Awesome
This repository contains code to build cosponsorship networks from bills passed in the lower/national chamber of the Austrian Parliament.
HOWTO
Replicate by running make.r in R.
The data.r script downloads information on bills and sponsors. There is not much data to download in both cases, so the loops should execute pretty quickly. All photos should download fine.
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 < 0.03) and show an odd case of dissortative mixing: due to the "grand coalition" formed by the SPÖ and the ÖVP in the post-Haider era, members of the two parties cosponsor with each other as if they formed a single entity.
DATA
Bills
legislature-- legislature id (coded in Roman numbers)date-- date of introduction of the bill (yyyy-mm-dd)title-- titleref-- unique identifierurl-- bill URLsponsors-- bill sponsors, using their numeric ids
Sponsors
The sponsors data have multiple entries for each sponsor (more or less one per legislature in which the sponsor sat), which allows to keep track of party transitions like the move from FPÖ to BZÖ for several MPs after the 1999 legislative election (Haider). Sponsors who sat in multiple legislatures without changing parties are grouped on a single row.
id-- numeric id, preceded by "id_"; points to the profile URLname-- full name, with titles like "Dr." or "Mag."legislature-- legislature of activityparty-- political partymandate-- semicolon-separated mandate years, used to compute thenyearsseniority variablesex-- gender (F/M), imputed from title ("Abgeordnete/r")born-- year of birth (stored as character)photo-- photo URL, a variation of the profile URL stored inid
Note -- constituency is not scraped because it is present only for the current legislature.