Results

DHM280 · Project 2

The project is finished, so this page looks at actual results: what the archive delivered to each of its audiences, and how that lines up with its goals and values.

Screenshot of the archive page with a banner saying the project has concluded
“This project has now concluded.” Screenshot, Social Media Lab

Where things ended up

Collection ran from June 2020, with the last monthly dataset covering October 2022. The final archive holds about 4.17 billion tweet IDs, each monthly file has a permanent DOI on figshare, and the site is now preserved as a read-only archive. Nothing new is being added, but nothing is being taken down either.

Results for each audience

Researchers

Goal: large, reusable datasets.

Result: they got a free, citable record of the pandemic conversation that no single researcher could have collected alone. The project team used this kind of data to publish their own studies on how misinformation spreads (see Research Articles).

Values: openness, transparency

Journalists and the public

Goal: show the trends in public.

Result: a free dashboard, with no login needed, that shows what people were tweeting about COVID each month, where they were tweeting from, and which accounts and links were driving the conversation.

Values: openness, public engagement

Researchers with fewer resources

Goal: make huge files manageable.

Result: the Tweets Sampling Toolkit lets someone on an ordinary laptop pull a random sample from a 100-million-ID file. It is free under an open license.

Values: openness, lowering barriers

The result that changed

One result did not hold up over time. When the archive launched, a researcher could download IDs and rebuild the full tweets for free with Hydrator. After Twitter’s 2023 API changes, Hydrator stopped working and access became expensive. So for most of the audience, the result today is a permanent, well-documented list of IDs that is much harder to turn back into tweets than it used to be.

What it means for the audience going forward

A historical record with a catch

For future researchers, the archive is a fixed historical record of how the pandemic was discussed on Twitter, and the IDs and documentation will outlast the project. For the public, it remains a window into the infodemic through the dashboard. The catch is that getting the actual tweets now depends on Twitter, not on the lab. That is why the Evaluation page flags sustainability as the project’s weakest point.

← Research ArticlesEvaluation →