Guidance for researchers planning to do internet-mediated research (IMR)

See also Dundee guidance for researchers planning to use social media for recruitment

Introduction

UKRI definition of internet-mediated research taken from their research ethics guidance on IMR 

‘Internet-mediated research (IMR) can take place in a range of settings, e.g. email, chat rooms, web pages, social media (Tik Tok, Facebook, Reddit, Twitter for example) and the many forms of instant messaging.’

‘Information provided in forums, social media or spaces on the internet that are intentionally public would be considered ‘in the public domain’, but the public nature of any communication or information on the internet or through social media should always be critically examined.’

The key considerations for researcher embarking on IMR are explored in greater depth in the Ethics guidelines for internet mediated research created by the British Psychological Society and are introduced in the checklist advice below.

Checklist steps

Researchers conducting IMR should complete the following steps when planning, writing and submitting their application to the School Research Ethics Committee (SREC):

1. In the application provide the SREC with a statement describing the proposed techniques including:
   1. A list of the sites to be used and the method used to access data
   2. An account of how potential participants will be identified and how their privacy will be maintained.  
   3. If it is feasible to get informed consent from participants in the community being researched, describe how this will be achieved. If not, justify this decision.
   4. An explanation of how issues of participant suitability are addressed in the methodology. Consider whether the characteristics of participants can be established to ensure the research is valid, and to inform issues of consent e.g.  demographic information, such as age, may be relevant. Data from those under the age of 16 should not be used without parental consent. 

2. Ensure that the research methodology complies with relevant data protection legislation. If there is any doubt over whether the proposed strategy complies with relevant data protection legislation, researchers should always refer to or take advice from the University information governance team at [email protected]. Additional information can be found at the Data Protection webpages. 

   Country-specific legal requirements (for example, for data protection) apply in internet-mediated research that crosses national boundaries due to the location of the data or the researchers on the team.

3. The website is a stakeholder in your research, they effectively mediate your access to data subjects. In 2024 the UK Research Integrity Office (UKRIO) hosted a webinar presented by Professor Nicholas Gold, in which he described social media companies as gatekeepers. This means that it is necessary to seek their consent. This can be passive (where your use of publicly available data posted on that platform expressly matches their terms and conditions) or active (where the moderator is contacted for permission to access data e.g. a semi-public forum for a particular community). 
   
   Mitigate risk - consider whether you can continue with your proposed methods and at the same time respect the rights of all stakeholders, and meet copyright restrictions.  If you are unable to inform individuals that they are part of your research study and the website/platform has promised users control over their data in the future, their public data may not be suitable for research. Accept that your research question or methodology may be restricted by the terms and conditions of the platforms you are using.
4. Whilst data can technically be scraped from websites it does not follow that the practise is legitimate. At time of writing, more ubiquitous social media platforms, such as Facebook and YouTube, bar the scraping of data, and TikTok requires any research publications are submitted to the platform prior to publication.
   1. You will need to provide the SREC with a statement certifying compliance with the current policies and terms of use of relevant websites.
   2. Regular checks need to be made to ensure the proposed or ongoing research is in compliance with current website policy.
5. Ensure that the proposed data collection strategy respects all relevant ethical norms, including:
6. Does not involve deception or the fabrication of online identities
7. Does not involve the researcher ‘lurking’ or ‘creeping’ on social media sites in ways members of the site are unaware of.
8. Does not involve approaches or contact that could embarrass or stigmatize potential participants.
9. Does not involve the researcher guiding or shifting the content of the social media exchange (e.g. by suggesting a conversational topic) for research purposes without transparency.
10. Ensures confidentiality e.g. in addition to anonymisation of user names and time stamps, it may be necessary to alter the wording of direct quotes to avoid backwards searching compromising participants anonymity.

11. At all times be mindful of the unique issues presented by using the internet to gather information on or from data subjects. 

    Key considerations defined by British Psychological Society (BPS) are reproduced below with some additional explanation in italics:

12. public-private domain distinction online; (use of private data that is publicly available will require the researcher considers an individual’s feelings and reactions to the planned research)
13. confidentiality and security of online data; (researchers must be aware that multiple sources of openly available information relating to an individual may, in combination, reveal their identity)
14. participant anonymity and procedures for obtaining valid consent; (consent given to a website or social media platform to publish personal data does not necessarily imply consent to be the subject of research, direct quotes should not be used to ensure participant anonymity)
15. observing procedures for ensuring withdrawal rights; (in addition to protecting anonymity it may be preferable to paraphrase quotes, to avoid generating requests to withdraw personal data)
16. levels of researcher control; (lack of control over variables, such as browser or device used, or the true identity of the participants, could ultimately affect the validity of data);  and
17. implications for scientific value and potential harm of both participants and researchers.

These considerations are further explained in greater detail in the Ethics guidelines for internet mediated research created by the British Psychological Society. See British Psychological Society Ethics guidelines for internet-mediated research.

Further information and advice

See the full UKRIO webinar on Social Media and Ethics.

See UKRI guidance on internet-mediated research, where there are links to further information.

See the Association of Internet Researchers guidelines.