youtube-stat2data

youtube-stat2data is a small research utility for collecting public video metadata from a YouTube channel and exporting the results as a CSV file. It was created for computational media research and uses the YouTube Data API v3.

Project status: This repository is a research prototype last developed in 2022. The script is preserved for transparency and reuse, but YouTube API behavior, quotas, and available fields may have changed. Review the current API documentation before using it in a new study.

Output

The script exports one row per video, with fields for:

• title
• video ID
• description
• publication date
• like count
• favorite count
• view count
• comment count

Some statistics may be unavailable because of video settings or API changes.

Requirements

• Python 3
• pandas
• google-api-python-client
• a Google Cloud project with the YouTube Data API v3 enabled
• one or more YouTube API keys

Install the Python dependencies with:

python -m pip install pandas google-api-python-client

Setup

1. In the Google Cloud console, create or select a project.
2. Enable the YouTube Data API v3.
3. Create an API key and restrict it to the APIs and environments needed for your project.
4. Open youtube-stat2data-for-public.py and replace the example placeholders in DEVELOPER_KEY, channel_id, and the list of years.

Never commit API keys to GitHub. For new or shared research code, prefer loading credentials from environment variables or another secret-management system.

Finding a channel ID

A channel ID is not always the same as the channel name shown in a YouTube URL. Open a video from the channel, select the channel name, and inspect the channel page URL. In a URL such as:

https://www.youtube.com/channel/UCupvZG-5ko_eiXAupbDfxWw

the final segment is the channel ID. The screenshots below illustrate the workflow used when this project was created:

1. Open the channel page.

   

2. Open a video and select its channel name.

   

3. Read the channel ID from the resulting URL.

   

How collection works

The script queries the channel one month at a time for the requested years, follows paginated search results, retrieves video statistics in batches, and writes the combined table to youtube-data.csv.

For large collections, consult Google's current quota documentation. Multiple API keys should only be used in accordance with Google Cloud and YouTube API policies; do not use key rotation to evade quota or usage restrictions.

Research-use notes

• Record the collection date, channel ID, requested date range, query settings, and software version in your research documentation.
• Treat counts as time-dependent observations rather than permanent facts.
• Check applicable platform terms, institutional requirements, and research ethics obligations before collecting, storing, or sharing data.
• Do not commit downloaded datasets if they contain restricted, sensitive, or unpublished research material.

Limitations

• Search results and metadata availability are controlled by the YouTube API.
• Deleted, private, unlisted, region-restricted, or otherwise unavailable videos may be absent.
• Metrics can change after collection.
• The current script is configured by editing placeholders in the source file; it does not yet provide a command-line interface or automated tests.

License

Distributed under the MIT License.

Author

Heesoo Jang — website