Skip to content

Voter Psychology

ElectoralSim models voter psychology using established political science frameworks.

Big Five Personality (OCEAN)

Each voter has five personality traits that influence their ideology:

Trait Range Political Correlation
Openness 0-1 Higher → more liberal
Conscientiousness 0-1 Higher → more conservative
Extraversion 0-1 Higher → more politically engaged
Agreeableness 0-1 Higher → more cooperative voting
Neuroticism 0-1 Higher → more responsive to threat appeals

Accessing Traits

model = ElectionModel(n_voters=10_000, seed=42)
voter_df = model.voters.df

# View traits
print(voter_df.select([
    "openness", "conscientiousness", "extraversion", 
    "agreeableness", "neuroticism"
]).head())

Research Basis

  • High openness correlates with liberal positions (Carney et al., 2008)
  • High conscientiousness correlates with conservative positions
  • These are implemented as soft influences on ideology generation

Moral Foundations (Haidt)

Five moral foundations that shape political reasoning:

Foundation Description Political Association
Care Protection from harm Liberal emphasis
Fairness Justice, equality Liberal emphasis
Loyalty In-group solidarity Conservative emphasis
Authority Respect for hierarchy Conservative emphasis
Sanctity Purity, disgust Conservative emphasis

Accessing Foundations

foundations = voter_df.select([
    "mf_care", "mf_fairness", "mf_loyalty", 
    "mf_authority", "mf_sanctity"
])

Research Basis

  • Haidt's Moral Foundations Theory (2012)
  • Liberals prioritize Care and Fairness
  • Conservatives value all five more equally

Affective Polarization

Measures emotional distance between in-group and out-group.

polarization = voter_df["affective_polarization"].to_numpy()
# Range: 0 (no polarization) to 1 (highly polarized)

Effects: - High polarization → stronger party loyalty - High polarization → less responsive to policy changes

Political Knowledge

Voter awareness of political facts and processes.

knowledge = voter_df["political_knowledge"].to_numpy()
# Range: 0-100

Effects: - High knowledge → more consistent ideology - High knowledge → more strategic voting

Misinformation Susceptibility

Vulnerability to false information.

misinfo = voter_df["misinfo_susceptibility"].to_numpy()
# Range: 0-1

Correlates with: - Lower political knowledge - Higher neuroticism - Lower education

Media Diet

Each voter has a media source preference:

model = ElectionModel(n_voters=10_000, seed=42)

# Media source (0=Left, 1=Center, 2=Right)
media_source = voter_df["media_source_id"]

# Media bias (-0.5=Left, 0=Center, 0.5=Right)
media_bias = voter_df["media_bias"]

Media Sources

ID Label Bias
0 Left-leaning -0.5
1 Centrist 0.0
2 Right-leaning +0.5

Selection Mechanism

Voters are more likely to consume media aligned with their ideology (selective exposure).

Economic Perception

Sociotropic vs Pocketbook voting distinction:

perception = voter_df["economic_perception"].to_numpy()
# 0 = Pocketbook (personal finances)
# 1 = Sociotropic (national economy)

Research: - Higher education → more sociotropic - Used by SociotropicPocketbookModel