Political Hyperlinking in South Korea: Technical Indicators of Ideology and Content
by Han Woo Park, Mike Thelwall and Randolph Kluver
YeungNam University; University of Wolverhampton; Nanyang Technological University
Sociological Research Online, Volume 10, Issue 3,
<http://www.socresonline.org.uk/10/3/park.html>.
Received: 10 Nov 2004 Accepted: 13 Jul 2005 Published: 30 Sep 2005
Abstract
As the Internet has become a more important source of information for citizens and consumers, politicians in a number of nations have employed the Web as a tool to facilitate contact with constituents and supporters. One of the least understood phenomena in the new ecology of political communication, however, is the use of hyperlinks to build public recognition, to demarcate ideological spaces and to reflect political alliances. The purpose of this paper is to examine the political hyperlinks established by the National Assembly members in South Korea, in order to ascertain their functions. This paper examines the communicative agendas of politicians as represented by their (out)linking practices. Hyperlink data obtained from the homepages of South Korea's National Assembly members during June of 2003 was collected. A hyperlink network analysis revealed that outlinks to political parties were the most common type of link. The next most preferred target websites were those maintained by the National Assembly, local governments such as city hall, and central government bodies, including Ministries. Websites rarely hyperlinked to civic and advocacy groups compared to other categories. In summary, there were more navigational outlinks for informative content than ideological affiliations. The results are discussed from the perspective of Asian values underlying political communication as well as online culture. Finally, this study performs an important role in contributing to the small but growing literature on how the Internet is affecting the practices of nations outside of the established democracies of the West.
Keywords: Link Analysis, South Korea, Technical Indicator, Political Communication, Asia, National Assembly, Hyperlink Network Analysis
Introduction
1.1 As new technologies such as the Internet have permeated society, they have become a driving force to change the organizational forms of political communication systems (Norris, 2002). With the increasing use of the Internet among the general public, candidates, parties, groups, media, and other political actors create websites and use them for online communication. The Web seems set to be a primary political communication channel of the next decade.1.2 As a consequence, studies of political communication on the Internet have proliferated as scholars explore the multiple potential impacts upon practice. Previous studies of the Internet in political communication have tended to focus on the way in which content is generated, and the likely impact of the migration of traditional forms of political communication. For example, a number of studies have conducted content analysis of political actors' websites, such as the sites maintained by parties and politicians (Davis, 1999; Sunstein, 2001). Others have examined the deployment of websites during specific periods such as election campaigns (Benoit & Benoit, 2000; Gibson, Ward, and Nixon, 2002; Schneider and Foot, 2002). Moreover, other studies have examined how the Internet can facilitate the political actions of minor parties and interest groups and its function in fostering greater participation among supporters (McCaughy & Ayers, 2003). Scholars have mostly agreed with the Internet is a rich and valuable resource for political organizations, candidates, and citizens to provide access to political information and to engage in political action.
1.3 One significant gap in this burgeoning literature is an examination of how latent features of the technology are used to establish or cement political alliances between various political players, in other words, the relational aspect of politics. The purpose of this essay is to examine the political hyperlinks established by the National Assembly members in South Korea in order to ascertain the functions of hyperlinks. There are two significant reasons for this. First, although multiple studies have examined the content and features of political websites, the relational patterns of Internet-mediated communication networks between political actors have rarely been examined. Since modern representative democracy is largely about the role of interest groups, advocacy organizations, media organizations, and other large, complex social groupings, this is a significant gap in the literature of the political consequences of the Internet. This study will broaden this area of study by arguing that not only to hyperlinks help to establish the structure and boundaries of political communication on the Web, but by themselves have an important communicative function (Soon and Kluver, 2005). This essay also furthers a line of research that has been examining the ways in which hyperlinks reveal deeper relational patterns between organizations through "hyperlink network analysis" (Park, 2003; Park & Thelwall 2003). Although a few studies have used hyperlink analysis to systematically explore the structure of political communication on the Internet, (Foot et al., 2003; Park, Kim, and Barnett, 2004) these studies have focused primarily on describing how links are used, rather than seeking to identify underlying patterns of communication that the links may be intended to reflect. Finally, this study contributes to the small but growing literature on how the Internet is affecting the political practices of nations outside of the established democracies of the West, which is a critical gap in our understanding of how the Internet is likely to affect politics in nations around the globe (Kluver, 2005; Kluver and Banerjee, 2005).
Theory & Literature
Web sphere and hyperlink analysis
2.1 Cyberspace is traditionally considered a digital world constructed by a computer network (such as the Internet) where people can engage in relationships of mutual communication that are not tightly tied to their physical location (Rheingold, 2003). A political web sphere is a particular type of cyberspace, comprised of websites within which political actors are given the opportunity to act using features such as links and text (Schneider and Foot, 2002). The structure of a political web sphere may reflect a set of relationships among a set of political entities such as citizens, politicians, parties, presses, or interdependent civic organizations (Park, Kim, and Barnett, 2004), serving as a "parallel universe" to the offline world, separate but connected. Thus, it is reasonable to investigate whether the relationships that exist among individual actors may be examined through the configurations of interlinking web elements. Moreover, the depth and complexity exhibited through interlinking may demonstrate patterns or regularities that may not be found if individual actors are analyzed. The web thus facilitates large-scale analyses by making data easily available to research politicians through their websites.2.2 Previous analysis of the political use of the Internet tends to focus on features or content items of politically oriented websites, or web components that individual actors commonly employ for political purposes in cyberspace. For example, the Internet and Election Project (2004) provides a comprehensive list of those elements. Information features include: election content, biography/about us, endorsement, issue position, speech, calendar/list of events, comparison of issue positions, information about campaign process, information about voting process, image, audio or video files, privacy policy, and terms of use. Engagement features include: contact information, join a member, register to vote, get email from website, donation, communication space, offline distribution of political materials, send a link, public support statement, e-paraphernalia for download, and volunteer.
2.3 Although these content and technological features are certainly a significant part of the political orientation of the Internet, the focus of this paper is on the way in which hyperlinks enable political actors to demonstrate relational and/or issue affinity, by linking one specific website to another (Rogers and Marres, 2000; Rogers, 2002, 2004). One of the unique features is that a variety of web networking activities produced by political actors can be inscribed on the deployment of links. In other words, the links among website producers may reflect political affinity, (in)formal tokens of mutual communication, and/or potential avenues of coordination (Park and Thelwall, 2003; Wellman, 2001). Recall that there are two perspectives for hyperlinks: inlinks are the links a website receives from the other sites, while outlinks are the links originating from the site. Since outlinks are initiated and modified by the actors who produce websites, outlinking practices may reflect their communication choices and agendas. This paper examines the communicative agendas of politicians as represented in their outlinking practices.
2.4 The relational aspect of political websites is often overlooked, but is especially critical, as part of the political power of the Internet lies in the ability of large numbers of participants to galvanize attention on critical issues, thus challenging the official discourse of politicians and more mainstream press outlets (Rheingold, 2003). For example, in the 2004 US Presidential elections, blogs provided one of the more unusual and unpredictable dynamics, and made it difficult for candidates or parties to control their message (Rogers, 2005). The relational aspect of these blogs was critical, as the voices of a few individuals quickly became amplified by the power of relational networking (Adamic and Glance, 2005).
2.5 Some have argued that hyperlinks are a potential new nexus of places and connections, enabling people to create and expand a certain type of space and ties. Foot et al. (2003) performed an exploratory analysis of outlinking practices exhibited on websites produced by US congressional candidates during the 2002 campaign season. They focused on the extent and development of outgoing links from the websites to other types of political sites such as civic & advocacy groups, political parties, government agencies, and press organizations. Further, they examined the effect of the characteristics of candidates-party type, incumbency status, level of competitiveness, and office sought-on the deployment of linking practices.
2.6 The configuration of hyperlink networks can be a source for conveying useful overall information about the political landscape of certain issues in a society. Adamic (1999) explored communication lines for the issue of abortion, in cyberspace. Her results revealed that the websites belonging to a pro-life community were more tightly connected than pro-choice sites in terms of the number of hyperlinks amongst them. The network of pro-life sites was not only more densely connected, but also better organized in terms of content structure. Rogers and Marres (2000) and Burris et al. (2000) found that the political, ideological, and issue oriented organizations, from white supremacists to environmentalists, produced symbolic representations of their alliances through their selection of hyperlinks. The density of hyperlink connections is stronger among groups with similar ideological affinity or goals and weaker among organizations with different interests. From the perspective of network ethnography, Hine (2000) described the way hyperlink relations are formed and maintained between actors related to the case of British nanny Louise Woodward. More recently, Garrido and Halavais (2003) described the hyperlink network structure of NGO websites related to the Zapatista movement and examined their role in the global NGO networks.
2.7 Morover, Adamic and Glance (2005) recently note that an analysis of the political "communities" revealed through hyperlinks of blogs demonstrate a sort of "cyberbalkanization" of the Internet, and more specifically, the sources of information that people use to make informed political choices. Their analysis of hyperlinks within the "blogosphere" reveals two significantly different political communities, with little overlap in terms of news sources or points of reference. Moreover, within the context of the US, liberal and conservative blogs have significantly different linking patterns, with conservative blogs linking to greater numbers of other blogs, with greater frequency (p. 3). Moreover, among both liberal and conservative blogs, the great majority of links are to ideologically similar blogs or organizations.
2.8 To date, most studies of political communication have focused on the advanced democracies of the West, which share a long history of representative democracy and free and open debate. In the Asian context, however, there are not only radically different histories, but more significantly, different values, believes, and mythologies underlying political communication (Kluver, 2004a; Pye, 1985). More research is needed in a variety of cultural contexts to gain a wider sense of the way in which the Internet affects political behavior and outcomes (Kluver, 2005).
2.9 Park, Barnett, and Kim (2000) have analyzed the pattern of inter-linking connectivity among political parties and National Assemblymen in South Korea. They found that the structure of their hyperlink network was significantly related to party membership. In other words, political actors form a community of "birds of a feather." More recently, Park, Kim, and Barnett (2004) conducted a follow-up study and found that the hyperlink communication network rapidly evolved into a denser, more centralized and integrated network in one year. They suggested that the nation's move to informatisation was a factor that, in part, determined the hyperlink structure among South Korean political actors' websites.
2.10 Within the political context of Singapore, Soon and Kluver (2005) have demonstrated that hyperlinks form an even more important role in that semi-authoritarian nation, allowing marginalized political actors to link with one another in a way that suggests greater organizational strength. Hyperlinks often seem to serve a "bonding" function, binding together groups that are already ideologically akin to one another, rather than a "bridging" function, of connecting political and ideological divides. Even within this tightly controlled political environment, hyperlinks seem to enable political bodies to have a significantly greater ability to suggest policy alternatives.
2.11 To sum up, prior studies suggest that the web has created a space in which political members can make communication connections and hyperlinks have become a channel for signifying relations such as political affinity and as well as connecting to relevant information. Although it is clear that hyperlink analysis has important implications for understanding the political consequences of the Internet, so far, few studies have examined this, particularly in the Asian context. There are a number of significant theoretical reasons why this study would be helpful, including the fact that Asian democracies, such as South Korea, Japan, and Taiwan, are often subject to very different regulatory requirements than nations in the West, substantially altering the way in which campaigns work on the Internet (Tkach-Kawasaki, 2003). Moreover, although the "Asian values" debate has lost much of its luster, it is clear that often very different political values drive electoral politics in Asian nations, and the shifting nature of alliances and relational connections in a nation as highly attuned to relational affiliation deserves further scrutiny.
Hyperlinks as a mirror of political communication
2.12 An early speculation in the field of information science was that hyperlinks between two different websites could be comparable to academic citations, mainly representing friendly connections between their website producers (Ingwersen, 1998). Thus counts of hyperlinks to academic websites could be used as indicators of value or impact in the same way that citation counts are used. This is also close to Google's hyperlink based ranking algorithm better known as "PageRank", which assumes that more linked-to websites will be more useful. On the other hand, academic hyperlinking is very different to citation (Bar-Ilan, 2004; Thelwall & Harries, 2004; Thelwall, 2004). Links can be created for purposes unrelated to the contents of the target site, such as to acknowledge a collaborative venture (Wilkinson et al., 2003). Furthermore, since anyone can create hyperlinks to anywhere, even at random, they may not carry any meaning at all (Thelwall, 2003). Conversely, there is no reason why collaboration should always result in a hyperlink, or why academics will create links to relevant resources even when they have a website and such a link would be beneficial to them (Beaulieu, 2005).
2.13 In a similar vein, it is likely that political hyperlink creation motivations are not always attributable to the authority, reputation, or quality of linked websites. The existence of hyperlinks among political websites is not necessarily a mirror of the positive relevance of sites involved, or the relationships between actors. Political hyperlinks may be based upon hostile relations rather than cooperative alliances among website producers (Sunstein, 2001; Balkin, 2004). The hyperlinks can be used to call for negative or antagonistic action, as especially exemplified in satirical political websites.
2.14 As an emerging communication medium, communicational representations of the Internet can be analyzed only to a limited degree. Nevertheless, the examination of hyperlink distributions on the Internet may have important applications for communication research, especially if they can later be used as a means to study communication between political actors.
The role of the Internet in South Korean politics
2.15 Internet development in South Korea has been increasing very rapidly and its surprising growth has attracted much attention in academia as well as business. According to the International Telecommunication Union (ITU) (2004), Korea was in fourth place in the world and first in the Asia-Pacific region for the 2002 digital access index, a measurement to gauge access to information and communication technologies such as the Internet. Furthermore, the ITU announced that Korea is the runaway leader for broadband penetration with 21.9 percent of the population signing up for this facility. According to a recent report published by the Korea Network Information Center (2004), 65.5 percent of all Koreans aged 6 and older access the Internet at least once a month, and Korea's Internet population is estimated to be 29.22 million in December 2003, with 13.2 million having high-speed Internet access. The population of females online was 59.2% of all Internet users. Thus, Korean websites are an interesting sample for research because of the extensive Internet penetration in this non-Western nation.
2.16 South Korea in many ways has been seen as an ideal test case for understanding the political impact of the Internet. Not only has Internet penetration been high, but many political actors have seen the Internet as a powerful tool to wrest politics away from the traditionally powerful groups (Kluver and Banerjee, 2005; Yeom, 2003). For instance, during the 2000 Korean National Assembly election, the Internet emerged as a powerful political channel for the diffusion of information. Citizens' groups published a "black list" of 86 "unfit" candidates on the Internet, disclosing details of their backgrounds such as serious criminal records and tax evasion. Fifty-eight of the 86 candidates lost in the election, including several important figures. However, the civic campaign against targeted candidates was ruled illegal by the Supreme Court in 2001. Other legal issues have been raised, including libel, freedom of expression, and the publication of politically sensitive information, issues which the Korean government has not yet adequately addressed (Yeom, 2003, p. 167).
2.17 The Internet continues to play a critical role in providing information about candidates. Since 2000, the National Election Commission (NEC) has put detailed information about the candidates, including records on their wealth, military service and tax status, on the Internet. A number of scholars have argued that the 2003 victory of President Roh Moo-Hyun owes much to the Internet (Kluver and Banerjee, 2005). The majority of his supporters were younger voters between 21-39 years old. Given that young adults are the predominant users of the Internet in Korea, the coalition made up of several pro-Roh support groups, including the Internet-based supporter club Nosamo, People's Power, the Internet-based Surprise and Radio 21, had urged young people to take part in the presidential polls via the Internet. In the past, younger adults rarely bothered to vote. Judging from the results, the civic groups' strategy seemed to have had a significant influence on voting choices as well as participation. This suggests the increased use of the web for political discourse in Korea.
Method
Hyperlink network analysis
3.1 Social network analysis focuses on patterns of relations among people, organizations, or nation states (Wasserman & Faust, 1994). Social networks, that is, the interconnections between social entities (often called "nodes") can be used to understand their interactions. Hyperlink network analysis is the extension of traditional social network analysis in that it focuses on the structure of a social system based on hyperlinks, with the assumption that they often represent shared ties among communication partners. The difference between hyperlink and traditional network analysis is the use of hyperlink data from websites. Two nodes (websites) are connected in through their interlinking. In the hyperlink network analytical framework, social actors and their hyperlinking activities may be analyzed as a whole. This paper explores hyperlink networks themselves as a source of information that can illuminate a web communication environment in which many political actors are connected by hyperlinks. The National Assembly of South Korea was selected as a first order representation of complexity of political choice in this research.Data collection
3.2 The primary data used are outlinks contained in National Assembly members' websites during June 2003. The websites of Assembly members are listed in the National Assembly's site (http://www.assembly.go.kr). The Assembly website includes some content in English so that it can be internationally accessible. The site is particularly useful for an international audience in listing in English all (previous and current) members and parties. At the time of data collection, there were 272 Assembly members, but one seat was absent. We discovered that 222 out of 272 Assembly members maintained their own website. We developed a 2-mode link connectivity matrix (222 x 14), in which the rows represent political actors, that is, the websites of the Assembly members, and the columns represent 14 types of outlinked political site. The value of cell (i,j) is the presence of an outlink from the Assemblyman i to the political site j. This matrix was loaded into the social network software UCINET for Windows (Borgatti et al., 2002, http://www.analytictech.com/downloaduc6.htm).
3.3 The types of the outlinking Assemblymen's sites were classified based on their party affiliation. The parties included are the ruling MDP (Millennium Democratic Party, http://www.minjoo.or.kr), the major opposition GNP (Grand National Party, http://www.hannara.or.kr), and another opposition party, the ULD (Union of Liberty and Democracy, http://www.jamin.or.kr). In addition, there were 5 minor parties and 2 independent Assembly members. Political sites were categorized into five broad types: political parties; government agencies; legislative bodies; press organizations; and civic & advocacy groups. These were subdivided into the following smaller categories: 1. Political party; 2. Unofficial research group; 3. Parliamentary committee; 4. Central government; 5. Local government; 6. National assembly; 7. Local assembly; 8. Broadcasting; 9. Internet broadcasting; 10. Newspaper; 11. Internet newspaper; 12. Magazine; 13. Internet magazine; and 14. Civic & advocacy group. These categories are illustrative rather than exhaustive or exclusive. This taxonomy for outlinked political site categorization has been used in other studies (Foot, et al., 2003).
Results
4.1 The numbers in Table 1 reveal that, within Korea, the degree of digitalization does not vary greatly from party to party, which is significantly different to another Asian democracy, Singapore (Kluver, 2004b).| Table 1: Assemblymen of each party with and without a website | |||||
| Items | MDP | GNP | ULD | Others | Total |
| Website | 80(78.43%) | 131(85.06%) | 6(66.67%) | 5(71.43%) | 222(81.62%) |
| No website | 22(21.57%) | 33(14.94%) | 3(33.33%) | 2(28.57%) | 50(18.38%) |
4.2 As illustrated in Table 2, website ownership is consistent across the level of campaign experience of politicians, which we define as the number of times an Assemblyman has been elected, including in the current election. The more elections a candidate has stood in, the more likely that candidate is to be relatively older (the mean of age of the sample Assemblymen is 58.75 with a standard deviation of 8.45). Based on a presumed positive relationship between the number of wins and age, this result suggests that older politicians have not shunned technological innovations. Given the popularity of the Internet among the Korean voters and politicians' increasing use of the Internet for political communication and electoral campaigning, the proportion of Assemblymen running websites is expected to continue to increase.
| Table 2: The ratio of politicians who maintain websites by campaign experience | ||
| Campaign Experience (# of elections) | Website | No website |
| 1 | 99(85.34%) | 17(14.66%) |
| 2 | 66(83.54%) | 13(16.46%) |
| 3 | 25(78.12%) | 7(21.88%) |
| 4 | 20(83.33%) | 4(16.67%) |
| 5 | 9(64.29%) | 5(35.71%) |
| 6 | 3(60%) | 2(40%) |
| 7 | 0(0.00%) | 0(0.00%) |
| 8 | 0(0.00%) | 1(100.00%) |
| 9 | 0(0.00%) | 1(100.00%) |
| Total | 222(81.62%) | 50(18.38%) |
4.3 Fifteen Assemblymen were isolated, linking to none of the 14 categories. However, 41 Assembly members linked to more than 7 types. One Assemblyman, Yoon Soo LEE, had set up links to 10 categories and 10 members outlinked to 9 types. They were Ki Woon BAE, Han Chun CHO, Byoung Gug CHOUNG, Sung Soon KIM, Oh Eul KWON, Won Chang LEE, Chae Jung LIM, Jae Kwon SIM, Hoon SUL, and Sang Hyun KIM. Some of those websites (e.g., http://www.ilovecj.or.kr) can be traced through the Wayback Machine at the Internet Archive (http://www.archive.org).
4.4 Outlinks to political parties were the most common type of link (Table 3). The next most preferred target websites were those maintained by the National Assembly, local governments such as city hall, and central government bodies, including Ministries. The websites produced by civic & advocacy groups, regular newspapers, regular broadcasting stations, local assemblies, parliamentary committees, Internet-based newspapers and magazines were the next most important target. There was not a big difference between the ruling MDP party, with a reputation for being more favorable to citizens' groups (Park, Kim, and Barnett, 2004) and the opposition GNP party (MDP: 38.8 percent, GNP: 32.1 percent). Most peripheral in the network were regular magazines, Internet-based broadcasting stations, and unofficial research groups.
4.5 Figure 1 illustrates the hyperlink network connectivity of the 272 Assemblymen's websites to 14 political websites, using the visualization program NetDraw attached to UCINET. An arrow represents a hyperlink and its head the direction of the hyperlink. Circle nodes are Assembly members' websites and square nodes represent political websites. The Kamada and Kawai (1989) algorithm has been used to position the nodes so that the websites placed in a central position in the diagram tend to represent highly linked political websites. Figure 2 represents the data at a larger level of aggregation.
|
| Figure 1. Network diagram of links from 272 Assemblymen's websites to 14 political websites |
|
| Figure 2. Network diagram of 14 political websites after grouping 272 Assemblymen's websites into their affiliation parties |
| Table 3: Fourteen types of outlinked websites based on the number of Assembly Members who linked to them. | |||||
| Items | MDP | GNP | ULD | Others | Total |
| Political parties | 70 | 115 | 2 | 4 | 191 (86.0%) |
| National assemblies | 62 | 100 | 4 | 0 | 166 (74.8%) |
| Local governments | 54 | 97 | 2 | 1 | 154 (69.4%) |
| Central governments | 54 | 78 | 2 | 1 | 135 (60.8%) |
| Civic & advocacy groups | 31 | 42 | 1 | 0 | 74 (33.3%) |
| Newspapers | 25 | 40 | 1 | 0 | 66 (29.7%) |
| Broadcasting companies | 18 | 27 | 1 | 0 | 46 (20.7%) |
| Local assemblies | 13 | 19 | 0 | 0 | 32 (14.4%) |
| Parliamentary committees | 14 | 17 | 0 | 0 | 31(14.0%) |
| Internet newspapers | 16 | 14 | 0 | 0 | 30 (13.5%) |
| Internet magazines | 7 | 18 | 0 | 0 | 25 (11.3%) |
| Magazines | 0 | 6 | 0 | 0 | 6 (2.7%) |
| Internet broadcasting companies | 3 | 1 | 0 | 0 | 4 (1.8%) |
| Unofficial research groups | 1 | 1 | 0 | 0 | 2 (0.9%) |
| Total | 368 | 575 | 13 | 6 | 962 |
4.6 The fourteen types of outlinked sites were collapsed into the five broader categories. In narrowing 14 smaller categories into 5 bigger ones, multiple counts of links from some Assembly members can occur. As shown in Table 4, government agency sites were the most common target type and civic and advocacy groups the least common. Figure 3 illustrates the network at this level of aggregation.
| Table 4: Five types of outlinked websites based on the number of links from Assembly members | ||||||
| Source Party | Political parties | Government agencies | Legislative bodies | Press organizations | Civic & advocacy groups | Total |
| MDP | 70 | 108 | 90 | 69 | 31 | 368 |
| GNP | 115 | 175 | 137 | 106 | 42 | 575 |
| ULD | 2 | 4 | 4 | 2 | 1 | 13 |
| Others | 4 | 2 | 0 | 0 | 0 | 6 |
| Total | 191 | 289 | 231 | 177 | 74 | 962 |
|
| Figure 3. Network diagram of 5 political websites after grouping 272 Assemblymen's websites into their affiliation parties |
Discussion
5.1 Our results suggest that there are no substantial differences in the use of websites amongst Assembly members of different party affiliations, winning status, and constituency. In the past, not all political actors have benefited equally from the Internet, and there has been a so-called 'digital divide', generally defined as the inequality in accessing and using new information equipment and services such as computers and the Internet. In this context, it has been argued that the Internet enables marginalized or less significant political actors to increase their communicational reach and organizational power. Some have seen in the Internet the potential to provide significant benefits to minority or marginalized politicians by potentially leveling the playing field for smaller political actors (Nixon, Ward, and Gibson, 2003;Norris, 2001) while others have argued that it is primarily the most organized and powerful parties who are best able to deploy the Internet (Kluver, 2004). The former does not seem to be the case in South Korea, at least for elected political parties.
5.2 From a different perspective, there have been investigations into the way in which the Internet is used to make major political actors and discourses more prominent and salient in cyberspace (Owen, Davis, and Strickler, 1999). Our results indicate that gaps in Internet usage appear to be diminishing around traditional demographic variables such as age or career status. Korean politicians' material access to new digital technologies is clearly widespread. This may have been the result of proactive government information society policies leading to increased use of the Internet among the population.
5.3 We found more content-neutral navigational outlinks than self-expressive and/or ideological affiliations in the Assembly member websites. In other words politicians linked more for informative content than for similarity of ideology. This suggests that the websites function as a credible online source of political information more than creating perceived community boundaries. As society moves into the information age, politicians are positioning their websites as political portals for the general public, albeit in an uncontroversial and conservative way.
5.4 There was more outlinking to the websites of the existing press organizations than to those of independent online presses, except for magazines. We assume that this resulted from the perceived political influence of the news dot-com sites. Some Internet press sites such as OhMyNews.com (in South Korea) and Slate.com (in the USA) largely exist as independent entities on the web with little presence in physical space. On the other hand, there are news organizations that sell an offline product (newspapers, television or radio shows) and distribute their own news content online. Generally, the latter news sites are perceived as more credible than the former. A politician's website's credibility may increase through associating with a credible site.
5.5 South Korean politicians infrequently outlink to civic & advocacy groups. This provides counter-evidence to Sunstein's (2001) argument that hyperlinks are more likely to target issue-specific affiliated sites. It is possible that Assembly members' websites put more emphasis on attracting new supporters, rather than mobilizing existing supporters, which is different to the Howard Dean campaign in the US, for example (Jacobs, 2005). Thus, it seems that Korean political linking demonstrates a cautious, conservative approach rather than an active, issue-campaigning approach. Outlinking to only one-type of website for certain issues could have a negative effect on getting more visitors. The findings showed only weak ties with unofficial research groups and parliamentary committees. This may be caused because of the dominant role of their affiliated political party.
5.6 The analysis above suggests is that there indeed is some merit to assuming that national political culture plays a role in how the Internet is deployed (Kluver, 2005). Korean politicians are using the Internet, and especially hyperlinks, in ways that demonstrate stronger commitments to parties than to issues, stressing organizational ties rather than issue oriented constituencies. There are a number of cultural values that might facilitate this, including a tendency within Korean society, and Confucian societies generally, to stress lineage and descent. Pye (1985), in particular, argues that political identity in Confucian societies is often analogous to family membership, and this study suggests that the political boundaries drawn by the hyperlinks of Korean politicians reflect organizational ties, or lineage, more strongly than they reflect political identity based on issue affiliation. Of course, this assertion needs much more rigorous testing than we are able to provide here, and would need validation through comparative studies.
5.7 In conclusion, there is significant value in viewing hyperlinks as communicative devices, rather than just technical devices. By investigating the ways in which hyperlinks are produced within the context of political behavior, we argue that this suggests that hyperlinks do indeed help to define political communities
Limitations and suggestions for future research
6.1 There are a number of limitations to this study in the way in which the data was collected and analyzed. First, judgments about site category were subjective, made by one researcher and could contain errors of interpretation. These should not have a significant impact on the data, due to the large sample size, unless there was some kind of systematic misunderstanding, which seems unlikely due to the relatively straightforward categories used.6.2 Second, the web is an environment in which imitation is common (Pennock et al., 2002). This means that significant differences between groups, particularly political parties, may represent imitation rather than fundamental underlying differences. Similar caution has been expressed for the results of comparative studies of web linking between academic disciplines (e.g., Harries et al., 2004). In addition, the creation and maintenance of a website can be outsourced to a professional design service, and it is possible that the links generated by the contracted professionals could significantly account for the absence or presence of certain types of links. In particular, a hired professional is more likely to think a party link is valid, but would be much more cautious about choosing which issues an Assembly member would choose to highlight in their websites. This affects the degree of confidence with which differences in hyperlinking practices can be generalized. In particular, we cannot attribute motives based on practice, but it does provide a useful starting point for a more in-depth study of the motives behind web linking. In particularly, we would propose that follow-up studies engage more qualitative measures, such as interviewing politicians, political staff or advisors, and the contracted design personnel to gain more insight into how the selection of hyperlinked sites are made.
6.3 How politicians use the Internet is an important area of study, especially given the increasing importance of the web as a means of gathering political information. Other critical areas of future research would include projects such as using content analysis to classify the political stances of the outlinked sites, such as press organizations and civic & advocacy groups, would give a more fine-grained picture of linking practices. Moreover, there is value in investigating the extent to which political parties hyperlink to allies and adversaries (see Sunstein, 2001). Presumably this would show a tendency of like to link to like, but nothing should be taken for granted on the web and a different result would allow fruitful comparisons with other countries.
6.4 We have already argued, and the literature demonstrates, that hyperlinking practice can reveal inherent relationships between political actors. But, there are a number of motives for why this might be the case, which we would suggest as future research. One of these is the value and role of reciprocity, or the extent to which politicians provide default back links in the same way that bloggers do. It would be useful to investigate the extent to which hyperlinks, particularly reciprocal links, reflect cooperative alliances among website producers, or even a formal linking agreement.
6.5 There is also a need for a further exploration of the role of political values and beliefs in hyperlinking behavior, at both the national and ideological level. It may be the case that more authoritative or credible sites, in the sense of attracting large numbers of inlinks, use links in different ways to other sites, perhaps choosing them more carefully, or linking to more official sites. Content analysis may be able to reveal whether or not this is the case.
6.6 Finally, we have not explored the extent to which links reveal sympathetic or hostile intent. Metaphorical or some other form of analysis might reveal the extent to which hyperlinks are provided with "editorial commentary," or input to the user about how to approach or receive the information from the linked site.
Acknowledgements
The first author acknowledges that this work was partly supported by a Korea Research Foundation Grant (KRF-2004-042-H00004). He is particularly grateful for his assistants in New Media & Society Laboratory.
References
ADAMIC, L. A. (1999). 'The small world Web'. Proceedings of 3rd European Conference of Research and Advanced Technology for Digital Libraries, ECDL. <http://www.hpl.hp.com/shl/papers/smallworld/smallworldpaper.html>.ADAMIC, L. and Glance, N. (2005). The Political Blogosphere and the 2004 U.S. Election: Divided They Blog. <http://www.blogpulse.com/papers/2005/AdamicGlanceBlogWWW.pdf>.
BALKIN, J. (2004). 'Balkinization Blog'. <http://balkin.blogspot.com/2004/01/what-i-learned-about-blogging-in-year.html>
BAR-ILAN, J. (2004). A microscopic link analysis of academic institutions within a country - the case of Israel, Scientometrics, Vol. 59, Issue (3), pp. 391-403.
BEAULIEU, A. (2005). 'Sociable Hyperlinks: an ethnographic approach to connectivity.' In: Christine Hine, (ed.) Virtual Methods: Issues in Social Research on the Internet. London: Berg.
BENOIT, W. L., and Benoit, P. J. (2000). 'The virtual campaign: presidential primary Web sites in campaign 2000.' American Communication Journal, Vol. 3 Issue (3), pp. 1-22.
BORGATTI, S.P., Everett, M.G. and Freeman, L.C. (2002). UCINET for Windows: Software for Social Network Analysis. Harvard, MA: Analytic Technologies.
BURRIS, V., Smith, E., and Strahm, A. (2000).'White Supremacist Networks on the Internet,' Sociological Focus, Vol. 33 Issue (2), pp. 215-235.
DAVIS, R. (1999) The web of politics: The Internet's impact on the American political system. New York: Oxford University Press.
FOOT, K., Schneider, S., Dougherty, M., Xenos, M., and Larsen, E. (2003). Analyzing linking practices: Candidate sites in the 2002 US electoral Web sphere. Journal of Computer Mediated Communication, Vol. 8 Issue (4). <http://jcmc.indiana.edu/vol8/issue4/foot.html>
GARRIDO, M., and Halavais, A. (2003). 'Mapping networks of support for the Zapatista movement.' In McCaughy, M., and Ayers, M. D. (Editors.), Cyberactivism: Online activism in theory and practice. London: Routledge.
GIBSON, R.; Ward, S., and Nixon, P.(Editors). (2003). Net Gain?: Political Parties and the Impact of the New Information Communication Technologies. London: Routledge.
HARRIES, G., Wilkinson, D., Price, E., Fairclough, R. and Thelwall, M. (2004). 'Hyperlinks as a data source for science mapping,' Journal of Information Science, Vol. 30 Issue (5), pp. 436-447.
HINE, C. (2000). Virtual ethnography. Thousand Oaks, CA: Sage.
INTERNATIONAL TELECOMMUNICATION UNION (2004). Digital Access Index. <http://www.itu.int/ITU-D/ict/dai/index.html>
INTERNET AND ELECTION PROJECT (2004). An International Project for the Comparative Study of the Role of the Internet in the Electoral Process. Unpublished manuscript. <http://www.ntu.edu.sg/sci/sirc/internet&elections.html>
INGWERSEN, P. (1998). The calculation of Web impact factors. Journal of Documentation, Vol. 54 Issue (2), pp. 236-243.
JACOBS, D . (2005). 'Internet Activism and the Democratic Emergency in the U.S.' Ephemera, Vol. 5, Issue (1), pp. 68-77.
KAMADA, T., and Kawai, S. (1989) 'An algorithm for drawing general undirected graphs', Information Processing Letters, Vol. 31, Issue 1, pp. 7-15.
KOREA NETWORK INFORMATION CENTER. (2004). Survey on the Computer and Internet Usage <http://isis.nic.or.kr/eng_report_down/upload/summary200312_eng.pdf>
KLUVER, R. (2004a). Editor's Introduction: Political Communication in Asia. Asian Journal of Communication,Vol. 14 Issue (2), pp. 117-120.
KLUVER, R. (2004b). Political Culture and Information Technology in the 2001 Singapore General Election. Political Communication, Vol. 21. pp. 435-458.
KLUVER, R. (2005). Political culture in online politics. In M. Consalvo and M. Allen (Eds.) Internet Research Annual, Vol 2. Newbury Park, CA: Sage Publications, pp. 75-84.
KLUVER, R. and Banerjee, I. (2005). 'Political Culture, Regulation, and Democratization: The Internet in Nine Asian Nations.' Information, Communication, and Society, Vol. 8, Issue 1, pp. 1-17.
MCCAUGHY, M., & Ayers, M. D. (Editors) (2003). Cyberactivism: Online activism in theory and practice. London: Routledge.
NIXON, P., Ward, S. and Gibson, R. (2003). Conclusions: the net change. In R Gibson, S Ward, and P Nixon.(Editors). Net Gain?: Political Parties and the Impact of the New Information Communication Technologies. London: Routledge.
NORRIS, P. (2001) Digital Divide: Civic Engagement, Information Poverty, and the Internet Worldwide.Cambridge: Cambridge University Press.
OWEN, D., Davis, R., and Strickler, V. J. (1999). Congress the Internet. Harvard International Journal of Press-Politics, Vol. 4 Issue (2), pp. 10-29.
PARK, H.W., G.A. Barnett and C.S. Kim (2000). Political Communication Structure in Internet Networks - a Korean Case, Sungkok Journalism Review, 11, 67-89.
PARK, H. W. (2003). Hyperlink network analysis: A new method for the study of social structure on the web. Connections, Vol. 25 Issue (1), pp. 49-61. <http://www.insna.org/Connections-Web/Volume25-1/7.Hyperlink.pdf>
PARK, H. W., and Thelwall, M. (2003). 'Hyperlink analyses: A review,' Journal of Computer-Mediated Communication, Vol. 8, Issue (4), <http://www.ascusc.org/jcmc/vol8/issue4/park.html>
PARK, H. W., Kim, C. S.,and Barnett, G. (2004). 'Socio-communicational structure among political actors on the Web.' New Media and Society. Vol. 6, Issue (3), pp. 403-423.
PENNOCK, D.M. Flake, G.W. Lawrence, S., Glover, E.J. Giles, C.L. (2002). 'Winners don't take all: Characterizing the competition for links on the web,' Proceedings of the National Academy of Sciences, Vol. 99, Issue (8), pp. 5207-5211.
PYE, L. (1985). Asian power and politics: the cultural dimensions of authority. Cambridge, MA: Harvard University Press.
RHEINGOLD, H. (2003). Smart mobs. Cambridge: Perseus Publishing
ROGERS, R., and Marres, N. (2000). 'Landscaping climate change: A mapping technique for understanding science and technology debates on the world wide web.' Public Understanding of Science, Vol. 9, pp. 141-163.
ROGERS, R. (2002). 'Operating Issue Networks on the Web,' Science as Culture, Vol 11, Issue 2, pp. 191-214.
ROGERS, R. (2005, forthcoming). "Old and New Media: Competition and Political Space," Theory & Event. <http://www.govcom.org/publications/full_list/rogers_blogs_theoryevent.pdf>.
SCHNEIDER, S. M., and Foot, K. A. (2002). 'Online structure for political action: Exploring presidential Web sites from the 2000 American election.' Javnost (The Public), Vol. 9, Issue (2), pp. 43-60.
SOON, C. W. T. and Kluver, R. (2005, May). 'The Internet and Online political communities in Singapore.' Paper presented to the International Communication Association, New York.
SUNSTEIN, C. R. (2001). Republic.Com. Princeton, NJ: Princeton University Press
THELWALL, M., and Harries, G. (2004). Do better scholars' web publications have significantly higher online impact? Journal of the American Society for Information Science and Technology, Vol. 55, Issue (2), pp. 149-159.
THELWALL, M. (2003). What is this link doing here? Beginning a fine-grained process of identifying reasons for academic hyperlink creation, Information research, Vol. 8, Issue (3). <http://informationr.net/ir/8-3/paper151.html>
THELWALL, M. (2004). Link analysis: An information science approach. San Diego: Academic Press. <http://linkanalysis.wlv.ac.uk>
TKACH-KAWASAKI, L. (2003) 'Clicking for votes: assessing Japanese political campaigns on the Web', in K. C. Ho, R. Kluver, and K. Yang (Editors) Asia.com: Asia Encounters the Internet,London: Routledge.
YEOM, K. H. (2003). The Internet and Democracy in South Korea. In I. Banerjee, (Editor) Rhetoric and Reality: the Internet challenge for democracy in Asia. Singapore: Eastern Universities Press, pp. 141-178.
WASSERMAN, S., and Faust, K. (1994).Social network analysis: Methods and applications. Cambridge, NY: Cambridge University Press.
WELLMAN, B. (2001). 'Computer networks as social networks,' Science, Vol. 293, Issue (14), pp. 2031-2034.
WILKINSON, D., Harries, G., Thelwall, M. and Price, E. (2003). 'Motivations for academic web site interlinking: Evidence for the web as a novel source of information on informal scholarly communication,' Journal of Information Science, Vol. 29, No. (1), pp. 59-66.
