SM&Society Day 3: Keynote

Desiring numbers: when social media data are ordinary

It was interesting to listen to Helen Kennedy talk once more about how visualising data is not merely a cognitive process, but also an affective one. Her research in the Seeing Data project reveals the extent to which we react on an emotional level to data, especially those presented visually. These responses can be both positive and negative, so it becomes crucial for those interested in disseminating data to be aware that they are not simply passing on information in a neutral way, but may be affecting the recipients emotionally.

I wondered in what ways other forms of visual media invoke similar or different responses, if the images don’t have to carry with them the numerical burden that visualisations do? The talk yesterday discussing how gifs might invoke meaning making perhaps has parallels; they doubtless also prompt an emotional response, yet the playfulness with which they’re (usually) created perhaps has a different outcome.

When asked whether artists might usefully contribute to the field of visualising data, Helen of course welcomed the possibility. Projects like the WWI ceramic poppy memorial at the Tower of London a couple of years ago used multimodal, multi-sensory channels to remind us of the figures behind the events being commemorated. What this talk particularly helped me with was in reminding me to keep the affective issues in mind. As I shift towards a more ontological sensibility, and am keen to explore my participants’ realities, their emotional responses to professional learning may warrant a closer consideration.

SM&Society Day 2: Poster session

flickr photo by ianguest https://flickr.com/photos/ianinsheffield/28165208722 shared under a Creative Commons (BY-NC-SA) license

Day 2 closed with the poster session; a wonderfully warm, welcoming and informative session, made all the more hospitable by the delightful buffet.

Many of the posters summarised talks which had already (or were about to be) presented, so it made for a useful refresher and a chance to capture some of the details which might have been missed. More importantly however, it was the chance to talk with presenters who, due to the tight schedule, weren’t provided with much time to answer questions at the close of their talk. From a personal perspective, I often need a little more time after a presentation to process what’s been said before I feel in a position to seek further clarification – if a concept or data being presented isn’t clear, my default response is that I’ve missed something (‘Imposter syndrome’ still doing its job!). So having the chance to return to an issue and discuss it under more relaxed circumstances was most welcome.

Although I talked with a number of people, the person I really wanted to catch was Claudine Bonneau who was presenting a poster summarising her research findings on ‘Working out Loud.’ I was grateful for the opportunity to discuss some of her findings at greater length, especially how ‘backstage’ work and activities normally invisible to external gaze are made more visible. I’m particularly intrigued as to what the effects of making this work visible might be, both in terms of the learning of the person making their work manifest, and on those who form the audience for it. How does this fit with professional learning? Claudine’s work is still in progress, so it will be interesting to see what arises from the forthcoming interviews.

SM&Society Day 2: Visual Social Media

I attended this session, like some of the others, to find out more about issues in areas with which I’m less familiar. I know that visual media form part of the data i encounter, but until my assignment for the Discourse Analysis module I recently completed, had given little thought to their significance. In the opening paper, Dianne Rasmussen Pennington posed the idea that sources of data other than text might offer potential, but in order to access the meaning they convey, we may need to look to methods we’re less familiar with.

Ivo Furman described the situation regarding suppression of communications media by the government in Turkey. Facebook and Twitter tend to be ‘turned off’ during times of crisis; at other times, trolls are employed by the government to create disinformation and discourage others. Instagram, for the moment, presents an unfiltered channel through which to communicate, as indicated during the Pride march last year. The use of images, memos and innovative hashtag behaviour have all been employed to circumvent tools of repression.

In a paper with plenty of amusing anecdotes, Kate Miltner and Tim Highfield presented their research into ‘reaction GIFs’ as performative tools through which to respond in online exchanges. So, if you wanted t indicate how your thesis was progressing you might respond with:

And finally, Yimin Chen answered the question that had been on all our lips ‘When does the narwhal bacon?’ by looking at memes, how they spread and how we can use them as a sort of group identity marker.

These were all well presented papers; interesting, informative … and witty. Although I couldn’t see a direct connection with my study, what they nevertheless served to do was to ensure that I keep the media they were discussing in mind. They may prove to be significant in my data, and had I not attended the session, might have missed them completely. For me, sessions like this are about reminding me to attend to detail and remain open to possibilities.

SM&Society Day 2: Academia

flickr photo by ianguest https://flickr.com/photos/ianinsheffield/28234738596 shared under a Creative Commons (BY-NC-SA) license

In the post break session,  four papers, each a work-in-progress, on the broad topic of social media in academia were given. I couldn’t help but notice two things: how they all drew from higher education contexts (perhaps that’s simply what ‘academia’ is) and how they were oriented towards the activities which lead to the production of data.

The first looked at the individual and collective factors contributing to use of social media by/in research teams. The second considered how imagined audience influenced social media participation. Next we heard how iSchool faculty members are connected by and participate in social media. Finally Sian Joel-Edgar explained the part played by social media in engineering student design teams.

Each seemed to be concerned, to different degrees, with what data were being produced. Additionally, how the data were produced, what the reasons for that were and in some cases, what outcomes there might be for the producers. All valuable information, but I was left wondering why the subjects (apologies for that term) might engage in the activity they do, and what the effects or impact might be on the recipients or audiences of their activity. Reflecting in this way helped me recognise and acknowledge my research philosophy which leans far more towards the interpretivist paradigm and exploring why a phenomenon is as it is, rather than what is occurring or how.

I know that conference attendees are largely from higher education contexts, but again I wonder where the studies are which, whilst still from educational contexts, focus their attention on different phases. The work is out there, but clearly not coming to the conference. Given what I said earlier, I’m pondering why that is? Are the topics presented in the conference from a higher educational context because that’s where conference audiences are from, or simply that’s where the researchers are? What (or where) are the audiences for research arising in different contexts? Now it’s occurred to me, I’ll be wanting to see whether that continues throughout the remainder of the conference.

SM&Society Day 2: Identity, Professions, Institutes and Culture

2c

Finola Kerrigan and Kathryn Waite opened this session with a talk on a rather different research strategy involving filmic methods in exploring online identity. Essentially a competition was launched in which participants were to produce a short film illustrating their online behaviours, but also explaining their rationale, choices and techniques to the research team. This could then be used as a reflective device for those conducting research using similar methods.

Marc Miguel-Ribé’s explanation of the differences which arise in different Wikipedias as a result of cultural differences was fascinating. Although aware that there are different language versions of Wikipedia, I’d never really considered the implications of different authorship on the knowledge that’s constructed. The knowledge regarding a particular phenomenon that’s assembled on one Wikipedia might express a completely different view from that on another. This is not only because the authors on the two pages will be different, but because they’re also likely to be from different cultures with different worldviews. This certainly hit home for me in terms of the way that’ll reflect on the ‘knowledge’ that emerges from participants in my own studies.

In the final presentation of the session, I was introduced to the concept of the ‘greedy organisation’ by Kim van Zoest and Sietske de Ruijter and the additional factors police officers are obliged to consider have when using social media. Their employers impose certain expectations, both explicitly and implicitly. This took me back to my life as an employee, where signing my contract of employment included agreeing to never conduct myself on online in a way which might embarrass my employing organisation. As Kim and Sietske pointed out, social media and digital connectivity means organisations are increasingly reaching into our personal spaces.

Although each paper was interesting in its own right, I’m not sure how well they sat together; but then it can’t be easy for conference organisers to gather all the submitted papers into coherent themes.

 

SM&Society Day 2: Challenging Social Media Analytics

flickr photo by Steve Burt https://flickr.com/photos/sburt/3167655269 shared under a Creative Commons (BY-SA) license

Susan Halford provided the opening keynote and reminded us that ‘data never sleeps’ and is being generated at ever increasing scales in real time and over time. Whilst this may constitute an ‘unexpected gift’, it’s meant we’re also ‘building the boats as we row’ in terms of the way we’re gathering and analysing those data.

Susan challenged us to consider three questions:

  1. What are social media data?
  2. Where are the data produced?
  3. Why does this matter?

There is genuine concern that much of the current evangelism around Big data may have done more harm than good, leading to inflated expectations about what is possible and what we can learn. If we’re not careful, our reliance on the platforms through which we access the data may unduly influence what we find, in a host of different ways, and in ways which vary over time. Demographic and geographic data especially need treating with caution, or at least with care and in full knowledge of their limitations. Perhaps we should go beyond demographics and make a virtue of the biases, limitations and specificities inherent in the data.

I hope in a sense that is what my research is doing, where I’m focusing on a particular,  self-selecting sample, engaged in a specific activity. For me, the demographics are in some senses pre-defined – teachers who using Twitter. What their gender, religion or ethnicity is, will be of no consequence since I’ll not be classifying my results using those criteria. Or at least I never intended to, until I though about location. I’ve assumed my participants will be teachers drawn from a global population, though due to my linguistic limitations, from the english-speaking world. The keynote has encouraged me to revisit my thinking; in different places (with different cultures?), might teachers have a different view of, and approach to, professional learning using Twitter?

Susan asked Les Carr, her colleague from Southampton, to join her on stage. Amongst other things, Les pointed out that vivas inevitably ask us to justify our methods and the data they generate, and how they are appropriate for the research questions we pose. I was grateful for that reminder as I begin to think about my RF2 submission. Duly noted!

SM&Society Day 1: Small data and big data controversies

This workshop session  had us split into groups to each consider one of the six ‘V’s of big data: variety, value, volume, velocity, veracity and variability. The three hours were split in two, each part session opened by a number of speakers presenting the findings of the papers they had contributed to the forthcoming Sage Handbook of of Social media Research Methods. We were asked to consider our ‘V’ (we had Variety) in the  context of any tension between Big and small data, if indeed there was any. Our table, as it transpired, consisted of social scientists rather than computational scientists, so unsurprisingly tended to focus on the positive aspects of small data.

I found the opening presentation by Claudine Bonneau and Mélanie Millette on their ‘small’ data projects spoke to my research – very much a hands-on,  immersive, participatory approach, where tweets were collected and analysed manually. The approach within the long-term observation was described as ‘agile’, following the conversations from place to place. There’s a resonance for me in the way teachers shift between Twitter, #chats on Twitter and blogs when discussing their practice. I yet to grasp how or if I can or should incorporate the offline places where these discussions occur. There are clear sites of interest where teachers gather to discuss and share practice (TeachMeets etc); my problem however, will be whether I have the scope to chase them down.

I found the topic of ‘Working out Loud’ practice on Twitter had a close fit with my own research, although I was surprised that other professions also engaged in this practice (how insular am I?!). However, the most compelling aspect was how we ‘thicken’ small data, perhaps reducing its breadth whilst enhancing its depth.

SM&Society Day 1: Conceptual Challenges in Interdisciplinary Social Media Research

flickr photo by ianguest https://flickr.com/photos/ianinsheffield/28144230192 shared under a Creative Commons (BY-NC-SA) license

Following welcomes and the opening address by Evelyn Ruppert to set the scene, the first day consisted of workshops.

Although billed as a workshop, the unforeseen absence of a key player meant this session became more of a panel discussion. With such esteemed and knowledgeable panellists as Evelyn Ruppert, Susan Halford and Les Carr, and chaired by Mark Carrigan, we were nevertheless unlikely to be short-changed.
Evelyn opened the batting, providing a brief history of ‘Big Data’ and how the term became accepted by and incorporated into the academy. The field of practices which address it are still emerging, but invariably demand a range of capabilities, hence the need for interdisciplinary teams. When the data are gathered, analysed and patterns begin to emerge, it is often the social scientist who helps to provide the interpretation. Whilst the computational or algorithmic elements of this often need to be black-boxed by the social researcher, perhaps the reverse might be true for a data scientist who is unaware of the social issues (even if unwittingly?). This then encouraged us to consider whether there is too much black-boxing and what the effects of this might be on our analyses.

Susan and Les shared what they learned from their interdisciplinary experiences of setting up Web Science doctoral and other programmes. The substantive theoretical commitment required in a venture of this sort can potentially generate an initial set of hurdles to overcome; people from different disciplines inevitably bring to the table very different ontologies and epistemologies. Computational specialists for example have a much narrower theoretical base from which to draw, but in addition to the multiplicity that social scientists have to contend with, there’s also the degree to which different theories may be applied depending on the questions asked. For many computational specialists, the work often ends with the user; for the social scientist, that is where it begins.

When opened to the floor, the discussion ranged far and wide, but the difficulty of attempting interdisciplinary work was made only too clear. In particular the significance of power imbalances between disciplines and how they may be competing for cultural capital within the academy. How some disciplines are blessed with apparently greater status historically because of the research publication processes or the ease with which they’re able to draw down funding for research. The silo mentality which then arises makes interdisciplinarity so much more difficult. This left me wondering whether there might be a case for addressing this at an earlier stage in the academy? At undergraduate level perhaps? I appreciate my naivety is doubtless getting the better of me, but I can’t help being taken back to my undergraduate years at the end of the 70s. Materials Science, the subject I chose, was very much the infant (interdisciplinary) sibling to the big brothers and sisters of metallurgy, ceramics and polymer science. I get the impression that, with the decline in heavy industry the UK has experienced in the intervening time, and the extent to which new composites have become significant, interdisciplinary Materials Science may have come of age. Perhaps other disciplines might have something to learn?

‘Teaching Skills for Doctoral Students’

After thirty-plus years teaching and supporting teachers, it would be easy to feel you’ve pretty much got it nailed. But, if you’ll forgive me dipping into the often saccharine realm of online quotes for a moment, the more you learn, the more you realise how little you know. So I’m happy to share I had been looking forward to the start of the ‘Teaching Skills for Doctoral Students’ (TSDS) course. This is an obligatory course for doctoral students who are required to undertake a teaching load as part of their commitment. I remember at my PhD interview, being told that I would ‘have to’ do this course, despite my teaching experience … as though I might be likely to offer up an objection. Although I’m sure some might, for me, this was a marvellous opportunity to learn about teaching in a phase of education with which I have no experience. What’s not to like?!

flickr photo by Jirka Matousek https://flickr.com/photos/jirka_matousek/8481205571 shared under a Creative Commons (BY) license

The TSDS course, comprising only four afternoon sessions, serves only as an introduction. It isn’t assessed and carries no study credits, so I assume it’s there to make sure post-grads don’t do too much damage to the learning of those for whom they will have responsibility. But that makes it sound like a hurdle to be overcome, rather than an opportunity to be embraced. Our tutor Sylvia clearly held the latter view as she enthusiastically welcomed us to the opening session yesterday. We were quite a disparate group; although all of us are conducting doctoral research, we are at different stages and from an eclectic range of academic disciplines. Sylvia’s opening activity, ‘Name Game,’ intended to help us get to know one another consequently made perfect sense. It of course modelled a technique we could use when we encounter our students for the first time and reminded us of the importance of knowing someone’s name as the first stage in forming a relationship. It also highlighted ‘verbal chaining’ as a memory technique. As a secondary teacher, I never remembered my students names this way; there I had the luxury of meeting them two or more times a week over the duration of (at least) a year. In higher ed., it’s quite possible of course that you might meet your class just once a week for a single semester … or as in the case of TSDS, even less.

During the course of the session, we were put into different groups several times and also set the room up differently. In addition to helping us become aware of the different ways we could use the space and structure the learning, we were asked at various points to reflect on how that felt as learners, and share our thoughts with each other. We were being encouraged in a sense to be both reflective and self-critical, and consider the consequences for our future learners of the choices we make as teachers. Experienced teachers do this quite naturally, though perhaps rarely articulate it to others in the way we did during the session. We also were asked to provide (on three sticky notes) the concerns/fears we had of embarking on our teaching. Some of these were shared, whilst others will be revealed later in the course. My concerns included a worry that my subject knowledge might not be adequate, that I might be bringing inappropriate secondary school teaching strategies into a tertiary arena, and that my digital expectations of the students will be misaligned with their needs and expectations. (I could unpick each of those with a blog post(!), but for the moment will leave them to simmer and see how things develop).

We also covered the more mundane, but nonetheless still important areas, like punctuality, attendance and access to teaching rooms. Ooo, I’ve just thought of another concern! Going into a room after someone else if: a) they’re late finishing or b) leave the room in a poor state. I know I’m not very understanding of other people’s shortcomings in this regard, feeling that that kind of behaviour is disrespectful of colleagues and sets poor expectations of our students. I suspect I’m going to have to develop a more tactful approach, or risk initiating a diplomatic incident. I will certainly have to keep reminding myself of my junior status and that my future prospects may depend on me not upsetting anyone. Another admin-related area I hadn’t yet thought about is that of health and safety, which for a former science teacher is perhaps surprising. When working with students in a lab, safety issues are always at the forefront of your mind, but I had spared little thought to the particular circumstances in a seminar room or lecture theatre. It was helpful therefore to be reminded that our students may be joining us in a room or even building they’ve never been in before, and should therefore, at the very least, be informed about evacuation procedures.

As Sylvia drew the session to a close by recapping what we’d covered and looking forward to the remaining sessions, she asked us each to share something positive we’d be taking away from the session. For me that was easy; I never fail to feel a sense of gratitude for the opportunity to watch someone else teach. I always learn something, whether the teacher is as experienced, skilled and capable as Sylvia, or whether they’ve just embarked on their career. I think we’re getting better as a profession at providing opportunities to observe one another, even if as a formal undertaking, observation can be a complex issue requiring careful consideration. What I found fascinating though was how Sylvia either told us the reason she had done something in a particular way, or asked us to discuss why she had chosen a particular strategy and the consequent effects on us as learners. It’s a rare privilege to have access to that degree of insight, and not something I ever recall experiencing as a trainee teacher, other than on a few rare occasions after observing an experienced teacher when they found the time to share the rationale behind what I’d just seen. Perhaps there’s value in that sharing for both sides? Certainly for the less-experienced practitioner, but also for the more experienced one in articulating the choices they had made and making them open to scrutiny by someone whose relative naivety might provide a powerful lens?

Can’t wait until next week 🙂

MoreThan … NVivo Training

flickr photo by Raul P https://flickr.com/photos/rolexpv/15111554846 shared under a Creative Commons (BY-NC-ND) license

Earlier this week, I had an appointment  cancelled, which free’d up Wednesday. This mattered because an NVivo training course was scheduled for that day, and I’d been unable to attend because of my earlier commitment. As luck would have it, there was one place left, which i immediately snapped up … what luck! But at that stage, I didn’t realise quite how lucky.

The session was delivered by Ben Meehan of QDA Training, an experienced NVivo trainer, who promised us a rather different training experience to those we were perhaps familiar with. Although not quite as radical as we might have been led to believe, there were definitely some distinct differences between this session and conventionally-structured computer application training sessions. It’s not often you’re on such a course lasting only a day, where you don’t even touch the keyboard until after lunch! Nevertheless, that proved, for me, to be exactly the right strategy.

My context.

I wasn’t attending the conference as a complete novice. Given the approach I’ve taken to my research, at some stage I was going to need to collect, manage, analyse and interpret a wide range of qualitative data. With multiple methods (often within digital domains) generating several types of data, it made sense to me to seek computer assistance. Since the package to which the University subscribes is NVivo, I decided quite early on to begin to get to grips with it. There is plenty of helpful literature out there1, and a good number of YouTube videos; unsurprisingly some are better than others. I found this one particularly helpful:

Together, these resources have helped me become familiar with the the structure, features and a range of uses of NVivo; I could doubtless bring it to bear on my research and make (fairly) effective use of it. Although I’ve given it a good test drive whilst researching the literature and as part of my pilot methods study, I kept hitting snags and was convinced I wasn’t getting the most from it. More worryingly, I was concerned that I might set up a poor structure that would later require me to dismantle it and thereby increase my workload unnecessarily. The question was though, whether a course titled ‘Introduction to NVivo’ might be aimed at people with less experience. I needn’t have worried.

Structure of the day.

With its full title of ‘Introduction to NVivo: Building your database’ we can see how important this course will be in addressing the fundamental aspects of using NVivo. It was with this in mind that Ben devoted the initial session to discuss why we might want to use a computer to assist our qualitative analysis in the first place, and what the implications of that choice might be. He then asked to consider what data we were likely to be collecting and how they might be thought of as ‘cases’ so that we would be in a position to design a structure to organise them within NVivo. Time invested in this initial planning would pay dividends in later stages when we needed to conduct and test different levels of analysis across and through our data. Another crucial element in the design and build of our database is the analytical strategy we intend to adopt. This is where I became a little unstuck, as this isn’t an area I’ve yet devoted sufficient thought to, beyond using open coding, developing categories and themes, and synthesising the emergent concepts. I used constructivist grounded theory in an earlier study, but hadn’t really moved on from there. This session above all else forced me to confront my choices and whether framework analysis, content analysis, discourse analysis, interpretative phenomenological analysis, or indeed grounded theory are consistent with my epistemology. At the moment, I see virtues in several approaches and have to resolve whether it’s possible to draw what might be useful from each, or whether they are completely incompatible. More thinking still to do.

Having collected, analysed and interpreted our data, the next stage will be to report our findings. Within that reporting, in addition to the insights we developed, we will need to present and justify the decisions we made; once more NVivo can help with that process and help us to construct a more robust and credible thesis. By annotating, memoing and reliably entering pertinent metadata, NVivo can help us maintain a transparent audit trail connecting the findings we present back through the decisions we took, to the data where those seeds were sown.

We finally launched NVivo after lunch, when we were given the opportunity to put into practice much of what we’d discussed earlier. Using the data from an actual piece of research, Ben introduced us to all of the features and capabilities within NVivo which would allow us to implement the ideas we had developed in the morning. With the ‘why’ in place, the ‘how’ now made far more sense.

What I came away with.

Although my awareness of the NVivo environment and the tools within it didn’t advance far, instead I gained something of far greater value. I now have a much better appreciation of what my database could look like and how to begin setting that up. I know that my planning will be key and that that will have to be informed by the analytical strategies I choose.

As one might expect, we were provided with a set of resources we can refer back to; a workbook (printed and pdf) and data we can use to practice our technique. Less usual was the offer of access to ongoing support for the duration of our projects. We can get additional personal advice and assistance when setting up our projects or if we encounter any problems, either by email or when necessary through Skype and remote desktop sharing. Incredibly generous, a potential lifesaver, but precisely the kind of safety net many of us will need as our research unfolds.

What other people thought.

It was clear from the initial introductions that people were at a range of different places in both their research and their capability with NVivo. That’s very difficult for the person at the front to cater for, and yet I got the impression from those I talked with and from the gratitude people expressed at the close of the course, that most people, like me, had gained a great deal. Ben obviously knew his stuff and was able to draw from a wide range of experiences when answering queries from the audience. He never once sugar-coated things, nor over-evangelised the virtues of NVivo, but what he did do was set out the questions we need to answer and a path to follow when we have those answers.

1BAZELEY, Patricia, and JACKSON, Kristi, (2013). Qualitative data analysis with NVivo. Second edition / Pat Bazeley and Kristi Jackson.. ed., London, SAGE.
BAZELEY, Patricia, et al. (2000). The NVivo qualitative project book. London, SAGE Publications; SAGE.
GIBBS, Graham (2002). Qualitative data analysis: Explorations with NVivo (Understanding social research). Buckingham: Open University Press.
RICHARDS, Lyn (1999). Using NVivo in qualitative research. Sage.