254,257 transcribed interactions between citizens and hotline operators recorded on the Chinese government service hotline system in 2019. The dataset was created by Zhao, Xinghua for a study examining how citizens' discursive strategies influence bureaucrats' prioritization decisions. Discursive strategies were identified through a supervised machine learning algorithm.
Use Cases
- Analyze citizen complaint prioritization based on identified discursive cues
- Train machine learning models to classify discursive strategies in citizen complaints
- Study the effectiveness of different citizen voice behaviors in government-citizen interactions
Strengths
- 254,257 transcribed interactions provide a substantial corpus for analysis
- Data is from a specific year (2019) and system (government service hotline)
- Discursive strategies were identified using a supervised machine learning algorithm
Limitations
- Column-level documentation is absent; field semantics must be inferred after download
- Row count is known, but specific features and file formats are unknown
- Data may reflect temporal and geographic bias inherent to the specific hotline system
Provenance
- Source
- Journal of Public Policy Dataverse
- Collection Method
- Transcribed interactions from the government service hotline system
- Time Range
- 2019
- Freshness
- Last updated 2026-04-13 01:08:29; freshness should be verified
- Geography
- China