This repository contains code to build cosponsorship networks from bills passed in the Estonian parliament.
Replicate by running make.r in R.
The data.r script downloads information on bills and sponsors. Information on sponsors from past legislatures are downloaded as single-page summaries, whereas sponsors from the current legislature have more detailed individual biographies.
The build.r script then assembles the edge lists and plots the networks, with the help of a few routines coded into functions.r. Adjust the plot, gexf and mode parameters to skip the plots or to change the node placement algorithm.
The networks are very sparse (0.01 < d < 0.02), basically because there are very few nominally cosponsored bills per legislature. It is much more common for bills to be introduced by the executive or by 1+ parliamentary committee than by 1+ nominal sponsors.
legislature-- legislature number (numeric)session-- legislature id (text)number-- legislature numberdate-- date of introduction of the bill (dd.mm.yyyy)title-- bill titleurl-- bill URLauthors-- bill sponsors, using their full nameslinks-- bill sponsors, using their URLs (when available)n_a-- number of sponsors (from full names)n_l-- number of sponsors (from URLs)
Note -- given that only sponsors present in legislature 13 (the current one) have a URL, the sponsor information should be derived from their full names rather than from their URLs, except if one is working on legislature 13 alone.
legislature-- legislature of activityname-- full nameborn-- year of birth (int)party-- political party (abbreviated)constituency-- constituency (using generic names)nyears-- seniority (multiple of term length, 4 years)committees-- semicolon-separated committee membershipsphoto-- photo URL (legislature 13 only)sex-- gender (F/M), imputed from first names
Note -- committees are stored either as acronyms or as full committee names.