GLMP Documentation
The GLMP is designed as a systematically documented and reproducible research infrastructure. This section provides information on the organisation of the corpus, metadata, identifiers, data structure, and quality control.
Data model
GLMP distinguishes between documents and document uses.
A document is a unique source text identified by its SHA-256 content hash. A document use records how that document is associated with a municipality, political actor, election year, or other observation in the corpus.
This distinction allows the same source document to be associated with several municipalities or observations without duplicating the underlying document.
Document types
The current release distinguishes three document types:
- Kommunalwahlprogramm Gemeinde — municipal election programme
- anderes kommunales Wahlprogramm — other municipal political programme
- Grundsatzprogramm — basic programme
Programme scope
The scope of a programme is recorded separately from the municipality with which a document is associated. Possible scopes include:
- municipality
- county
- county and municipalities
- state
- region
- higher-level scope
This distinction is particularly important for programmes that are used across several municipalities.
Municipality identifiers
The Amtlicher Gemeindeschlüssel (AGS) is provided as a current municipality reference based on the municipality structure as of 31 December 2024.
Historical AGS values are not reconstructed for earlier election years. The 2024 AGS therefore provides a consistent municipality identifier across the current GLMP corpus.
Population data
Population figures refer to the municipality associated with a document and use the 2024 municipality reference state.
Population data are not applicable to documents that are not associated with a municipality, such as basic programmes. These documents are therefore not treated as missing population data.
Quality control
The GLMP corpus is subject to systematic quality-control procedures covering file integrity, metadata consistency, municipality and political-actor identification, election-year information, and duplicate observations.
The project aims to provide a transparent and reproducible corpus while preserving the original source documents.