• Kate Kavanagh

    Hi Reem, thanks for sharing your poster! I don’t see any particular questions from you for my attention, so I was wondering if you could explain something for me about how the different Functions for the corporate tweets were identified (for example, what disinguishes ‘information’ from ‘advertisment’)? Cheers!

  • Reem Al Madani

    Hi Kate,
    Thank you for having a look at my poster and for your comment.
    I mention in the (Next Steps) section on the poster what I am planning to do next in the research so your feedback and input on this would be appreciated.
    Many of the functions were identified based the apparent communicative function of the tweet. For example, the advertisement is a direct marketing message about a certain product or a call for action to buy or do something. On the other hand, information is basically telling the audience something about the daily life of the users and the companies (for example, tweeting about an important visit by an official) so the aim is to give the audience information for the general public relations of the brand but does not require any action from the audience. The functions I used were also compared to previous work on multilingualism online.
    I hope this answers your question.
    Thank you again :)

  • Kate Kavanagh

    Great, thanks :)
    I think your plan to conduct the same analysis across the other tweet sub-corpora and then compare them sounds good! What were the reasons for having different proportions of Corporations/Influencers/Normal user accounts, and will this influence the comparison? The interview element is interesting too… Were you asking the account representatives to reflect on particular tweets, or about their brand strategy in general? Cheers!

  • Mashael Assaadi

    Hi Reem, I really enjoyed reading through your poster, it shows clear argument. The most thing that grabbed my attention is figure 4, which shows how the languages are used differently based on different speech functions. Do you find any reason why is the percentage of the language used in the questions is mostly Colloquial Arabic, whereas the language used in advertisements is mostly Modern Standard Arabic?

  • Kate Steel

    Hi Reem, really interesting! Following on from Kate K’s question above: You refer to the tweet’s ‘main function’, and it seems to me that there may be more blending between functions in the influencers’ tweets (i.e. their advertising tweets may be composed in such a way that they are less easy to categorise as ads). Do you have any insight into that at this stage?

  • Reem Al Madani

    @Kate K
    Thank you so much for comment.
    The proportion is divided as 50% for corporations, which most of them employ professional branding to personalise the identity of the brand, and 50% for personal accounts (influencers and ordinary users) to look at self-branding as a tool to achieve visibility and influence. The same categorisation was also used by Page (2012) on an investigation of hashtags on Twitter.
    For the interviews, I asked about the general social media communication strategy and how they use the different languages in this strategy. I also shared with them some of the analysis I did on their accounts during the interview.

    Thank you so much for your comment!
    Based on my observation from the data and the outcomes of the interviews, Colloquial Arabic is mostly used in asking questions because the account holder wants to interact with the audience and wants them to respond to the tweet and as many of these questions imitate a daily conversation such as (what are you doing today?) or (what are you having for breakfast?) they may prefer to use Colloquial Arabic.
    On the other hand, advertisements are a one way communication and therefore Modern Standard Arabic may be dominating. Although using the colloquial variety is advertising is a widely spread strategy to attract the attention of the consumers, but this data shows that Modern Standard Arabic is still the most used variety.

    @Kate S
    Yes, Kate you are right. There are tweets that serve more than one function (for example, a greeting and a question). I prioritized the function based on the content of the tweet.
    Regarding advertising tweets, I tried to group all the functions that appeared in the data to five main categories (formulaic purposes, culturally specific genres, reported speech, dialogical interrelation, and marketing). In the Marketing category, I included information, advertisement, awareness campaigns, corporate social responsibility, and job recruitment). I hope this answers your question.
    Thank you so much!

  • Gerard O'Grady

    Hi Reem, This is a topic I know very little about but I wondered how you identify language function. For instance, in relation to question what are your criteria for identifying them? Are there inter textual cues you can rely on or all your coding decisions made intra tweet? Others have also mentioned that particular uses can have mutiple functions and I wondered how that is accounted for.

  • Amanda Potts

    Hi Reem! This is an interesting project. I think it would be helpful in posters to be clear about how the various account types are determined (ticks, followers, etc). I agree with the others that it’s difficult to say that tweets have a single function. Might it be interesting to distinguish between tweets (which would largely be informative/advertising) vs. replies (which would be more interactive)?

  • Matthew Coombes

    Hi Reem – an interesting choice indeed, and my line of questioning is a follow up from Gerard (his being my tutor being of no coincidence I’m sure!). Interesting balance between the colloquial vs modern standard in advertising – so this is where my train of thought went in terms of the language used. Had you also thought of branching further into the data by perhaps considering the product/service being advertised? How the product or service relates to the social media environment? And not that I know much at all about Twitter, but perhaps factoring in the response rate (as Amanda has suggested) and the respective sub-strata of the language used by the users.

  • Keighley

    Hi Reem,

    Thanks for sharing your work. It was really interesting to read how different Twitter users use social media. I’d be very interested to see how the language of different types of users varies.

    I’m curious about how you found Herring’s computer-mediated discourse analysis? Did you find it useful?


  • Zeen Al-Rasheed

    Hi Reem, this is very interesting topic and have enjoyed reading through your poster. As Kate mentioned your next step plan seems good to me too. Reading through your research questions and the whole sections of your poster, made me think about the way you’d interpret the outcomes (whether related to the functions and/or patterns), and which of the approaches you’d follow for that purpose?
    All best wishes

  • Reem Al Madani

    @Dr O’Grady
    Thank you for your comment.
    My criteria for identifying the functions is not based on textual cues alone in the tweets as I must look at the content and overall meaning. For example, I can’t rely on textual cues only to identify questions as some of them are rhetorical questions that are used to create an effect. One of the fast food restaurants started the tweet with (Are you hungry?) so in this case it is part of an advertisement rather than a question to be answered by the audience. In cases of multiple functions, I prioritised the function based on the content of the tweet.

    @Dr Potts
    Thank you very much for going through my poster.
    I actually have a list of criteria on how the accounts were selected but I didn’t mention it in the poster because of the space limitation. Thank you for highlighting the importance of this so I can include it in future posters. Regarding distinguishing tweets, I only included timeline tweets in my analysis without the replies.

    Thank you for your input.
    The corporations selected for this research represent different businesses and I observed different patterns of how the languages are being used. It would be interesting to link these patterns to the product/service being advertised. I agree with you that the response of the audience to the tweets (likes/retweets/interaction) is a very interesting area to look at and I hope to be able to research it in the future based on the outcomes of this research.

    Thank you so much for reading my poster.
    Yes, Herring’s CMDA was very useful for me as it is based on discourse analysis and shaped by technological features of digital communication and that was suitable for my research design.

    Thank you so much, Zeen!
    The research analysis approach is to explore what is happening in the data and then use qualitative interpretations to assist the quantitative analysis.

Leave a Reply

Your email address will not be published. Required fields are marked *