Twitter network of members of the 19th German Bundestag

December 2018 and July 2019, Markus Konrad markus.konrad@wzb.eu

Wissenschaftszentrum Berlin für Sozialforschung / WZB Social Science Center

This repository contains R scripts for

1. scraping links to social media accounts of members of the 19th German Bundestag (called deputies here);
2. fetching the "following" list for those deputies with a Twitter account (i.e. which Twitter accounts does a deputy follow);
3. processing and visualizing this data as network.

See the following blog posts:

• A Twitter network of members of the 19th German Bundestag – part I
• A Twitter network of members of the 19th German Bundestag – part II

The respective downloaded and processed data also resides in the data directory.

Data sources

Data on German representatives in different parliaments can be found on abgeordnetenwatch.de, which also provides an API. The list of deputies of the current (19th) German Bundestag is obtained from:

https://www.abgeordnetenwatch.de/api/parliament/bundestag/deputies.json

Unfortunately, links to social media profiles cannot be obtained via this API, although the data is available on the profile pages for individual deputies, see for example this profile. These links are extracted via scraping.

Scripts

At first, the file deputies.json from the above link must be downloaded. The process of obtaining the social media data is divided into the following scripts:

1. scraper.R – scrapes the abgeordnetenwatch.de profile page of each deputy from data/deputies.json in order to extract the links to social media platforms; saves the result in data/deputies_custom_links.csv
2. twitter_profiles.R – extracts the Twitter handles (where present) from the social media links for each deputy and combines that information with the deputies' profile data from abgeordnetenwatch.de; saves the result in deputies_twitter.csv
3. fetch_friends.R – fetches the "following" list (called "friends" in Twitter API terminology) of each deputy Twitter profile using the rtweet package; because of Twitter API's rate limiting, this takes quite some time; saves the result – consisting of Twitter user IDs – in data/deputies_twitter_friends_tmp.RDS
4. lookup_friends.R – fetches Twitter profile data (like user name, bio, location, latest tweet, etc.) for each Twitter user ID that was obtained via fetch_friends.R; again, this takes quite some time; saves the result in data/deputies_twitter_friends_full.RDS

There is a Makefile which allows calling the scripts directly and running them in the background from command line. They write their output in the respective file in the logs folder.

The datasets deputies_twitter.csv and deputies_twitter_friends_full.RDS can be joined resulting in a dataset with deputies and a list of Twitter profiles that they follow.

The script friends_network.R uses this dataset to create and visualize the Twitter network between deputies (i.e. who follows whom / who is followed by whom).

Data and plots

All collected data resides in data, generated plots in plots and HTML files for the interactive network visualizations are in the root directory named dep_visnetwork_XXX.html.

Data and plot files are suffixed (_XXX) by the two points in time when the data was collected: _20181205 for Dec. 5 2018 and _20190702 for July 2 2019.

• data/deputies_XXX.json: full data on members of the 19th German Bundestag downloaded from the abgeordnetenwatch.de API
• data/deputies_custom_links_XXX.csv: URLs from the "further links" section scraped from each deputy's profile page on abgeordnetenwatch.de (including links to Twitter, Facebook, etc. for many profiles)
• data/deputies_twitter_XXX.csv: dataset of deputies data from abgeordnetenwatch.de combined with Twitter user names (where listed on the profile page)
• data/deputies_twitter_friends_full_XXX.RDS: RDS file (load with readRDS()) containing data frame that for each deputy Twitter user name contains information about her/his Twitter followings (aka "friends")
• data/deputies_twitter_friends_tmp_XXX.RDS: tempory dataset that for each deputy Twitter user name contains the Twitter user IDs of her/his Twitter followings