There are three major stories in this campaign:

1. Who will form government? The incumbent Conservative / Liberal Democrat coalition are fighting the election separately, and frequently in opposition to each other. Experts and polls suggest that a whole range of outcomes are possible, including both majority government for the Conservatives and Labour, however the smart money suggests either a minority government (7/4 Labour, 5/2 Conservatives) or a coalition. Amongst the coalition options, a renewal of the Conservative / Liberal Democrat coalition is possible (6/1), as is a partnership between the Liberal Democrats and Labour (13/2). The Scottish Nationals are another possible coalition partner for Labour (16/1 or 20/1 with the Lib Dems also included), while UKIP and the Democratic Unionists may still be needed to get either party over the line. Unusually for the UK then, we have a lot of electoral calculus in play.
2. How will Scotland vote? As above, the SNP may play a role in a coalition, but Scotland seems set to vote on very different grounds from the rest of the UK. A recent poll suggested that almost half of both Labour and Conservative voters would consider tactically voting for the other major (English) party, if it prevented the SNP being elected in their seat – something that would be unheard of in basically any other seat, in a hangover from the Scottish Referendum. That said, they are still expected to gain a large number of seats in Scotland, and thus put themselves in a position of power at Westminster.
3. Finally, there’s the UKIP factor. As the graphs below show, UKIP are one of the most talked-about parties of the campaign, well ahead of their expected representation in parliament (although, we should acknowledge, the UK has a first-past-the-post system, and UKIP’s representation would likely be higher in a proportional system). Paddy Power suggest that they will receive fewer than 3.5 seats (8/11), compared to 43.5 for SNP, yet the two seem to be roughly equivalent in terms of discussion on Twitter. Of course, pure volume of conversation doesn’t tell the full story, and much of the conversation around UKIP may be negative in nature – yet, our overall sentiment gauge shows little difference between UKIP and the other political parties in that regard. So, this will be interesting to watch — will interest in UKIP die down as their political prospects recede, or will we see a surprise on election day?

Aside from these, our live graphs of Twitter activity around the election tell a story of their own and are designed to be explored on a daily or even hourly basis as the social media conversation shifts in volume, tone and topic during the campaign. Previously, we have seen that major shifts in conversation do not just make for interesting journalism, but can also be cause of speculation about how the election might actually play out. We are expecting to see a more exciting representation of the conversation during this election, compared to our previous work in Queensland and new South Wales, due to the higher volume of election-related social media conversations across the UK.

The most notable change compared to our previous election coverage is the addition of the pie charts, which we think provide a much clearer visualisation of the share of conversation, and clearly compare the whole-of-campaign trends with developments over the past 24 hours. Also included are sentiment breakdowns per party, alongside currently trending hashtags, and an analysis of the total conversation volume over time.

Overall Conversation Share

Sentiment

Volume of Conversation

We plan to release some further graphs as the 7 May election date approaches – including a look at particular battleground seats, and a breakdown of the “positive” conversation around parties.

Hypometer is also undergoing some significant development in the lead-up to our first launch product – also to be released in May. You can follow the progress on the Hypometer project blog.

The New South Wales Election Social Index (NSWESI) tracks a combination of the key hashtags (#nswpol, #nswvotes, etc.), keywords relating to the parties and politicians, and tweets by and @mentions of candidates across all parties, to the extent that we have been able to identify their Twitter accounts. We’re combining the headline figures from our analysis of this dataset into three live graphs which are embedded below, and which are available for embedding on other sites. The graphs are updated with new data every five minutes.

In addition to the analytics for the various parties, and for the two major parties, this time around we are also adding some experimental sentiment analysis. As is by now well-established, sentiment analysis of individual messages – especially when they are as short as tweets – is very difficult, as standard sentiment solutions are struggling considerably with rhetorical devices like irony and sarcasm. However, in aggregate this analysis may still generate some useful overall patterns – but please take them with a grain of salt for now.

NSWESI: Overall patterns across the parties

NSWESI: Major party contest

NSWESI: Sentiment and trending topics

To embed these graphs on your own site, please use the following code:

<img src=”http://dev.thehypometer.com/images/election-allparty.png” style=”width:600px”> <img src=”http://dev.thehypometer.com/images/election-2party.png” style=”width:600px”> <img src=”http://dev.thehypometer.com/images/election-sentiment.png” style=”width:600px”>

Hypometer is a QUT-based commercial start-up which tracks and ranks social media activity around major topics, events and brands with assistance from qutbluebox, the University’s innovation and knowledge transfer company. For more information on The Hypometer, please contact Katie Prowd (k2.prowd@qut.edu.au). For more information about the principles behind Hypometer technology, see the Telemetrics Project.

As in previous elections, the core focus for our social media analytics activities remains on Twitter, though this time we’re also adding Instagram to the mix. For the purposes of our analysis, we are tracking a combination of the key hashtags (#qldpol, #qldvotes, etc.) and keywords relating to the major parties and their leaders, as well as tweets by and @mentions of all of the Twitter accounts associated with local candidates across all Queensland electorates.

But while in earlier elections we’ve posted weekly updates of the major social media trends, this time we’re moving to a real-time visualisation format. Working with the team behind last November’s G20 Hypometer, which was featured extensively on television and in online publications during the course of the event, we’ve now launched the Queensland Election Social Index (QESI):

The QESI Hypometer (click here for full size) combines the Twitter and Instagram data to show the focus of current social media discussions about the election, aggregated by party. Mentions of the major parties within popular hashtags such as #qldpol, #qldvotes and #qldvotes2015 are incorporated in real time, while mentions of the candidates of each political party are added on a slightly delayed basis. Percentage changes are shown on a day-by-day basis, while the volume graph is updated each hour.

A second QESI Hypometer focusses specifically on the two major parties, and also compares mentions of the party leaders, Campbell Newman and Anastacia Palaszczuk, also in real time. This Hypometer (full size here) shows the top trending hashtags within the election discussion.

To date, there has been a notable shift in conversation, from a strong focus on the LNP during the first week of the campaign (at times commanding  more than 70% of the total conversation) towards a greater level of debate about the ALP: towards the end of the second campaign week, the LNP is capturing a little over 54% of the total conversation since the election was declared. Together, Labor and the LNP clearly dominate the conversation, however, with the minor parties recording less than 5% between them. Campbell Newman also remains a far more frequent topic of discussion than Anastacia Palaszczuk, with a share of over 79% of the conversation. In total, at the time of writing the QESI Hypometer has tracked almost 85,000 posts across Twitter and Instagram since 6 January.

Refresh this page every hour or so to see the latest updates on how these numbers are developing. We’ll also post further analysis of key trends and developments over the remainder of the election campaign, and hope to deploy a similar election Hypometer again for the New South Wales election in a couple of months.

The two QESI Hypometers are also available for embedding:

<iframe style="overflow: hidden; height: 700px; width: 100%;" src="http://dev.thehypometer.com/election-allparty/embed" 
width="100%" height="1000px" frameborder="0" marginwidth="0" marginheight="0" border="0" scrolling="no"></iframe>

<iframe style="overflow: hidden; height: 700px; width: 100%;" src="http://dev.thehypometer.com/election-2party/embed" 
width="100%" height="1000px" frameborder="0" marginwidth="0" marginheight="0"></iframe>

Hypometer is a QUT-based commercial start-up which tracks and ranks social media activity around major topics, events and brands with assistance from qutbluebox, the University’s innovation and knowledge transfer company. For more information on The Hypometer, please contact Katie Prowd (k2.prowd@qut.edu.au). For more information about the principles behind Hypometer technology, see the Telemetrics Project.

‘Big Data’

• Jean Burgess: A Decade of ‘Social Media’ and What to Do about It: Methodological Challenges of Historicising the Proprietary Web, presented at the Association of Internet Researchers Conference pre-conference on Internet Histories, Denver, 23-26 Oct. 2013.
• Axel Bruns and Jean Burgess: Easy Data, Hard Data? Twitter Research and the Politics of Data Access, presented at the Compromised Data symposium, Toronto, 28 Oct. 2013.
• Darryl Woodford: Slicing Big Data: Gambling, Twitter & Time Sensitive Information, presented at the Association of Internet Researchers Conference, Denver, 23-26 Oct. 2013.
• Darryl Woodford: Lightning Talk: Social Media Methods & Ethics, presented at the Association of Internet Researchers Conference pre-conference, Denver, 23-26 Oct. 2013.
• Sky Croeser and Tim Highfield: Mapping Movements: Social Movement Research and Big Data: Critiques and Alternatives, presented at the Compromised Data symposium, Toronto, 29 Oct. 2013.

Crisis Communication

• Jean Burgess, Axel Bruns and Avijit Paul: Cumulative and Comparative Social Media Analytics for Crisis Communication, presented at the Association of Internet Researchers Conference, Denver, 23-26 Oct. 2013.
• Axel Bruns: Social Media Issue Publics in Australia, presented at the IBM Research Colloquium 2013, Melbourne, 12 Nov. 2013.
• Avijit Paul: Extracting Important Information from Social Network Stream during Crisis, presented at the Association of Internet Researchers Conference, Denver, 23-26 Oct. 2013.
• Axel Bruns: Social Media in Times of Crisis: The Australian Perspective, presented at a Project EPIC symposium, Boulder, 23 Oct. 2013.

News and Politics

• Axel Bruns, Tim Highfield and Theresa Sauter: #ausvotes Mark Two: Twitter in the 2013 Australian Federal Election, presented at the Association of Internet Researchers conference, Denver, 24 Oct. 2013.
• Theresa Sauter and Axel Bruns: Exploring Emotions on #auspol: Polarity and Public Performance in the Twitter Debate on Australian Politics, presented at the Association of Internet Researchers conference, Denver, 26 Oct. 2013.
• Tim Highfield: Appropriating Breaking News? The Evolving Twitter Coverage of the Lance Armstrong Doping Scandal, presented at the Association of Internet Researchers Conference, Denver, 23-26 Oct. 2013.
• Axel Bruns: Twitter and the 2013 Australian Election, presented at Political Imperatives: Media and the 2013 Election Campaign, Brisbane, 27 Sep. 2013.
• Wilfred Wang: Defending Cantonese Weibo and Geo-Identity Politics in Guangzhou China, presented at ‘The Impact of War on Modern Chinese Society’, University of Queensland, Brisbane, 18-19 Oct. 2013.
• Axel Bruns and Theresa Sauter: Anatomie eines Trending Topics: Retweet-Ketten als Verbreitungsmechanismus für aktuelle Ereignisse, presented at the DGPuk Innovative Methods for the Study of Public Communication workshop, Vienna, 8 Nov. 2013.

Popular Culture

• Ana Vimieiro and Renato Vimieiro: Oscar Pistorius and the Death of Reeva Steenkamp: Mapping Public Opinions Using Frame Analysis and Data Mining, presented at the Association of Internet Researchers conference, Denver, 23-26 Oct. 2013.
• Katrin Weller and Axel Bruns: Friends or Followers: German Soccer Clubs and Their Fans on Twitter, presented at the Association of Internet Researchers conference, Denver, 25 Oct. 2013.
• Tim Highfield: News via Voldemort: The Role of Parody and Satire in Topical Discussions on Twitter, presented at the Association of Internet Researchers Conference, Denver, 23-26 Oct. 2013.
• Theresa Sauter and Axel Bruns: #Pontiff-Ex: The Twitter Community’s Reaction to the Papal Resignation, presented at the Association of Internet Researchers conference, Denver, 25 Oct. 2013.

Platforms

• Kim Osman: Wikipedia as a New Media Institution: Issues of Diversity, Regulation and Sustainability in an Open Encyclopaedia, poster presented at WikiSym & OpenSym 2013, Hong Kong, 5-7 Aug. 2013 (paper at QUT ePrints).
• Wilfred Wang: Identity, Locality and Social Media in Guangzhou China, presented at the Fifth Postgraduate Symposium: Identities, Ideas, Ideologies, Griffith University, Brisbane, 1 Nov. 2013.

That should be enough for 2013! See you next year at a conference somewhere…

As I’ve explained in the previous posts, what we’re doing in our election research is to track all tweets by and @mentions of sitting members and candidates in the 2013 federal election. As more (especially minor party) candidates have become known, we’ve progressively extended our list as far as possible.

In this post, we’re relating such activity to the electorates in which the candidates are standing (this necessarily excludes Senators and Senate candidates, therefore, who don’t have local electorates). The resulting maps show the local electoral races which have received especially much attention on Twitter, because of local issues or because of the national prominence of their local candidates. They also make for interesting reading alongside Guardian Australia’s map of where the leaders have been making campaign stops so far.

In the maps which follow, we’ve shown Twitter activity since the start of July. The stronger the red colouring of an electorate, the more @mentions its candidates have received – and just to make this absolutely clear, that Twitter attention may have come from anywhere in the country (and even from overseas), not just from Twitter users who are actually based in the local electorate itself. (Incidentally, for those of you who are interested in such things: colouring is applied on a logarithmic scale – otherwise only Rudd’s and Abbott’s electorates would show up in bright red.) Click to enlarge the images.

First, then, here’s a national overview, which largely reflects where the major population centres are: there’s more activity in the Sydney, Melbourne, and Brisbane / Gold Coast regions than anywhere else in the country. However, a few outliers appear even from this way out – Bob Katter’s north Queensland seat of Kennedy is more active because Katter has received significant national attention (or notoriety, take your pick)  as the leader of a fledgling political party. Similarly, the northern and central NSW seats of retiring Independents Tony Windsor and Rob Oakeshott are receiving attention because both of them have been such important figures in Australian politics over the past legislative period, with further attention on New England (just south of the Queensland border) because LNP Senator Barnaby Joyce is now a National Party candidate for that seat.

Further south in NSW there’s also the famous bellwether seat of Eden-Monaro, held by comparatively active Twitter user and Labor minister Mike Kelly – and across the border to Victoria, the seat of Indi, held by controversial Liberal member Sophie Mirabella is also receiving substantial attention. This may be related to recent reports suggesting that Mirabella is faced with a credible challenge from her Independent opponent Cathy McGowan this time around.

Also notable on this overall map, however, are the mostly rural and remote electorates in which Twitter does not yet play any role at all (in grey) – these are electorates whose candidates either don’t have Twitter accounts, or haven’t yet received a single @mention since July. This also provides a reflection on where the use of Twitter may have an impact during the campaign of course.

Most of the major centres are difficult to make out from a distance, however – time, then, to zoom in for a closer look at the various state capitals and surrounding areas. (The numbers behind each electorate name indicate the combined number of @mentions its candidates have received since the start of July.) In Queensland, the electorate of Griffith (home to PM Kevin Rudd) necessarily leads the way, but a number of other interesting flashpoints also emerge. These include Rankin, where retiring MP Craig Emerson has chosen to go out with a bang rather than a whimper; Forde, where former state Premier Peter Beattie has been recruited to reclaim the seat for Labor; as well as Lilley, held by Deputy PM-turned-backbencher Wayne Swan. On the Sunshine Coast, former Speaker Peter Slipper’s seat of Fisher is as hotly contested on Twitter as it is elsewhere – and just to the north of it, billionaire Clive Palmer is making his well-publicised run for parliament.

In Sydney, a number of city seats predictably show up in red mainly because they are held by leading federal politicians including Opposition Leader Tony Abbott (Warringah), Joe Hockey (North Sydney), Scott Morrison (Cook), and Malcolm Turnbull (Wentworth). Prominent Labor members from the area include Deputy PM Anthony Albanese (Grayndler), Tony Burke (Watson), Tanya Plibersek (Sydney), and Ed Husic (Chifley). The substantial level of activity around each of these seats should not surprise us, therefore.

In Melbourne, sole Greens MP Adam Bandt is the centre of much attention in the eponymous electoral division as he attempts to hang on to his seat. Next door in Maribyrnong, Labor powerbroker Bill Shorten also attracts considerable @mentions, and opposition climate change spokesman Greg Hunt’s seat of Flinders flares up not least because of the considerable number of tweets by smaller parties which have been targetting him over the Coalition’s climate change policy (from both scientific and denalist perspectives, no less). Most active is Lalor, however, where former PM Julia Gillard, retiring at the election, is still receiving substantial @mentions even though she has tweeted only once since 1 July. The darker red area in the north-east is the lower end of Indi, incidentally.

The most active electorates in the Adelaide area are Port Adelaide, held by Mark Butler, and Adelaide, held by Kate Ellis – both of them Labor ministers. As far as I can tell, there’s only a fake Chris Pyne on Twitter, and we’re not counting his @mentions towards the electorate of Sturt, so his electorate remains inactive to date – though we may not have picked up on the Twitter activity around the other local candidates there.

The Perth area shows even more grey as soon as we get further away from the city itself. Here, opposition foreign policy spokeswoman Julie Bishop’s Curtin electorate is most active, unsurprisingly; this is followed by Dennis Jensen’s Tangney division and – more interestingly – by Australia’s most marginal Liberal seat of Hasluck, held by the nation’s first indigenous MP, Ken Wyatt.

The remaining state and two territories can be covered very quickly – nothing much to see here. In Tasmania, only Independent Andrew Wilkie’s seat of Denison shows any signs of life (and then, not much):

In the ACT, Labor’s Gai Brodtmann isn’t generating much activity for the Canberra electorate, while Fraser (with its strange coastal exclave, the Jervis Bay Territory – hence the equidistant placement of its label on our map) is held by Labor MP Andrew Leigh, an active blogger and Twitter user who therefore attracts considerably more @mentions:

Finally, neither of the two Northern Territory electorates are earthshatteringly active. However, the fact that Labor MP Warren Snowdon’s immensely large electorate of Lingiari generates any Twitter activity at all is itself noteworthy, perhaps, compared to the poorer performance of many of the other rural and remote electorates across WA, SA, and Queensland.

So much for a first glimpse at the geographic distribution of Twitter attention across the federal electorates, then. We’ll check in again later in the campaign to see whether Twitter activity provides any indication that new electorates have come into play as the parties’ fortunes wax or wane.

This post continues my analysis of Twitter data concerning the WA state election, held on 9 March 2013: see this post for an introduction to this project. Previous posts have looked at election-related hashtags, candidates’ posting patterns, and the networks formed between candidates through @mentions, replies, and retweets.

It’s now been over three weeks since the WA state election, which was ultimately won convincingly by the Liberal Party, with a continued swing towards them following their victory in the 2008 state election. Although the analytical posts have been absent since election day, the research has been continuing behind the scenes – as mentioned previously, Axel Bruns and I will be co-authoring a paper about the election as seen from Twitter for the ANZCA conference, to be held in Fremantle in July.

For now, then, a brief return to the datasets to have a look at what happened on election day. In this post, I’m focusing primarily on the #wavotes hashtag, as the main marker used during the second half of the election campaign – later analysis will return to the other campaign-related hashtags and keywords tracked, though. As always with these posts, the usual caveats apply here: the users and views collected are not representative of the general electorate, and not all relevant tweets will have been archived here. The yourTwapperkeeper data capture does not have access to the unlimited stream of tweets from the Twitter API, and in particular, ‘button’ retweets are not always archived. In addition, the network maps do not imply any agreement or support of the users connected (or that the context is an election-specific tweet), but merely that a connection exists through one or more tweets.

Activity using the #wavotes hashtag saw a gradual increase as the election approached, with a previous spike noted during the televised leaders’ debate in February. This is in keeping with other election campaigns covered on Twitter, from Queensland to Scandinavia; similarly, election day itself led to the largest spike in activity, following patterns established in other studies of elections on Twitter. The graph below shows the intensity of the election day spike in #wavotes tweets – jumping from under 2,000 tweets the day before to over 10,000 tweets collected on 9 March itself (see too the similarly dramatic drop-off the day after the election).

If we then look at election day itself, though, tweeting #wavotes does not follow a regular pattern across the entire day. Instead, breaking the day down into hour-long sections, we see that there is somewhat uniform levels of tweeting during the early polling period itself, from 8am onwards, and then increased activity as the close of polls and announcement of results approaches. The results and analysis phase of election day accounts for the main #wavotes coverage here; this is, however, fairly short-lived (decreasing activity after 9pm), as the overall election result itself was called early on the night. Rather than hours of speculation about the possible results, the Liberal Party’s success was evident soon into the count, and commentary instead turned to the extent of this victory.

Not only did election day provoke increased #wavotes activity, but these tweets were also from a greater range of Twitter users. Earlier in the election period, the #wavotes network was focused more directly on the candidates and parties involved in the campaign (with some participation by said candidates, although not to the same extent as in the candidate-specific maps featured in previous posts). However, the heightened #wavotes activity on election day (and the days immediately surrounding it) also generates a network that, in a sense, is more generally covering the election; as the results come in, observers from WA and interstate comment on the trends without necessarily discussing (or needing to discuss) the individual electorates involved. Below is a section of the network map from the #wavotes tweets (based on tweets published between 8 and 10 March, filtered to only include nodes with a degree range of ten or more). As with previous maps, I’ve coloured nodes based on their party affiliation, although only for state candidates. However, the activity by these accounts, with the primary exception of the WA Liberals Twitter account, was relatively minor on election day itself.

To show more clearly the presence of these accounts within the map, then, I’ve set the edge colour to reflect the target, rather than the source, of the edge: a red edge is then directed at a WA Labor account, not necessarily from it (e.g. a tweet mentioning @MarkMcGowanMP). What is then apparent is that, while there was some party-related activity, a lot of the #wavotes connections link users who were not directly involved in the campaign (this includes federal politicians, such as Julia Gillard, Tony Abbott, and Julie Bishop (who was a commentator for the ABC on election night)). Instead, further clusters of users are identifiable within the network; one of these I have highlighted below in yellow, which refers to accounts of ABC journalists, and ABC radio and television stations and programmes. Given the timing of this activity around election night, the centrality of analyst Antony Green is not unexpected here; however, the mass of yellow around Green highlights not just that the ABC was a primary source of information for election coverage, but also the rebroadcasting role of many of these ABC accounts – Green’s own tweets were subsequently retweeted by other ABC accounts, and spread further by other users following these additional, regional or town-specific accounts.

While the extent of the swing to the Liberal Party was the story of the election night (if the success of the Liberals overall was not unexpected, the results and close races in individual electorates such as Perth and Midland were), during election day there was concerted Twitter activity on another aspect of the voting process; the democratic act of going to a local polling place, voting, and then partaking in a sausage sizzle. The night before election day, Twitter user @bogurk floated the idea of a central hashtag – #democracysausage – in order to collate information about polling places with or without food options. During election day, as users tweeted the sausage sizzle status of their local schools, churches, and community centres, the relevant details were then used to generate and update a #democracysausage Google Map.

Unsurprisingly, the #democracysausage activity follows a different pattern to #wavotes on 9 March; where the latter hashtag spiked during the coverage of the results, #democracysausage – as an election day activity – peaks around midday (lunchtime!), and tails off as booths close, as seen in the graph below. In particular, the number of original tweets decreases after mid-afternoon – while booths remained open until 6pm, the possibility of sausage sizzles or the need to provide information about previously-tweeted sites may decrease in the last few hours of voting. However, it should also be noted that this was an experimental hashtag with a few hundred relevant tweets in the archive, which was only suggested the night before election day and with further aspects developed over the course of the polling day (many of the retweets here, then, were spreading the initial idea behind the hashtag). The people behind the hashtag and map have expressed their keenness in repeating the project at the national scale for the federal election, which will then provide the opportunity to see whether or not these patterns are replicated. [For more, see the @DemSausage Twitter account.]