Everybody Hates Partisans — Comprehensive Notes
4. Everybody Hates Partisans
Big-picture aim: Explore how the motivations that push people to identify as independents might have consequences beyond mere partisan labels, including how those motivations affect judgments of others and the desirability (or undesirability) of partisanship.
Framing: Move beyond Chapter 3 findings to ask how dislike of partisans shapes everyday judgments and social behavior, including housing and neighborly interactions.
Core takeaway (overall chapter): There is little to no positive impression of partisanship; instead, reminders of partisan disagreement amplify negative visualizations of partisanship, making independents seem more socially desirable by contrast.
The social logic: Independence is seen as socially beneficial; partisanship is perceived as negatively affecting impression management and social interaction.
Real-world implications: Partisanship can shape who people want to work with, live near, socialize with, and even how attractive they find partisan faces; reminders of disagreement can push people away from engaging with partisans and toward viewing independents more favorably.
Meta-theme: The undesirability of partisanship acts as a mechanism that explains why people may prefer independence as a social signal, with broad consequences for political participation and civic life.
4.1 Assessing Perceptions of Partisans
Key question: How do people feel about partisans, regardless of which party they support?
Measurement approaches used in this chapter:
Affective measures of inter-party attitudes:
Thermometer method: rate Democrats and Republicans on a 0–100 scale (the higher the warmer score, the more favorable the attitude).
Open-ended/descriptive measures: list everything you like and dislike about both parties (Konda & Sigelman 1987).
Affective questions about close relationships: e.g., how would you feel if someone close to you married a person from the opposing party (Iyengar, Sood, & Lelkes 2012).
Conceptual scope: Address not only feelings toward the opposing party but also feelings toward strong partisans within one’s own party.
Study design (context manipulation): Participants were randomly assigned to read one of three news clip types:
Partisan unity information
Partisan disagreement information (elite-level)
Groundhog Day (nonpolitical control)
Objective: See how different news contexts affect willingness to work with a partisan coworker who shares the participant’s presidential vote.
Sample: N = 156 (demographics in Table A1.4).
Core finding: Exposure to partisan disagreement (versus unity or control) increased negative affect toward partisans and reduced willingness to work with a partisan colleague, illustrating how context shapes impressions of partisanship.
4.2 Imagine Me and You as Partisans
Central idea: How people visualize the concept of partisanship affects their evaluation of partisans and partisanship as a social phenomenon.
Background on visualization:
Visual imagery can influence political judgments; vivid images affect choice and evaluation (Petersen & Aarøe 2013).
People store knowledge both verbally and visually; visualization may reflect feelings about a concept.
Measurement challenge: Capturing images people visualize when hearing the term "partisanship" without relying solely on verbal descriptions.
Innovative method (Study 4.2):
192 Internet-savvy adults asked to find online images that best fit their perception of partisanship, and then provide the image URLs.
Two independent coders, blind to purpose, coded each image for content: anger, fighting, negativity, etc. (Table 4.1 provides sample codes).
Experimental manipulation (same treatment structure as Chapter 3, with a twist): four news clipping conditions
Partisan unity
Elite partisan disagreement
Mass partisan disagreement
Groundhog Day (control)
Rationale for study design: Move beyond elites to capture general perceptions of partisanship; examine how information about disagreement affects visualizations of partisanship in everyday life.
Self-monitoring note: Absent from this study; focus is on perceptions, not self-description or behavior regulation.
Data handling: After image collection, researchers retrieved each image, coded by two independent coders for attributes such as anger, divisiveness, fighting, and overall negativity.
Implication: Visualizations of partisanship shift toward hostility and negativity when readers are reminded of partisan disagreement, suggesting an automatic, visceral response to disagreement cues.
4.2.1 Capturing an Image (Coding Details)
Process: Images sourced online; coders rated attributes including anger, fighting, negativity, and divisiveness.
Coding table: Table 4.1 provides example image typings (Angry, Fighting, Negative, Cheerful).
Theoretical link: Visualizations are expected to correlate with attitudes toward partisanship and its perceived social costs.
Key note: The study underscores the challenge of measuring visual representations and demonstrates a scalable method for capturing broad visual impressions via web-sourced images.
4.2.2 Visualizing the Party (Results)
Analysis: Compare control vs each treatment group on image attributes.
Core finding: Reminders of partisan disagreement significantly shift images toward negativity:
Anger: more likely with elite and mass disagreement vs unity/control (p-values vary by comparison; some p < 0.10; see Fig. 4.1a).
Divisiveness: higher in mass disagreement vs unity/control; differences noted (p < 0.10).
Fighting: greater prevalence in mass disagreement vs neutral/unity (p < 0.05 or better; exact p-values given in Figure 4.1c).
Overall negativity: higher in both disagreement conditions vs control/unity (p < 0.10 to p < 0.05 depending on comparison, see Fig. 4.1d).
Base-rate consideration: Researchers tracked base rates by analyzing the prevalence of certain image types on the web (e.g., fighting images) using the URLs participants provided to assess whether treatment effects exceeded baseline web availability.
Overall trend: Across the sample, negative imagery dominated when participants were reminded of disagreement, with unity conditions showing comparatively more positive imagery.
Interpretation: Visualizations of partisanship reflect broader attitudes toward disagreement; negative imagery is more likely when disagreement is salient.
4.3 These Are the People in My Neighborhood
Big question: Do negative views of partisanship spill over into where people want to live and with whom they want to associate socially?
Theoretical link: The Big Sort argument (Bishop 2008) about residential clustering, contrasted with Nall & Mummolo (2014) arguing practical constraints limit sorting by politics.
Study design (Study 4.3): Nationally representative sample of 513 Americans. Procedure:
Read one of two news stories: partisan unity or partisan disagreement.
After reading, view two photographs of neighborhoods (Neighborhood A and Neighborhood B) that are otherwise similar in aesthetic and social attributes.
Rate two neighborhoods on quality (1–10 scale), decide how willing to live in each neighborhood, and indicate whether they would attend a social event with residents of each neighborhood.
Neighborhood manipulation: Two photos were baseline (no signs) and second neighborhood with modifications:
Sign-free baseline: No yard signs.
Political signs: Yard signs that are political but do not indicate party.
Nonpolitical signs: Yard signs for a graduation party (neutral).
Six experimental groups arise from combining article type (Unity vs Disagreement) with sign type (No signs, Political signs, Nonpolitical signs) across a 2x3 design (Table 4.2).
Pre-test: Independent group (n=90) rated baseline neighborhoods for desirability and quality to establish comparability before photo manipulation.
Key result: Presence of political signs lowers the second neighborhood’s attractiveness relative to the baseline neighborhood, and this effect is amplified when participants read about partisan disagreement. Primary metric: difference score = Rating(Second) − Rating(Baseline).
Specific findings (Table 4.3 and narrative):
Without signs, article content has little effect on neighborhood quality differences.
With political signs, the perceived quality gap widens in the disagreement condition; the second neighborhood is judged significantly less desirable when political signs accompany partisan disagreement (p < 0.05 for the relevant comparisons).
Willingness to live in the second neighborhood shows similar trends: political signs reduce desirability, especially after exposure to partisan disagreement. In Unity conditions, effects are smaller or non-significant; in Disagreement conditions, political signs produce a notable drop in “very much” wanting to live there (roughly a 15 percentage-point decrease).
Social-event preference: participants’ willingness to attend a social event with residents of the second neighborhood declines when political signs are present, particularly after disagreement exposure (e.g., Unity vs Disagreement comparisons; p < 0.05 or stronger in several contrasts).
Interpretation: Negative views of partisanship spill into everyday life domains (neighborhood desirability and social interaction), with political cues (signs) intensifying these effects when disagreement is salient.
Nuanced findings: On the Unity condition, participants are more flexible about associating with the second neighborhood; in the Disagreement condition, political signs create a strong anti-partisan social signal that reduces social and residential interest.
4.3.1 Comparing Neighborhoods (Table 4.3) – Key Takeaways
Formation of difference score highlights that political information interacts with signs to shape evaluations.
No-sign vs political-sign contrasts show the clearest effects in the Partisan Disagreement condition.
Sign type alone can alter perceptions, but its impact is magnified when participants receive negative information about partisan disagreement.
4.4 Hey, Good-Looking
Question: Do social norms about beauty and attractiveness extend to judgments about partisans and independents?
Rationale: Prior chapter experiments suggested that negative views of partisans might color social judgments; this study introduces a human face element to assess whether people assign partisan labels to faces and whether disagreement changes these assignments.
Study design (Study 4.4):
Faces generated by computer algorithm (FaceGen 3.1; Todorov & Oosterhof typology) with validated ratings on attractiveness, competence, trustworthiness, and likability.
163 participants assigned to two groups: control (mere mention of partisanship) vs partisan disagreement.
Faces labeled as Democrat, Republican, or Independent.
Participants rate and rank faces on the four attributes (1–9 scale; higher = more favorable).
Baseline: Faces are median on all dimensions; the partisan labels are the only manipulated variable.
Goal: Determine whether partisan disagreement shifts the perceived social value of partisan vs independent faces.
4.4.2 Who Is Attractive and Likable?
Partisan faces: After disagreement information, a larger share of participants rate partisan faces as unattractive, unlikable, incompetent, and untrustworthy (relative to control). Specific changes include:
Attractiveness: 10% in control vs 22% unattractive after disagreement.
Likability: similar increases; trustworthiness and competence also show declines in partisan faces after disagreement, with significance on attractiveness, likability, and trustworthiness (p < 0.05); competence changes reach p < 0.1.
Independent faces: After disagreement, independents are judged more positively across attractiveness, likability, and competence; trustworthiness shows a smaller, non-significant change in many cases.
Partisan independence interaction: Independents become more positively rated after seeing partisan disagreement, even among participants who themselves identify as partisans.
Robustness check: Results persist even when analyzing only participants who self-identify as partisans; a meaningful minority still rates independents more positively after disagreement.
Overall interpretation: Partisan labels combined with disagreement cues taint partisan faces (making them seem less attractive, less likable, less competent, less trustworthy) and boost perceived value of independents’ faces.
Practical upshot: Even with neutral, computer-generated faces, social judgments about who is “partisan” vs “independent” shift dramatically once disagreement is part of the narrative.
4.4.1 This Is What a Partisan Looks Like (Methodological point)
Rationale for using FaceGen faces: Control for visual features while manipulating social categorization (Democrat/Republican/Independent).
Faces are designed to have median levels on key social attributes to isolate the effect of partisan labeling and disagreement information.
This design tests whether social judgments about political affiliation extend beyond verbal descriptions to facial perception.
4.4.2 Implications of Facial Judgments
Partisan disagreement reduces perceived attractiveness, likability, competence, and trustworthiness for partisan faces, even when base faces are equivalent on these dimensions.
Independents gain in perceived social value after disagreement exposure across multiple attributes.
Within-partisan groups, partisans still show some bias against their own party when disagreement is salient, but independents receive the strongest positive shifts.
Takeaway: Social judgments about political affiliation extend to facial perception; disagreement cues can make independents look more favorable and partisans less favorable, on average.
4.5 Dirty Rotten Scoundrels
Theoretical framing: Links to Hibbing & Theiss-M Morse’s (2002) stealth democracy; distrust of government by many Americans who favor independent expertise over party-led governance.
Core argument: The social desirability of independence is reinforced by a simultaneous undesirability of partisanship; people view parties as sources of negativity, conflict, and social cost.
Key connections across the chapter:
The collective pattern shows that partisanship is broadly off-putting across social domains: work relationships, neighborhoods, social events, and even perceptions of attractiveness and trustworthiness.
Disagreement cues amplify these effects, suggesting that the social costs of partisan conflict extend beyond policy outcomes to everyday social interactions.
Philosophical and ethical implications:
If partisanship erodes social cohesion, there may be ethical concerns about political polarization and its impact on democratic participation.
The findings raise questions about how political systems and institutions might mitigate social costs of partisanship while preserving healthy political deliberation.
Practical implications:
Campaigns and political communication should be aware of the social costs associated with partisan labeling and public disagreement cues.
Non-partisan or cross-partisan messaging could reduce social frictions and facilitate more constructive civic engagement.
The visual and social perception effects imply that reducing visible partisan signaling (e.g., yard signs) could improve neighborly relations and participation in shared communities.
Summary stance: Identifying with any party is a social liability, and independent identity is socially desirable due to the perceived negativity of partisanship itself.
4.6 Synthesis and Implications
Across studies, independence is framed as a socially favorable identity, whereas partisanship is associated with negative social signals, anger, and confrontation.
The chain of effects runs from cognitive/affective reactions to behavioral outcomes in social life (work, neighborhoods, dating/romance-adjacent judgments, and even facial perceptions).
The recurring theme is that reminders of partisan disagreement heighten negativity toward partisans and simultaneously elevate the perceived value of independents, shaping both attitudes and potential actions.
Broader implications for political participation:
If people associate partisanship with social costs, engagement (beyond voting) may wane, or participants may seek out independent social networks and activities to avoid partisan conflict.
The findings suggest a potential mechanism by which political moderation or nonpartisanship could increase cross-cutting social ties and participation in democratic processes.
Connections to Prior Lectures and Foundational Principles
Ties to Chapter 3 findings on the social benefits of being independent and the costs of displaying partisan identity.
Aligns with classic views of party politics as a sorting mechanism (Aldrich 1995; Schattschneider 1942) and the role of parties in political participation (Verba, Schlozman, & Brady; Holbrook & Krosnick 2010).
Builds on affective polarization literature (Iyengar, Westwood, & pop; Iyengar et al. 2012) by extending affective measures to partisans’ self-representation and cross-cutting social cues.
Integrates visualization theory (Valentino et al. 2002; Petersen & Aarøe 2013) to demonstrate how mental imagery interacts with political judgments.
Uses The Big Sort debate as a backdrop (Bishop 2008) and contrasts it with empirical findings showing social distance can be driven by perceived negativity of partisans, not only by geographic sorting.
Methodological Notes and Key References
Sample sizes and design details:
Study 4.1: N = 156; three news clipping conditions (unity, disagreement, Groundhog Day control).
Study 4.2: N = 192; image-search task with four treatments (unity, elite disagreement, mass disagreement, control).
Study 4.3: N = 513; two news stories (unity vs disagreement); two neighborhoods manipulated with sign types (six groups total).
Study 4.4: N = 163; FaceGen faces rated for attractiveness, likability, competence, trustworthiness; two groups (control vs partisan disagreement).
Key formulas and data representations:
Difference score for neighborhood study:
Positive values indicate the second neighborhood is preferred; negative values indicate the baseline is preferred.
Thermometer scales and Likert-type scales used across studies (0–100 for warmth toward parties; 1–9 for facial trait ratings).
Important citations and concepts referenced:
Parties and democracy: Aldrich (1995); Schattschneider (1942).
Political participation and activity: Verba, Nie, & Kim; Holbrook & Krosnick (2010).
Affective polarization and inter-party attitudes: Iyengar, Sood, & Lelkes (2012).
Visualization and decision-making: Valentino et al. (2002); Petersen & Aarøe (2013).
Visualization measurement and environment (prior work): Paivio (1971); Prior (2014).
Visual perception of faces and social judgments: Oosterhof & Todorov (2009).
Big Sort debate and geographic clustering: Bishop (2008); Nall & Mummolo (2014).
Stealth democracy and distrust of parties: Hibbing & Theiss-Morse (2002).
Observed patterns to remember:
Partisan disagreement consistently increases negative visualizations of partisans in studies using image and facial judgments, and reduces affinity for politically signaled neighborhoods.
Independents tend to gain in perceived attractiveness, likability, competence, and trustworthiness after exposure to partisan disagreement.
Even partisans themselves show more negative judgments toward partisans when disagreement is salient, but may increasingly view independents more positively.
Practical and ethical implications:
Social norms and political signaling (e.g., yard signs) have tangible effects on social cohesion and neighborhood dynamics.
Interventions aimed at reducing partisan signaling in local environments could foster greater cross-partisan interaction and participation.
Understanding these dynamics can inform strategies for civic engagement initiatives, community planning, and political communication.
Quick Reference: Summary of Key Points by Study
4.1 Perceptions of partisans: Disagreement cues increase negative affect toward partisans and reduce willingness to engage with partisan individuals; independence is perceived as socially beneficial.
4.2 Visualizing the party: Disagreement cues drive more negative visualizations of partisans; unity cues mitigate negativity; images chosen under disagreement conditions show more anger, divisiveness, fighting, and negativity.
4.3 Neighborhood decisions: Political signs amplify negative evaluations of neighborhoods when participants read about partisan disagreement; people are less willing to live with politically signaling neighbors and less willing to attend social events with them under disagreement cues.
4.4 attractiveness and faces: Partisan disagreement makes partisan faces appear less attractive, less likable, less competent, and less trustworthy; independents become more positively evaluated on these dimensions; effects persist across partisan affiliation.
4.5 Synthesis: Dissatisfaction with partisanship is pervasive and spills into social life; independence gains as a socially preferable identity; partisanship is socially costly and may undermine broader democratic participation.
Potential Exam Topics to Review
Why independence is socially desirable and how partisanship is socially undesirable across multiple domains (work, housing, social life, faces).
How even neutral or unclear cues (e.g., nonpolitical signs) compare to political cues in shaping social judgments.
The role of disagreement information in altering both affective responses and behavior towards partisans and independents.
Methodologies used to study subjective perceptions (thermometer scales, image-based judgments, and randomized news cue experiments).
The interplay between social norms, beauty/attractiveness judgments, and political identities.
Real-world implications for political engagement, social cohesion, and policy communication strategies.
Notes on Citations and Housekeeping
Core authors and works cited include: Aldrich (1995); Schattschneider (1942); Verba et al. (1995); Holbrook & Krosnick (2010); Iyengar, Sood, & Lelkes (2012); Paivio (1971); Prior (2014); Valentino et al. (2002); Petersen & Aarøe (2013); Groenendyk (2013); Hibbing & Theiss-Morse (2002); Bishop (2008); Nall & Mummolo (2014); Todorov & Oosterhof (FaceGen, 2009).
Footnotes and base-rate considerations in 4.2 emphasize that image availability on the web can influence observed frequencies, and replication across time may yield different base-rate distributions.