Moldova Disinformation Narratives: 70 Texts Coded for Characters and Frames

70 texts from Moldova have been manually coded for narrative elements like characters, settings, and moral judgments. This dataset was created by Cristina Panaguta as part of an undergraduate thesis in Government at Colby College. The analysis focuses on disinformation narratives and their framing within a specific regional context.

Use Cases

• Train or test NLP models for narrative frame detection based on the coded frames.
• Analyze character roles and moral judgments in disinformation campaigns based on the coded attributes.
• Study regional patterns in political communication and media analysis based on the Moldova-specific texts.
• Conduct comparative studies of narrative structures in disinformation across different regions.

Strengths

• 70 manually coded texts provide a foundation for qualitative and quantitative analysis.
• Coding includes specific narrative elements: characters, setting, moral judgments, and frames.
• Dataset is openly licensed under CC-BY-4.0, facilitating reuse and sharing.

Limitations

• Row count and column-level documentation are unknown, limiting suitability assessment.
• The dataset is small at 404.4 KB, offering limited scope for large-scale ML training.
• Data is provided as a PDF, which may require extraction and preprocessing for computational analysis.

Provenance

Source

Cristina Panaguta, Colby College.

Collection Method

Texts were coded for narrative analysis as part of an undergraduate thesis.

Time Range

null

Freshness

Last updated 2026-04-17 19:04:13; freshness should be verified.

Geography

Moldova