Healthcare Representations

Find further details of each talk in the Book of Abstracts here.

Those marked with are eligible for nomination to a student researcher award. Find the full list of awards here.

You are welcome to use the comment function at the bottom of the page to comment on papers you have seen and/or submit questions that you would like to see raised in the discussion panel. If replying to an individual paper, please specify who you are talking to.

Panel chaired by Daniel Hunt (@mynameisdanhunt).

A corpus-based study of representations of social care during the 2019 UK General Election campaign

Carmen Dayrell & Elena Semino Lancaster University

c.dayrell@lancaster.ac.uk
@carmendayrell
https://www.lancaster.ac.uk/people-profiles/carmen-dayrell

e.semino@lancaster.ac.uk
@elenasemino
https://www.lancaster.ac.uk/linguistics/about/people/Elena-Semino

[long paper]

This image has an empty alt attribute; its file name is Screenshot-2020-06-03-at-15.17.47-1.png

A genuine fantasy? Comparing distinctive language used to represent people with schizophrenia in the UK tabloid and broadsheet press ★

James Balfour Lancaster University

[long paper]

This image has an empty alt attribute; its file name is Screenshot-2020-06-03-at-15.17.47-1.png

Army Deserters or Fat Kids?: A Corpus-assisted Critical Metaphor Analysis of Obesity News in Chinese Official Media ★

Xiang Huang Universitat Pompeu Fabra

xiang.huang01@estudiant.upf.edu

[long paper]

Please ensure HD is enabled (by clicking HD in the bottom right corner) on this video for a clearer picture.

This image has an empty alt attribute; its file name is Screenshot-2020-06-03-at-15.17.47-1.png

Dementia in the media: a corpus and multimodal analysis of how dementia is represented by non-profits and newspapers in Britain. ★

Emma PutlandUniversity of Nottingham

emma.putland@nottingham.ac.uk
@Emma_Putland
https://www.midlands4cities.ac.uk/student_profile/emma-putland/

[short paper]

Examining the use of reported speech in the PrEPUK corpus ★

Luke CollinsLancaster University

l.collins3@lancaster.ac.uk
@LukeCCollins

[long paper]

This image has an empty alt attribute; its file name is Screenshot-2020-06-03-at-15.17.47-1.png

Health, Diaries, and Digital Technologies ★

Justyna Robinson & Rhys Sandow University of Sussex

[short paper]

‘Social workers dismissed concerns’: A corpus-assisted discourse study of the portrayal of a profession in UK newspapers

Maria LeedhamOpen University

maria.leedham@open.ac.uk
@marialeedham
@ouwisp

http://www.open.ac.uk/people/mel245
http://www.writinginsocialwork.com/

[long paper]

12 thoughts on “Healthcare Representations

  1. Hello everybody,

    I’m looking forward to our discussion of these fantastic papers tomorrow. In the meantime, please do post any questions or reflections that you have here and we can build these into our discussion. Questions/comments can be related to specific papers or (even better!) address multiple papers or the theme of the panel as a whole.

    Best wishes,

    Daniel Hunt

    Like

  2. Re ‘health, diaries and digital technologies’ short talk – how did you decide if positive or negative affect was displayed? Was it a binary one or the other or more graded? I’m interested as I did this for newspaper texts on social work and it was often tricky to decide!

    Like

    1. Hi Maria,

      Thank you for your comment.

      We decided to focus specifically on positive/negative affect which meant that we discarded ‘neutral’ affective stances.

      In coding this data, my policy was that if I had to think about it at all, I discarded that example. Most examples had adjectives which clearly index positive/negative affect, e.g. fantastic, brilliant, good/ terrible, awful, bad.

      When we get more data we would like to develop a more graded coding system, e.g. strong positive affect/weak positive affect. However, we felt that as this was not feasible due to the relatively small data set in this pilot study. We would need to think about exactly how to make this coding system systematic but we haven’t given that much thought at this stage. It is definitely something we’ll be thinking about more in the near future, though.

      Best of luck with your work on newspaper texts!

      Best wishes,

      Rhys Sandow

      Like

      1. Hi Rhys,

        Thanks for the reply. And going with your gut reaction makes a lot of sense to me. I tried looking at collocates and going in through semantic categories but for each of these only got the ‘fail’ lemma. So ended up reading more around the search term ‘social worker’ to categorise.

        I guess you’d need several graders to make it more nuanced. I mentioned in the panel that an earlier study did a fine-grained analysis of newspaper articles around social worker. While not your area, you might be interested in their methodology as it involves multiple graders and agreeing on 10 points on an affect scale. (However, in their discussion they largely stick to ‘positive’ or ‘negative’ – and their inter rater reliability was I think only around 83%). Anyway, ref is here:
        Reid, W. J., & Misener, E. (2001). Social work in the press: a cross-national study. International Journal of Social Welfare, 10(3), 194-201. doi: 10.1111/1468-2397.00172

        Good luck with the next stage!

        Maria

        Like

  3. For Emma Putland – just a comment really. You could consider semantic tagging using WMatrix to see if that throws anything different up in your findings. I found this helpful as a ‘sense check’ even when it’s not the main method. It means words that aren’t on their own key might constitute part of a key semantic category.

    Like

    1. Thank you Maria Leedham, that’s a really useful suggestion! I was considering trying WMatrix for exactly that reason so it’s great to hear that you’d recommend it.

      Like

  4. Eloisa Lillywhite June 16, 2020 — 9:40 pm

    Question for Justyna and Rhys:- how did you decide where the geographical boundaries were? Did the affect scoring provide an obvious distinction between areas, or did you assign the boundaries first?

    Like

    1. Hi Eloisa,

      I answered this question in the panel yesterday but I don’t think that you were present so I’ll repeat my answer here.

      The data we received from the Mass Observation Project included a lot of meta-data. This included information regarding the contributor’s place of domicile, both on the level of the specific village/town/city, and region. The regions specified were quite broad, e.g. south-west, north-east.

      The more specific location could be used to distinguish rural/urban areas but we haven’t considered that here. We have focused on the level of region. Some regions had quite low token counts which would skew the percentage data on the heat-map. as a results, we had to cluster regions together in order to make our data suitable for quantitative analysis. We decided on the North/Midlands/South distinction largely because these are regional categories that are visible in public discourses as well as in government policy/rhetoric, e.g. the Northern powerhouse, the Midlands engine.

      When we get more data we would like to narrow down our regions to more specific locations which relate to specific NHS trusts.

      I hope this answers your question.

      Best wishes,

      Rhys

      Like

  5. Something that came to mind while watching Justyna and Rhys’ presentation was work looking at regional differences with respect to politeness, specifically a chapter written by Jonathan Culpeper and Mathew Gillings in the edited collection of studies of the Spoken BNC2014: https://www.routledge.com/Corpus-Approaches-to-Contemporary-British-Speech-Sociolinguistic-Studies/Brezina-Love-Aijmer/p/book/9781138287273

    There would be a bit of work to do to join the dots between the concepts of atittudes towards particular aspects of health and linguistic performances of politeness but it feels like there is something in a kind of regional outlook, concerning particular types of face (threat) and praise, imposition etc… if you’re investigating sociodemographic factors, it may inspire some ideas!

    Like

    1. Hi Luke,

      Thank you for this suggestion! I’ve managed to have a brief look and it looks super interesting. I’ll definitely explore this paper in more detail once the conference is over.

      Thanks again for the heads up!

      Best wishes,

      Rhys Sandow

      Like

  6. Thanks for your comments so far, folks. If you have any more questions for the panel’s speakers then do post them here and we can make sure they’re integrated into the discussion during the Zoom panel. Alternatively, you can hold onto your question to ask during the Zoom talk too.

    Like

  7. I really enjoyed James Balfour’s talk, especially the aspect of romantisising mental health issues in the context of art.
    Also, towards the beginning you mentioned having experimented with other keyness measures, finding that despite statistical differences, the broad topics tended to be similar. We found the same tendency when looking into association measures (unfiltered effect size measure vs. log likelihood vs. LR combined with a confidence interval). While the overlap in words was quite small, they performed quite similarly in terms of what broader topics and actors emerged. Broadly speaking, where LL found “meat” (reference to livestock farming in a healthcare context), LR found “raw sausage”.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this:
search previous next tag category expand menu location phone mail time cart zoom edit close