This month’s #OTalk Research will be hosted by Professor Sarah Pederson @SarahPedersen2, Robert Gordon University and will be supported by @preston_jenny and @SamOTantha on the @OTalk_ account.

Hello everyone and welcome to this week’s #OTalk on the ethical use of social media as research data. My name is Sarah Pedersen and I am Professor of Communication and Media at Robert Gordon University, Aberdeen.
Social media can be an amazing resource for researchers, really allowing us to access grassroots opinion and see how particular issues are discussed outside the classroom or more formal focus groups and interviews. It can also be used to engage participants in your research and to disseminate your findings beyond the usual suspects.

First do no harm

It can be a temptation to just leap in and collect lots of data from social media. However, we must remember that there are real people behind those posts and we must behave ethically. When I am assessing a student’s plans for engaging with social-media data, I want to see a thoughtful approach based on the practice of other scholars in the field and acknowledging the ethical tensions. A very good place to start are the ethical guidelines produced by the Association of Internet Researchers Ethics (aoir.org)

To some extent, your approach to social-media data will depend on where you are
collecting it. For example, I have an on-going relationship with the parenting forum Mumsnet. On occasion, Mumsnet has allowed me to run discussion threads about a project and I have encouraged forum users to chat with me and share their opinions. Here both Mumsnet and users are aware that their words may be used in my research and later publications so the ethical issues are less problematic. However, collecting data from social media when posters are not aware that this is happening is more problematic.

Is it publicly available?

One of the first questions we need to ask about social-media data is whether it is freely and publicly available. There is a clear difference between posts on Facebook, which you can mostly only see if you are a friend of the poster or member of a group, and posts on Twitter, for example. Unless the poster makes their tweets private, then it might be assumed that they are making use of Twitter to broadcast their thoughts to the world. We might therefore apply the same ethical standards to using this material as we would to any published material – acknowledging the source. However, of course we know that when people use social media they are not thinking about the rest of the world – just their friends. So I would tend to use Twitter as a data source but not Facebook. In the same way, I use Mumsnet because you don’t have to be a member to read the threads and because Mumsnetters are very aware that others make use of their words – Mumsnet itself has published books of baby advice taken from posts on the forums and journalists from the Daily Mail, etc, regularly run columns about a funny thread on the site. If I quote from blogs I will write to the blogger and ask for permission to quote. I usually receive an amazed response since the blogger has never been asked for permission before. But I consider this to be good practice.

Should I anonymise?

If you are using social-media data for qualitative research, you are going to want to quote from your corpus of texts. So should you try to anonymise your sources? Again, I would say that this depends on the particular source and your particular needs. I would expect you to discuss the reasons behind your decision in your methods section.
Over the years I have taken a number of different approaches here:

1. Yes, completely anonymise. I have taken away all identifying features from the text and just quoted ‘a poster’, ‘a user’, ‘a tweeter’.
2. When using source material from a tweeter/poster who is already using a pen name, I have quoted using that pen name. My thinking here is that they are already anonymised but deserve to have their authorship acknowledged. I often take this approach when using Mumsnet material, particularly when the name is relevant to my discussion. Thus two of my most recent works on Mumsnet have discussed users’ knowledge of the women’s suffrage campaign (so users with suffragette names are interesting) and users’ expression of anger during lockdown (some names demonstrated their anger very forcibly and so were an important part of the data). I am also aware that google can often find a quote on social media – so why not acknowledge its author?
   Sometimes I have demonstrated this when reviewing articles – googling the direct quotes used and showing the author that their attempts to anonymise the data have not worked.
3. I have named the author outright. I have tended only to do this if they are public figure. MPs, celebrities and other public figures often make use of Twitter and Instagram to make direct statements to the public or the media. In these circumstances I think they an be identified.

Some questions to think about:

1. Would you be comfortable with your social media posts being quoted by an academic researcher?
2. Do you agree with the distinctions I make between the different social medias?
3. Which sources would you be happy using and would you anonymise the data?
4. Social media is great for engaging a wider public with your research, both to engage participants and to disseminate published work. How might you do that?

POST CHAT 

Host:  @SarahPedersen2

Support on OTalk Account: @preston_jenny and @SamOTantha

Evidence your CPD. If you joined in this chat you can download the below transcript as evidence for your CPD, but remember the HCPC are interested in what you have learnt.  So why not complete one of our reflection logs to evidence your learning?

https://otalk.co.uk/reflection-logs/

HCPC Standards for CPD.

• Maintain a continuous, up-to-date and accurate record of their CPD activities.
• Demonstrate that their CPD activities are a mixture of learning activities relevant to current or future practice.
• Seek to ensure that their CPD has contributed to the quality of their practice and service delivery.
• Seek to ensure that their CPD benefits the service user.
• Upon request, present a written profile (which must be their own work and supported by evidence) explaining how they have met the Standards for CPD.