From Political ‘Surf’ to Political ‘Turf’?: Developing Website Analysis to Better Understand the Internet as a Political Catalyst
Paper prepared for the 6th Annual APSA Pre-Conference on Political Communication
Old Media, New Media: Political Communication in Transition
Co-sponsored by the Political Communication Section of the American Political Science Association and the JoanShorensteinCenter on the Press, Politics, and Public Policy
HarvardUniversity
Boston, Massachusetts
August 2008
Prof. Sarah Oates
Politics Department
University of Glasgow
Glasgow G12 8RT
United Kingdom
Email:
Phone: 44-141-330-5124
Fax: 44-141-330-5071
Website:
Abstract
One of the most puzzling issues to face those who study the internet and the political sphere is how online activity can translate into action in the offline world. This paper proposes a development of the Gibson and Ward model for categorizing and measuring the function of political party websites. Although the Gibson and Ward model has provided a very useful method for comparing the content of political party websites across both time and space, developments in Web design and technology call for modifications to the model. Even more importantly, it is key to develop a broader understanding of the role of the Internet not just as a data retrieval tool, but also as a new type of communication tool with functions that are distinct from other mass media. This paper discusses ways in which the Gibson and Ward scheme is still important to understanding the content of the online world. It suggests that information gathered about internet content become part of a more holistic, five-step approach to the analysis of the online impact on politics by adding four more levels of analysis: communities, catalysts, constraints, and control/cooptation. This should allow us to better define to what degree political party/candidate websites empower internet users to translate their interests into political action. This should allow us to better understand how the internetcan facilitate the shift from internet content consumer, to internet content producer, and possible political involvement in the offline world.
There is an old adage that dogs and owners often start to resemble one another. In the case of the internet, it would sometimes seem that the relative chaos and lack of structure on the internet is reflected in the study of the internet itself. Several decades into the existence of the internet as an important political communication tool, it would seem that there are a dearth of useful research theories and tools with which to study this social phenomenon. This is perhaps an unfair criticism, in particular as the internet has changed and evolved at a rate that makes it difficult to describe, much less analyze. That being said, it is time for a more systematic attempt to make order out of the perceived communication ‘chaos’ online, particularly as it becomes more and more apparent that the internet can equally become a powerful tool for the development of political culture or political chaos.
The struggle to analyze the internet relates to a broader concern about understanding of the role of information in shaping society. Unfortunately, there is not a great deal to guide us firmly from the ‘classic’ literature of comparative political science and political communication that provides a definite hypothesis about the relationship among leaders, the public, and information. There is a broad rejection of the ‘propaganda’ theory that was popular in the wake of World War II, the notion that a public could be led, by clever manipulation of the media, into supporting regimes as diverse as the Soviets, the Nazis, or even a particular U.S. political party in a presidential election. Rather, studies in the second half of the 20th century (mostly in the United States) suggested that information from the mass media is tempered and filtered through the socio-economic characteristics held by individuals (Berelson et al. 1954, Campbell et al. 1960). This is turn has led to a wealth of studies that have attempted to find that elusive formula between information in the form of political marketing and the political act of voting. The ideas that have arisen suggest that the media have the power to set the agenda (Iyengar and Kinder 1987, Patterson 1994); educate us into better citizens (Norris 2000); propagate capitalism (Herman and Chomsky 2002); and alienate us from the democratic process with negative advertising (Ansolabehere and Iyengar 1995). The academic literature on voting behavior and media effects – used here as an example of one of the most energetic areas of political communication research – has never emerged with a particular model that describes the relationship between information and political action. We generally understand that people are attracted to media that support, rather than challenge, their world views. We know that media socialization occurs rather early in life, which explains why the older generation continues to favor newsprint more than their middle-aged children who prefer television or their internet-savvy grandchildren. We also know that media campaign coverage can make or break a candidate and change election results in a tight race – but it cannot change entrenched preferences on the part of the electorate in the course of one campaign.
What can we take from studies of media and political behavior to apply to the internet? For the purposes of this paper, we will consider one small aspect of the information online to conceptualize approaches to studying the internet in general. For this exercise, we will consider the attempts by political parties and candidates to attract voters online. While this is a very narrow part of the online world, it will allow for a consideration of political communication tools to analyze the internet that could be used in a range of different cases. In addition, there is a fairly well-developed literature on party websites across a number of countries thanks to political scientists such as Rachel Gibson, Stephen Ward, Wainer Lusoli, and others. As such, it is possible to analyze methodology developed over a number of years, assess the value of the research, and suggest ways to augment the existing research tools developed to dissect political party efficiency online to more wider analytical use in studying the online sphere. In particular, it would be useful to find a more generalized way to gauge how online chat can translate into offline action. Party scholars have a range of ways to measure this, from gauging how much people are willing to donate online to whether they can persuaded via online methods to attend offline events such as meetings and rallies. I am interested in attempting to use methods developed via the party website literature to look at more general radicalization online, i.e. when people may move from talking about possible dissent to participating in protest actions offline.
There is a great deal of speculation that online activity leads to real-world activity (particularly in terms of fears about terrorism) but there is actually relatively little tangible evidence. In fact, one of the most-oft cited examples is the Chiapasrevolt in Mexico in the mid-1990s – now more than 10 years old. In addition, there is evidence that the internet has been important in augmenting recruitment efforts into alternative political movements, such as the international anti-globalization movement. At the same time, however, it is not really known how many people are drawn into anti-state or what could be termed anti-civil movements that promote terrorism, racial hatred, xenophobia, or other elements that could be seen as destructive to civil society. Nor is there much systematic evidence that the online information sphere is enlightening and empowering citizens. For example, despite large investment in E-governance, one could argue that British citizens are not measurably better informed or more active in their governance than in previous times. Certainly they weren’t overwhelming using the internet in the most recent parliamentary election in 2005 (Lusoli and Ward 2005). For a small number of information-hungry and politically active citizens, the internet provides a place for them to exchange information and hone their political skills, becoming what Linaa Jensen (2006) has termed ‘political gladiators’.
Into this dearth of information and confusion about internet effects come a number of well-designed studies of political party websites. In particular, a coding scheme designed by Rachel Gibson, Stephen Ward, and others (see Gibson and Ward 2000; Gibson et al. 2003) is useful at comparing political party websites both across time and among different countries. The scheme is outlined in full in the first appendix of Gibson, Margolis, Resnick, and Ward (2003) and is summarized here. Websites receive points for a variety of features under the following categories: information provision, resource generation, internal networking, external networking, participation, campaigning, delivery, access, navigability, freshness, and visibility. Each category is defined via a series of variables that can be measured completely via observation of the website online. For example, information provision has 16 subcategories, ranging from organizational history to article archive, and the website achieves a score of 1 for each element present. In resource generation, the website scores more highly for the ease with which people can join, donate or buy merchandise online. In networking, websites score points for internal links such as to local parties or a special web area for members only. For external links, websites obtain points via links to partisan sites, reference links (such as to media websites) and commercial links (such as booksellers). For participation, websites get points for their number of Email contacts, online forms for feedback, opinion polls, and the opportunity to debate with party leaders. For campaigning, websites score points for specific online tactics, such as by targeting a particular constituency being able to download leaflets.
The last five categories in the Gibson and Ward scheme have become slightly less useful as the central parties have moved beyond the static text-only sites of the 1990s. Websites score points for graphics, moving icons, and video but virtually everyone in the developed world has these features now. For the access category, websites accrue points for elements such as text-only options and software for the visually impaired. For navigability, websites score points for having a search engine, site map, etc. Finally, freshness is scored on how often the website is updated, while visibility is a single calculation of links in (counted by using a search engine such as Google).
One of the key elements to point out about the Gibson and Ward index is that it still works at drawing distinctions among political party websites in the United Kingdom. This may seem surprising in that the index was constructed several years ago and there have been massive changes to the capability of the internet. It is particularly effective at pointing out significant differences between party websites in countries such as Britain and ‘challenged’ democracies such as Russia (March 2006). While British voters may fret that Labour and Conservative websites are perhaps not the most exciting or dynamic places on the web, they do dutifully provide a range of information about party policies, party history, local chapters, and ways to join the political party. In the case of Russia, most party websites are amateur, limited in features and relatively uninformative (March 2006). This is unsurprising, in that web usage is much lower in Russia than in the United Kingdom (Cooper 2008). However, it also parallels political reality, in that parties have very little power to aggregate the interests of the masses in Russia as the dominant ‘party’ is essentially a propaganda front for the Kremlin’s interests (Oates 2006). Lusoli and Ward (2006) speculate that the use of the internet to find political information is relatively low in Britain because people are generally quite well-informed about their somewhat stable political system. Ironically, while Russians could arguably benefit far more from the ability of the internet to deliver relatively transparent and timely political information, there is much less of it online.
One way to add to the Gibson and Ward scheme would be to attempt to measure how online activity translates into offline action. This is a discussion that is more prevalent among ‘cyber-pessimists’, particularly those who fear that terrorists or others who seek to plot violence against society have an easy way to aggregate interests online. In particular, governments often voice concern that the online world provides the sort of specialized, yet accessible environment that allows those are alienated from society ways to gather together to challenge society in a violent way. The concerns range from the accessibility of information on the tools of terrorism (making bombs, finding training camps, etc.) to concerns over the power of online influence to persuade people into violent behavior. This highlights the point raised by Newhagen (1997), in that officials are simultaneously worrying about different things. They are concerned that the internet is a powerful data retrieval system that can provide the concrete ability for people to carry out violent threats. At the same time, they are concerned with the internet as a communication method and that its ability to collapse both physical space and time (with instant communication, video chat, etc.) will broaden and hasten the growth of terrorist groups. There is a great deal of classified and unclassified study of terrorist activity online (and much of it in the public sphere, such as publications from the Dark Web project at the University of Arizona, see but political science writing focuses on the fact that most terrorists are not able to take advantage of the potential of the internet. As Reilly (2008) points out, some of this is due to a lack of resources for sophisticated Web design and usage. More importantly, terrorist groups are highly constrained in their Web presence by national anti-terrorism laws and in cases such as Northern Ireland, use the Web far more routinely to disseminate information on their cause and historical interpretation than as a call to violence (Reilly). As Reilly points out, the internet serves as a ‘panoptican’, a devise in which the participants can be observed and tracked by security officials. The very openness of the internet becomes a constraining factor for terrorist groups. In authoritarian regimes, the relative transparency of the internet is a constraining factor for any alternative political voices online.
Is there any useful way to translate the work on terrorist websites to political party websites? Can the two approaches help each other? One of the key findings of the Dark Web project is that terrorist groups enjoy more use of interactive internet-based communication tools such as chat rooms and forums than their government counterparts (Qin et al. 2007). In addition, despite the difficulties that terrorist groups face in establishing permanent Web addresses, the terrorist groups provide more multi-media features than their government counterparts, which the Dark Web researchers find important in terms of the efficacy of propaganda/recruitment efforts online (Qin et al. 2007). How – if at all -- do these concerns translate from terrorist groups to political parties? Unlike terrorist groups, there is no real constraint on the part of political parties in democratic societies in attempts to spur online supporters into offline action. In particular, party websites are concerned with getting supporters to vote. In British elections, there is relatively little emphasis on real-world meetings, although they are reported on the websites during elections – as are the actions of party leaders the rest of the time. Democratic Presidential Candidate Barack Obama’s website has an emphasis on finding nearby off-line groups (with a search window to fill in ZIP codes, see last accessed 18 August 2008). This is certainly an area in which the Gibson and Ward scheme could be augmented, particularly in coding for how many events are listed, whether there is search engine by postal code/ZIP code, and what types of events are listed. In addition, is there an ability of users to post their own events onto the main website?
The notion of posting ‘up’ to an official party website is an interesting one. Most party websites are closed to this, although there are features that allow comments. On the British Labour website, a section with links to blogs (albeit with a disclaimer that the party does not control the content, which begs the question of the value of the link) has been added relatively recently. This addresses a broader concept of ownership and authority. Despite all the emphasis on the blurring of the line between content producer and content consumer in the online sphere, the content on a party website is still delivered almost completely from the top down.
What does the Gibson and Ward scheme tell us about key contemporary party websites in political environments such as the United Kingdom and the United States? In Britain, the websites of the two dominant parties (Labour and Conservatives) achieve a similar score on the Gibson and Ward code. Thus, while the Gibson and Ward scheme will highlight differences between the web presence of the Communist Party of the Russian Federation and the Labour Party, for example, it doesn’t show a marked difference between the two major political players in Britain. However, a qualitative review of the Labour and Conservative websites identifies some interesting differences. The tone of the Labour Party is very much that of a party in governance, highlighting events such as the Olympics to link the party in the minds of supporters with patriotic events. The Conservative website (in keeping with leadership of a relative young, media-friendly party leader David Cameron) has a more casual approach and highlights links to social networking sites on the top of its home page. In contrast, Labour continues to promote its own membership ‘zone’ in which people can chat. This provides more evidence from a qualitative review that the Conservatives are embracing the concept of Web 2.0, in which the distinction between content producer and content user of a web space is further blurred. There are a range of factors in the British political sphere that could arguably contribute to the different look and feel to the two websites, despite their similarities in types of features as recorded by the Gibson and Ward scheme. The Conservatives are not the party of power and, in the tradition of British ‘loyal opposition’, have more latitude than the ruling Labour Party to make sweeping policy statements and plans (as they won’t necessarily be called on to carry them out in the near future). In addition, leadership and personality politics are key parts of political marketing, even in the party-centric system of Britain (Langer 2008). While Gordon Brown is struggling with image in the wake of the more charismatic Tony Blair, Cameron continues to enjoy relatively good media coverage. He appears in a fairly casual pose on the Conservative home page on 14 August 2008, in his shirtsleeves and with a microphone on this hip. On the other hand, Brown was absent from the home page of the Labour Party on the same day.