VoterEcology Summary Report

As a new year just began, we thought it would be useful to look back at the one just past and put together a brief overview of our project’s findings so far. Given the relevance of Google searches to political communication scholarship as well as practice, the aim of this paper is to be informative but at the same time also concise and jargon-free. With examples from the U.S., the UK, Italy and Egypt, this paper points out both challenges and opportunities of embracing Google Trends as not only a platform for business analyses and economic forecasts, but also a ‘knowledge’ tool in much broader contexts.

To download a PDF copy of the VoterEcology Key Findings Report, click here.

Advertisements

New Publication – Forthcoming Article in First Monday

Filippo Trevisan’s article “Search Engines: From Social Science Objects to Academic Inquiry Tools” was accepted for publication in the inter-disciplinary Internet studies journal First Monday. This piece addresses the challenges and opportunities involved in integrating online tools such as Google Trends in social science research. Publication is scheduled for November or December 2014 – watch this space for a link once the article is available online!

Abstract:

This paper discusses the challenges and opportunities involved in incorporating publicly available search engine data in academic scholarship. In recent years, an increasing number of researchers have started to include tools such as Google Trends (http://google.com/trends) in their work. However, a central ‘search engine’ field of inquiry has yet to emerge. Rather, the use of search engine data to address social research questions is spread across many disciplines, which makes search valuable across fields but not critical to any one particular area. In an effort to stimulate a comprehensive debate on these issues, this paper reviews the work of pioneering scholars who devised inventive – if experimental – ways of interpreting data generated through search engine accessory applications and makes the point that search engines should be regarded not only as central objects of research, but also as fundamental tools for broader social inquiry. Specific concerns linked to this methodological shift are identified and discussed, including: the relationship with other, more established social research methods; doubts over the representativeness of search engine data; the need to contextualize publicly available search engine data with other types of evidence; and the limited granularity afforded to researchers by tools such as Google Trends. The paper concludes by reflecting on the combination of search engine data with other forms of inquiry as an example of arguably inelegant yet innovative and effective ‘kludgy’ design (Karpf, 2012).

Context, context, context!

— To watch a short video extract of Dounia Mahlouly’s presentation at the Citizen Media Workshop that inspired this blog post, click here. —

Truly valuable analyses of the socio-political implications of new media technologies –including search engines – ought to be rooted in the local context of the country or countries under scrutiny, with particular attention to pre-existing media systems and the relationship that citizens have with them. While some of us have focused elsewhere on the need for the Internet not to be analysed in isolation from systemic and circumstantial contextual elements, we were reminded again of the importance of this approach when we gave a presentation at a workshop on citizen media hosted by the University of Manchester a few weeks ago. This was a great opportunity to meet other researchers exploring emerging forms of political expression and engagement in Russia, Central and Eastern Europe, China, East Asia and the Arab world. Discussions at the event led us to consider some of the specific traits of the Egyptian case within the country case studies selected for the VoterEcology project, as well as on the usefulness of the comparative approach. Expanding our perspective on the use of citizen media helped us to reflect on the inadvertent assumptions and potential ‘ethnocentricity’ that often characterise comparative work on non-Western countries by Western – or more simply West-based – scholars.

Recent literature about the Arab Spring uprisings has veered increasingly towards a critique of Eurocentric and ‘Orientalist’ interpretations of the 2011 protests. Researchers such as Rabab El-Mahdi (2011) and activists such as the members of the ‘Mosireen’ collective have warned about Western analyses according to which social media and cyber-activism would constitute major forces for the democratisation of the Middle East. In their work, these alternative voices have criticised Western scholars for their deterministic approach to new communication technologies and for not paying sufficient attention to the context(s) in which the 2011 revolutions took place. In particular, they pointed out a tendency for Western accounts of the Arab Spring to promote a unilateral perception of democratisation and modernity based on the supposedly Western features of the ‘liberal’ youth in Middle Eastern and North African (MENA) countries. This approach – critics claimed – supported the rise of an ‘Arab awakening’ narrative that has influenced not only the ways in which traditional mass media reported the event, but also the discourse of Western governments, as exemplified by the speech that US secretary of State John Kerry gave in Cairo at the beginning of November 2013.

In our work so far, we found that a useful strategy to operate outside such constraints and account instead for fundamental cultural or ideological differences is to focus on how new media technologies are employed rather than asking which tools or devices galvanised the revolutions. While pretty straightforward, this approach has enabled us to avoid addressing the historical evolution of the Arab world by means of unsuited benchmarks and ‘foreign’ concepts. In other words, it is best to focus on the local particularities in order to avoid inappropriate assumptions when examining democratisation in the Middle East. Arguably, the notion of democracy that is prevalent in Western societies stems from a specific ideological framework and therefore fails to reflect the hopes and aspirations that motivate the Egyptian people in the quest for their own democratic development. Considering specific parameters in the study of search engine use helped us circumventing what Edward Said (1979) described as a distorted vision of the East as a ‘projection’ of the West.

Crucially, our study of key search trends in the Egyptian Internet during the presidential elections of 2012 and the coup of summer 2013, was consistent with the idea that democratisation involves the use of new communication technologies to raise issues that traditional mass media fail to express. Even more so, our findings suggested that – unlike in established democracies such as the UK and the U.S. – Egyptian citizens were especially likely to search for an alternative and emerging political discourse. This called for a reflection on the correlation between democracy and political awareness, as well as on the changing configuration of the public sphere in countries that have undergone dramatic regime changes and whose citizens used to be entirely reliant on state-controlled news outlets until recently. In other words, our comparative approach revealed that in Egypt Internet search fulfilled a profoundly democratic function by enabling informational practices aimed at overcoming the structural limitations of the local mass media context. In turn, such a context-oriented use of technology provided a new perspective from which to examine the question of citizenship in Western democracies.

Our observations add to a growing body of evidence that has shown how the main contribution of social media and cyber-activism in the 2011 protests was to facilitate logistics, circumvent censorship and draw the attention of the international community to citizen-powered change in the MENA region (Breuer, 2012). In addition, authors such as Gerbaudo (2012) argued that the online activist campaigns that brought down authoritarian leaders in countries such as Egypt and Tunisia were highly interdependent with street activism and constituted but one of the channels through which citizens recovered their freedom of expression. Another useful example is that of the role of Twitter in the Tunisian revolution. Whereas Western news media narratives enthusiastically portrayed the Tunisian revolts as a ‘Twitter revolution,’ Salem and Mourtada (2011) demonstrated that in fact the number of Twitter users in the country was lower than other social networking platforms. Since then, however, Howard and Hussain (2013) have noted that social media use has been on a steep rise following the revolution. Overall, this is testimony to the fact that, far from operating on a blank canvas, new media always grafts itself onto specific elements of the existing cultural, political and historical context.

Exciting cross-media transformations have occurred since the Arab Spring protests, with dissident bloggers such as Slim Amamou being appointed as a junior minister in the Tunisian interim government of 2011 and YouTube satirist Bassem Youssef offered a programme on a mainstream TV channel in Egypt. Having said that, in order to fully understand the pace and significance of this change, researchers ought to be wary of their own perspective, alert to the assumptions that come with it and capable of breaking free from the dominant narrative if necessary.

—————-

References:

Breuer, A. (2012). The role of Social Media in Mobilizing Political Protests, Evidence from the Tunisian Revolution, Bonn: German Development Institute.

El-Mahdi, R. (2011). Orientalising the Egyptian Uprising. Jadaliyya. http://www.jadaliyya.com/pages/index/1214/orientalising-the-egyptian-uprising [10.02.2014]

Gerbaudo, P. (2012). Tweets and the Streets: Social Media and Contemporary Activism. London: Pluto Press.

Howard, P. N. and Hussain, M. M. (2013). Democracy’s Fourth wave? Digital Media and the Arab Spring. Oxford: Oxford University Press.

Said, E. (1979). Orientalism. London: Vintage Books.

Salem, F. and Mourtada, R. (2011) The Role of Social Media in Arab Women’s Empowerment. Arab Social Media Report. 1(3). 1-26.

Viswanathan, G. (2004). Power, Politics and Culture, Interviews with Edward W. Said. London: Bloomsbury.

Egypt and the Symptoms of a Global Revolution: Alternative Politics

Last October, we presented some preliminary findings from our Egyptian case study at the Political Studies Association’s (PSA) “Challenges to Campaigning” conference at Edinburgh University. A key theme that dominated much of the event was that of low election turnout and decreasing interest in politics among young voters worldwide, especially in “established” democracies such as the UK and the U.S. In turn, this highlighted the uniqueness of the Egyptian case, suggesting that successful youth mobilisation in Egypt could outline the conditions for future manifestations of political engagement elsewhere in the Middle East and, possibly, further afield too.

In particular, our study of search trends during the 2012 Egyptian presidential election revealed that Google use at such a key political juncture was hardly correlated to the content of traditional mass media (i.e. established print and broadcast news providers). Instead, search engine users showed greater interest in alternative political agents such as famous activists, youth organisations and young martyrs of the revolution during the entire period of the Egyptian political crisis, from the 2011 revolt against Mubarak’s authoritarian regime to the army coup that ousted Mohammad Morsi earlier this year. Without being involved in a traditional political structure, these new opinion leaders tried to reach a significant audience by using blogs, digital footage and social media platforms. Crucially, this case study showed the potential outcome of a proactive application of social media and digital technologies by opposition movements when national mass media outlets and traditional political institutions are no longer regarded as reliable.

Internet use in Egypt as well as elsewhere in the Middle East increased exponentially over the past decade, connecting a new mass of radicalised youth. Indeed, online communication flows have not only provided opposition movements with the ability to share ideas and ideologies, but also enabled young users to circumvent social taboos (Wheeler, 2006) and experience unconstrained relationships. In light of this, one could reasonably argue that the Egyptian youth has become the most prolific and proactive contributor to the nation’s newly formed “online public sphere,” relaying street protests and promoting opposition movements when traditional mass media outlets were restricted by the government as well as the military.

These trends resonate with broad strands in communication and media theory. In particular, one of the common assumptions shared in contemporary media studies lies in the fact that participative digital technologies confirm Lazarsfeld’s theory of the two-step flow of communication (1955), which states that every individual acts as a potential opinion leader within their social circle. As they rediscover the significance of interpersonal relationships (Lazarsfeld: 1940), social scientists examine whether participative and connective interactions enable new forms of political engagement by increasing the influence of alternative opinion leaders. Such a hypothesis appears to be particularly relevant in the Egyptian context, as social media and digital technologies succeeded in leading the opposition against Mubarak’s former government and achieved efficient political action. Given that the Internet is applied as an alternative source of information and a tool for resistance, the use of search engines in Egypt is consonant with citizens’ interest for a new range of political stakeholders that clash with traditional representative structures. More broadly, this reveals the opening of a new type of political space that is extremely fluid and at the moment seems unlikely to become stabilised any time soon.

Check for updates here as we develop our Google Trends data analysis and further compare the role of search engines in shaping emergent versus established democracies. If you wish to read more about our initial considerations on Internet search in the new Egyptian media ecology, click here <Mahlouly_Google and Egyptian Elections> to download Dounia Mahlouly’s paper for the “Challenges to Campaigning” PSA conference.

—————–

References:

  • Berelson, B., Gaudet,H and Lazarsfeld, P.F. 1944. The people’s choice: how the voter makes up his mind in a presidential campaign. New York: Columbia University Press.
  • Kats, E. and Lazarsfeld, P. F. 1955. Personal Influence. New York: Free Press.
  • Wheeler, D. 2006. ‘Empowering publics: Information Technology and democratization in the Arab World-Lessons from Internet cafés and beyond.’ Oxford Internet Institute (11), pp. 1-18.

Under the Tip of the Iceberg

A couple of weeks ago we presented some of the Voter Ecology Project’s preliminary findings at the ECREA Communication and Democracy annual symposium in Munich, Germany. Among the many questions raised in the Q&A session, one went right at the heart of the challenges with which we have been getting to grips in the past couple of months, which led us to be wary of applying traditional political communication research methods in the context of a new media ecology. In particular, someone asked how exactly we were planning to “disentangle” what Internet users were interested in because they had seen it on TV from what traditional news media covered by responding to a rise in popular demand for information on a certain topic or event. At first, this may sound like a chicken and egg question. However, this is not really an issue of causation, but rather one that speaks to the challenge of mapping and understanding increasingly complex informational trajectories in a context in which online content and that of ‘legacy’ media are ever more intertwined.

Since writing about the need to develop a citizen-centric approach to study electoral information flows in our first blog update, we spent quite a bit of time dealing with methodological riddles. Having hypothesised the existence of a “search agenda” populated by issues that reflect the “true” interests of Internet users, a key problem was to find a viable system to map and compare that set of issues against those that dominated news media coverage of electoral affairs and official campaign messages respectively. As media scholars, our first intuition was to use content analysis to produce a comprehensive overview of the news content and campaign literature surrounding elections in each of the countries under scrutiny (the UK, the U.S., Italy and Egypt). However, we soon realised that this would have been not only extremely time-consuming and resource-intensive, but would have constituted also a potentially distortive application of old research frameworks in a new context.

As it is often the case in social science research, a breakthrough occurred when we realised that in fact we were asking the wrong question, starting from what was ‘old’ rather than what was ‘new’ in the informational environment that typically surrounds electoral contests. This newly found awareness inspired us to ‘reverse’ the traditional research process, using Google Trends data to identify the moments in which demand for online political information was particularly intense and then carry out an in-depth analysis of these informational ‘junctures.’ For example, we found that, although British Internet users were relatively uninterested in Gordon Brown during the campaign for the UK 2010 general election, search levels for information related to the then prime minister and Labour party leader suddenly exploded towards the end of April to then decline just as sharply a couple of days later. As it turned out, this search peak coincided with the ‘bigot-gate’ episode, which was arguably a watershed moment in the 2010 UK general election campaign and some described as one of the most over-reported stories of that year. Similarly, search levels for information related to Mitt Romney went through the roof in conjunction with the ‘binders full of women’ gaffe he made during the second televised debate in which he and Barack Obama took part during the 2012 U.S. presidential campaign.

Such extreme search patterns invited further investigation to expose in full the informational transactions that surrounded each of these sharp rises in voter interest in political content. Indeed, shining the spotlight on such intensely mediated political events while at the same time hypothesising the existence of a specific – but not necessarily independent – ‘search agenda’ may sound rather paradoxical. However, a true contradiction in terms would arise here only if one were to mistake this for the arrival point in the research process, when in fact it is quite the opposite. Identifying the moments in which citizens are most likely to be interested in political information with Google Trends constitutes a first level of investigation that ought to be followed by a number of other questions. In this framework, finding that the same episodes had generated a vast amount of interest across both online and traditional mass media did not necessarily mean that each of these events also occupied exactly the same position on the ‘search agenda’ as it did on the agenda of traditional news media. Rather, as researchers and innovators our job is to go beyond the tip of the iceberg, reaching underwater to identify the deep ramifications of key socio-political events.

Thus, for each of the episodes listed above as well as for others, we are in the process of asking what role(s) they played and how long they featured in the agendas of search engine users, traditional news media, and campaign managers respectively. In addition, and perhaps more importantly, we are also testing whether voters used such seemingly trivial episodes as ‘springboards’ to complete “thematic leaps” (that is, to start researching or spreading information on a policy issue connected to the events in question). Although this is still very much work in progress, it has been interesting to see how Google Trends provides opportunities for researchers to revisit events that have become crystallised in the academic literature on the mass media and elections with a view to finding different angles from which to interpret them.

To browse the complete series of presentation slides from the ECREA Communication and Democracy 2013 conference click here:

Turning the Pyramid Upside Down

By Filippo Trevisan, 6th Aug. 2013

Joe Trippi, campaign manager for U.S. presidential hopeful Howard Dean in 2003-4, once explained that in the era of online campaigns “What you’ve got to do, is you have to have two-way communication. It’s the bond, to be able to talk to each other about you, that is important.” At the time, such words may have sounded like heresy to those used to traditional “top-down” campaigns. Ten years on, they make instead for basic textbook remarks on democratic electioneering. Campaigning has come a staggeringly long way in a very short amount of time. Politicians want citizens to get in touch through online media because it helps them build a winning campaign. Yet, can we be sure that citizens are equally as eager to hear from politicians as Trippi’s remarks seem to imply? Couldn’t it rather be that today’s voters, with so many online resources at their disposal, are building their own “election agenda” instead of relying on “managed” campaign information? After all, one could be forgiven for thinking that being able to track Barack Obama’s favourite dining spots in real time can only prelude to a better-than-ever-informed citizenry.

The idea of emancipated voters who make the most of their technological proficiency to break free of electoral information hegemony is a captivating one, but the truth is that we still don’t quite know whether this actually happens. In fact, doubts on the pluralising effects of online campaigning started to emerge in the very early days of Web 2.0 technology. Shockingly for cyber-enthusiasts, scholars such as Philip Howard (2006) pointed out that by “redlining some constituents and communities and then narrowcasting political content, hypermedia campaigns diminish the amount of shared text in the public sphere” (p. 183). In other words, personalised information feeds and tailored messages would restrict, not enhance, pluralism and participation. From a pragmatic point of view, this is perhaps not as much as a surprise as it may seem at first. As Kreiss (2012) explained in his work on the Dean and Obama campaigns, “campaigns simply are not designed to be the training grounds of radical democratic participation that many desire” (p. 183). Thus, “many supporters not only accept but embrace this, given […] the objective is to defeat rivals, not remake democracy” (ibid.). That said, our knowledge of information flows at election times remains confined to what Lilleker and Vedel (2013) have termed “supply-side studies,” which focus on campaign messages rather than asking where else citizens get their information from before heading to the polls.

To use a metaphor, it’s time for researchers to turn the pyramid upside down. That means, start by asking how voters navigate and interact with the virtually limitless “sea” of online information, as well as mapping those practices against campaign messages and traditional news coverage of election issues more broadly. To this effect, some have analysed how people use Twitter before, during and after televised debates and current affairs programs. That has yielded some interesting results, but at the same time it’s a bit like putting the cart before the horse or preparing the icing before even looking for a cake recipe. Despite increasing in numbers, being particularly vocal and arguably influential, Twitter users remain a minority among internet users. What, then, about the broader picture? Some have sounded internet users through traditional surveys, but one has to wonder whether there are better ways of capturing informational practices in the digital age. We believe that a good place to start in this process could be mapping user-interests using Google Trends: What types of information do voters search for in times of elections? How do they search? Are they guided by established mass media templates and official campaign messages, or capitalise on the relative freedom provided by search engines to create their own “search agenda”? And finally, what role does search occupy in the nascent “new media ecology”?

As search contends with email to be the most popular online activity in the U.S. and remains the primary way through which British users access information on the internet, acquiring a detailed overview of search patterns would ensure as representative a base as possible for comparing the interests of online citizens to the agendas of both politicians and traditional mass media outlets, which can be mapped using more traditional methods. For example, Google Trends suggests that British internet users became increasingly interested in knowing about a “hung parliament” as the 2010 general election approached despite politicians doing their best to avoid talking about this issue.

Fluctuations in search volume for “hung parliament” on Google.co.uk, March-April 2010

Fluctuations in search volume for “hung parliament” on Google.co.uk, March-April 2010

Was it traditional mass media coverage that inspired voters to search for information on this possible scenario, or were they rather using search to remedy the lack of information on this issue in mainstream channels? Indeed, unravelling this type of issues is bound to throw up some important methodological challenges, as I discussed recently in a dedicated paper, and require an inductive approach, as search trends are likely to be uncovered by testing for a wide range of potentially relevant keywords rather than merely anticipated on the basis of theoretical assumptions. That said, this promises to be an exciting journey in which, once overarching trends as well as standout search “events” have been identified through Google Trends, multiple methods can be employed to map how key messages are generated, shared, and re-framed across different media platforms in times of elections in a process that could be conceptualised as “nexus analysis.”

——————

References:

  • Howard, P. (2006). New Media Campaigns and the Managed Citizen, Cambridge: Cambridge University Press.
  • Kreiss, D. (2012). Taking Our Country Back: The Crafting of Networked Politics from Howard Dean to Barack Obama, New York: Oxford University Press.
  • Lilleker, D. and Vedel, T. (2013). “The Internet in Campaigns and Elections,” in Dutton, W. (ed.), The Oxford Handbook of Internet Studies, Oxford: Oxford University Press, pp. 401-420.