In May 2019 we published an analysis of data relating to the upcoming EU elections, "some facts and figures about the choice voters have about who represents them in Brussels and Strasbourg."
The article was based on a range of analysis, including combining ONS data with information on MEP wages; a Python-based Twitter scraper and analysis in R; collating names and using a names gender API; and generating some visualisation using R.
- European Parliament: Salaries and pensions; European elections: What pay can UK MEPs expect?
- BBC: 2019 European elections: List of candidates
- Spreadsheet: MEPs standing and not standing
- ONS: Employee earnings in the UK: 2018
- Spreadsheet: Matching earnings by region against MP salaries
- CSV: MEP pay versus regional average
- European Parliament: Turnout and gender factsheet (PDF)
- CSV: Turnout by country
- CSV: Female MEPs by country
- Spreadsheet: MP names' gender - API results, analysis and checking
- Spreadsheet: Checking numbers of candidates by party and region
- CSV: UK MEP Twitter accounts
- CSV: Word frequency comparison, lower case versus original case - this was used to check whether terms like 'EU' and 'UK' were from URLs or being used in reference to geography
- Two datasets - 180,001 scraped tweets, and a dataset of all tweets by MEPs in the last 12 months - are too large to include here
- Spreadsheet: analysis of election spending data from Electoral Commission (not used in the final story)
- CSV: Donations during 2014 European election period (unused, as it is not possible to distinguish between donations for EU or other elections)
- Julie Girling, independent MEP for the South West
- Mary Honeyball, Labour Party MEP for London
- Table: Number of parties standing vs number of seats for each region
- Table: MEPs standing down by region
- Bar chart: MEPs' pay compared with regional averages
- Bar chart: Most common words tweeted by UK MEPs
- Bar chart: Percentage of MEPs that are female by member state
- Bar chart: Turnout in European elections 2014 by member state
- Word cloud: word frequency in MEP tweets
- Twitter scraper: the first part collects accounts from a list; the second part scrapes the most recent 3,300 tweets by each account on a list
- R Markdown: analysing MEP tweets
- Shell script: generating a word count from a text file (unused). Note: the text file is created by changing the .csv extension to .txt