Researchers welcome easier access to TikTok data ‘goldmine’

Researchers in Europe have welcomed an interface that will allow easier access to the social media platform TikTok, although concerns remain about their ability to study how it is used.

ByteDance, the Chinese company that owns the app, said it would open its application programming interface (API) to European academics, ahead of tougher regulations under the European Union’s Digital Services Act that are scheduled to come into force in August.

Interfaces allow other software to pull large volumes of linked data from an application, such as video clips, captions, comments and audience interactions.

The company, which will also allow researchers to browse a library of adverts on the platform, said the changes were “designed to enhance transparency”. TikTok added that it had made the process simpler since it allowed similar access to US academics in February.

The change comes after major data access tussles between social media platforms and researchers, particularly after the 2018 Cambridge Analytica scandal, when Facebook shut down many academic apps. Twitter’s introduction of fees for its API in February this year has created a barrier for many.

Daniela Jaramillo-Dent, a researcher at the University of Zurich who studies minority communities on TikTok and Instagram, said opening the API was a “very significant move” that could draw more researchers to study TikTok.

Zoetanya Sujon, who researches social media use at University of the Arts London, said it was “promising and quite strategic in the current landscape”, but added that there were still “really serious limitations”, such as researchers in the US needing to have papers approved by the company before publication.

“It’s a good gesture in the right direction, but there’s still a lot of counter-research ethos, monitoring and potential interference. If you want to do something that’s critical of TikTok, you might not be ever able to release that.”

Without an API, researchers can still do smaller, qualitative studies on social media platforms. They can also write software that scrapes data from the app, but this takes considerable time, and its use often requires programming skills, said Dr Sujon.

Christian Ilbury, a sociolinguist at the University of Edinburgh who studies young people’s language use on social media, scrapes data from TikTok. He said he hoped the API would give him “fine-grained” metadata on how videos were shared, which he cannot get at present, calling it “a goldmine of an opportunity”.

Christian Montag, who studies molecular psychology at Ulm University, said the workarounds that did exist were not ideal. “There are ways, but whatever we do at the moment, it’s very difficult,” he said.

He said it was important to combine biological and self-reported data to get a full picture of issues such as addiction. “It is at the moment fiercely debated what a healthy social media platform would look like,” he said, adding that self-reports were questionable when used alone. “People often have time distortions when they use these platforms, and you don’t get exact estimates.”

ben.upton@timeshighereducation.com