Use of Social Media Data in Research
Data obtained from social media platforms is often used for research in the Social, Behavioural & Educational Research (SBER) domain. As these identifiable data are often publicly-available online, and there are usually no intervention or interaction with the poster (i.e. content creator), researchers often misunderstand that ethics approval is therefore not required.
Researchers are to note that using data from social media for research will usually require IRB review. Please refer to the table below for examples.
| Scenario | Examples | IRB review |
|---|---|---|
| Social media content that is usually considered non-identifiable. | Reddit. whereby the poster (i.e. content creator) is usually pseudo-anonymised. | Review Not Required (RNR): If you are only accessing data from openly-available channels or groups, and no traceable content will be published. |
Expedited Review: | ||
| Social media content that is usually considered identifiable. | LinkedIn, Facebook, Instagram, X-Twitter, etc. whereby the poster (i.e. content creator) is usually identifiable. | Exempt (Category 4): Provided poster will not be contacted, and no traceable content will be published. |
Expedited Review: (ii) If you will be joining closed channels or groups to access the data within. Note that permission from the platform owner should be sought before conducting your research. | ||
Full Board Review: |
In addition, meeting any of the conditions below may necessitate IRB reviews:
- The data on social media is identifiable.
- Private forums (or channels or groups) instead of public forums.
- Any interaction (e.g. messaging, posting, recruitment) with posters (i.e. content creators).
- Harvesting or scrapping of data.
- Usage of traceable content (e.g. quotes, screenshots, images, usernames, verbatim) whereby the poster (i.e. content creator) can then be easily traced or re-identified.
- Research on sensitive topics (e.g. mental health, sexual harassment, racial violence).