It’s been a little quiet again here, as I’ve taken February and March off on Long Service Leave. That’s all about to change, though, because two major new research projects are about to start now – more of these soon.

For the moment, here’s my first conference presentation for 2014, from the Media Talk symposium at Griffith University in Brisbane. I used this to work through the three layers of communication on Twitter which Hallvard Moe and I have identified in our chapter in Twitter and Society, and to provide some examples for how these layers operate in practice.

This is also the first time I’m trying Penxy as a tool for archiving my slides with audio recordings, since Slideshare has made the unfortunate decision to discontinue its slidecasts and remove any audio recordings from its site. Most of my past slidecasts are therefore also on the Penxy site now, and I’ll try to update the existing links to recorded presentations on this site when I get a chance.

Here’s my talk:

Layers of Communication: Forms of Talk on Twitter

Now that I’m back in Australia from my extended conference trip, I immediately got back on a plane to travel to a freezing Melbourne, to present our social media research in crisis communication and beyond at the 2013 IBM Research Colloquium. Below are my slides and audio – many thanks again to Jennifer Lai and her team at IBM Research Australia for the invitation!

The next speaker at Digital Methods is Irmgard Wetzstein, whose interest is in social media monitoring tools. Social media monitoring is an increasingly important research area, of course, in both scholarly and commercial research, as the rise of 'big data' demonstrates. A social media monitoring industry has now emerged, providing a range of tools and services across various platforms. There are even systematic evaluations of the various tools, documenting this diversity.

If many such tools have emerged from commercial contexts, are they nonetheless also useful for scholarly purposes? Irmgard's study examined some 100 tools, focussing on tools which provide analytics for multiple social media platforms, and examined them across a range of parameters. Most of these tools are from the US or Canada, and focus on English-language content and user interfaces; largely, they focus on consumer, customer, and brand relations, but often offer broader thematic analysis, too.

The next paper at Digital Methods is by Nina Springer, who continues the framing theme. Classical framing research examines the impact of frames on audiences, while it is obvious that audiences aren't simply passive recipients, but actively engage with media frames; is this the case especially in new and social media spaces?

This is also a question about the plurality of opinion - the multiplicity of frames may point to the presence of diverse opinions in media coverage. This can also be understood from the perspective of interpretative repertoires: within any repertoire, multiple interconnected frames may exist and involve logics of evaluation, logics of action, and typical actors and structures.

The next Digital Methods panel starts with a presentation by Gianna Haake, whose interest is in framing analysis in social media. Social media texts are often very short, of course, making interpretation and framing analysis very difficult; attempts to analyse content may be inherited from other media forms, too, which may not always be appropriate. Where social media and mainstream media content intersect (e.g. in the form if user comments in news sites), user-generated content could be analysed as media content, or as a reaction to media content.

Frames can be described as persistent patterns of interpretation, and can be broken down into a number of elements which show that there are more than simply thematic contexts for debates; frames support interpretation and action, for example. How may we find indicators for frames, or frame elements, in the short user-generated texts of social media?

The final speakers in this Digital Methods panel are Jürgen Grimm and Christiane Grill. They're interested in moving beyond the analysis of individual tweets to the aggregation of Twitter data which can be used reliably in media research. This requires the use of transparent and clear search or tracking strategies, and a further manual reduction of the data to weed out irrelevant material; further, the intertextual connections of tweets need to be identified and examined, both between each other and with external texts (e.g. from mass media).

The idea in this is to move from an atomistic Twitter universe, based on individual tweets, to a conversational and/or intermedial Twitter universe (variously recognising tweet relationships through @mentions and retweets, or through links and other pointers to external media texts). In the context of the Salzburg state election in Austria, for example, the former means focussing on conversations rather than individual tweets; the latter means identifying all links being shared by Twitter users and generating a hybrid network including tweets and other resources.

The second speaker this morning at Digital Methods is Andreas Jungherr, who shifts our focus back to Twitter: he is interested in how we may use observations from this platform to understand what happens in society as such. What, if anything, may we read out of, for example, the patterns around an election which could help us predict the outcome of the election?

In the German election 2009, for example, Andreas found substantial activity around the Pirate Party, but this is an artefact of the specific demographics of Twitter in the country at the time rather than a sign of genuine pandemic interest in the party. In the same campaign, the volume of political news being shared during the campaign clearly shows the gradual growth of interest ahead of Election Day, and pinpoints key moments like debates and state elections in the run-up.