Networks and Communication Studies NETCOM, vol. 16, n° 3-4, 2002 p. 155-174

Interorganizational Hyperlink Networks among Websites in South Korea Han Woo Park1, George A. Barnett2 and In Yong Nam3

Abstract.— This paper examines the interorganizational networks of websites. The interorganizational network is defined by the shared hyperlinks among the organizations’ websites. Individual websites network with others for the purpose of strengthening their credibility (trustworthiness, expertise, and safety) on the web. This paper examines the association among hyperlink network structures and the number of visitors and the perceptions of the website’s credibility. These relations are examined with a sample composed of the 50 most frequently visited Korean websites obtained from the Internet Metrix for June 2001. The results indicate that a site’s centrality in the hyperlink network is significantly related to visiting behavior and perceived website credibility. Keywords.— Interorganizational relations, Hyperlink networks, Website credibility, Korea, Internet, Social network analysis Résumé .—“Réseaux d’hyperliens interorganisationnels entre sites Web de Corée du Sud”. Le réseau interorganisationnel est défini par les hyperliens partagés parmi les sites Web d’organisations. Cet article examine les relations dans un échantillon composé des 50 sites Web coréens les plus fréquemment visités (Metrix, juin 2001). Mots-clés.— Relations – Organisations – Reseaux – Hyperliens – Internet – Website – Corée – Reseaux sociaux

1. Networked Research and Digital Information (Nerdi), Royal Netherlands Academy of Arts and Sciences (NIWI-KNAW), Joan Muyskenweg 25, PO Box p. 39, 1090 HC Amsterdam, The Netherlands. [email protected], [email protected], http://www.niwi.knaw.nl/ nerdi. 2. School of Informatics, State University of New York at Buffalo, Buffalo, New York 14260. [email protected]. 3.Department [email protected].

of

Advertising

and

Public

Relations,

Silla

University,

Pusan,

Korea.

156

NETCOM, vol. 16, n° 3-4, 2002

STATEMENT OF PROBLEM

As new communication technologies such as the Internet have permeated society (Castells, 1996), they become a driving force changing the (inter)organizational forms and relations (Dewett and Jones, 2001). Because organizations are open systems surrounded by an external environment, they interact with a variety of elements in the society (Barnett and Thayer, 1997). Regarding the role of communication technology in interorganizational settings, studies of interorganizational networks mainly focus on the way in which interorganizational communication linkages operate, by studying the patterns of relationships within and between organizations in the context of complementing human networks (Contractor and Eisenberg, 1990 ; Kettinger and Grover, 1997). Or they examine how organizations make use of technology for reducing transaction costs (Hart and Estrin, 1991 ; Malone et al., 1987 ; Steinfield et al., 1995). There has been insufficient research examining how network positions in interorganizational networks are associated with audience attitudes or behaviors such as people’s expectation or intention to invest in the organizations. This is especially true with to the Internet. Little research has been done to investigate the relational structures among organizations in terms of their organizations’ communication activities on the web. Telecommunications such as the Internet are space-adjusting technologies. They permit instantaneous communication between people and organizations that are geographically distant from each other, thereby making physical space fungible. On the Internet, all points on the earth may be considered equivalent. In other words, electronic communication alters the geometry in which interaction takes place (Barnett and Choi, 1995). The Internet changes the proximity of places by improving the connections between them. It has made it easier for individuals and organizations to have more contacts with those that are nearby and to communicate with others at a distance. This has led to the process of globalization and the presence of a worldwide economy in which organizations efficiently communicate with others regardless of physical location (Giddens, 1990 ; Waters, 1995). The purpose of this study is to investigate what influences interorganizational network formation on the web and how organizations make use of the web for increasing communication effects on their organizational audiences. Specifically, it examines how interorganizational network structures based upon shared hyperlinks among websites are associated with the number of visitors and the perceptions of website’s credibility. 1. HYPERLINK NETWORKS AS INTERORGANIZATIONAL NETWORKS

Although researchers have conceptualized the Internet differently, it is generally characterized as the network of networks (Berners-Lee, 1999). The network attribute of the Internet is altering interorganizational relations, from a

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

157

mechanism of hierarchy or power to a variety of network forms (Achrol and Kotler, 1999). Most organizations run their own wesbites, regardless of whether their activities, services, or products are concerned with the Internet. Some organizations such as Yahoo.com largely exist as an independent entity on the web with little presence in physical space. Thus, websites may be regarded as organizations themselves. Websites in transition toward greater competition can get into managerial difficulties if they do not satisfy various users’ needs or concerns such as quality content or transaction security. Nonetheless, it is difficult for a website in the early stage of development to take all the steps necessary to meet online audience’s expectations, because of the Internet’s complex infrastructure. This situation is accelerating networking among websites for the purpose of either complementing the weakness of individual websites or strengthening their competitive positions. According to Park et al. (2002), websites are creating a hyperlink-network that links their partners, in order to enhance their efficiencies in terms of quality contents, technological sophistication, brand reputation, and customer management. Through hyperlinks, websites are getting connected with others, seeking to accomplish a number of goals. For example, websites can communicate purposively with other websites whose contents are more appropriate to geographically distinct audiences. Assuming that interorganizational communication occurs at the organization’s websites, then the basic structural element of the Internet is the hyperlink. A hyperlink may be defined as a technological capability that enables one specific website (or webpage) to link directly with another. The Internet as a hyperlink system has let organizations running a website on the Internet expand their social relations by making possible easy and direct contacts with other organizations anywhere in the world. In other words, using hyperlinks, organizations are able to facilitate bilateral communication and coordination that crosses and/or strengthens off-line boundaries within and between organizations regardless of their physical location. They can be linked together, exchange information, and maintain cooperative relationships by means of a hyperlink centered around common background, interest, or project. In other words, a network structure linking organizations can be found on the Internet. Shared hyperlinks among websites can be discussed from the perspective of interorganizational communication networks. The websites are treated as nodes (or organizations) and the individual hyperlinks as communication links. According to Eisenberg et al.’s (1985) distinction among interorganizational relations, there are two types and three levels of interorganizational linkage. Type may be divided into information and material exchanges. Level is classified as institutional, representative, and personal linkages. From this perspective, interorganizational relations via shared hyperlinks on their websites are information exchanges and institutional linkages. Because information flows occur through

158

NETCOM, vol. 16, n° 3-4, 2002

hyperlinks among organizations’ websites and the hyperlink represents a relational connection or link among organizations. Thus, hyperlinks may be the formalized bridge among organizations. Hyperlinks serve as symbols or signs of institutional interorganizational linkage. The greater is the number of hyperlinks between any two websites, the more proximate is the organizations. 2. DETERMINANTS OF INTERORGANIZATIONAL HYPERLINK NETWORKS

As described above, the organizational environment leads the organizations (or websites) to be interdependently interacting with each other on the Internet. What facilitates interorganizational network development ? In other words, what motivates an organization to have on its website hyperlink with another website ? The expectations organizations have on one another (e.g., some social characteristics or attributes) may be associated with interorganizational relations (Mitchell, 1973). Given that many websites are still at an early stage of development, a partner website’s credibility may be a critical element in deciding with whom to hyperlink. Thus, websites would be central in the emergent hyperlink network if they are perceived as more credible. Organizations pursue particular interests. The main purpose of a website is to acquire a number of visitors. An Internet user’s behavior of visiting (inquiring information or purchasing a product) a specific website may be understood as a form of persuasion. It entails a change in one’s attitude or behavior over time. A communicator’s credibility, in this case, website’s credibility, has been regarded as one of the most influential factors in the persuasion process (Berlo et al., 1969 ; Hovland et al., 1953 ; Hovland and Weiss, 1951 ; McCroskey and Teven, 1999 ; Sundar and Nass, 2001). Recently, researchers have turned their attention to website (or Internet) credibility (Flanagin and Metzger, 2000 ; Johnson and Kaye, 1998 ; Park et al., 2002 ; Schweiger, 2000 ; Tseng and Fogg, 1999). Various definitions can be summarized as the degree to which people believe a website in terms of its trustworthiness, expertise and safety. Each dimension of website credibility may be defined as follows : trustworthiness refers to the truthfulness of a website’s contents and the site’s reputation. Expertise is the website’s apparent competence : how complete, useful, and timely the services of the website are compared to others. Safety is defined as how secure or reliable the website’s technical systems are for online payment and personal information. Once users perceive that a website lacks credibility, they are likely to stop visiting it or performing financial transactions (Gefen, 2000 ; Park et al., 2002 ; Tseng and Fogg, 1999). Thus, one can infer that websites would prefer those websites with high credibility in selecting their partners in cyberspace (Evans and Wurster, 1999 ; Hagel and Armstrong, 1997 ; Shapiro and Varian, 1999). Particularly, websites get connected with others by forming a hyperlink. Past studies corroborate the

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

159

credibility theory4. Ching et al. (1996) examined networks among information technology firms and found that the reputation of a company had a strong relationship to the selection of current and potential partners. Website credibility may be positively related with interorganizational linkage formation among websites. In relation to hyperlink networks, Terveen and Hill (1998) examined the number of hyperlinks between websites as an indicator of the quality of sites and found that hyperlink connectivity had a significant relationship to the expert quality judgments of sites. Also, the indegree connectivity of a site (the number of sites that are linked to a given site) was positively correlated with these judgments. Palmer et al. (2000) used the number of inward hyperlinks to a website as an indicator of the trust of Internet firms. The results revealed that the number of incoming links was strongly related with the use and prominence of TTPs (Trusted Third Parties) and privacy statements that may be regarded as another trust indicator5. Further evidence can be found in more recent studies. A series of studies conducted by Persuasive Technology Laboratory at Stanford University found that having a partner’s website hyperlinked may influence the perceived credibility of the websites (Fogg et al., 2001). Thus, a website that intends to increase its credibility adds hyperlinks to credible websites. A website perceived as highly credible receives more links from others. The websites would be more central in an interorganizational network if they are more credible. In other words, interorganizational hyperlink networks are formed based upon perceived website credibility. 3. CONSEQUENCES OF INTERORGANIZATIONAL HYPERLINKAGE

What are the consequences of interorganizational networks via shared hyperlinks ? Considering that communication links or relationships are likely to be reciprocal, a website’s credibility increases when associating with a credible site because the credibility is partially transferable. A website may enhance its credibility by forming social networks with credible sites. From the perspective of information retrieval (Henzinger, 2001 ; Fogg et al., 2001), a particularly important aspect is whether websites furnish their users with relevant contents and quality 4. Regarding the factors that facilitate interorganizational network development, most research has been devoted to tangible and economic exchange relationships : organizational size, resource needs, buyer-seller, or supplier-customer. There has been relatively little attention to non-economic exchange relationships in the study of interorganizational networks. 5. One might ask the question, are trust and credibility synonymous concepts ? In other words, how different (or similar) are the two concepts ? According to Tseng and Fogg (1999), trust generally indicates a positive belief about the perceived dependability of a person, object, or process. For example, it is different from credibility when involving the effectiveness of technological capability, like a “trust system” frequently used in computer technology (Stefik, 1999). But, it can be used synonymously with credibility when referring to the psychological construct such as people’s beliefs or expectations.

160

NETCOM, vol. 16, n° 3-4, 2002

information in the shortest path. In other words, the perceived credibility of a website may be contingent on how much it strengthen its visitor’s navigating position. A relatively incompetent website benefits from having a hyperlink network with competent sites, since it offers visitors a broader range of navigation opportunities. Also, the configuration of social networks conveys reliable information about the quality and prestige of the central actors in a network (Burt, 1992 ; Granovetter, 1985 ; Mizruchi and Galakiewicz, 1993) so that a central site’s credibility would accrue as a result of the frequency or intensity of hyperlink relations. A hyperlink from website A to site B is a recommendation of site B by the author of site A (Henzinger, 2001 ; Kleinberg, 1999a ; Kleinberg, 1999b). Interorganizational alliances could affect the perception of the organization’s performance, in this case, website credibility perception. Past research suggests a positive relationship between interorganizational network centrality and the perceptions of organizational power or influence (Boje and Whetten, 1981). Further evidence can be found in more recent studies, in relation to interorganizational endorsement. Stuart et al. (1999) studied young biotechnology companies. Their research revealed that young companies “endorsed” by prestigious associates are perceived as having better performance than ventures that lack prominent associates. This research examines interorganizational networks among websites defined by shared hyperlinks among the websites. It investigates the determinants (and consequences) of who is hyperlinked to whom on the web and the role a website’s credibility plays in forming the network. It suggests that the hyperlink network on the web is likely to be among those sites that can leverage an individual website’s credibility (trustworthiness, expertise, and safety synergies). Based on the literature presented above, the following research question is raised : how is interorganizational network, defined as the shared hyperlinks among the websites, associated with the number of visitors and the perceptions of individual website’s credibility ? 4. METHODS

This paper uses multiple methods : 1) a survey to measure the perceptions of website credibility by Internet users and 2) a network analysis of interorganizational hyperlinkage. The results of the two methods are combined. This methodological strategy has the advantage for identifying hyperlink networks among websites, by examining why and how websites are interconnected. 4.1. Sample

The sample for this research is the 50 most frequently visited Korean websites obtained from the Internet Metrix for June, 2001. The most popular website list of the Internet Metrix has been used in earlier research (Park et al., 2002). Also,

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

161

the list included the number of visitors per website (Table I). Further, Table II provides a description of the type of service and ownership of each of the 50 sites. Most of them are dot-com companies since the names of companies and those of websites are the same. In this case, websites best represent organizations. The information of companies was obtained from the Alexa (www.alexa.com). Alexa.com provides a variety of statistics about individual websites. Their data have been frequently used for Internet research because they are useful in terms of wide coverage and ease access (Kumar et al., 1999 ; Palmer et al., 2000). The number of Koreans using the Internet has increased rapidly : 0.14 million in 1995, 1.6 million in 1997, 10 million in 1999, 19.04 million in 2000, and 22.23 million in September 2001 (Korean Network Information Center, 2002). Korea ranks first in the Asia-Pacific region in the penetration of broadband access for 2001. 95 percent of home Internet users, or 15.8 million people, have highspeed connections such as cable modem (Korea Herald, 2001). According to Nielsen/NetRatings (2001), 23 Korean websites are included in the list of world’s 100 biggest locally accessed web properties. This suggests that Korean websites represent an adequate sample for the Internet research.

Table I.— 50 Most Frquently Visited Websites (Number of visitors is expressed in thousands). ID

Website

Number of Visitors

ID

Website

Number of Visitors

ID

Website

Number of Visitors

1

daum

14457

18

iloveschool

5353

35

hompy

3718

2

yahoo

13235

19

chosun

5289

36

korea

3592

3

naver

12674

20

msn

5232

37

sayclub

3519

4

lycos

11355

21

netsgo

5214

38

mycgi

3390

5

netian

11308

22

bugsmusic

5211

39

soribada

3380

6

dreamwiz

10487

23

kbs

4911

40

hitel

3234

7

hanmir

10338

24

imbc

4877

41

joins

2735

8

chollian

9074

25

simmani

4646

42

channeli

2718

9

superboard

8839

26

easynara

4611

43

VRML24

2625

10

hihome

7754

27

fireball

4443

44

sportsseoul

2609

11

empas

7751

28

cgiserver

4414

45

skylove

2513

12

shinbiro

7132

29

weppy

4265

46

megapass

2493

13

hananet

7117

30

x-y

4148

47

donga

2479

14

com.ne.

6804

31

hankooki

4021

48

auction

2472

15

freechal

5814

32

interpia98

3941

49

okcashbag

2352

50

stoo

2341

16

sbs

5564

33

cgiworld

3856

17

dreamx

5423

34

damoim

3730

162

NETCOM, vol. 16, n° 3-4, 2002

4.2. The data 4.2.1. Hyperlinks

The hyperlinks among 50 websites were traced using the Altavista.com6. The number of hyperlinks between two websites was entered into each cell, sij, in an 50 x 50 matrix S1. Because we are concerned with the directionality of the hyperlinks, S1 was made asymmetrical (sij ≠ sij). To examine dichotomous relations, another asymmetrical matrix S2 was created. In this case, sij is a simply a zero or a one depending if there is a hyperlink between node VRML and j. The data were gathered during a 7-day period, from August 1 to 7, 2001. 4.2.2. Website credibility perceptions

A survey of Internet users in Korea was conducted, to rate the 50 websites based on three dimensions of credibility (trustworthiness, expertise, and safety) and to compare their centralities in the interorganizational network with perceived credibility. 4.3. Instrument construction

The survey for Internet users consisted of six topics : trustworthiness, expertise, safety, overall judgment of credibility, age, and gender. The three dimensions of credibility were measured by asking about specific items comprising each dimension. Trustworthiness : (a) truthfulness of services such as content, (b) reputation of website. Expertise : (a) usefulness of services such as contents, (b) timeliness of services such as contents, and (c) competency of services such as contents. Safety : (a) security of privacy protection, (b) reliability of technical systems such as online payment. Each item was measured using a fractionation scale (Barnett et al., 1981). For example, trustworthiness was defined as the truthfulness of services such as contents. Subjects were asked to judge the website’s services in terms of truthfulness. Question asked respondents to let 50 represent the average judgment of a given topic and zero represent none of the topic. Subjects were asked to quantitatively estimate how truthful websites’ services in relation to this standard.

6. For the detailed search command of AltaVista, refer to the advanced query of altavista.com. In the past, several researchers used AltaVista to trace the hyperlinks among websites (Adamic and Adar, 2001 ; Brunn and Dodge, 2001). However, there is a criticism about the academic use of search engine for hyperlink analysis (Snyder and Rosenbaum, 1999). They cast a question about the reliability of the results from search engine. It should be noted here that there must be a systematic error in tracing hyperlinks using AltaVista. The error may be significantly related with results. But, it is very difficult to estimate what the error is. At this point, AltaVista is only search engine to trace incoming and outgoing links between websites (Park, in press).

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

163

Table II- Type of Service and Ownership of each of the 50 Websites (www.Alexa.com). (Lievrouw, Rogers, Lowe, & Nadel, 1987). Note : ISP = Internet Service Provider. ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

Website daum.net kr.yahoo.com naver.com lycos.co.kr netian.com dreamwiz.com hanmir.com chollian.net superboard.com hihome.com empas.com shinbiro.net hananet.net com.ne.kr freechal.com sbs.co.kr dreamx.net iloveschool.co.kr chosun.com msn.co.kr netsgo.com bugsmusic.co.kr kbs.co.kr imbc.co.kr simmani.com easynara.com fireball.co.kr cgiserver.co.kr weppy.com x-y.net hankooki.co.kr interpia98.com cgiworld.co.kr damoim.net hompy.com korea.com sayclub.co.kr mycgi.co.kr soribada.com hitel.net joins.com channeli.net VRML24.com sportsseoul.com skylove.com megapass.net donga.com auction.co.kr okcashbag.com stoo.com

Name of Company Daum Communications Yahoo Korea Naver.com Lycos Korea Hancom Net Dream Wiz Korea Telecom Dacom Superboard Technophil Knowledge Power Plant Onse Telecom Hanaro Telecom Kil Information Consulting Freechal Seoul Broadcasting System Dreamline Human Communications and Consulting The Digital Chosun-Ilbo Microsoft Corporation SK Telecom Bugstech Korean Broadcasting System Internet Media Broadcasting Company Simmany Easynara Daum Communications Korea Networks Unitel New Korea HK Internet Interpia98 Holy Net Damoim Hompy Korea Thrunet Neo Wiz Mycgi Sorinara Korea PC Telecom Joins.com LG InterNet Simplex Internet Sports Seoul 21 Skylove Korea Telecom Dong-A Ilbo Internet Auction SK Corporation Stoo.com

Service Portal/Search Portal/Search Portal/Search Portal/Search Portal/Search Portal/Search Portal/Search ISP Community/Computer Community/Internet Portal/Search ISP ISP Community/Computer Community/Chat Media/Broadcasting Community/Chat Community/Chat Media/Newspaper Portal/Search ISP Entertainment Media/Broadcasting Media/Broadcasting Portal/Search Community/Internet Community/Chat Computer/Internet Community/Chat Computer/Internet Media/Newspaper Community/Computer Community/Computer Community/Chat Community/Internet Portal/Search Community/Chat Community/Internet Entertainment ISP Media/Newspaper Community/Chat Community/Chat Media/Newspaper Community/Chat ISP Media/Newspaper Shopping/Auction Community/Chat Media/Newspaper

164

NETCOM, vol. 16, n° 3-4, 2002

4.4. Survey questionnaire

Based on instruments above, the survey questionnaire was originally written in English. In order to conduct a survey of Internet users in Korea, standard back-translation procedures were employed to translate the questionnaire from English to Korean. Specifically, the back-translation procedure used in this research operates as follows : First, the survey questionnaire is written in English. Next, it is translated into the second language, in this case, Korean, by a second person and then, a third person translates the questionnaire back into English. Lastly, the English and Korean versions are compared by a bilingual who modifies the Korean version based on the comparison. The back-translation procedure has shown the best results in past research, by eliminating the variance between original language and translation (Brislin and Sinaiko, 1973). 4.5. Data collection

First, the selection of Korean subjects was obtained from those who are taking courses in communication theory at an urban university in a large Korean city. The survey was conducted during August 2001. Specifically, the date of the survey was announced in advance during class and students voluntarily participated. The announcement also indicated that the participants would receive research credit. Thus, a researcher confirmed whether students filled out the questionnaire at the time of redemption. Each subject was asked to evaluate five of the 50 websites. The five websites were randomly selected. Considering the objects measured are websites, subjects were given about 7 days to complete survey. During the period, they were able to visit websites if preferred. 164 students participated in the survey. 4.6. Analysis 4.6.1. Network analysis

The hyperlink connectivity matrix was analyzed using the UCINET-XVRML computer program (Borgatti et al., 1999). Network analysis is a set of research procedures for identifying structures in social systems based on the relations among the system’s components rather than the attributes of individual cases (Rogers and Kincaid, 1981 ; Richards and Barnett, 1993). The method may be generalized to describe the patterns of communication among different social systems or individuals in the specific system. For the structure of the hyperlink network, the density was measured. Density is the actual number of links divided by the number of possible links, [n(n-1)] (for directional data). It is a good measure of group cohesiveness or knittedness (Wasserman and Faust, 1994). With density indicator, network centralization is also useful to examine overall characteristics of the network. Centralization

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

165

(or global centrality) indicates the extent to which a network is organized around its most central point (Freeman, 1979). For the 50 websites, centrality of each site was determined. Centrality refers to the position of actors relative to other actors in a network. Freeman’s degree measure was used to measure centrality (Freeman, 1979). Freeman’s Degree centrality consists of ingoing and outgoing degree. Indegree refers to the number of links a node receives from the other nodes, while outdegree is the number of links originating from a node. Indegree centrality can be interpreted as the credibility of a site in the hyperlink network. Outdegree is also a good measure of expertise credibility in that it indicates a website’s expansiveness. 4.6.2. Pearson’s correlations

Pearson’s correlations were used to examine the relationship between network centrality, the number of visitors and the perceptions of individual website’s credibility. This research expects to find that the more credible the website, the more central the website’s position in the network, and the more central the website’s location in the network the more visitors it receives. 5.RESULTS

5.1.Description of network

Among the 50 websites, the system density is 0.567. There are about 1,389 links, of a possible number of 2,450. This network is very dense. More than 50 % of the potential hyperlinks among the websites are present. The degree centralities for the 50 top websites are presented in Table III. The centralities are comprised of ingoing and outgoing degree. The indegree centralities reveal that portal websites or search engines (daum, yahoo, naver, lycos, netian, dreamwiz, hanmir, dreamx, simmani, and korea) are most central group in the hyperlink network. Next, Internet service providers (chollian, shinbiro, hananet, netsgo, and hitel), mass media (sbs, chosun, imbc, joins, and donga), homepage service (hihome) and cybercommunity (freechal) are the next most central websites. Most peripheral in the network are easynara and mycgi that provide programs related to computer and Internet. They do not readily allow for easy interpretation. The outdegree centralities also show that portal websites and search engines occupied the most central positions. Overall, the centralities measures are quite similar to indegree’s results. They correlate .733 (p = .000), indicating that centrality based solely on the pattern of inward linkage and that of the outgoing links between the nodes ranked the websites similarly. Given that the goal of portal sites and search engines is to provide users with a variety of services including contents, it is no wonder that they are connected to

166

NETCOM, vol. 16, n° 3-4, 2002

Table III.- Degree Centralities. Network Centralization (Outdegree) = 45.111 %. — Network Centralization (Indegree) = 17.474 ID

Website

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

daum yahoo naver lycos netian dreamwiz hanmir chollian superboard hihome empas shinbiro hananet com.ne. freechal sbs dreamx iloveschool chosun msn netsgo bugsmusic kbs imbc simmani easynara

OutDegree

InDegree

ID

Website

47.000 49.000 49.000 49.000 49.000 49.000 49.000 49.000 32.000 49.000 49.000 48.000 46.000 45.000 46.000 29.000 0.000 0.000 22.000 8.000 48.000 0.000 23.000 5.000 46.000 7.000

35.000 35.000 36.000 33.000 36.000 36.000 33.000 35.000 27.000 31.000 29.000 34.000 35.000 27.000 30.000 31.000 32.000 23.000 32.000 25.000 33.000 20.000 29.000 30.000 32.000 14.000

27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

fireball cgiserver weppy x-y hankooki interpia98 cgiworld damoim hompy korea sayclub mycgi soribada hitel joins channeli cafe24 sportsseoul skylove megapass donga auction okcashbag stoo Mean S.D.

OutDegree

InDegree

1.000 44.000 36.000 33.000 26.000 39.000 9.000 0.000 12.000 33.000 0.000 13.000 0.000 39.000 40.000 44.000 0.000 0.000 0.000 45.000 33.000 25.000 5.000 19.000 27.780 27.780

15.000 28.000 26.000 24.000 23.000 29.000 26.000 19.000 20.000 31.000 26.000 14.000 20.000 34.000 31.000 31.000 24.000 22.000 23.000 24.000 34.000 28.000 22.000 22.000 19.190 5.934

the most sites in the network. Likewise, their high indegree centrality scores can be interpreted as the degree of involvement of major websites in information gathering and distribution activities : the more information they seek to provide, more likely that the portal and search sites play an important role in the network. The centralities of mass media websites such as chosun and donga may be explained in a similar vein. Despite the same value of the means for indegree and outdegree, 27.78, the standard deviations (indegree 5.93, outdegree 19.19) and centrality measures (indegree 17.5 %, outdegree 45.1 %) are quite different. Centralization (or global centrality) shows the extent to which a network is organized around its focal point (Freeman, 1979). A centralized network (the higher percentage, the more centralized) may reflect an uneven distribution of resources (in this case, hyperlinks) such that resources are concentrated in the central points of the network. Thus, the difference could be explained by the fact that major portal and search engine websites link heavily to other sites within the group but they (that it, other sites) are more likely to send hyperlinks to sites outside the group.

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

167

Consumers online tend to want a variety of services. They would like to check the price, quality, and information about products before buying from a site. Thus, in order for websites to satisfy digital shoppers, sites need to reinforce their competency by strengthening their visitor’s navigating position (Evans and Wurster, 1999). Powerful navigation comes from various dimensions of extension : inter-website connections, cooperation between different types of Internet businesses, and quality and quantity of product information. In this process, networks among websites become organized around a core website (Hagel and Armstrong, 1997). A relatively incompetent website will benefit from a larger network, since a core website offers visitors a broader range of navigation opportunities. Once such a network is built, a core website’s expertness also increases because it is more comprehensive. In other words, social capital represented by the relationships of hyperlink is the final arbiter of competitive success (Burt, 1992). Websites gain resources through their hyperlink networks. Further, portal and search sites have a number of visitors so that strategic hyperlink alliances with the sites can encourage many people to visit a website more often. In other words, other websites are positioning themselves as gateways to prominent portal sites. 5.2. Results of survey

164 students participated in the survey. Of these, 133 (81.1 %) were male and 31 (18.9 %) female. Since each subject was assigned to evaluate five of the 50 websites, a total number of 820 questionnaires were collected. 2 out of 820 questionnaires were discarded because they were not reliable. The average age of subjects was 21.94 years (s.d. = 2.09). The reliability of the instrument using Cronbach’s alpha was .90 : They are truthfulness, reputation, usefulness, timeliness, competency, security of privacy protection, reliability of technical systems such as online payment, and overall credibility perception (the perceived credibility of the website as a whole). Theoretically, three underlying dimensions (trustworthiness, expertise, and safety) are assumed among seven variables (truthfulness, reputation, usefulness, timeliness, competency, security, and reliability) to measure the website credibility. Factor analysis was employed to Table IV.- Results of Factor Analysis. determine if there are three dimensions. Factor analysis was done using the prinComponent Loadings cipal factors extraction method and a Items 1 2 3 VARIMAX rotation. Prior communalities Truthfulness .873 .299 .288 were set at 1.0. The criterion for retainReputation 0.07962 .826 .218 ing factors was a minimum Eigenvalue Usefulness .244 .848 .216 of .05. The component matrix is summaTimeliness .449 .705 .125 rized in Table IV. Competency .248 .810 .363 Only one item, truthfulness, Security .263 .214 .859 loaded on the Factor 1 (Trustworthiness Reliability .117 .268 .887

168

NETCOM, vol. 16, n° 3-4, 2002

Factor). Four items (reputation, usefulness, timeliness, and competency) loaded on the Factor 2 (Expertise Factor). Security and reliability loaded on the Factor 3 (Safety Factor). Thus, three underlying dimensions were found. 5.3. Correlation analysis

Relationships among a website’s network centrality, the number of visitors, and the perceptions of site’s credibility were examined. The number of visitors was significantly correlated with all other variables. Particularly, it showed a fairly strong relationship with outdegree (r = .557), indegree (r = .572), truthfulness (r = .486), reputation (r = .501), usefulness (r = .523), competency (r = .524), and overall judgment (r = .596) at the .000 level. Indegree centrality had a moderate correlation with truthfulness (r = .243), reputation (r = .355), safety (r = .333), reliability (r = .291), and overall judgment (r = .358) at the .01 or .05 level. In addition, it was found that outdegree centrality was not significantly related to any of the credibility variables. 6.DISSCUSSION Table V.— Correlations Matrix (* Correlation is significant at the 0.05 level (1-tailed).— ** Correlation is significant at the 0.01 level (1-tailed).— N = 50). Items

Visitor

Visitor

1.00

OutDegree

InDegree

OutDegree

.557** .000

1.00

InDgree

.572** .000

.733** .000

Truthful

.486** .000

.089 .270

.243* .044

Reputation

.501** .000

.090 .268

.355** .006

Useful

.523** .000

.090 .267

.224 .059

Timeliness

.289* .021

-.113 .218

.085 .279

Competency

.524** .000

.083 .283

.219 .063

Safety

.395** .002

.185 .185

.333** .009

Reliability

.327* .010

.157 .138

.291* .020

Overall

.596** .000

.214 .068

.358** .005

1.00

Hyperlinks between two websites, which was initiated by Amazon.com, is becoming the representative form of interorganizational cooperation among websites (Shapiro and Varian, 1999). This form of networking is increasing rapidly. In this context, the purpose of this paper was to examine the structure of an emerging hyperlink network among websites and the factors affecting this pattern. To do this, this paper used interorganizational network theory and multiple methods : network analysis and survey. In this study, portal websites (or search engines) occupied the most central positions in the network. Also, Internet service provider, mass media, homepage, and cybercommunity websites were fairly central. Interestingly, outdegree centralization is higher than indegree. Another important finding was

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

169

a relatively dense system of websites. One of the more important findings is that the indegree centralities of the hyperlink network are influenced by the number of visitors and the credibility perceptions of partner websites to which other sites are hyperlinked. Given the probability that there is a hyperlink between two randomly chosen websites is nearly 0 (Terveen and Hill, 1998), the density of those websites is exceptionally high. However, Albert et al.’s (1999) research shows that the web has the flocking nature. According to them, you can get a document to the other document by clicking on hyperlinks on average 19 times, if you select two webpages at random. Terveen and Hill’s (1998) research also reveal that topically related sites are tightly knitted together by inter-site link. While the web as a whole is a very sparely connected network, one can start with a seed site and navigate a large majority of related sites by clicking just one or two times. The high density reported here was most likely due to the selected sample. The 50 most frequently visited Korean sites may be regarded as potentially good examples of seed sites for navigation on the web. Had a greater variety of sites been selected for evaluation, the density of the network would probably have been lowered. There was a higher outdegree centralization relative to indegree. Most of the sample sites receive hyperlinks from sites within the network but they tend to send few links to the sites. This may be due to the competition among popular sites. Commercial sites that are in competition with one another do not link (Terveen and Hill, 1998). Further, Kleinberg (1999b) found that prestigious websites may not link to other famous sites. Generally, the results are consistent with these findings. While portal sites and/or search engines send many links to other sites, they give relatively few links back. In past research, the positions and roles of websites having high indegree centrality were theoretically assumed and partially analyzed as if there were a causal association between the results and website’s intention to increase credibility (Henzinger, 2001 ; Kleinberg, 1999a ; Kleinberg, 1999b ; Park et al., 2002). The more credible is the website, the more incoming links the website have. The more incoming links, the more visitors the website has, and vice versa. The results of this research support these contentions. This may indicate that the website credibility functions as a causal factor or antecedent influencing the hyperlinkage pattern among websites. One interesting finding in the factor analysis is that truthfulness had a single high loading on Factor 1 (trustworthiness) while reputation was highly loaded on the Factor 2 (expertise) differentiating itself from truthfulness. This reveals that the reputation of website is more influenced by its expertise rather than trustworthiness or morality.

170

NETCOM, vol. 16, n° 3-4, 2002

7. LIMITATIONS AND FUTURE RESEARCH

The research found significant associations among hyperlink network structures, the number of visitors and perceptions of individual website’s credibility. However, this should be viewed skeptically. Because the sample is only the 50 most frequently visited websites, the websites in the sample may all be considered as credible. Had a greater variety of sites been included in the sample, there would have been greater variance in the crediability of the sites. In addition to website’s credibility, such factors like advertising (Hoffman and Novak, 2000), homophilous attributes (Park et al., 2000), interface (Park, 2001), or interpersonal communication may account for hyperlink networks among websites. A second problem with the research is the use of students to examine the perceptions of the credibility of the individual websites rather than having website managers evaluate the sites. An email survey of website managers was conducted. However, in spite of numerous follow-ups, the return rate was too low to justify an analysis of their responses. A third problem with this research was that the data came from a single point in time. This network is in the process changing over time. Thus, the snapshot provided by this research may provide a distorted picture of the interorganizational hyperlink network due to the particular circumstances. The above discussion suggests the need for future research to examine how the hyperlink network changes over time. Also, future research should include a greater variety of websites to determine the relation among a site’s network centrality, visiting behavior and perceived credibility. In this way, the impact of a website credibility on other sites and the public may be determined. REFERENCES ACHROL R.S. & KOTLER P. (1999). “Marketing in the network economy”. Journal of Marketing, vol. 146. ADAMIC L.A. & ADAR E. (2001). “You are what you link”. The 10th annual International World Wide Web Conference [Online], Hong Kong, Available : http://www10.org/ program/society/yawyl/YouAreWhatYouLink.htm ALBERT R., JEONG H. & BARABASI A.L. (1999). “Diameter of the world wide web”. Nature, vol. 401, n° 9, p. 130-131. BARNETT G.A. & CHOI Y. “Physical distance and language as determinants of the international telecommunications network”. International Political Science Review, vol. 16, n° 3, p. 249-266. BARNETT G.A., HAMLIN D.M. & DANOWSKI J.A. (1981). “The use of fractionation scales in communication audits”. in BURGOON M. (ed.), Communication Yearbook 5. New Brunswick (NJ) : Transaction, p. 455-471. BARNETT G.A. & THAYER L. (eds.) (1997). Organization communication : Emerging perspectives (The renaissance in systems thinking) 5, Norwood (NJ) : Ablex.

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

171

BERLO D., LEMERT J. & MERTZ R. (1969). “Dimensions for evaluating the acceptability of message sources”. Public Opinion Quarterly, vol. 33, n° 4, p. 563-576. BERNERS-LEE T. (1999). Weaving the web : The original design and ultimate destiny of the World Wide Web by its inventor. New York : HarperCollins Publishers. BOJE D.M. & WHETTEN D.A. (1981). “Strategies and constraints affecting centrality and attributions of influence in interorganizational networks”. Administrative Science Quarterly, vol. 26, p. 378-395. BORGATTI S., EVERETT M. & FREEMAN L. (1999). UCINET-XV : Network Analysis Software. Columbia, SC : Analytic Technologies. BRISLIN R.W. & SINAIKO W.H. (1973). “Evaluating language translations : Experiments on three assessment methods”. Journal of Philosophy and Rhetoric, vol. 57, p. 328-334. BRUNN S.D. & DODGE M. (2001). “Mapping the ‘Worlds’ of the world wide web : (Re)Structuring global commerce through hyperlinks”. American Behavioral Scientist, vol. 44, n° 10, p. 1717-1739. BURT R.S. (1992). Structural holes : The social structure of competition. Cambridge (MA) : Harvard University Press. CASTELLS M. (1996). The rise of the network society. Cambridge (MA) : Blackwell Publishers. CHING C., HOLSAPPLE C.W. & WHINSTON A.B. (1996). “Toward IT support for coordination in network organizations”. Information & Management, vol. 30, n° 4, p. 179-199. CONTRACTOR N.S. & EISENBERG E.M. (1990). “Communication networks and new media in organizations”. in FULK J. & STEINFIELD C.W. (eds.), Organizations and communication technology. Newbury Park (CA) : Sage, p. 143-172. DEWETT T. & JONES G.R. (2001). “The role of information technology in the organization : A review, model, and assessment”. Journal of Management, vol. 27, n° 3, p. 313-346. EISENBERG E.M., FARACE R.V., MONGE, P.R., BETTINGHAUS E.P., KURCHNERHAWKINS R., MILLER K.I. & ROTHMAN L. (1985). “Communication linkages in inter-organizational systems : Review and synthesis”. in DERVIN B. & VOIGT M.J. (eds.), Progress in communication sciences. Norwood (NJ) : Ablex, p. 231-261. EVANS P. & WURSTER T.S. (1999). “Getting real about virtual commerce : Strategies for success in electronic and physical retail commerce”. Harvard Business Review, vol. 77, n° 6. FLANAGIN A.J. & METZGER M.J. (2000). “Perceptions of internet information credibility”. Journalism & Mass Communication Quarterly, vol. 77, n° 3, p. 515-540. FOGG B.J., MARSHALL J., LARAKI O., OSOPOVICH A., VARMA C., FANG N., PAUL J., RANGNEKAR A., SHON J., SWANI P. & TREINEN M. (2001). “What makes web sites credible ? A report on a large quantitative study”. The Computer-Human Interaction Conference. FREEMAN L.C. (1979). “Centrality in social networks : conceptual clarification”. Social Networks, vol. 1, p. 215-239. GEFEN D. (2000). “E-commerce : The role of familiarity and trust”. OMEGA-International Journal of Management Science, vol. 28, n° 6, p. 725-737. GIDDENS A. (1990). The consequences of modernity. Stanford (CA) : Stanford University Press.

172

NETCOM, vol. 16, n° 3-4, 2002

GRANOVETTER M. (1985). “Economic action and social structure : A theory of embeddedness”. American Journal of Sociology, vol. 91, p. 481-510. HAGEL J. III. & ARMSTRONG A.G. (1997). Net gain : Expanding markets through virtual communities. Boston (MA) : Harvard Business School Press. HART P. & ESTRIN D. (1991). “Inter-organizational networks, computer integration, and shifts in interdependence : The case of the semiconductor industry”. ACM Transactions on Information Systems, vol. 9, n° 4, p. 370-398. HENZINGER M.R. (2001). “Hyperlink analysis for the web”. IEEE Internet Computing, vol. 5, n° 1, p. 45-50. HOFFMAN D.L. & NOVAK T.P. (2000). “How to acquire customers on the web”. Harvard Business Review, vol. 78, n° 3, p. 179-188. HOVLAND C. & WEISS W. (1951). “The influence of source credibility on communication effectiveness”. Public Opinion Quarterly, vol. 15, n° 4, p. 633-650. HOVLAND C., JANIS I. & KELLEY H. (1953). Communication and Persuasion. New Haven (CT) : Yale University Press. JOHNSON T. & KAYE B. (1998). “Cruising is believing ? : Comparing the internet and traditional sources on media credibility measures”. Journalism & Mass Communication Quarterly, vol. 75, n° 2, p. 325-340. KETTINGER W. & GROVER V. (1997). “The use of computer-mediated communication in an interorganizational context”. Decision Sciences, vol. 28, n° 3, p. 513-555. KLEINBERG J.M. (1999a). “Hubs, authorities, and communities”. ACM Computing Surveys [Online], vol. 31, n° 4, Available : http://www.cs.brown.edu/memex/ACM_HypertextTestbed/papers/10.html KLEINBERG J.M. (1999b). “Authoritative sources in a hyperlinked environment”. Journal of the ACM, vol. 46, n° 5, p. 604-632. KOREA HERALD (2001). “Broadband penetration nears saturation in Korea” [Online], Available: http://www.koreaherald.co.kr/SITE/data/html_dir/2001/09/27/200109270007.asp KOREA NETWORK INFORMATION CENTER (2002). “Official statistics in relation to Internet in South Korea” [Online], Available : http://www.nic.or.kr KUMAR R., RAGHAVAN P., RAJAGOPLALAN S. & TOMKINS A. (1999). “Trawling the Web for cyber communities”. The 8th World Wide Web Conference. MALONE T.W., YATES J. & BENJAMIN R. (1987). “Electronic markets and electronic hierarchies : Effects of information technology on market structure and corporate strategies”. Communications of the ACM, vol. 30, n° 6, p. 484-497. MCCROSKEY J.C. & TEVEN J.J. (1999). “Goodwill : A re-examination of construct and its measurement”. Communication Monographs, vol. 66, p. 90-103. MITCHELL J.C. (1973). “Networks, norms and institutions”. in BOISSEVAIN J. & MITCHELL J.C. (eds.), Network analysis studies in human interaction. Hague : Mouton, p. 15-35. MIZRUCHI M.S. & GALAKIEWICZ J. (1993). “Networks of interorganizational relations”. Sociological Methods and Research, vol. 22, p. 46-70. NIELSEN/NETRATINGS (2001). “Korea claims 23 of world’s 100 biggest locally accessed web properties”. Press Release [Online], Available: http://www.nielsennetratings.com/news.jsp

INTERORGANIZATIONAL HYPERLINK NETWORKS AMONG WEBSITES IN SOUTH KOREA

173

PALMER J.W., BAILEY J.P. & FARAJ S. (2000). “The role of intermediaries in the development of trust on the WWW : The use and prominence of trusted third parties and privacy statements”. Journal of Computer-Mediated Communication [Online], vol. 5, n° 3, Available : http://www.ascusc.org/jcmc/vol5/issue3/Palmer.htm PARK H.W. (in press). “What is hyperlink network analysis ? : New Method for the Study of Social Structure on the Web”. Connections. PARK H.W. (2001). “Older adults and persuasive technology - Developing persuasive principles and integrating into artifacts”. The annual conference of International Communication Association. PARK H.W., BARNETT G.A. & NAM I.Y. (2002). “Hyperlink-Affiliation network structure of top websites : Examining affiliates with hyperlink in Korea”. Journal of the American Society for Information Science and Technology, vol. 53, n° 7, p. 592-601. PARK H.W., BARNETT G.A. & KIM C.S. (2000). “Political communication structure in Internet networks - A Korean case”. Sungkok Journalism Review, vol. 11, p. 67-89. RICHARDS W.D. Jr. & BARNETT G.A. (eds.). (1993). Progress in communication science 12. Norwood (NJ) : Ablex. ROGERS E.M. & KINCAID D.L. (1981). Communication networks : Toward a new paradigm for research. New York : Free Press. SCHWEIGER W. (2000). “Media credibility - Experience or image ? A survey on the credibility of the World Wide Web in Germany in comparison to other media”. European Journal of Communication, vol. 15, n° 1, p. 37-59. SHAPIRO C. & VARIAN H.R. (1999). Information rules : A strategic guide to the network economy. Boston (MA) : Harvard Business School Press. SNYDER H. & ROSENBAUM H. (1999). “Can search engines be used as tools for web link analysis ? : A critical review”. Journal of Documentation, vol. 55, n° 4, p. 375-384. STEFIK M. (1999). The internet edge : Social, technical and legal challenges in a networked world. Cambridge : The MIT Press. STEINFIELD C., KRAUT R. & PLUMMER A. (1995). “The impact of interorganizational networks on buyer-seller relationships”. Journal of Computer-Mediated Communication [Online], vol. 1, n° 3, Available: http:// www.asusc.org/jcmc/vol1/issue3/steinfield.htm STUART T.E., HOANG H. & HYBELS R.C. (1999). “Interorganizational endorsements and the performance of entrepreneurial ventures”. Administrative Science Quarterly, vol. 44, n° 2, p. 315-349. SUNDAR S.S. & NASS C. (2001). “Conceptualizing sources in online news”. Journal of Communication, vol. 51, n° 1, p. 52-72. TERVEEN L. & HILL W. (1998). “Evaluating emergent collaboration on the web”. The Conference of Computer Supported Cooperative Work. TSENG S. & FOGG B.J. (1999). “Credibility and computing technology”. Communications of The ACM, vol. 42, n° 5, p. 39-44. WASSERMAN S. & FAUST K. (1994). Social network analysis : Methods and applications. Cambridge (NY) : Cambridge University Press. WATERS M. (1995). Globalization. New York : Routledge.