamine prior findings, and new potential research programs.