Replication material for "Searching for the Cartel Party: Inter-Party Collusion and Legislative Activity" as presented at EPSA 2018.

The code turns legislative cosponsorship data into a panel dataset of political parties nested into country-specific parliamentary chambers. Some parties are observed at several points throughout the period under examination. Some aspects of the panel data related to party-specific cosponsorship behaviour are then modelled via (one-part) fractional response regression models.

Read

The working paper and conference slides are available from the authors on email request.

Data

Details on the data can be found in DATA.md.

Code

Initially written in R 3.3.3 in strict mode. Currently being updated for R 4.1.0.

Make

source("1_extract_cosponsorships.r")  # get the data
source("2_create_unbalanced_panel.r") # panelize it
source("3_fractional_regressions.r")  # estimate FRMs
source("4_export_and_plot_models.r")  # plot results

# session info
sink("session_info.log")
devtools::session_info()
sink()

Package dependencies are listed at the top of each script.

Last updated: June 2021.