Awesome

This repository contains code to build cosponsorship networks from bills (and motions) passed in the Danish Parliament.

• interactive demo
• static plots
• more countries

HOWTO

Replicate by running make.r in R.

The data.r script downloads information on bills and sponsors. The code is pretty straightforward and excludes only a handful of bills for which the scraper gets the columns wrong.

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.

DATA

Bills

The data contains three types of legislative items: "motions" (Beslutningsforslag, "B"), "bills" (Lovforslag, "L"), and resolutions (Forslag_til_vedtagelse, "V"). Only the first two are passed to the network building routine (resolutions are nonbinding). The networks are not radically different if resolutions are included.

• title -- bill title
• uid -- a unique identifier of the form "YYYYS-TNN"
  • "YYYY" is the year of introduction
  • "S" is the session number ("1" or "2")
  • "T" is the type of item ("B", "L" or "V")
  • "NN" is a number
• ministry -- the ministerial area (empty for resolutions)
• year -- the year of introduction, deduced from uid
• status -- the outcome of the bill (not clean)
• authors -- when the bill comes from the executive, the ministerial sponsor
• links -- semicolon-separated identifiers of bill sponsors
• vote -- detailed voting report on the item
• summary -- a summary of the item

Sponsors

• name -- sponsor name
• func -- political function(s), e.g. "member,exsted,exmin"
• party -- political party, abbreviated
• partyname -- political party, full name
• constituency -- constituency, stored as the string to its Wikipedia Dansk entry
• mandate -- semicolon-separated mandate years, used to compute the nyears seniority variable
• job -- occupation
• url -- a unique identifier almost always composed of just uppercase letters
• photo -- photo URL, shortened to its unique identifier (identical to url or missing)
• bio -- short sponsor biography

Note -- due to how missing sponsors are handled, the version of sponsors.csv that is saved to the data folder contains only an intermediary version of the sponsors data. To retrieve the finalized dataset with the variables listed above, export the sponsors object s after running data.r in full.

THANKS

Thanks to Niels Erik Rasmussen and Thomas Leeper for useful feedback on a preliminary version of the code.