Homophily in Voting Behavior: Evidence from Preferential Voting

This is a dataset for the paper "Homophily in Voting Behavior: Evidence from Preferential Voting" (co-authored by Lucie Coufalová and Michal Ševčík) that allows for the replication of all regression tables and descriptive statistics.

There are four files in the dataset:

• "rdata.csv" is the main file with data used in regressions
• "homo_occupation.csv" is an auxiliary file with data used to calculate some descriptive statistics
• "municipalities_age.csv" is an auxiliary file with data used to calculate some descriptive statistics
• "replication_data_20220910.RData" is a file that contains all tables packed and compressed for use in R.

 Tables
 rdata...main data table used in regressions
 homo_occupation...auxiliary data table used in descriptive stats
 municipality_ages...auxiliary data table used in descriptive stats

Variables
 Outcome:
 pref...number of preferential votes

 Variables of interest:
 homo_municipality...indicator variable for a candidate running in the municipality of his/her residence
 homo_(education/occupation/age/gender)...percentage of the population sharing the characteristic of the candidate

 Other variables:
 KOD_OBEC...municipality ID in CISOB classification
 total_fe...ID of candidate-election pair
 maxPORCISLO...number of candidates on the ballot
 cluster_ID...ID for error term clustering
 POC_HLASU...total number of votes cast for the party in the given municipality
 small_municipality...indicator variable for a municipality with a population below the median (defined separately 
 for each year and constituency)
 year...election year (election ID)
 VOLKRAJ...constituency ID
 PORCISLO...position of the candidate on the ballot
 VEK...age
 agecat...age category
 MANDAT...indicator variable for elected candidates
 tertiary_educ...indicator variable for candidates with tertiary education
 gender...male/female
 ger_share...indicator variable for municipality being dominated by ethnic Germans in 1930