Birdwatch Guide Birdwatch Guide

Downloading data

We can’t wait to learn with you!

All Birdwatch contributions are publicly available on the Download Data page of the Birdwatch site so that anyone in the US has free access to analyze the data, identify problems, and spot opportunities to make Birdwatch better.

If you have questions or feedback about the Birdwatch public data or would like to share your analyses of this data with us, please DM us at @Birdwatch.


Working with the Birdwatch data

Data snapshots

The Birdwatch data is released as three separate files: one containing a table representing all Birdwatch notes, one containing a table representing all Birdwatch note ratings, and one containing a table with metadata about notes including what statuses they received and when. These tables can be joined together on the noteId field to create a combined dataset with information about notes and their ratings. The data is released in three separate tables/files to reduce the dataset size by avoiding data duplication (this is known as a normalized data model).

Currently, we release one cumulative file each for notes, notes status history, and note ratings. However, in the future, if the data ever grows too large, we will split the data into multiple files as needed.

A new snapshot of the Birdwatch public data is released daily, on a best-effort basis, and technical difficulties may occur and delay the data release until the next day. We are not able to provide guarantees about when this may happen. The snapshots are cumulative files, but only contain notes and ratings that were created as of 48 hours before the dataset release time. When notes and ratings are deleted, they will no longer be released in any future versions of the data downloads, although the note status history dataset will continue to contain metadata about all scored notes even after they’ve been deleted, which includes noteId, creation time, the hashed participant ID of the note’s author, and a history of which statuses each notes received and when; however, all of the content of the note itself e.g. the note’s text will no longer be available.

The data download page in Birdwatch displays a date stamp indicating the most recent date of data included in the downloadable files.

File structure

Each data snapshot table is stored in tsv (tab-separated values) file format with a header row. This means that each row is separated by a newline, each column is separated by a tab, and the first row contains the column names instead of data. The note and note rating data is directly taken from the user-submitted note creation and note rating forms, with only minimal added metadata (like ids and timestamp). The note status history file contains metadata derived from the raw notes and ratings, and contains the outputs of the note scoring algorithm. Below, we will describe each column’s data, including the question or source that generated the data, data type, and other relevant information.


Updates to the Data

As we iterate and improve Birdwatch, we will occasionally make changes to the questions we ask contributors in the note writing and note rating forms, or additional metadata shared about notes and rating. When we do this, some question fields / columns in our public data will be deprecated (no longer populated), and others will be added. Below we will keep a change log of changes we have made to the contribution form questions and other updates we have made to the data, as well as when those changes were made.

  • Added entirely new note status history dataset

Updated Columns

  • notHelpfulArgumentativeOrInflammatory - Changed name to notHelpfulArgumentativeOrBiased

Added Columns

  • helpfulUnbiasedLanguage
  • notHelpfulOpinionSpeculation
  • notHelpfulNoteNotNeeded
  • Note Helpfulness question now has 3 response categories (Yes, Somewhat, No), rather than 2 (originally: Yes, No)
  • We have removed the ‘Agree’ note rating question
  • We have updated the set of categories contributors can use to describe why a note is helpful or unhelpful. (Note: both helpful and unhelpful descriptors can be selected for notes that are rated as ‘Somewhat’ Helpful)

Deprecated Columns

  • helpful - Replaced with helpfulnessLevel
  • notHelpful - Replaced with helpfulnessLevel
  • helpfulInformative
  • helpfulEmpathetic
  • helpfulUniqueContext
  • notHelpfulOpinionSpeculationOrBias
  • notHelpfulOutdated
  • notHelpfulOffTopic

Added Columns

  • helpfulnessLevel
  • helpfulAddressesClaim
  • helpfulImportantContext
  • notHelpfulIrrelevantSources

Field Type Descripton Response values
noteId Long The unique ID of this note
participantId String A Birdwatch-specific user identifier of the user who authored the note. This is a permanent id, which remains stable even if the user changes their username/handle.
createdAtMillis Long Time the note was created, in milliseconds since epoch.
tweetId Long The tweetId number for the tweet that the note is about.
classification String User-entered multiple choice response to note writing question: “Given current evidence, I believe this tweet is:” “NOT_MISLEADING” “MISINFORMED_OR_POTENTIALLY_MISLEADING”
believable String User-entered multiple choice response to note writing question: “If this tweet were widely spread, its message would likely be believed by:” “BELIEVABLE_BY_FEW”, “BELIEVABLE_BY_MANY”
harmful String User-entered multiple choice response to note writing question: “If many believed this tweet, it might cause:” “LITTLE_HARM”, “CONSIDERABLE_HARM”
validationDifficulty String User-entered multiple choice response to note writing question: “Finding and understanding the correct information would be:” “EASY”, “CHALLENGING”
misleadingOther Int User-entered checkbox in response to question “Why do you believe this tweet may be misleading?” (Check all that apply question type). 1 if “Other” is selected, else 0.
misleadingFactualError Int User-entered checkbox in response to question “Why do you believe this tweet may be misleading?” (Check all that apply question type) 1 if “It contains a factual error” selected, else 0.
misleadingManipulatedMedia Int User-entered checkbox in response to question “Why do you believe this tweet may be misleading?” (Check all that apply question type) 1 if “It contains a digitally altered photo or video” selected, else 0.
misleadingOutdatedInformation Int User-entered checkbox in response to question “Why do you believe this tweet may be misleading?” (Check all that apply question type) 1 if “It contains outdated information that may be misleading” is selected, else 0.
misleadingMissingImportantContext Int User-entered checkbox in response to question “Why do you believe this tweet may be misleading?” (Check all that apply question type). 1 if “It is a misrepresentation or missing important context” is selected, else 0.
misleadingUnverifiedClaimAsFact Int User-entered checkbox in response to question “Why do you believe this tweet may be misleading?” (Check all that apply question type). 1 if “It presents an unverified claim as a fact” is selected, else 0.
misleadingSatire Int User-entered checkbox in response to question “Why do you believe this tweet may be misleading?” (Check all that apply question type). 1 if “It is a joke or satire that might be misinterpreted as a fact” is selected, else 0.
notMisleadingOther Int User-entered checkbox in response to question “Why do you believe this tweet is not misleading?” (Check all that apply question type). 1 if “Other” is selected, else 0.
notMisleadingFactuallyCorrect Int User-entered checkbox in response to question “Why do you believe this tweet is not misleading?” (Check all that apply question type). 1 if “It expresses a factually correct claim” is selected, else 0.
notMisleadingOutdatedButNotWhenWritten Int User-entered checkbox in response to question “Why do you believe this tweet is not misleading?” (Check all that apply question type). 1 if “This Tweet was correct when written, but is out of date now” is selected, else 0.
notMisleadingClearlySatire Int User-entered checkbox in response to question “Why do you believe this tweet is not misleading?” (Check all that apply question type). 1 if “ It is clearly satirical/joking” is selected, else 0.
notMisleadingPersonalOpinion Int User-entered checkbox in response to question “Why do you believe this tweet is not misleading?” (Check all that apply question type). 1 if “It expresses a personal opinion” is selected, else 0.
trustworthySources Int Binary indicator, based on user-entered multiple choice in response to note writing question “Did you link to sources you believe most people would consider trustworthy?” 1 if “Yes” is selected, 0 if “No” is selected
summary String User-entered text, in response to the note writing prompt “Please explain the evidence behind your choices, to help others who see this tweet understand why it is not misleading” User entered text explanation, with some characters escaped (e.g. tabs converted to spaces).
Field Type Descripton Response values
noteId Long The unique ID of this note.
participantId String A Birdwatch-specific user identifier of the user who authored the rating. This is a permanent id, which remains stable even if the user changes their username/handle.
createdAtMillis Long Time the note was created, in milliseconds since epoch.
timestampMillisOfFirstNonNMRStatus String The timestamp, in milliseconds since epoch, of when the note got its first status besides “Needs More Ratings”. Contains -1 if the note never left “Needs More Ratings” status. 1 if “Yes” is selected, 0 if “No” is selected
firstNonNMRStatus String The first status the note received when it got a status besides “Needs More Ratings”. Contains -1 if the note never left “Needs More Ratings” status. “”, “CURRENTLY_RATED_HELPFUL”, “CURRENTLY_RATED_NOT_HELPFUL”
timestampMillisOfCurrentStatus String The timestamp, in milliseconds since epoch, of when the note got its current status, including “Needs More Ratings”. 1 if “Yes” is selected, 0 if “No” is selected
currentStatus String The current status of the note. “NEEDS_MORE_RATINGS”, “CURRENTLY_RATED_HELPFUL”, “CURRENTLY_RATED_NOT_HELPFUL”
timestampMillisOfLatestNonNMRStatus String The timestamp, in milliseconds since epoch, of when the note most recently received a status of either “Currently Rated Helpful” or “Currently Rated Not Helpful”. This value will be the same as timestampMillisOfFirstNonNMRStatus if the note has never switched status after receiving its first non-”Needs More Rating” status. Value is -1 if the note never left “Needs More Ratings” status. “NEEDS_MORE_RATINGS”, “CURRENTLY_RATED_HELPFUL”, “CURRENTLY_RATED_NOT_HELPFUL”
latestNonNMRStatus String The latest status the note received, when it got a status besides “Needs More Ratings”. Value is -1 if the note never left “Needs More Ratings” status. “”, “CURRENTLY_RATED_HELPFUL”, “CURRENTLY_RATED_NOT_HELPFUL”
Field Type Descripton Response values
noteId Long The unique ID of the note being rated.
participantId String A Birdwatch-specific user identifier of the user who authored the rating. This is a permanent id, which remains stable even if the user changes their username/handle.
createdAtMillis Long Time the note was created, in milliseconds since epoch.
agree Int Binary indicator, based on user-entered multiple choice in response to note rating question “Do you agree with its conclusion?” 1 if “Yes” is selected, 0 if “No” is selected
disagree Int Binary indicator, based on user-entered multiple choice in response to note rating question “Do you agree with its conclusion?” 1 if “No” is selected, 0 if “Yes” is selected
helpful Int Binary indicator, based on user-entered multiple choice in response to note rating question “Is this note helpful? ” - Deprecated as of 2021-06-30. 1 if “Yes” is selected, 0 if “No” is selected
notHelpful Int Binary indicator, based on user-entered multiple choice in response to note rating question “Is this note helpful?” Deprecated as of 2021-06-30. 1 if “No” is selected, 0 if “Yes” is selected
helpfulnessLevel String User-entered multiple choice response to note rating question: “Is this note helpful” Added as of 2021-06-30. “NOT_HELPFUL” “SOMEWHAT_HELPFUL” “HELPFUL”
helpfulOther Int User-entered checkbox in response to question “What about this note was helpful to you?” (Check all that apply question type). 1 if “Other” is selected, else 0.
helpfulInformative Int User-entered checkbox in response to question “What about this note was helpful to you?” (Check all that apply question type). Deprecated as of 2021-06-30. 1 if “Informative” is selected, else 0.
helpfulClear Int User-entered checkbox in response to question “What about this note was helpful to you?” (Check all that apply question type). 1 if “Clear and/or well-written” is selected, else 0.
helpfulEmpathetic Int User-entered checkbox in response to question “What about this note was helpful to you?” (Check all that apply question type). Deprecated as of 2021-06-30. 1 if “Nonjudgmental and/or empathetic” is selected, else 0.
helpfulGoodSources Int User-entered checkbox in response to question “What about this note was helpful to you?” (Check all that apply question type). 1 if “Cites high-quality sources” is selected, else 0.
helpfulUniqueContext Int User-entered checkbox in response to question “What about this note was helpful to you?” (Check all that apply question type). Deprecated as of 2021-06-30. 1 if “Offers unique information or context” is selected, else 0.
helpfulAddressesClaim Int User-entered checkbox in response to question “What was helpful about it?” (Check all that apply question type). New as of 2021-06-30 1 if “Directly addresses the Tweet’s claim” is selected, else 0.
helpfulImportantContext Int User-entered checkbox in response to question “What was helpful about it?” (Check all that apply question type). New as of 2021-06-30 1 if “Provides important context” is selected, else 0.
helpfulUnbiasedLanguage Int User-entered checkbox in response to question “What was helpful about it?” (Check all that apply question type). New as of 2021-12-15 1 if “Neutral or unbiased language” is selected, else 0.
notHelpfulOther Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). 1 if “Other” is selected, else 0.
notHelpfulIncorrect Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). 1 if “Incorrect information” is selected, else 0.
notHelpfulSourcesMissingOrUnreliable Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). 1 if “Sources missing or unreliable” is selected, else 0.
NotHelpfulOpinionSpeculationOrBias Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). Deprecated as of 2021-06-30. 1 if “Opinion, speculation, or bias” is selected, else 0.
notHelpfulMissingKeyPoints Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). 1 if “Misses key points or irrelevant” is selected, else 0.
notHelpfulOutdated Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). Deprecated as of 2021-06-30. 1 if “Outdated information” is selected, else 0.
notHelpfulHardToUnderstand Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). 1 if “Hard to understand” is selected, else 0.
notHelpfulArgumentativeOrBiased Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). Variable name changed from notHelpfulArgumentativeOrInflammatory in 2021-12-15 1 if “Argumentative or biased language is selected, else 0.
notHelpfulOffTopic Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). Deprecated as of 2021-06-30. 1 if “Off topic” is selected, else 0.
notHelpfulSpamHarassmentOrAbuse Int User-entered checkbox in response to prompt “Help us understand why this note was unhelpful” (Check all that apply question type). 1 if “Spam, harassment, or abuse” is selected, else 0.
notHelpfulIrrelevantSources Int User-entered checkbox in response to prompt “What was unhelpful about it?” (Check all that apply question type). New as of 2021-06-30 1 if “Sources do not support note” is selected, else 0.
notHelpfulOpinionSpeculation Int User-entered checkbox in response to prompt “What was unhelpful about it?” (Check all that apply question type). New as of 2021-12-15 1 if “Opinion or speculation” is selected, else 0.
notHelpfulNoteNotNeeded Int User-entered checkbox in response to prompt “What was unhelpful about it?” (Check all that apply question type). New as of 2021-12-15 1 if “Note not needed on this Tweet” is selected, else 0.
Submit feedback Note Ranking