Tweets on the Russia-Ukraine War with Botometer Scores, February-March 2022

5,203,746 tweets, retweets, quotes, and replies containing hashtags supporting Russia or Ukraine were collected via the Twitter API V2 from February 23 to March 8, 2022. The dataset includes Botometer automation probability scores for a random sample of 483,100 English-posting accounts. Joshua Watt created this dataset, which is associated with a paper on arXiv.

Use Cases

• Analyze public sentiment and discourse based on tweets containing pro-Russia or pro-Ukraine hashtags.
• Study bot network activity and coordination based on Botometer Complete Automation Probability (CAP) scores.
• Model temporal patterns of online discussion based on data collected over a two-week period after the invasion.
• Compare language and engagement patterns between accounts flagged with different bot likelihood scores.

Strengths

• Large-scale collection of 5,203,746 tweets from a defined two-week period.
• Includes Botometer bot-detection scores for a substantial random sample of 483,100 accounts.
• Data collection predates the invasion by one day, providing a baseline for temporal analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Data is limited to tweets containing six specific hashtag pairs, which may not represent the full conversation.
• Botometer scores are only available for a subset (26.5%) of English-posting accounts.

Provenance

Source

Twitter API V2

Collection Method

API query for tweets containing case-insensitive versions of 12 specific hashtags related to the conflict.

Time Range

February 23, 2022 00:00:00 UTC to March 8, 2022 23:59:59 UTC