Australia New Zealand Communication Association 2013

#wavotes: Tracking Candidates' Use of Social Media in the 2013 Western Australian State Election

Tim Highfield and Axel Bruns

• 5 July 2013 – Australia New Zealand Communication Association conference, Fremantle

Social media have become an established part of online political debate, being used by politicians, broadcasters, journalists, and activists alike, from tweeting about televised panel shows and debates to incorporating Twitter and Facebook into protest movements. These patterns are being replicated internationally, and social media use within national politics has been studied in contexts from Australia (Bruns & Burgess, 2011; Grant, Moon, & Busby Grant, 2010) to Austria (Maireder, Ausserhofer, & Kittenberger, 2012).

Federal politics, though, is not the only level of government that affects citizens’ lives or shapes political discussion; while in Australia social media, in particular Twitter, have attracted greater interest from federal members of Parliament, for example, than other web-based technologies, such as blogging (Highfield & Bruns, 2012), their adoption at the state level is more varied. While Twitter a widely-used platform for discussing national politics and breaking news, state politics is not as popular a topic – and there is variable interest in, and participation by the political actors discussed by, tweeting about state politics (Highfield, Bruns, & Harrington, 2012).

In this paper, we analyse Twitter activity about, and during, the 2013 Western Australian state election campaign, in order to evaluate how social media are used at the state political level by candidates, parties, journalists, and voters alike. This examination of the role of social media within state politics and campaigning builds on previous research into both the 2010 Australian federal election (Bruns & Burgess, 2011)and the 2012 Queensland state election (Bruns, Harrington, & Highfield, 2012); the methods and findings of this analysis will also inform the study of the federal election in September 2013 by the research team.

For the Western Australian state election, we track uses of Twitter during the campaign from its unofficial start at the beginning of 2013 through to the election day of 9 March 2013. The Twitter activity covers election-specific hashtags, such as #wapol, #wavotes, and #wa2013vote, as well as keywords around major policy announcements and candidates. We also track the tweets published by candidates, sitting politicians, and the main parties, as well as the interactions between these accounts and other users, by following more than 100 Twitter parliamentarian and candidate accounts. This approach allows us to provide a more rounded analysis of the campaign period as it happened on Twitter, rather than focusing on a single hashtag for the data collection. The methods used for the data collection and analysis for this paper have been developed by the research team for the ongoing study of large Twitter datasets across a variety of contexts, from political debate and crisis communication to live commentary on media events, such as sports and television broadcasts.

Key aspects to be explored in this analysis include the level of participation by official candidate accounts in the continuing hashtag conversations; the degree and nature of intra- and inter-party interaction between candidate accounts; and the overall network of communicative interconnections between Twitter accounts discussing the Western Australian election. Examining these aspects of the campaign allows us to determine how the election is covered on social media, identifying tweeting rhythms, including which events provoke spikes in tweeting activity and how these compare with previous studies of election campaigns on Twitter. The connections created within tweets mentioning different users allow us to also evaluate whether candidates and parties cross partisan lines, analysing, for example, the extent to which candidates tweet at their opposition. Such connections also enable an examination of any bridging effect, by candidates or other users, connecting any partisan clusters of Twitter users.

The tweets published by official candidates during the campaign are used to identify different uses of social media by political candidates and parties. Previous research has established various approaches by politicians on Twitter, from simply broadcasting their messages to repeatedly engaging with citizens’ comments (Broersma & Graham, 2012). Within the Western Australian context, we examine the tweeting patterns of active candidates to examine whether Twitter is being used primarily for sloganeering, communicating solely with other candidates or party members, or for engaging in public discussion. Looking at the users tweeting at, and being replied to by, the different candidates, we also investigate whether these other users follow the model of the ‘political junkies’ described by Coleman (2006), especially militant supporters of major parties, or if they are more representative of a wider range of views and voters.

The Western Australian context also provides an important comparison with both the Queensland and federal elections, as Twitter has not been universally adopted by all parties involved in the campaign; in Western Australia there is no official Twitter account for the Premier and leader of the Liberal Party, Colin Barnett. Indeed, the Liberal Party has appeared resistant to social media ahead of the election, with increased management of the party’s messages to avoid gaffes and unnecessary incidents attracting attention – especially when the party is expected to win the election. Previous research into online political communication, from blogging to tweeting, has investigated whether discussion is divided along partisan lines, where the people involved primarily link to and respond to others of similar opinions or party affiliations to themselves. However, the absence of prominent WA Liberal accounts from Twitter means that the tweeted discussions during the state election may take a different form than in other states; without these focal conservative accounts to tweet at or respond to, what forms do the campaigns and commentary from other parties, and voters, take during the campaign?

The analysis for this paper then forms part of ongoing research into political campaigning and debate on social media, both within Australia and worldwide. It provides new insights into approaches to social media campaigning, and acts as an important comparison between state-level politics and national political debates on Twitter.