Nine representative democracies, including Austria, Germany, the UK, and others, are covered in this corpus of over 6.3 million parliamentary speeches. The data spans periods between 21 and 32 years and includes metadata on date, speaker, party, and partially agenda items. It was created by Christian Rauh of the WZB Berlin Social Science Center.
Use Cases
- Train speech classification models based on speaker party metadata.
- Analyze political discourse trends over time based on speech dates.
- Study cross-national legislative language patterns based on the nine-country scope.
- Conduct speaker attribution or stylistic analysis based on speaker metadata.
Strengths
- Contains over 6.3 million full-text speech records.
- Covers nine countries over periods of 21 to 32 years, providing longitudinal data.
- Includes structured metadata for date, speaker, and party.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Last update date is unknown; freshness unverified.
- The description does not specify the completeness of agenda item metadata across all records.
Provenance
- Source
- WZB Berlin Social Science Center
- Collection Method
- Likely compiled from official parliamentary records and transcripts.
- Time Range
- Covers periods between 21 and 32 years (specific years not provided).
- Geography
- Austria, Czech Republic, Germany, Denmark, Netherlands, New Zealand, Spain, Sweden, United Kingdom