Thread 3: A corpus-assisted discourse analysis of linguistic transphobia on Mumsnet

The dataset goes back to 2008 apparently, so there will be lots of FWR posters in the scrape who have since been banned by Mumsnet (eg LangCleg).
They can’t log in to post on a thread.

we can only sign in under our own identity/username

which gives some the opportunity to sign in as many times as their shifting identities!

I should think that it's sufficient that we set up a thread and make a statement there. After all, we can only sign in under our own identity/username so that should be sufficient verification.

I'm more than happy to set up a email account to do this, or set up email but print it out and send, snail mail. Or just do it online? But then online might not be the best idea.

I dunno but my head hurts so I'll shelve the idea until someone else thinks of the pros cons

I was thinking about that this morning. Does anyone know if an FOI has been done asking about this? (Sorry I need to look after DC right now).

I wonder if we could make some kind of mass FWR letter (signed using screen names) making it clear we do not consent to Aston’s use of data?

Anyone got a ticket for that half day event next week?

All very well Aston so kindly agreeing to delete some of our data - I repeat, our data - but what about all the third parties who downloaded it from Aston? And how much money has Aston made from our data?

And most of all, what is the ICO take on this?

I wonder what the logic is for agreeing to delete the older, larger scrape but still to be “in dialogue” on the smaller, more recent scrape of, presumably, this board.

That rationale is ripe for speculation but it may not be helpful while the issue is ongoing, so I shall hold my water until it’s resolved.

Update just now from Justine on the Site Stuff thread:

https://www.mumsnet.com/talk/site_stuff/5057903-mumsnet-corpus?reply=135094282

'Thanks for your patience. I’m pleased to say that Aston has agreed to delete the historic dataset and make no further use of it. (It’s worth saying again that they insist that they haven’t breached any copyright laws because of exemptions that exist for the purposes of research.)

We are still in dialogue with them about the second, much smaller, dataset and the PHD research project and will keep you posted.

When an organisation acquires an individual’s personal data from a publicly-available source they automatically become what’s known as the Data Controller of that data and are responsible for processing that data in compliance with GDPR. If anyone has any particular concerns about personal information Aston may hold then do contact their Data Protection Officer: https://www.aston.ac.uk/about/statutes-ordinances-regulations/publication-scheme/data-protection.

GDPR has strict rules ensuring that personal information provided for the purposes of submitting a Subject Access Request are treated as confidential by the Data Controller in question.Likewise if anyone has concerns about particular posts on Mumsnet that contain personal identifiable information please do [email protected] and we’d be happy to delete them.

We’ll keep you posted on further developments of course.'

Thank you so much for the explanation. I read several pages of the second thread and couldn't work out wtf was going on. I assumed it was some kind of massive in-joke that was beyond me, but now I understand how potentially serious it is. Good on all of you who identified what was happening and understood the ramifications. Will follow and take action if required.

@ArabellaScott that was v interesting reading the information you copied from the Aston publication extracts; as part of my new job I'm about to get training to learn how to use Nvivo. I wonder if this will give me more context on what it was used for in this situation? At the moment my training will be to identify themes in quantitative data but I'm going to do a big load of training on the whole Nvivo system once budget is approved by my work, so maybe this will give me further insight on what Aston has been doing! I don't work in data scraping, although I know that a different team in my organisation does.

I would surmise that the PhD smaller number of posts was from FWR. The format of the board has changed as we know so I am not sure how complete it would be from 2008. For example, I think the posts on what used to be feminist support have disappeared completely now - but I don’t know or remember exactly when that happened.

You raise a valid point.
I was going by the content of Justine’s post which I have extracted below. Also in the 2019 presentation the doxy dudes state that they got the funding (or they had started the project) a year previously, so 2018 (I might be mid-remembering about the funding, I would need to go back and check). But you are right that this includes data on the site which goes back many years before then. For example, with the material from the adoption board, the identifiable post from Tim Grant’s book chapter was 2016.
So the data scrape would capture posts from however far back they go, not only 2019 and 2024.

Nonetheless, Jai’s PhD data is from 2014 and completed before the data scrape was done. They are two separate projects, which is the point I was trying to clarify.

edited - I forgot the image, I will post separately

Jai MacKenzie’s data was not part of the data scrape, as the data scrape was done in 2019 and then again in 2024. Jai collected her material in 2014

Off-topic for this researcher but the data scrapes date back to 2008 iirc (EP's research had the range 2008 to 2023 with the one in Jan 2024 for her benefit).

Ah, I didn’t realise Jai McKenzie had done her PhD at Aston. I posted extracts from something she had written on the Site Stuff thread as an example of how MN could be used ethically. She had written a chapter on digital interactions in an edited collection if I recall. I said it should be used as a primer for the doxy dudes at Aston. She was considerate of issues of anonymity and consent.

Jai MacKenzie’s data was not part of the data scrape, as the data scrape was done in 2019 and then again in 2024. Jai collected her material in 2014 and contacted users where she wished permission to use their material in publications and also contacted MN (MN are thanked in the acknowledgements of the book above as well and the copyright permissions noted).

Not all use of MN threads is unethical - however, I guess the thing is most people were not thinking about research use when they were posted, they were thinking about writing for and to a digital community of other women, mainly but not only mothers.

I see that thesis linked above is dated 2016, so around eight years ago. The work it's based on would have been done even earlier. So they used to know how to do things sensitively but it seems they've forgotten in then meantime. Like barbarians living in the ruins of a formerly great civilization because they don't know any better.

This was an individual who made what I think was a sensible, ethical, and considered choice. Including adjusting her.methods when she realised the potential for harm.

Aston ethics department worked with her on it.

They should ask her to train the rest of the staff.

which proves it can be done. It’s obviously a lot more work for the researcher but the resulting data is a lot stronger for it.

In this instance.

Re: above 2 posts, from Arabella's extracts, it seems as if the author handled the MN posts and posters appropriately and with due permission and consents.

Interesting.

(excerpts from the paper linked above)

'Special thanks must go to the Mumsnet users who allowed me to reproduce their posts, originally written with quite a different audience in mind, in my study. Thanks especially to xxxxxx and xxxxxx, whose generous engagement with my research cannot be underestimated in the development of my approach to internet research ethics.'
'
'Some key insights garnered through observation, and developed through subsequent reflection and analysis, include my growing understanding and appreciation of the norms of participation, interaction and sharing within the Talk forum. Observing the site through the eyes of a participant,bfor example, I have been able to recognise that my research has the potential to cause harm through violation of users’ expectations. Such insights have had a significant effect on the ethical decision-making process for this study'

'I use QSR International’s qualitative data analysis software, NVivo 10 (2012), to preserve sampled threads in a static format, which can be accessed for later reference and analysis. Using NVivo’s ‘NCapture’ function, I save threads in their original format and transfer them into an NVivo project space. By using this function, rather than copying and pasting text into a word file or using screenshots, for example, I am able to retain threads from the Mumsnet discussion forum in full, including all the idiosyncrasies of their original context. I integrate new threads into my project space over time, so that it gradually comes to resemble the kind of ‘log’ of interaction that Herring(2004) places at the centre of her ‘Computer Mediated Discourse Analysis’. I describe this collection as a ‘corpus’, which I define as a structured set of naturally occurring texts, albeit quite a small one in this instance. Once threads have been captured and saved in this way, I am able to access, annotate and code them within my NVivo project space'

't became clear to me that a lot of Mumsnet users value their sense of privacy and anonymity very highly, with many exercising their autonomy and agency in imaginative ways to control and shape the accessibility of their posts and the degree to which they are identifiable as single users. As an observer-participant, I was able to recognise the potential for my research to cause harm through violation of such norms and to re-evaluate my ethical choices accordingly. One of the most significant changes I made as a result of these considerations was to contact all of the Mumsnet users whose words I wished to quote, and/or analyse in detail, and ask for their informed consent. By seeking consent, I gave potential participants the power to decide for themselves whether or not they were happy for their posts to be used for research purposes. Subsequent amendments to my research ethics approval form reflect this shift in my approach and the steps I took to protect my research participants. For example, I state in my second version of this form (see Appendix I) that ‘Although the Mumsnet Talk forum is explicitly public, I believe that there is uncertainty regarding what users perceive to be ‘public’, and how they anticipate their interactions will be used. I therefore… seek informed consent from participants whose contributions will be analysed in detail, and potentially quoted in publications or presentations’. These amendments were also approved by the ethics committee at Aston University'

'I make it very clear in this message that silence will not be interpreted as consent; in other words, contributors who did not respond were not included in
the study. Samples of consent from participants included in this study can be found in Appendix K. My message also gives contributors the option to have their usernames anonymised. Several participants have chosen this option, despite already using pseudonyms, and have subsequently been given new names that retain the spirit of their original username. The fact that many
participants made this choice shows that ‘anonymity’ is a complex matter of protecting individuals’sense of privacy and dignity, and respecting the steps they have taken to control which selves they present to the world. However, guaranteeing anonymity is unfortunately not a straightforward process when data are collected online because quotes can often be traced back to the original source through a web search'

I'm a bit behind on all this, but am surprised to see that Aston hasn't removed publications like this while the legality of the situation is resolved?
https://publications.aston.ac.uk/id/eprint/33358/

Is this all as a result of the same data scrape?

Usernames have been redacted, but I guess posters might still recognise their own contributions?

I see Aston says:
If you have discovered material in Aston Research Explorer which is unlawful e.g. breaches copyright, (either yours or that of a third party) or any other law, including but not limited to those relating to patent, trademark, confidentiality, data protection, obscenity, defamation, libel, then please read our Takedown policy and contact the service immediately ([email protected])

Have only just caught up having last posted somewhere back in the middle of thread 2.

Appreciate that Justine is following this through and keeping us up to date.

And although in a jokey way I said on thread 2 it was shocking that no MSM papers had caught up on this situation, I am wondering why.

I assume (but might be wrong) some universities have done this to other sites, where posters may have different concerns than those on FWR.

But it seem this is an issue that maybe should be more widely discussed or made known.

For instance if not all, but parts of MN became private discussion would that mean it could not be scrapped.

If so, although that is a benefit, and in fact some may feel better able to bring up issues than they do now on a public forum, what about the negative to that. ie the FWR gives women who are just readers of threads evidence that their experience, concerns, aren't individual to them, and may get comfort from seeing this.

It is, like so much of the internet, that the positive potential of being able to reach out and make contact and share and support, is then undermined by those who use or exploit that innocent motive, for their own (often financial) interests.

Even if some sort of techie news outlet doesn't think this is a serious issue that they should report on, I would like to think that some academic, university focus news outlet, would want to write up what has happened here, to instigate discussion and more importantly inform academics what is or is not appropraite behaviour.

Maybe now is too soon and we need to wait until Justine has finished her interrogation of Aston, but I hope this isn't then just brushed off as a one off.

Have only just started on thread 3 and was wondering if it would be worth while as the other had drifted off.

But thank you so much for posting this.

Really helps to get an IRL perspective on how this could have happened.