Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Political Identity and Partisan Attachment

Abstract. Political identity and partisan attachment describe the psychological affiliations through which people connect their self-understanding, emotions, judgments, and political conduct to organized parties and movements. This fundamental research monograph reconstructs the election-study and social-identity traditions, distinguishes instrumental party evaluation from expressive group belonging, and examines family socialization, identity stability and change, partisan sorting, negative partisanship, cognitive cue processing, emotional mobilization, interparty judgment, and comparative electoral institutions. It critically synthesizes foundational research, longitudinal panel analyses, survey and identity-salience experiments, cross-national psychometric work, behavioral measures, and contemporary intervention evidence. The analysis distinguishes identity strength from salience and content, and separates party identification from ideology, outparty animosity, democratic attitudes, and actual conduct. Findings support a historically situated and context-sensitive account in which identity and substantive political evaluation can jointly shape political action, while causal estimates and generalization depend on research design. The work is a critical narrative research synthesis; it reports no new primary data or independently conducted systematic review.

Keywords: political identity; partisan attachment; party identification; expressive partisanship; instrumental partisanship; social identity; negative partisanship; partisan cues; political socialization; partisan sorting; identity salience; partisan emotion; electoral participation; partisan discrimination; multiparty systems; democratic norms; political identity measurement; Artificial Era

Introduction: Political Attachment as a Psychological Research Object

Political identity denotes the aspect of self-understanding through which a person recognizes a political party, electoral coalition, or organized movement as a meaningful social affiliation. Partisan attachment is the degree and character of identification, commitment, emotional investment, and perceived belonging associated with that affiliation. A party can become part of a person's self-description even when the individual holds no membership card, attends no meetings, and disagrees with some of its policies. The psychological object therefore extends beyond current vote choice. It includes the meaning of belonging, judgments about who represents the group, readiness to defend its standing, and the conditions under which identification guides decisions. The original party-identification tradition of Campbell et al. (1960), the social-identity account of Green et al. (2002), and the experimental program of Huddy et al. (2015) illuminate distinct elements of this object.

The central scientific problem is to explain how partisan attachment forms, becomes accessible in a situation, acquires varying strength, persists or changes across elections, and structures cognition, emotion, and action. Political institutions provide organizations and symbols, while families, peers, media environments, and political events supply occasions for acquiring and interpreting those symbols. Individual citizens may treat party affiliation as a cognitive shortcut for comparing candidates, as an evaluative commitment to shared goals, as a group-based identity, or as some combination. These possibilities generate different testable predictions. The same vote for a party can follow from strong psychological belonging, weak preference between candidates, protest against another party, or temporary policy agreement. Observing a ballot alone cannot establish which process occurred.

This monograph offers a critical narrative synthesis of foundational books, original experiments, longitudinal panels, cross-national comparisons, measurement-validation papers, and contemporary research on political identification and partisan behavior. Its evidence is organized around conceptual precision, mechanisms, developmental pathways, measurement, causal inference, contextual variation, and consequences for democratic conduct. Primary research abstracts and accessible publisher records were checked for bibliographic identity; claims are limited to the methods and findings reported in these records. The synthesis does not claim a preregistered systematic search, original datasets, independent statistical reanalysis, or a new causal discovery. The subject is affiliation with politically organized groups; affective polarization, political misinformation, generic social categorization, and human–artificial political persuasion appear where they help clarify its boundaries rather than replacing its focus.

The Conceptual Structure of Partisan Identification

A political party can be described by its formal organization, electoral candidates, ideological reputation, social constituency, historical narratives, and contemporary policy platform. Partisan identification refers to a person's psychological relationship with that category, which need not coincide with formal party registration or recent voting. A partisan identity can have direction, intensity, centrality, emotional positivity, perceived similarity to typical members, and expectations of solidarity. Negative partisanship adds a durable orientation against a disliked party; this is analytically separate from positive identification with a preferred one. Abramowitz & Webster (2016) treats rising hostility to the opposing party as relevant to electoral loyalty, whereas Huddy & Bankert (2017) shows why the positive and negative directions of attachment and the structure of electoral competition must be considered together.

Four constructs that are often conflated require separate operationalization. Party identification is a categorical or graded affiliation; partisan identity strength is the importance and investment attached to the group; ideological orientation describes beliefs about how public life should be ordered; and affect toward party groups measures positive or negative feelings toward those groups. These components can covary without having equivalent meaning. A respondent who says 'strong Democrat' on a conventional US survey might express emotional pride, habitual voting, certainty about policy preferences, or the confidence of a long-standing political family history. Bankert et al. (2017) demonstrated the additional information recovered by a multi-item identity measure, and Mason (2015) demonstrates why stronger alignment of multiple social and political identities can generate social division without requiring equal expansion of ideological distances.

Analytically, identity strength should also be distinguished from identity salience. A highly committed supporter may seldom think about politics during an ordinary workday, whereas a less committed supporter may become temporarily attentive when a party is mentioned. Salience describes the accessibility and situational relevance of the category; strength concerns how much it matters to the self. A third dimension, identity content, specifies what membership means: competence, moral character, national belonging, social status, ideological consistency, or opposition to an adversary. Two citizens can report equal strength yet respond differently to the same leadership statement because the content of their identities differs. Huddy (2001) warned that choice, subjective meaning, gradation, and stability must be included when social identity theory is applied to real political categories.

Historical Formation of the Research Tradition

The systematic investigation of party identification developed with modern election studies, organized survey research, and longitudinal attention to citizens' political orientations. Campbell et al. (1960), based in part on the Michigan Election Studies, placed partisan attachment at the center of an explanatory account of electoral choice. The conceptual innovation was to treat durable psychological identification as an organizing orientation through which more proximate candidate evaluations and issues could be interpreted. This approach made the time dimension central: a current preference could have a history that predates the campaign. As Green et al. (2002) later argued, a social-identity understanding emphasizes the slow formation and relative persistence of partisan ties while allowing long-run transformations when a party's social meaning changes.

The evolution of social psychology introduced a second, complementary vocabulary. The minimal-group experiments of Tajfel et al. (1971) showed that even experimentally created group distinctions could alter allocation decisions under specific conditions. Their contribution to political inquiry is an account of how categories become psychologically consequential. They did not themselves demonstrate the origins of durable electoral loyalties, the operation of election institutions, or the inevitability of outgroup hostility. Huddy (2001) subsequently clarified that political identities have historical meaning, variable choice, and differentiated strength. A party may be woven into family culture, occupational experience, regional social structures, or national transformation in ways that laboratory assignments cannot reproduce.

A third line asked whether partisanship primarily records reasoned evaluation of parties or expresses membership in them. The instrumental account emphasizes performance, policy proximity, competence, and issue agreement; the expressive account emphasizes emotional connection and the maintenance of group standing. Huddy et al. (2015) moved this question into comparisons of measured identity, policy and ideology variables, campaign participation, and experimental emotional reactions to party outcomes. Subsequent research extended these questions to countries with coalition governments and multiple parties Bankert et al. (2017) Huddy et al. (2018). The historical trajectory is therefore not a simple replacement of one model by another. It is the development of research designs able to ask when each mechanism explains identifiable behavior and when their predictions diverge.

Explanatory Models: Instrumental Commitment and Expressive Belonging

An instrumental model conceives party allegiance as a judgment about which organization is most likely to implement desired policies or govern effectively. Voters compare party positions, issue priorities, candidate competence, or retrospective performance. Under conditions of clear policy change, a strictly instrumental account expects corresponding reassessment of the preferred party, subject to information and switching costs. An expressive model instead gives psychological importance to the party as a social group: its prestige, continuity, perceived prototypical members, and threat from rivals can evoke loyalty or mobilization. Huddy et al. (2015) found in four studies that a multi-item identity scale explained campaign involvement better than several issue and ideology alternatives, and that experimental party-status messages elicited emotional responses related to identity strength.

These models imply different causal pathways without requiring different types of people. Instrumental considerations can motivate initial selection, after which repeated identification creates a self-relevant attachment; expressive identification can in turn shape which policy information receives attention. A party can also be both a representative of a social community and a policy vehicle for that community's interests. This creates difficult mediation problems. Finding that identity predicts participation more strongly than self-reported issue strength does not demonstrate that all issue motives are absent; measurement reliability, mobilization opportunities, and omitted experiences may differ. Bullock (2011) provides a vital counterexample to accounts in which party labels overwhelm content: two experiments found that when people were supplied with policy information, that information affected attitudes at least as much as party cues.

Expressive identification becomes more visible when the party's standing appears threatened or confirmed. In Huddy et al. (2015), experimentally induced expectations of party loss and victory changed anger and positive emotion especially among stronger identifiers, while comparable issue-centered threats produced weaker responses. The measurement and experimental manipulations support specific claims about emotional reactivity under campaign conditions. They do not show that all electoral engagement is caused by identity, nor do they identify a timeless trait that operates independently of institutions. In multiparty settings, parties form coalitions, compete across multiple dimensions, and may be assessed through both long-term membership and current policy bargains Huddy et al. (2018). A credible synthesis treats instrumental and expressive motives as mechanisms whose relative contribution must be estimated rather than declared in advance.

Group Categorization, Identity Content, and the Meaning of Belonging

Parties can serve as social categories that distinguish 'us' from 'them', but the psychological consequences depend on the perceived meaning of membership. The minimal-group paradigm Tajfel et al. (1971) demonstrates the causal significance of categorization under intentionally reduced social circumstances. Real political groups add historical loyalties, campaign communication, institutional boundaries, and asymmetrical political resources. Huddy (2001) notes that political identity research must investigate why a person chooses or inherits a category and how that category is understood. An identity may be organized around shared goals, solidarity in an excluded group, civic obligation, distrust of a governing elite, or a symbolically valued history. Treating these motivations as interchangeable loses explanatory detail.

Identity content can be studied through open-ended self-descriptions, perceived party prototypes, issue associations, traits ascribed to copartisans, and measures of collective emotions. Group identification may produce loyalty toward the in-party without equal hostility toward every out-party. Distinct opponents can evoke unequal dislike, especially in multiparty systems with ideological proximity and coalition possibilities. In four-country research, Westwood et al. (2018) reported substantial partisan discrimination and variation associated with ideological proximity, while showing that conflicting social and partisan ties can weaken party-based distrust. The presence of category boundaries is therefore compatible with conditional recognition of cross-cutting relationships.

Partisan symbols also provide signals of membership to other people. Displaying a slogan, choosing an information source, or describing oneself through a partisan label can communicate social location. Such behaviors should be distinguished from private allegiance because their public expression depends on social approval, perceived sanctions, local networks, and the anticipated audience. Klar (2014) experimentally examined social settings and found that interpersonal context changed how party identification shaped political preference formation. The implication is methodological: an identity questionnaire completed privately and an opinion voiced in a politically homogeneous conversation are different observations. A comprehensive account must study both durable belonging and the situation in which belonging becomes expressed.

Family Transmission and Early Political Socialization

Partisan attachments have developmental histories. Children encounter political classifications through family talk, responses to public events, local cultural meanings, and the affiliations of adults whose judgments matter to them. Transmission is a process of selective learning rather than exact copying: a child may acquire a parental label, the emotional meaning associated with that label, a general trust in political institutions, or none of these. The multigenerational longitudinal analysis of Jennings et al. (2009) found greater correspondence between parent and offspring political orientations when families were politically engaged and parents delivered consistent cues across time. The study's longitudinal design supplies stronger developmental evidence than a single cross-sectional parent–child resemblance because it follows political orientations over more than one historical period.

Family influence is mediated by communication opportunities, the credibility of political role models, and the child's emerging ability to interpret party differences. Strong parental identification may transmit only weakly when conflict, migration, competing institutions, or a young person's peer environment alter the meaning of the party. Conversely, a politically engaged household can supply social reinforcement that helps an emerging attachment persist. Jennings et al. (2009) reported a robust direct-transmission pattern after extensive controls, yet observational panels cannot assign a political family history at random. Shared neighborhood, education, social class, and group identity remain relevant to causal interpretation. The developmental question is which family practices transmit a party label, which transmit political values, and how each pathway interacts with later experience.

An adequate developmental model differentiates party recognition, affiliation, psychological investment, and habitual action. A child can recognize the parties before possessing a stable preference; an adolescent may state an affiliation before voting is legally available; an adult may retain a family label while modifying issue positions. Such trajectories should be observed with developmentally appropriate measures rather than assuming that a single 'strength of partisanship' item has identical meaning at every age. The intergenerational research tradition establishes the plausibility of durable transmission while leaving open how much of its apparent stability arises from social identity, policy socialization, network continuity, or wider historical structures Jennings et al. (2009) Green et al. (2002).

Campaigns, Political Events, and Identity Crystallization

Political socialization unfolds unevenly because salient events can make parties psychologically vivid. Sears & Valentino (1997) interviewed adolescents aged 10–17 and their parents in three waves spanning a presidential campaign and its aftermath. The study reported campaign-associated gains in information, affective expression, and stabilization of partisanship for political objects prominently involved in the campaign, with fewer changes in domains peripheral to it. This research supports an episodic model in which events intensify the acquisition or consolidation of political orientations. It also illustrates the value of specifying developmental timing and the content made salient: an electoral campaign may change the stability of party feelings without restructuring every political belief.

An event may operate by increasing attention to party categories, inviting public expression, introducing perceived competition, or connecting an issue to a group. A sudden leadership change can alter who appears typical of a party; a realignment can change which communities perceive a party as their representative. Historical shocks can therefore change identification without altering the basic survey label. Green et al. (2002) emphasizes that major shifts in partisan attachment may occur when a party's social imagery changes, while Levendusky (2009) describes the role of increasingly distinct elite party positions in alignment between party and ideology. Both approaches direct attention to structural opportunities for identity change rather than treating the momentary campaign message as the only influence.

Researchers should distinguish campaign-level change from maturation, period effects, and panel conditioning. Individuals interviewed repeatedly may become more practiced at articulating their identities. Their expressed certainty may increase because questions induce reflection as well as because political events produce genuine developmental change. Stronger inference requires suitable comparison groups, repeated measures beyond election periods, documented exposure to events, and separate outcomes for knowledge, emotional investment, and behavior. The most informative temporal design asks whether a newly salient identity remains central after political attention decreases. Sears & Valentino (1997) contributes a model of how to link events to the timing and domain specificity of political socialization, while subsequent panel work can investigate whether effects recur across different political systems.

Stability, Change, and Party Switching Across the Life Course

Identity stability is an empirical proposition about persistence under specified circumstances, not a property inferred from a label. A person may identify with the same party across decades while the party changes its issue commitments, electorate, and symbolic associations. A stable survey answer can therefore conceal substantial transformation in substantive meaning. Conversely, a change of affiliation may reflect altered partisan imagery, new institutional choices, or disengagement from the old group rather than a complete reversal of political values. Green et al. (2002) argues that partisan identities are often durable and that major changes can follow transformations in the social constituencies associated with parties. This account highlights the distinction between individual-level continuity and collective-level historical change.

Longitudinal analyses must resolve measurement error, temporary ambivalence, genuine identity transition, and changes in the available party system. Repeated single-question party identifiers can overstate mobility if respondents fluctuate around a category boundary. Multi-item scales, latent-state models, and repeated observations of emotion and behavior can help distinguish noise from sustained change. But adding a longer scale does not guarantee invariance when party meanings change. The measurement literature Bankert et al. (2017) shows advantages of multi-item identity measurement for assessing intensity, and this evidence motivates careful evaluation of longitudinal equivalence before interpreting a score difference as a change in underlying attachment.

Switching pathways may be asymmetrical. An individual may first change issue positions and then shift party identification, or may identify with a new party before adopting its policy profile. Levendusky (2009) uses National Election Study panel evidence to examine these alternatives in the context of US ideological sorting and places substantial explanatory weight on adaptation of ideological positions to party attachments. The historical and causal interpretation depends upon the alignment of political elites and the availability of plausible alternatives. A multiparty electorate may permit movement between ideologically adjacent parties that would be difficult to detect in a binary two-party account. The general research program therefore requires repeated identity content, policy position, exposure, and institutional context measurements.

Social Sorting and the Alignment of Political Identities

Partisan sorting describes the increasing correspondence between party identification and ideological or other social classifications. In the US context studied by Levendusky (2009), party elites became more ideologically differentiated, and citizens increasingly aligned partisan labels with ideological positions. Sorting is analytically distinct from an increase in the extremity of every citizen's policy preferences. A population can become more uniformly classified into opposing party camps even when its average issue locations change little. This distinction is central to Mason (2015), which argues that the alignment of identities intensifies social polarization, bias, activism, and anger without requiring equally large shifts in issue polarization.

The psychological mechanism involves convergence among categories that once crossed party lines. When ideological, religious, ethnic, geographic, or lifestyle identities overlap with party membership, an attack on a party may be experienced as an attack on multiple personally valued affiliations. The strength of that association depends on how identities are distributed and represented, and on whether cross-cutting relationships remain available. Westwood et al. (2018) reports that party-based distrust can be weakened when partisan and social identities collide, supporting the importance of overlapping versus cross-cutting group boundaries. But it would be a causal overreach to infer the size of identity-convergence effects from observational sorting trends alone; the distribution of exposure, institutional rules, and historical events can move alongside those trends.

Research on sorting needs to measure changes at two levels: the statistical correspondence between identities in the population and the salience or centrality of identities for individual people. The same national association can arise from people changing parties, changing beliefs, selective social participation, differential mobilization, or the replacement of older cohorts. Mason (2015) analyzed American National Election Studies data to relate social alignment to polarization, while Levendusky (2009) examined longitudinal evidence about how citizens sort. Their contributions are strongest when read as complementary rather than interchangeable: one describes consequences for political social relations, and the other interrogates paths by which alignments form. Country comparisons must also specify whether a party system makes every identity divide map onto the same two organizations.

Negative Partisanship and Identity by Opposition

Some citizens orient politically through opposition to an adversarial party rather than affection for any one party. Negative partisanship describes durable dislike, distrust, or rejection of a party as a political outgroup. It may coincide with positive loyalty, but it can also structure behavior among people who are dissatisfied with the party they vote for. The distinction matters because a person can vote consistently for an organization they barely admire if the alternative is perceived as worse. Abramowitz & Webster (2016) documents a negative shift in feelings toward rival parties among US major-party supporters and relates that development to party loyalty and the nationalization of election outcomes. The study addresses an observed electoral pattern without demonstrating that hostile affect is the sole underlying cause.

Positive and negative identification may have different emotional trajectories. Party victory can evoke pride and reassurance for an emotionally invested supporter, while the prospect of an opposing party's victory can evoke fear or anger even when one's own party generates little enthusiasm. Huddy et al. (2015) examines emotional reactions to perceived party threat and reassurance, and Ward & Tavits (2019) relates positive and negative party affect to perceptions of ideological extremity and the stakes of political competition. Negative attachment should consequently be measured with independent instruments for rejection of the outparty, social distance, and perceived threat, alongside positive identification. Subtracting a dislike score from a liking score can conceal respondents who simultaneously dislike all available parties.

The concept also depends on electoral structure. Opposition to a single dominant adversary in a two-party system has different implications from opposition distributed across several parties, some of which may enter coalition governments together. An individual may detest one party yet tolerate or cooperate with another whose policies are distant. Huddy & Bankert (2017) identifies negative identity and multiparty coalitional contexts as important directions of inquiry, while Westwood et al. (2018) finds that ideological distance conditions party-based distrust across different national systems. The critical scientific question is how negative identification influences choice when the relative availability of parties, institutions, and democratic norms changes. Negative feelings about a party, support for democratic rule, and willingness to engage in harm must remain independently measured outcomes.

Partisan Cues and the Architecture of Political Judgment

A partisan cue conveys information about the party affiliation or endorsement of a candidate, leader, organization, or policy. Cues can save cognitive effort by supplying evidence about positions in a complex political environment; they can also evoke allegiance and opposition independently of detailed policy evaluation. The causal question is what a partisan label changes when identical substantive information is presented under different partisan sponsorship. Nicholson (2012) employed election-period survey experiments and reported that outparty candidate endorsements polarized responses whereas copartisan endorsements did not uniformly persuade. The findings caution against treating all cues as equivalent: the person associated with the position, and the audience's relationship to that person, can influence responses.

Bullock (2011) used experiments that supplied participants with concrete policy information as well as elite-party cues. Under those conditions, policy content influenced attitudes at least as strongly as cues, and the presence of cues did not eliminate deliberation about substance. This result challenges a universal account of partisan citizens as automatic followers of labels. The findings are conditioned on the specific information participants received, the policy setting, and experimental design. In environments where party programs are ambiguous, issues technically complex, or evaluation costly, a cue may have a different role. The experimental literature therefore supports a conditional rather than a universal cue-following hypothesis.

Party cues are often manipulated in survey vignettes because random assignment can isolate their effect while holding message content constant. Such experiments offer strong internal identification of the immediate response to particular endorsements. The dependent variable may nonetheless be an expressed opinion rather than a consequential vote, donation, or willingness to compromise. Repeated exposure, elite credibility, unequal media environments, and the costs of publicly contradicting a group may change the effect outside the experiment. Bakker et al. (2020) tested explanations for partisan cue receptivity and found mixed support for expressive utility and little support for simple low-cognitive-resource predictions. Researchers must ask which cues elicit belonging, which convey useful information, and which trigger opposition.

Identity-Protective Cognition and Motivated Evaluation

Political information processing occurs within social expectations about whose claims are credible and which conclusions affirm the group. Van Bavel & Pereira (2018) proposes an identity-based model in which partisan affiliations can affect attention, memory, reasoning, and evaluation of evidence. This integrative account provides a theoretical architecture rather than a single experiment establishing a specific neural mechanism for every partisan belief. The psychological proposition is that the personal significance of a group can shift the goals brought to an information task, influencing what counts as persuasive evidence. A study must distinguish the effect of an identity manipulation on judgment from the broader causal origins of that identity.

Directional motivation and accuracy motivation can coexist. A citizen may prefer a conclusion favorable to their party yet also care about policy effectiveness, professional competence, or factual truth. An information environment that offers credible details about a policy can reduce dependence on a partisan endorsement, as Bullock (2011) found in experimental comparisons. Conversely, a credible opponent's endorsement can activate suspicion or resistance Nicholson (2012). Neither pattern proves that the participant possesses an invariant intellectual deficiency. The construct concerns how goals, available evidence, and institutional signals jointly organize a judgment in a defined situation. Stronger evidence requires experiments that vary both substantive information and the social identity implied by its source.

The role of expertise is likewise contested. If political cues only compensate for low information, they should be strongest among respondents with the fewest relevant cognitive resources. Bakker et al. (2020) tested bounded-rationality and expressive-utility accounts across survey experiments, finding no support for the predicted low-resource pattern and mixed results for the identity-expression account. These results make it necessary to distinguish ability, motivation, issue familiarity, and identity centrality. The effects of partisan affiliation on reasoning can vary by the domain of knowledge, respondent engagement, threat context, and measurement of the outcome. This dossier treats identity-protective cognition as a family of testable mechanisms whose scope remains open rather than as an explanation that automatically accounts for every erroneous political belief.

Partisan Emotion, Threat Appraisal, and Mobilization

Political identification provides an emotional reference point because events can be appraised as successes or threats to a valued group. Anticipated electoral defeat can evoke anger or anxiety; news of possible victory may evoke enthusiasm, pride, or reassurance. The expressive-partisanship studies in Huddy et al. (2015) tested this pattern using partisan identity measurement and experimental exposure to threat and reassurance. Strong identifiers showed differentiated emotional responses to anticipated outcomes, and these emotions were linked to campaign activity. Such findings concern the activation of group-oriented emotion in electoral contexts. They do not equate strong attachment with continuous anger or assume that every electoral disappointment produces mobilization.

Appraisal matters because the emotional significance of an event cannot be inferred from whether it is objectively favorable or unfavorable to a party. A supporter may view a concession as wise governance or as an unacceptable betrayal of identity. Group status, perceived fairness, anticipated efficacy, and trust in party leadership can shape the emotional response. Ward & Tavits (2019) found that positive and negative feelings toward parties were associated with perceptions of party ideological positions and of high-stakes political competition. The direction of causation in observational relationships remains a question: perceived differences may intensify emotional attachments just as attachments may exaggerate perceived differences.

Mobilization follows a further step. Emotional arousal can increase the intention to participate, but actual participation depends upon resources, safety, opportunities, social networks, and institutional conditions. Huddy et al. (2015) compares identity, ideology, and issue variables for campaign involvement, suggesting a substantial role for group-based emotional attachment. A complete design would examine whether identities predict observed turnout, volunteering, or donations after the effects of previous participation and socioeconomic constraints are included. It would also distinguish benign political activism from coercive action and evaluate whether threat produces participation, withdrawal, or avoidance under different contexts. Political identity research explains conditions of engagement; conduct and its democratic quality remain separately observable variables.

Identity Signaling in Social Interaction and Media

An expressed party affiliation is simultaneously a self-description and a social signal. It informs listeners about probable alliances, values, and expected responses, although these inferences can be stereotyped or inaccurate. The choice to communicate identity depends on the anticipated audience. In a homogeneous gathering, announcing a partisan label may reinforce belonging; in a diverse workplace, the same act may threaten relationships or professional expectations. Klar (2014) reported that social settings altered the influence of partisanship in preference formation, showing why privately measured dispositions cannot fully account for expressed judgments. Public political expression consequently requires situational analysis alongside surveys of durable attachment.

Digital environments provide new opportunities to display, infer, reinforce, or contest identity. A party symbol placed in a social profile differs from an answer given in a confidential survey because its viewers may impose social rewards and costs. Repeated interaction with like-minded networks can increase the salience of an affiliation, although identifying the independent effect of platform architecture requires interventions or detailed longitudinal designs rather than simple comparisons of users. Overton & Cunningham (2025) associates politically homogeneous everyday groups with stronger political identities and greater political engagement in survey and panel analyses, while reporting more disengagement among people in heterogeneous groups. These observational patterns suggest testable social-context pathways and should not be interpreted as experimental proof of inevitable radicalization.

Communication researchers must differentiate the volume of exposure to political content from the degree to which that content makes a party relevant to the self. An individual may consume news intensely without adopting an expressive partisan identity; another may receive little political media yet inherit a powerful family affiliation. The causal pathway between media, identity, and action can therefore be reciprocal. Measuring only time spent on media or declaring all exposure an algorithmic effect overlooks content, source credibility, network pressure, and political predisposition. The best integrated studies pair media trace data with validated self-identification measures, pre-exposure baselines, and behavioral outcomes. The Artificial Era adds human–artificial conversational environments to the research agenda, but their effects must be evaluated through direct tests rather than assumed from the general identity literature.

Group Boundaries, Outparty Perceptions, and Discrimination

Partisan identity can structure perceptions of political opponents through group boundaries. In the experimental and survey evidence examined by Iyengar & Westwood (2015), party labels affected evaluations and discriminatory decisions extending beyond explicitly electoral judgments. Westwood et al. (2018) reported party-based distrust and differential treatment across Great Britain, the United States, Belgium, and Spain, with the effect conditioned by ideological proximity and cross-cutting social identities. These results make it plausible that partisan attachments can operate as consequential social categories in modern democracies. The comparative findings do not imply identical magnitudes across every country or social group, nor do they establish that discrimination follows inevitably from all identity strength.

Identity strength, outparty dislike, ideological distance, and partisan discriminatory behavior are distinguishable outcomes. A person may feel strongly attached to a party without discriminating in a nonpolitical allocation task, and a person with little positive attachment may nevertheless distrust a rival. The experiments in Iyengar & Westwood (2015) demonstrate how randomized partisan information can support causal claims about specific decisions. Their validity depends upon the measure and context: hypothetical judgments, incentivized allocations, reported social distance, and consequential institutional discrimination are not equivalent. The independent dossier on affective polarization examines the downstream structure of animosity more extensively; here the analytic concern is how the meaning and salience of an affiliation contribute to those outcomes.

Research on mechanisms should examine whether discrimination is mediated by perceived values, competence, threat, norm expectations, or the simple availability of group categories. A party label may convey real political disagreement along with social membership, making the distinction challenging. Cross-cutting relationships provide a potential natural constraint on categorical judgments, as Westwood et al. (2018) suggests, while broader identity research Huddy (2001) underscores the role of history and context. Interventions that encourage shared identities or correct beliefs about rivals require direct measurement of whether they change identity salience, outparty affect, or behavior. A reduced feeling-thermometer gap alone establishes an attitudinal outcome, not durable changes in democratic institutions.

Partisan Attachment Beyond the United States

An explanatory model based on the Democratic–Republican competition must be tested, not simply transported, when parties operate under proportional representation, coalition politics, changing alliances, or weak organizational continuity. In a multiparty system, voters may identify with one party while supporting an alternative coalition partner or a local candidate from another organization. The meaning of party loyalty can vary with historical party age, institutional trust, programmatic differences, and the distribution of social cleavages. Bankert et al. (2017) studied the measurement of partisan identity in the Netherlands, Sweden, and the United Kingdom, finding that an eight-item measure offered more information about identity intensity than a conventional single item and showing comparable measurement properties across the studied national settings. This gives stronger evidence for cross-national applicability of the measured construct within those cases, without establishing universality.

Huddy et al. (2018) evaluated expressive and instrumental accounts in several European democracies, focusing on whether party loyalty takes the form of defended social identity or calculated political evaluation. The work shows why the same partisan attachment construct must be situated in local institutions rather than defined exclusively by US practice. A party's role as member of a coalition changes the relevant opponent and the significance of a party victory. Public party affiliation may also be linked to ethnic or regional organization, changing the risks of expression. Comparative research needs both an identity measure and a map of the choices available to the citizen. Otherwise cross-national differences in political behavior may be mistakenly attributed to psychology when institutions alter which actions are possible.

The cross-national experiments and surveys of Westwood et al. (2018) add a second dimension by measuring party-based distrust across four democracies and considering its relation to other social divisions. The finding that party attachments can be a strong category even where other cleavages are salient motivates further comparative inquiry. However, the four-country design cannot establish an invariant hierarchy of all group identities across democracies, autocracies, transitional regimes, or nonparty forms of politics. For broad generalization, researchers require translated and cognitively tested instruments, multiple samples, checks of metric and scalar invariance, and contextual data on party systems. Local political meaning should be measured rather than assumed from a shared English translation of 'party identity'.

Multiparty Coalitions, Movement Affiliation, and Weak Partisanship

Political identity can attach to a movement or coalition whose organizational structure differs from a conventional electoral party. The psychological questions concern shared symbols, collective expectations, identification with leaders, and the perceived capacity to pursue public goals. Membership in a movement may overlap with party identification without being reducible to it. Party attachment may also be weak, conditional, or intermittent: an individual may vote strategically, support policy campaigns without affiliation, or abstain while maintaining political commitments. The boundary of this dossier includes political-movement identity when it functions as an affiliation tied to political representation and collective contests over public authority; collective action mechanisms in general belong to their own established research literature.

In multiparty politics, identity can be layered. A supporter may identify with a specific party, a broader ideological family, and a temporary governing alliance. A ballot choice can express any combination of these orientations. The review by Huddy et al. (2018) contrasts expressive identity and instrumental evaluation across European democracies; Bankert et al. (2017) provides psychometric evidence for measuring party identity in systems with more than two choices. Their combined implication is that researchers must not operationalize partisan identity solely as an arbitrary binary category. The strength of attachment to a party can be evaluated alongside coalition acceptability, rejection of specific competitors, and willingness to switch in the presence of feasible alternatives.

Political movements present an additional challenge: boundaries, leadership, and goals can change quickly, and formal membership may be absent. Repeated field studies should follow the emergence of group labels, track when participants begin to self-identify, and distinguish organizational commitment from agreement with a movement's latest statements. A researcher should not infer high identity strength from attendance at one protest or from online use of a hashtag. Attendance can be motivated by a single issue, social networks, perceived efficacy, or immediate events. An identity-centered explanation must show that affiliation predicts behavior or emotional reactions beyond these alternative mechanisms. The methodological standard is construct discrimination supported by repeated observations and, when possible, experimentally manipulated identity salience.

Party Identity, Political Participation, and Electoral Conduct

Party identification is related to turnout, campaign work, persuasion, volunteering, public expression, and financial support, but each behavior has its own opportunities and costs. Huddy et al. (2015) found that expressive partisan identity was a strong predictor of campaign involvement relative to measured issue positions and ideological identification in the studies examined. The emotional-reaction experiments strengthen an interpretation in which concern for the group's electoral standing can mobilize members. Yet a predictor's relative strength depends upon the reliability of the instruments and the participation measures available. A complete explanation of electoral conduct also requires the legal context, electoral competition, recruitment, efficacy, available time, and the social ties through which participation is coordinated.

One central empirical difficulty is reciprocal causation. Taking part in a campaign can strengthen a person's identity by exposing them to copartisans, shared rituals, and repeated declarations of commitment. Stronger identification can then make further participation attractive. Cross-sectional associations do not establish the order of these processes. A panel with repeated measures of involvement and identity could estimate their temporal sequence but must still address omitted events and selective attrition. An experimental mobilization message can isolate the immediate effect of identity salience, whereas long-term organizational participation is harder to assign randomly. The inference should match the intervention: increased stated willingness to volunteer does not automatically imply more verified volunteering.

Abramowitz & Webster (2016) relates growing negative partisan sentiment to strong partisan loyalty and increasingly aligned election results across offices. This observation suggests that durable hostility toward an alternative may reinforce consistent support even when enthusiasm for the chosen party is limited. The causal chain is harder to identify than the descriptive trend because institutional nationalization, elite behavior, media circulation, and voter sorting may change together. Comparative tests can examine whether the same negative identity predicts stable electoral support in systems with viable third parties and coalitions. Research can also ask whether strong attachment increases factual political knowledge, civic efficacy, or sustained attention under circumstances that permit deliberation. These outcomes should be measured individually rather than inferred from the valence of partisan emotion.

Civic Tolerance, Democratic Norms, and Political Conflict

Partisan attachment can motivate political engagement while also making competition emotionally consequential. Whether strong identity is compatible with tolerance, procedural fairness, compromise, and acceptance of electoral defeat is an empirical question with distinct outcomes. The research reviewed in Finkel et al. (2020) describes political sectarianism as an intensification of opposing-group differentiation and aversion in the United States, with potential democratic implications. However, social hostility and support for specific undemocratic acts are different constructs. A respondent can express anger toward opponents while supporting lawful transfers of power, or can endorse restrictions for reasons not exhausted by interpersonal dislike.

The large intervention study Voelkel et al. (2024) directly tested this separation. It assigned 32,059 US participants across 25 treatments intended to reduce partisan animosity or antidemocratic attitudes. Many treatments reduced animosity, while fewer reduced support for undemocratic practices or partisan violence. The reported structure of treatment effects suggests that improvements in interparty feelings do not automatically establish improvements in democratic conduct. This is a crucial limitation on theories that posit a single identity-based hostility factor behind all political harms. The study's sample, online intervention setting, and measurement periods also delimit the strength of its conclusions about durable public behavior.

Research needs to model not only allegiance but also norms governing what a loyal partisan may do. Strong identification with a party that publicly endorses fair procedures could plausibly support principled adherence to those procedures, whereas a threatening political context could activate different group expectations. This is a hypothesis requiring studies of norm perception, leadership cues, real institutional decisions, and behavior over time. Experimental manipulation of perceived opponent intentions may identify short-term changes in attitudes; it does not substitute for institutional evidence about censorship, electoral obstruction, violence, or democratic resilience. Voelkel et al. (2024) provides an example of measuring animosity, undemocratic practices, and partisan violence separately, which should remain the norm in future partisan-identity research.

Changing Political Environments and Emerging Research Questions

Party identity is embedded in communicative and institutional environments that change over time. Political affiliation may be expressed through local associations, broadcast news, online networks, private messaging, or conversations with artificial systems. These environments can amplify, conceal, or reframe partisan cues. The identity-based framework of Van Bavel & Pereira (2018) suggests that politically relevant categories can influence cognition and interpretation of information. Its application to newly designed communication settings requires new experiments because the causal features of an artificial interlocutor or recommendation system cannot be inferred from an older party-cue vignette. Technology may change how affiliation is signaled and perceived without proving a transformation in the psychological mechanism itself.

Two empirical developments help define the emerging agenda. Overton & Cunningham (2025) associates everyday group political homogeneity with stronger identities and more role-congruent political behavior, raising questions about peer environments and disengagement in heterogeneous settings. Voelkel et al. (2024) shows that interventions can shift several partisan attitudes and that different democratic outcomes require distinct experimental endpoints. Both strands motivate designs combining repeated naturalistic behavior, controlled social-interaction manipulations, and long-term follow-up. The goal is to identify whether changes in the expression of political identity persist after an intervention and whether changes in emotional attachment mediate actual conduct.

Human–artificial political communication creates a further research problem for the Artificial Era. A conversational agent may present political information, simulate party-affiliated viewpoints, or adapt to a user's previous statements. The relevant psychological outcomes include identity salience, perceived source affiliation, willingness to revise beliefs, and the perceived authenticity of political belonging. These are questions about human responses to artificial interlocutors; additional research would be required to investigate the independent organization or agency of artificial systems. The conceptual separation protects against importing untested assumptions about Artificial Sapiens into the empirical analysis of human partisanship. Any future claims of persuasion, attachment reinforcement, or durable political influence require direct randomized or longitudinal evidence in the actual deployed environment.

Measurement: Party Labels, Identity Scales, and Latent Structure

Conventional party-identification measures ask respondents to name a preferred party and indicate the strength of that affiliation. Their practical value lies in simplicity and historical continuity, especially in US election surveys descended from the Michigan tradition Campbell et al. (1960). Their limitation is construct ambiguity. A single intensity response can mix emotional involvement, certainty of vote intention, policy coherence, habit, and social identification. Bankert et al. (2017) tested an eight-item partisan identity measure in surveys in the Netherlands, Sweden, and the United Kingdom and used latent-variable modeling to assess its psychometric properties and predictive information. The authors reported that this scale offered greater information about partisan intensity than a conventional single-item indicator, supporting the measurement of attachment through several indicators rather than one label alone.

An identity battery should sample more than stated party preference. Depending upon the research question, indicators can assess the importance of group membership to self-description, feeling connected to other supporters, emotional responses to criticism, perceived shared fate, and willingness to represent the group. Construct validity requires distinguishing these indicators from political knowledge, ideological consistency, affect toward opponents, and current electoral action. A measure built only from campaign participation risks making the relationship between identity and participation tautological. Huddy et al. (2015) explicitly compares a multi-item identity measure with ideological and issue-based alternatives, exemplifying an approach that tests discriminant and predictive validity rather than assuming them.

Cross-national and longitudinal measurement pose additional difficulties. A party label can refer to a stable organization, a fluid electoral alliance, or a historical movement, and its social meaning may change between survey waves. Factor structure, response thresholds, and item interpretation require evaluation across countries, languages, ages, and political contexts. A scale that predicts activity within one group may not have equivalent predictive validity in a party system with little opportunity for participation. Bankert et al. (2017) provides evidence for cross-context measurement in three examined countries; generalization beyond them requires new invariance tests, cognitive interviewing, and independent outcome validation. Scientific reporting should supply scale wording, coding, missing-data handling, reliability estimates, fit indices when relevant, and the specific population in which these properties were evaluated.

Experimental, Panel, Survey, and Behavioral Research Designs

The methods of partisan-identity research answer different questions. Repeated representative surveys estimate the distribution of identification within a population and trace changes in observed affiliation. Panel studies can examine whether earlier identities predict later judgments or switching after baseline attitudes are controlled. Laboratory and survey experiments can randomize leader endorsements, policy descriptions, identity salience, or apparent party outcomes to estimate short-run causal effects. Nicholson (2012) and Bullock (2011) illustrate experimental identification through manipulated elite endorsements and policy information, while Jennings et al. (2009) illustrates the value of long-term multigenerational observation. Each design supplies evidence at a particular level and time horizon.

Experimental internal validity depends on successful randomization, clear treatment contrasts, valid measures, and appropriately handled exclusions. A party-label manipulation can show that the assigned label changes a judgment, but it cannot explain how an individual acquired their allegiance across decades. A panel association can establish temporal ordering and rule out some alternatives, yet identification and political events are not randomly assigned. Experiments using hypothetical vote choices may be weak proxies for actual electoral action. Behavioral measures such as recorded volunteering or incentives allocated to party-labeled recipients increase ecological relevance but may introduce selection into observed participation. Combining methods allows claims about mechanism to be tested against real outcomes without disguising a design's limitations.

Political identity is socially embedded, so interference between experimental participants is an additional consideration in network and field studies. An intervention that changes what one person communicates may influence untreated friends, violating the assumptions of simple individual randomization. Treatments may affect party identification, issue perception, and emotion simultaneously. Preregistered outcomes and transparent analysis plans reduce the opportunity to reinterpret one significant result among many tests. The intervention megastudy Voelkel et al. (2024) illustrates both the strengths of preregistered comparative testing and the importance of distinct outcomes: changes in partisan animosity were more frequent than changes in attitudes toward undemocratic practices or violence. Its conclusions apply to the interventions, sample, and measures actually tested.

Evidence Appraisal, Replication, and Inferential Boundaries

The evidence base includes historical monographs, theory papers, representative and quota-based surveys, panel analyses, controlled survey experiments, and comparative studies. These sources differ in the strength and type of claim they can support. Huddy et al. (2015) reports multiple studies that combine partisan identity measures with experimental emotional responses. Bullock (2011) demonstrates a competing explanation under conditions of supplied policy information, while Bakker et al. (2020) reports mixed support for expressive utility and little support for a low-resource cue-following hypothesis. A critical synthesis should preserve these divergent results instead of averaging them into a claim that party identity always dominates knowledge or policy judgments.

Replicability requires that the target outcome, design, and political context be held in view. A party endorsement that polarizes opponents in one election Nicholson (2012) may behave differently when issued by another political figure or when an issue is more familiar to participants. A result concerning emotion after a simulated electoral loss need not reproduce as an effect on observed volunteering six months later. Formal statistical replication, cross-sample confirmation, and conceptual replication test different aspects of robustness. Reviews should also account for nonrepresentative online recruitment, attrition in panels, historical changes in party systems, researcher degrees of freedom, publication selection, and measurement noninvariance. Absence of a significant experimental effect is not automatically evidence of a universal zero effect.

Several limitations are structural. Party attachment, social networks, elite strategies, and policy preferences influence one another, making single-direction causal models difficult to identify observationally. The same party label may signal different ideas across periods, undermining naive pooling of long time series. Cross-country samples may differ in available parties and democratic institutions, while conventional identity scales can privilege citizens who have formal partisan choices. The research synthesis therefore ranks evidence by identifiable design: randomized cue effects support local causal conclusions; longitudinal associations support temporally ordered relationships subject to confounding; national descriptive trends characterize change; and theoretical frameworks specify mechanisms requiring independent tests. Specific numerical effects should be reported only from inspected original results.

Integrating Instrumental, Expressive, and Contextual Explanations

The cumulative evidence supports a multilevel account in which partisan affiliation can be both a social attachment and an orientation toward political organizations capable of achieving goals. Green et al. (2002) emphasizes social identity and continuity, Huddy et al. (2015) demonstrates the emotional and participation-related relevance of expressive attachment, Bullock (2011) shows the substantial independent influence of policy information in experiments, and Levendusky (2009) identifies the reorganization of party–ideology associations over time. These are complementary empirical domains, not interchangeable proofs of one universal causal model. The key explanatory variables include identity strength and content, policy evaluations, available cues, institutional choices, social surroundings, and prior histories.

An integrated causal research design would first measure stable affiliation and issue commitments, then experimentally manipulate party cues and substantive information in separate conditions, track emotional responses, and observe consequential behavior when feasible. A longitudinal extension would examine whether repeated interactions alter the attachment itself. Contextual moderators should include political knowledge, social-network composition, coalition structure, perceived competition, and the alignment of other social identities. Klar (2014) illustrates the importance of immediate social setting, Westwood et al. (2018) the importance of cross-national and cross-cutting affiliations, and Bakker et al. (2020) the necessity of testing alternative cognitive-resource hypotheses. Convergence across these designs could establish the circumstances in which instrumental and expressive motives interact.

This synthesis also separates proximate activation from developmental formation. An experimental party cue may immediately shift a response without creating a partisan identity. Years of affiliation may organize expectations without becoming salient during a specific policy judgment. A political event can make a durable identity newly important Sears & Valentino (1997), while a sustained elite realignment can change the structure of available affiliations Levendusky (2009). Treating these temporal layers distinctly reduces causal overclaiming. The psychological organization of partisanship is most accurately studied as historically acquired affiliation with variable situational accessibility and behaviorally consequential meaning, constrained and enabled by institutions and collective interaction.

Unresolved Problems and a Program for Further Research

First, researchers need to establish how positive identification and negative rejection jointly affect civic conduct. Abramowitz & Webster (2016) associates negative party sentiment with electoral loyalty, but behavioral experiments and repeated panel measures can clarify whether opposition motivates action independently of positive group attachment. Second, the origins of party switching need better identification across life transitions, social mobility, changes in party programs, and institutional shocks. Third, the meaning of partisanship must be measured across diverse populations and political systems. The measurement gains reported in Bankert et al. (2017) should be extended to systems with volatile party brands, hybrid regimes, coalition politics, and weak party membership traditions.

Fourth, behavioral outcomes require direct observation. Surveyed antipathy, intention to vote, actual participation, discriminatory action, support for undemocratic practices, and political violence are distinct endpoints. Voelkel et al. (2024) provides experimental evidence that reducing negative partisan feelings does not automatically produce equally strong changes in all antidemocratic outcomes. Follow-up studies should test duration, treatment heterogeneity, replication in different elections, and effects on real conduct. Fifth, the intersection of political identity with cross-cutting social relationships needs careful causal testing. Westwood et al. (2018) suggests an important moderating role for conflicting ties, while Overton & Cunningham (2025) examines associations of local political homogeneity with stronger identities and activity; interventions and long-term observations can help determine when social relations attenuate or amplify affiliation.

Sixth, emerging political communication channels create new questions about identity activation and perceived source allegiance. Researchers can randomize the source, declared political viewpoint, and informational content of a human or artificial conversational message, then measure accuracy, identity salience, and consequential choice separately. Durable effects require longitudinal measurement and independent replication. Philosophical categories developed in the Aisentica corpus may guide questions about the relations between Homo and Artificial, but they cannot substitute for experiments about human political behavior. The research agenda is therefore defined by explicit constructs, identifiable outcomes, contextual moderators, and evidence capable of distinguishing institutional change from psychological change.

Conclusion

Political identity and partisan attachment are stable yet revisable relationships between persons and organized political categories. They acquire meaning through family histories, political events, social environments, institutional structures, and the symbolic representation of parties. A partisan identity can organize self-definition, emotional reactions, perceptions of allies and adversaries, and choices to participate in politics. Its effects depend on strength, content, accessibility, policy stakes, and the alternatives offered by political institutions. The Michigan party-identification tradition Campbell et al. (1960), the social-identity reconstruction Green et al. (2002), experimental research on expressive partisanship Huddy et al. (2015), and cross-national psychometric work Bankert et al. (2017) establish complementary foundations for understanding this complex phenomenon.

The strongest scientific synthesis preserves differences among observed outcomes and among research methods. Controlled experiments show conditional effects of partisan cues, the emotional consequences of party-status information, and the capacity of substantive policy descriptions to matter Nicholson (2012) Bullock (2011) Huddy et al. (2015). Longitudinal and historical analyses illuminate acquisition and gradual realignment Jennings et al. (2009) Levendusky (2009). Comparative evidence tests whether identity extends beyond the US two-party configuration Westwood et al. (2018). Research on polarization and intervention Mason (2015) Voelkel et al. (2024) further establishes that emotional attachment, dislike of opponents, issue polarization, and acceptance of undemocratic practices must be distinguished. The design of the evidence determines how far the interpretation can travel.

The resulting account identifies a clear scientific object: politically organized belonging with measurable developmental histories, identity investment, cognitive and emotional mechanisms, and consequences that are conditional on social and institutional environments. Its outstanding problems concern causal origins, identity change, comparative measurement, behavioral validity, and technological transformation. Political identity remains an indispensable psychological variable because the meaning of party affiliation often exceeds a single electoral decision. A rigorous research program must continue to connect affiliation to actual behavior while preserving the independence of evidence, theory, and the specific historical context in which citizens acquire and express their commitments.

References

  • Abramowitz, A. I., & Webster, S. (2016). The rise of negative partisanship and the nationalization of U.S. elections in the 21st century. Electoral Studies, 41, 12–22. https://doi.org/10.1016/j.electstud.2015.11.001
  • Bakker, B. N., Lelkes, Y., & Malka, A. (2020). Understanding partisan cue receptivity: Tests of predictions from the bounded rationality and expressive utility perspectives. The Journal of Politics, 82(3), 1061–1077. https://doi.org/10.1086/707616
  • Bankert, A., Huddy, L., & Rosema, M. (2017). Measuring partisanship as a social identity in multi-party systems. Political Behavior, 39, 103–132. https://doi.org/10.1007/s11109-016-9349-5
  • Bullock, J. G. (2011). Elite influence on public opinion in an informed electorate. American Political Science Review, 105(3), 496–515. https://doi.org/10.1017/S0003055411000165
  • Campbell, A., Converse, P. E., Miller, W. E., & Stokes, D. E. (1960). The American voter. Wiley. https://isr.umich.edu/timeline_year/1960/
  • Finkel, E. J., Bail, C. A., Cikara, M., Ditto, P. H., Iyengar, S., Klar, S., Mason, L., McGrath, M. C., Nyhan, B., Rand, D. G., Skitka, L. J., Tucker, J. A., Van Bavel, J. J., Wang, C. S., & Druckman, J. N. (2020). Political sectarianism in America. Science, 370(6516), 533–536. https://doi.org/10.1126/science.abe1715
  • Green, D. P., Palmquist, B., & Schickler, E. (2002). Partisan hearts and minds: Political parties and the social identities of voters. Yale University Press. https://www.jstor.org/stable/j.ctt1npp6m
  • Huddy, L. (2001). From social to political identity: A critical examination of social identity theory. Political Psychology, 22(1), 127–156. https://doi.org/10.1111/0162-895X.00230
  • Huddy, L., Mason, L., & Aarøe, L. (2015). Expressive partisanship: Campaign involvement, political emotion, and partisan identity. American Political Science Review, 109(1), 1–17. https://doi.org/10.1017/S0003055414000604
  • Huddy, L., & Bankert, A. (2017). Political partisanship as a social identity. Oxford Research Encyclopedia of Politics. https://doi.org/10.1093/acrefore/9780190228637.013.250
  • Huddy, L., Bankert, A., & Davies, C. (2018). Expressive versus instrumental partisanship in multiparty European systems. Political Psychology, 39(S1), 173–199. https://doi.org/10.1111/pops.12482
  • Iyengar, S., & Westwood, S. J. (2015). Fear and loathing across party lines: New evidence on group polarization. American Journal of Political Science, 59(3), 690–707. https://doi.org/10.1111/ajps.12152
  • Jennings, M. K., Stoker, L., & Bowers, J. (2009). Politics across generations: Family transmission reexamined. The Journal of Politics, 71(3), 782–799. https://doi.org/10.1017/S0022381609090719
  • Klar, S. (2014). Partisanship in a social setting. American Journal of Political Science, 58(3), 687–704. https://doi.org/10.1111/ajps.12087
  • Levendusky, M. (2009). The partisan sort: How liberals became Democrats and conservatives became Republicans. University of Chicago Press. https://doi.org/10.7208/chicago/9780226473673.001.0001
  • Mason, L. (2015). “I disrespectfully agree”: The differential effects of partisan sorting on social and issue polarization. American Journal of Political Science, 59(1), 128–145. https://doi.org/10.1111/ajps.12089
  • Nicholson, S. P. (2012). Polarizing cues. American Journal of Political Science, 56(1), 52–66. https://doi.org/10.1111/j.1540-5907.2011.00541.x
  • Overton, J., & Cunningham, G. (2025). Groups activate identities; identities activate behavior: How political homogeneity breeds extremism and apathy in American politics. Social Science Research, 132, Article 103234. https://doi.org/10.1016/j.ssresearch.2025.103234
  • Sears, D. O., & Valentino, N. A. (1997). Politics matters: Political events as catalysts for preadult socialization. American Political Science Review, 91(1), 45–65. https://doi.org/10.2307/2952258
  • Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. (1971). Social categorization and intergroup behaviour. European Journal of Social Psychology, 1(2), 149–178. https://doi.org/10.1002/ejsp.2420010202
  • Van Bavel, J. J., & Pereira, A. (2018). The partisan brain: An identity-based model of political belief. Trends in Cognitive Sciences, 22(3), 213–224. https://doi.org/10.1016/j.tics.2018.01.004
  • Voelkel, J. G., Stagnaro, M. N., Chu, J. Y., Pink, S. L., Mernyk, J. S., Redekopp, C., Ghezae, I., Cashman, M., Adjodah, D., Allen, L. G., Allis, L. V., Baleria, G., Ballantyne, N., Van Bavel, J. J., Blunden, H., Braley, A., Bryan, C. J., Celniker, J. B., Cikara, M., ... Willer, R. (2024). Megastudy testing 25 treatments to reduce antidemocratic attitudes and partisan animosity. Science, 386(6719), Article eadh4764. https://doi.org/10.1126/science.adh4764
  • Ward, D. G., & Tavits, M. (2019). How partisan affect shapes citizens’ perception of the political world. Electoral Studies, 60, Article 102045. https://doi.org/10.1016/j.electstud.2019.04.009
  • Westwood, S. J., Iyengar, S., Walgrave, S., Leonisio, R., Miller, L., & Strijbis, O. (2018). The tie that divides: Cross-national evidence of the primacy of partyism. European Journal of Political Research, 57(2), 333–354. https://doi.org/10.1111/1475-6765.12228