Embed Size (px)
Citation preview
Knowledge sharing in virtual communities: an e-business perspective
J. Koh*, Y.-G. Kim
Graduate School of Management, Korea Advanced Institute of Science and Technology (KAIST) 207-43 Cheongryangri-dong,
Dongdaemoon-gu, Seoul 130-012, Seoul, South Korea
Abstract
Thanks to availability of the Internet, virtual communities are proliferating at an unprecedented rate. In-depth understanding of virtual
community dynamics can help us to address critical organizational and information systems issues such as communities-of-practice, virtual
collaboration, and knowledge management. In this article, we develop a virtual community activity framework, integrating community
knowledge sharing activity into business activities in the form of an e-business model. We examine how the level of community knowledge
sharing activity leads to virtual community outcomes and whether such community outcomes are related to loyalty toward the virtual
community service provider. Based on a field survey of 77 virtual communities currently operating in Freechal.com, one of Korea’s largest
Internet community service providers, we found that the level of community knowledge sharing activity is related to virtual community
outcomes and such outcomes are significantly associated with loyalty to the virtual community service provider. These results imply that the
level of community knowledge sharing activity may be a proper proxy for the state of health of a virtual community. Implications of the
findings and future virtual community research directions are discussed.
q 2003 Elsevier Ltd. All rights reserved.
Keywords: Knowledge management; Virtual community; Knowledge sharing activity; Virtual community provider
1. Introduction
Owing to the limits of IT-driven knowledge management
for interactive innovation processes, a community-based
approach has been alternatively spotlighted (Swan, Newell,
& Robertson, 2000). Among a variety of approaches to
knowledge management in organizations (Choi & Lee,
2002; Lee & Kim, 2001; Wiig, Hoog, & Spek, 1997), the
community-based approach has been considered as one of
the most effective tools for knowledge creation and transfer
(Brown & Duguid, 1991; Wegner & Synder, 2000). The
approach emphasizes dialogue through social networks
(person-to-person contact) (Swan et al., 2000), and helps to
informally share knowledge which is obtained from
experienced and skilled people (Jordan & Jones, 1997).
As the Internet revolution has evoked an unprecedented
proliferation of virtual communities all over the world
(Fernback, 1999; Hiltz & Wellman, 1997), exchanging
information and knowledge inside virtual communities
rapidly has dramatically changed our lives. Now infor-
mation and knowledge are often sent directly from member
to member and any member is able to disseminate
information electronically without hierarchical channels
(Alavi & Leidner, 2001; Larsen & Mclnerney, 2002;
Liebowitz, 1999). A virtual community may be understood
as one of the knowledge community types via computer-
mediated communications (CMC). On the commercial
front, the most successful e-commerce initiatives turn out
to be the community-based ones such as Internet auction or
group purchasing. For instance, eBay has established an
Internet auction community of 16 million registered
members as of January 2001 (Sinclair, 2001) and outper-
forms virtually every type of e-commerce rival. On the non-
commercial front, growth of Internet community service
providers worldwide has been phenomenal. The ilove-
school.co.kr, an on-line alumni association support site in
Korea, attracted 7 million members in 12 months (Jan
2000–Jan 2001) and, as of July 2002, is hosting about 1
million virtual communities. The websites such as geocities.
com or iloveschool.co.kr are also trying to develop various
community-based business models.
What implications does this unprecedented growth of
virtual communities have on the information systems (IS)
community? First, understanding of virtual community
dynamics may facilitate virtual collaboration among
0957-4174/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0957-4174(03)00116-7
Expert Systems with Applications 26 (2004) 155–166
www.elsevier.com/locate/eswa
* Tel.: þ82-29583674; fax: þ82-29583604.
E-mail address: [email protected] (J. Koh).
organizations across their organizational boundaries (But-
ler, 2001; Espinosa, Cummings, Wilson, & Pearce, 2003;
Finholt & Sproull, 1990; Scott, 2000). Secondly, transform-
ing the traditional off-line communities-of-practice (CoPs)
(Brown & Duguid, 1991; Wegner & Synder, 2000) into on-
line virtual communities will greatly improve their com-
munity scope (within-site ) inter-site), interaction effi-
ciency (face to face communications ) on-line,
multimedia communications), and sharing of critical
information and knowledge (physical documents ) on-
line repository) (Abbott, 1988; Kim, Yu, & Lee, 2003;
Scott, 2000). Lastly, changing our views of an organization
from a hierarchy of command and control into a network of
competency-based virtual communities will lead us to a
radically different set of organizational design options (Lee
& Kim, 2001; Miles, Miles, Perrone, & Edvinsson, 1998;
Wiig et al., 1997). Sometimes, this may take the form of
Nonaka’s (1994) hyper-text organization where critical
organizational knowledge is created through multiple
modes and media of interaction among individuals and
groups across departmental boundaries and management
levels. In the cases of firms such as Dell Computer and Cisco
Systems, this has been materialized through their extremely
well-maintained global supplier and customer community
networks (Patel, 2002). By transforming suppliers and
customers into their corporate community members, Dell
and Cisco have been able to exchange valuable information
and knowledge with them while, using the same Electronic
Data Interchange (EDI) connection, other firms are proces-
sing orders and invoices (Kraemer & Dedrick, 2002;
Magretta, 1998).
Despite the virtual communities’ explosive growth and
non-trivial implications for the IS community, virtual
community service providers (e.g. geocities.com) that
mainly focus on offering to users their websites as the
place to build virtual communities for knowledge sharing
are searching for their unique profitable business models. In
Korea, major virtual community providers including the
well-known site Daum communications (http://www.daum.
net), try to develop various profitable strategies such as
selling avatars, cyber characters that symbolize a commu-
nity member’s identity in cyberspace, or charging a
community service fee. As the potential profits of the
Internet services for the virtual community providers are
being spotlighted, the link between the level of community
knowledge sharing activity and loyalty toward the commu-
nity service provider is stimulating the curiosity of IS
researchers as well as practitioners. Whether virtual
community services are profitable for the community
providers is still in question although Hagel and Amstrong
(1997) and more recently Rothaermel and Sugiyama (2001)
suggested the revenue potential of virtual communities. In
fact, many community service providers (portals) are
hesitating to invest their money in nurturing their commu-
nities owing to the lack of assurance that community
activation or knowledge sharing activity will finally lead to
a profit. In this study, we intend to examine whether the
level of community knowledge sharing activity predicts a
community service provider’s outcomes (e.g. loyalty) as
well as the virtual community’s outcomes. More specifi-
cally, we ask:
† Is the level of community knowledge sharing activity
associated with virtual community outcomes such as
community participation or community promotion?
† Is community knowledge sharing activity or commu-
nity stimulation really meaningful to virtual commu-
nity service providers? Are virtual community
outcomes related to loyalty toward the virtual com-
munity service provider?
Section 2 reviews the literature on the definitions of
virtual community, virtual community activity for knowl-
edge sharing, and the business value of virtual communities.
In Section 3, we introduce the research model and related
hypotheses of the study. Data collection and analysis
methods are described in Section 4. In Section 5, we report
the results of the statistical tests of the given hypotheses.
Finally, in Sections 6 and 7, we discuss our findings and
implications as well as the limitations of the study.
2. Conceptual background
2.1. Virtual community
A community is mainly characterized by the relational
interaction or the social ties that draw people together
(Heller, 1989). A community can also be seen as a group
where individuals come together based on an obligation to
one another or as a group where individuals come together for
a shared purpose (Rothaermel & Sugiyama, 2001). Gusfield
(1975) distinguished between two kinds of communities. The
first is the traditional territorial or geographic community. In
this sense, community refers to a neighborhood, town, or
region, thus sense of community implies the sense of
belongingness to a specific spatial setting (Obst, Zinkiewicz,
& Smith, 2002). The second is a relational community,
concerned with human relationship without reference to
location. For example, there are communities of interest such
as hobby clubs, religious groups, or fan clubs. These two
types of communities are not necessarily mutually exclusive;
many interest groups can also be location-based commu-
nities. Most of the communities sprouting in the Internet,
called virtual communities, seem to fall under the definition
of relational community since their members are not
physically bound together (Wellman & Gulia, 1999).
However, instead of just exchanging e-mail messages,
members of a virtual community actively interact with
each other for knowledge sharing on a specific site in
cyberspace, thus displaying the same kind of emotional
attachment to their site as people do towards their physical
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166156
place of relationship (e.g. house, workplace) (Brown &
Duguid, 2000).
Fernback and Thompson (1995) characterized the virtual
community as social relationships forged in cyberspace
through repeated contacts within a specified boundary.
Balasubramanian and Mahajan (2001) defined it as any
entity that exhibits all of the following characteristics: (1) an
aggregation of people, (2) rational members, (3) interaction
in cyberspace without physical collocation, (4) social
exchange process, and (5) a shared objective, property/
identity, or interest between members. Preece (2000) also
noted that a virtual community consists of four components:
people, a shared purpose, policies, and computer systems.
Regarding a virtual community as a class of group CMC,
Jones (1997) suggested a minimum set of conditions for
being a virtual community: interactivity, communicators,
sustained membership, and virtual space. In these defi-
nitions of a virtual community, we find common keywords
such as members (people), interaction, cyberspace, and
shared goals.
Despite subtle differences in focus, researchers agree on
the use of ‘cyberspace’ as the essential for the identification
of virtual communities. Some virtual communities exist
strictly in cyberspace. However, in a number of other virtual
communities, community members to engage in off-line as
well as on-line interactions (Weinreich, 1997). Rothaermel
and Sugiyama (2001) noted that a virtual community may
not be a complete substitute for personal, simultaneous, one-
to-one interaction, either vocally or face to face. In their
study, about 30% of the respondents communicated with
other TimeZone.com members via phone and in person, in
addition to their on-line participation within the virtual
community. This phenomenon is particularly visible in
virtual communities that originated from the off-line context
(e.g. star fan clubs, alumni associations, corporate CoPs,
etc.). Thus, to accommodate a broader range of the real
world virtual communities, in this study, we define the
virtual community as ‘a group of people with common
interests or goals, interacting for knowledge (or infor-
mation) sharing predominantly in cyberspace.’
2.2. Virtual community activity for knowledge sharing
2.2.1. Antecedents to virtual community activity
Most studies addressing virtual community activity (for
knowledge sharing) have been conducted at the conceptual
level. Godwin (1994), Kim (2000), Kollock (1998), Whi-
taker and Parker (2000), and William and Cothrel (2000)
proposed strategies for successfully managing virtual
communities. We find that there are two different
approaches that address virtual community activation
strategies. One is to approach it from a social perspective
and the other from a socio-technical perspective.
Kim (2000) suggested several characteristics of success-
ful and sustainable virtual communities: clear purposes or
vision (e.g. Jesus for Jesus club), flexible and small-scale
places, members’ roles (e.g. designing community activities
based on the membership life cycle: visitors, novices,
regulars, leaders), leadership of community moderators (i.e.
community leaders), and on-line/off-line events. For
example, regular off-line events (e.g. see the event calendar
of www.gamespyarcade.com/ or www.ivillage.com/chat/)
and rituals (such as handshakes, holidays, and rites of
passage) strengthen community members’ identification
and bonds among them. William and Cothrel (2000) also
proposed three virtual community managing strategies:
member development, community asset management and
community relationship management. More specifically, in
order to run a virtual community intelligently, a clear vision,
opinion leaders (e.g. see ‘mentors society’ in www.mentors.
play.net), off-line activities, rules/roles (e.g. AOL’s terms of
service and local ordinances), and useful contents based on
expertise are required (William & Cothrel, 2000; Yoo, Suh,
& Lee, 2001). Furthermore, Kollock (1998), bringing the
design principles for a traditional community (Axelrod,
1984; Ostrom, 1990) into the virtual world, argued that
communities succeed not because of flashy graphics, but
because they contain a number of requisite elements for a
successful community such as: identity persistence, a
coherent sense of space, and a sophisticated set of rituals,
which persistently remind community members of what
they have in common. Successful communities seem to
understand community dynamics from a social perspective.
In addition to these common social characteristics of
successful communities as mentioned above, Whitaker and
Parker (2000), based on Romm and Clarke’s (1995) model,
suggested four major categories for the stimulation of
Internet-based agricultural communities: technology,
motivation, task, and system factors. Technology factors
are general computer factors (such as consistent and
compatible software), which are very important in the
context of an IT-based community. Motivation factors
involve the member’s perceived benefits from community
membership, and task factors relate to perceived appro-
priateness of fit of the technology to the main task of the
community. Finally, system factors refer to fit between the
members’ way of doing things and that of the virtual
community. Godwin (1994) considered technical as well as
social factors for virtual community stimulation, such as
confronting the members with a crisis, encouraging the
members to resolve their own disputes, using software that
promotes good discussion for knowledge sharing (e.g. a
chatting room with a reserved name or private discussion
platform), and avoiding a length limitation on postings.
On the basis of above discussion, we believe that the
socio-technical approach (Pasmore, 1995) is useful to
address virtual community stimulation with since it allows
us to treat a virtual community as a dynamic process
(Fernback, 1999; Preece, 2000). The socio-technical
perspective emphasizes both sociability and usability, and
the fit between them. Community leaders work with
community members to plan and guide the community
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166 157
social evolution, and they develop sociability which is
concerned with planning and developing social policies that
are understandable and acceptable to members (Preece,
2000). Moreover, community developers should design
technology with good usability so that the members can
interact and perform their tasks easily and effectively. Good
usability of IT supports rapid learning and high productivity
so that the community can be stimulated. It is noteworthy
that the literature recognizes both of sociability and
usability to be essential for successful virtual community
stimulation. The two different perspectives, social and
socio-technical, for understanding the determinants of
virtual community activity (for knowledge sharing), are
summarized in Table 1.
2.2.2. Outcomes of virtual community activity
There are several outcomes which can result from virtual
community activity for knowledge sharing: group cohesion
and unity, members’ feeling of ownership of the virtual
community, members’ loyalty to the community, and
organizational citizenship behaviors (OCB) (Organ, 1988).
These outcomes are not directly related to commercial
performance but to social performance that may lead to
commercial performance in the long run whether the
commercial transactions are conducted by the community
itself or via the community portal. Although some outcomes
of virtual community activity (for knowledge sharing) have
been suggested by researchers as well as by practitioners,
there have been few studies that empirically examined the
relationship between virtual community activity (for knowl-
edge sharing) and the community service provider’s benefit.
That is, most of the previous studies principally discussed it
at the conceptual level, or did not see it from a portal’s point
of view. Thus, we need to empirically investigate whether
the level of virtual community activity (for knowledge
sharing) finally leads to loyalty toward the Internet-based
community service provider.
2.2.3. Measuring issues on virtual community activity
for knowledge sharing
Virtual community activity for knowledge sharing is an
important construct for explaining the dynamics of a virtual
community since without some forms of community
knowledge sharing activity, any virtual community will
fail to survive (Butler, 2001). We view virtual community
activity as an indispensable process by which community
members share information and knowledge among them
(Butler, 2001; Hare, 1976), and propose that there are three
different approaches to measuring the level of community
knowledge sharing activity. The first is to quantify the
‘traces’ left by community members in the log files of the
community’s web server. For example, archive data such as
page views, number of repeat visitors, average duration, e-
mail exchange count, or frequency and duration of chatting
can be used to measure community knowledge sharing
activity (Hansen, 1996). The second is to analyze chatting
content or e-mail exchange behaviors by textual analysis
(Bauer & Scharl, 2000). The third is to characterize
substantial community activities such as knowledge posting
and viewing activities (Butler, 2001). We believe that
knowledge posting and viewing activities are two major
knowledge sharing activities since in case of a virtual
community, the two activities not only appear visible as
well as frequent but even discussions also appear in the
forms of them. Further, it may be preferable to use objective
measures that combine quantitative with qualitative assess-
ment, considering measurement’s efficiency (e.g. speed) as
well as its relevance. In this study, we would see virtual
community activity as a knowledge creating and sharing (or
retrieving) process by community members (This may be
related to Nonaka and Konno (1998)’s concept of ‘Ba’).
Thus, we consider knowledge posting and viewing activities
the proper measures for community knowledge sharing
activity. Alternatives for measuring the level of community
knowledge sharing activity are given in Table 2.
2.3. The business value of virtual communities
Hagel and Armstrong (1997) pointed out that the
relationship building aspect of virtual community allows
people to engage in the exchange of information and
knowledge and to learn from each other. Rothaermel and
Table 2
Measuring virtual community activity for knowledge sharing
Target variables
or analysis
Characteristics Related studies
Traffic related counting
measure (page views,
number of repeat visitors,
count of e-mail exchange,
or frequency and duration
time of chatting)
Objective and
quantitative
Hansen (1996) and
Schubert and Selz
(1999)
Content of chatting or
e-mail: textual analysis
Subjective and
qualitative
Bauer and Scharl
(2000) and Bucy,
Lang, Potter, and
Grabe (1999)
Fundamental knowledge
sharing activities: the
mean number of know
ledge posts and viewings
Relatively objective
and quantitative as
well as qualitative
Butler (2001) and
Miranda and Saunders
(2003)
Table 1
Two different perspectives for determinants of virtual community activity
Perspective Description Related studies
Social perspective Focusing on sociability
(such as leadership, off-
line activities, persistent iden
tity, rules, or clear visions)
Kim (2000), Kollock
(1998) and William
and Cothrel (2000)
Socio-technical
perspective
Focusing on both sociability
and usability (useful contents
or IT system quality)
Godwin (1994),
Preece (2000) and
Whitaker and Parker
(2000)
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166158
Sugiyama (2001) addressed the relationship between virtual
community characteristics (such as membership size,
community scalability, and level of site management) and
commercial success, developing propositions at the com-
munity level as well as at the individual level through the
case of Timezone.com, a virtual Internet community
devoted to wristwatch hobbyists and enthusiasts. Thus,
communities are not only about aggregating information,
knowledge, or resources, but also about bringing people
together to meet some of their commercial needs as well as
social needs (Rothaermel & Sugiyama, 2001).
E-commerce entrepreneurs take a very broad view of
community (Preece, 2000). Any chatting system, bulletin
board or communications software program can be regarded
as an on-line community. For them, the important issue is
what draws people to and holds people on a website, a
concept known as stickiness, so that they will buy goods or
services, or view more advertisements. The successes of
America On-line (AOL) and Daum Communications have
proved that chatting on-line to friends, family, and new
acquaintances can be a big business. E-commerce entrepre-
neurs anticipate that virtual communities not only will keep
people at their sites, but will also play an important role in
marketing, as people tell each other about their purchases
and discuss banner advertisements.
Virtual community service providers (e.g. geocities.com)
are opening to users their websites as the space for users to
build and develop communities for knowledge sharing.
These providers, called virtual community portals also offer
their users other services such as personal e-mail systems,
useful content (e.g. news), and B2C shopping malls. In
Korea, there are several big community portals such as
Daum.net, Freechal.com, iloveschool.co.kr, or Sayclub.com
that, respectively, host more than 1 million virtual
communities each. Most of these portals were ranked
among the top 20 in the world by Alexa (http://alexa.com)
as of October 2002. The top managers of community portals
are now planning a change in strategy from a focus on
increasing the number of users to increasing user loyalty
toward their sites. Such loyalty to their site may be increased
by active community knowledge sharing. However, they are
wondering whether virtual community stimulation is related
to loyalty toward the community service portal and whether
it, in the end, results in profit for the portal. Positive or
optimistic answers from research such as this study would
trigger more investment in the community portals involving
community services.
Recently, Sayclub.com reported a profit because of the
‘avatar’ service model. Sales of Sayclub.com during 2001
were reported to be composed of three items: Avatar selling
(70%), advertisements (5%), and miscellaneous (25%). The
community service providers (portals) seem to believe that
when the virtual communities inside their sites are
stimulated, the sales volume of the providers will be
increased. For example, community members who enroll in
an activated community tend to purchase avatars to present
themselves to other members, or to make up the avatars
themselves in their own style, since they have the need to be
perceived as stylish or impressive persons in their commu-
nity. Thus, the sales volume of avatars may remarkably
increase when virtual community members interact with
each other actively, even after membership numbers
stabilize, which suggests that virtual community activity
for knowledge sharing may affect the service provider’s
profit. Furthermore, increased virtual community activity
will reinforce the number of advertisements viewed.
Taken in sum, IS researchers as well as practitioners are
interested in whether nurturing virtual communities is
beneficial for on-line portals. However, there have been
few studies to empirically validate whether it is true.
3. Research model and hypotheses
In the initial phase of this study, we focused on developing
a conceptual foundation to understand virtual community
dynamics related to knowledge sharing activity. First, we
may identify the factors that affect virtual community
activities for knowledge sharing. Also, the level of commu-
nity knowledge sharing activity is expected to predict virtual
community performance or explain the variance in the
community activity quality. These virtual community out-
comes may be related to loyalty to the virtual community
provider. Finally, such loyalty is expected to result in the
ultimate dependent variable, commercial performance of a
community portal (i.e. profits). In this study, we would
concentrate on the middle part of the full model, rather than
explicitly relating management strategies with monetary
outcomes. Further, owing to the effects of community size
and age on virtual community dynamics, two variables may
be controlled in the model. The overall research framework
and the research scope of this study are shown in Fig. 1.
Based on the literature and in-depth understanding of
virtual community behaviors through preliminary studies,
we predict that high community knowledge sharing activity
will lead to positive community outcomes, and that such
positive community outcomes will generate positive com-
munity provider outcomes. The rationale for each research
variable and related hypotheses follows next.
Bateman and Organ (1983) proposed the concept of
organizational citizenship behaviors (OCB) that means
informal behaviors contributing to organizations (or com-
munities) without formal rewards. That is, OCB can be
defined as the voluntary help-giving behavior (Barnard,
1938; Fernback & Thompson, 1995; Moorman, 1991;
Morrison, 1994; Williams & Anderson, 1991). We would
adapt the term of OCB to virtual community participation. If
a virtual community embodies an elaborate community
process of knowledge posting and viewing activities, it is
likely to lead its members to participate in OCB as well as
frequent visits or long stays. Community members can
provide valued messages (information and knowledge) for
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166 159
others or reply to the requests from help-seeking members.
For example, when one community member requests
specific tour information, we often find other community
members to promptly post kind and valuable tour guidance
on their community site. Such behaviors as knowledge
sharing (Abbott, 1988; Butler, 2001), disseminating their
ideas quickly (Finholt & Sproull, 1990), and providing
emotional support (King, 1994; Rice & Love, 1987) are
frequently observed in a virtual community in the form of
intensive postings and viewings by the members. That is,
intensive knowledge postings/viewings or frequent on-line
interactions all have the potential to support higher level of
help-giving behaviors and social support (Butler, 2001).
Thus, we expect that the more community activity for
knowledge sharing is, the more active community partici-
pation is in the form of OCB.
Hypothesis 1
There is a positive relationship between community
knowledge sharing activity and community participation.
Hypothesis 2
There is a positive relationship between community
knowledge sharing activity and community promotion.
Through virtual community participation, members are
likely to be exposed to a variety of user interfaces and
interventions offered by the virtual community provider.
Since various promotional services and events by the commu-
nity provider are more frequently exposed to the community
members who are committed to the virtual community, it is
likely that members with higher level of community
participation tend to form a more positive attitude toward
their virtual community provider. Eventually, community
members coincidentally interact with the community provi-
der through computer interfaces when they participate in their
community. It may generate community members’ loyalty
toward the community provider. Similarly, when community
members promote their community through positive word-of-
mouth behaviors (i.e. community promotion), they are
coincidentally promoting the community provider which
hosts their community. That is, community loyalty may be
indirectly or directly linked to loyalty toward the community
provider. Therefore, community participation and community
promotion are expected to be associated with the community
members’ loyalty to the virtual community provider.
Hypothesis 3
There is a positive relationship between community partici-
pation and loyalty to the virtual community provider (VCP).
Hypothesis 4
There is a positive relationship between community pro-
motion and loyalty to the virtual community provider (VCP).
4. Methods
4.1. Sample and data collection
The instruments of the study were developed based on the
relevant literature and the results of prior interviews with the
leaders of five virtual communities in Korea. A pilot test was
conducted with 90 members in 10 virtual communities
hosted by a community provider, ‘Netian (www.netian.net)’
in Korea. The pilot test was to evaluate the relevance of the
items related to our research variables. After modifying the
questionnaire based on the pilot testing, we were able to
secure ‘Freechal (www.freechal.com)’ as our research site.
‘Freechal’, with 11 million members, is one of the largest
community service providers in Korea and was reported to
have about 1.2 million virtual communities as of October
2002 (Freechal was ranked 14th community portal in the
world by Alexa.com in October 2002). Focusing on the
community-based Internet service business, Freechal is
hosting various types of virtual knowledge communities.
We decided to introduce the web-based (on-line) survey
method, gaining acceptance in IS research (Bhattacherjee,
2001), since it has some advantages over the traditional
paper-based survey: (1) lower costs, (2) faster responses, and
Fig. 1. Research model.
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166160
(3) geographically unrestricted sample (Bhattacherjee, 2001;
Tan & Teo, 2000). We installed a web-based questionnaire
system within the Freechal.com site in December 2001. A
banner advertisement about the web-survey was exposed to
every Freechal member for one week as a survey promotion.
Moreover, we mailed community masters (leaders) to
promote their community members to participate the web-
based survey. Memory disks ($20) were offered to actively
participated communities (which included more than 15
respondents) as an incentive to stimulate participation. Since
respondents were requested to fill out their community name
they considered in the survey, we could identify the
respondents’ communities. In total, 3450 members from
691 virtual communities participated in the web-survey
during the two weeks. Out of the 691 communities, the
number of virtual communities with at least three respon-
dents (including the community leader) was 82. Then, five
out of the 82 communities were dropped because of the
unacceptable inter-rater agreement level among members.
Thus, the final sample size was 77, and 641 usable web-
questionnaires from the 77 communities were used for
analysis. Moreover, we collected archive data such as
community size/age, and the average number of knowledge
postings/viewings per month of each community. We
counted the numbers of accumulated knowledge posting
activities and (logged) viewing activities on all the bulletin
boards of each community with the counting software
developed by the virtual community provider, Freecal.com.
Detailed descriptive statistics of the respondents’ charac-
teristics and their communities are shown in Tables 3 and 4,
respectively. The average size and age of participated
communities were 1124 members and 11 months, respect-
ively, and the community size was closely associated with
the community age (r ¼ 0:285; p , 0:05). The community
size ranged from 33 to 7896, while the age was in the range
from 2 to 26 months.
4.2. Measures
We developed outcome variables, which were measured
by a five-point Likert scale (1 ¼ strongly disagree, 5 ¼
strongly agree). Community promotion items were devel-
oped based on marketing research such as Srinivasan,
Anderson, and Kishore (2002) or Zeithaml, Berry, and
Parasuraman (1996). Community participation items,
adapted from Organ’s (1988) OCB study, included com-
munity members’ efforts to stimulate the community and
help-giving behaviors. Also, the items of loyalty to the
community service provider were developed based on
customer loyalty related literature (Lu & Lin, 2002;
Reinartz & Kumar, 2002). The questionnaire items for the
variables are given in Appendix A.
Finally, we considered knowledge posting activity and
viewing activity as the two major knowledge sharing
activities since in a virtual community they appear
frequently as well as visibly in terms of knowledge sharing
(e.g. even discussions appear in the forms of posting activity
and viewing activity). In order to measure the level of
virtual community activity for knowledge sharing, we
calculated the mean value of knowledge posting activity
and viewing activity for each community (i.e. divided
the summed value by two), using two elements of archive
data, the average number of knowledge postings per month
and the average number of knowledge viewings per month.
We note that before averaging them, we adopted a
logarithmic transformation to reduce the variance since
the distribution of the archive data was skewed (Kimberly &
Table 4
Descriptive statistics of participated virtual communities
Measure Items Frequency Percent
Community size (unit: members) ,100 18 23
100–500 27 35
501–1000 6 8
.1000 26 34
Total 77 100
Average 1124 –
Community age (unit: months) ,6 17 22
6–12 23 30
13–18 19 25
.18 16 21
Missing 2 3
Total 77 100
Average 11 –
Table 3
Descriptive statistics of the respondents’ characteristics
Measure Items Frequency Percent
Gender Female 270 42
Male 361 56
Missing 10 2
Total 641 100
Age ,16 24 4
16–20 149 23
21–25 172 27
26–30 163 26
31–35 100 16
.35 24 3
Missing 8 1
Total 641 100
Education level
or occupation
Students below
University
117 18
University students 163 26
Master students 22 3
Company employees 237 37
Others 72 11
Missing 30 5
Total 641 100
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166 161
Evanisko, 1981; Mosteller & Tukey, 1977). Here, the
average number of knowledge postings per month was
calculated as the cumulative number of postings divided by
the virtual community age (months), and the average
number of knowledge viewings per month was measured
as the cumulative number of viewings of the contents
divided by the virtual community age (months). Appendix B
provides the results of the principal component factor
analysis that confirmed the expected factor structure of the
outcome variables.
The internal consistency reliability of the variables was
assessed by computing Cronbach’s alphas. The Cronbach’s
alpha values of all the variables, ranging from 0.784 to
0.856, were well over 0.700, which is considered satisfac-
tory for measures (Nunally, 1978). Since only two items do
not allow Cronbach’s alpha value calculated, we just report
the Pearson correlation value for the variable of virtual
community activity (i.e. r ¼ 0:660).
Also, to confirm whether individual responses from
the same community can be aggregated to the commu-
nity level, we needed to estimate the agreement score for
the three variables with the index of within-group
agreement, rwgðjÞ; derived by James, Femaree, and Wolf
(1984). Within-group agreement among individuals could
theoretically qualify to aggregate the individual percep-
tions to the higher unit level and to use the mean to
represent this collective interpretation. We used the
within-group agreement statistic to examine whether the
community members’ views about the community
variables are in harmony or not. The average within-
group agreement level for the three variables such as
community participation, community promotion and
loyalty to the virtual community provider (VCP) were
0.877, 0.797 and 0.909, respectively. These agreement
levels are generally acceptable because they are well
above 0.700 (Hater & Bass, 1988; Schneider & Bowen,
1985).
5. Results
We examined the effect of community knowledge
sharing activity (i.e. knowledge posting activity and
viewing activity) on virtual community outcomes. First,
based on the Kolmogorov–Smirnov Z tests, it was not
rejected that the distribution of each variable (including the
knowledge sharing activity variable) is normal. Thus, we
conducted a Pearson correlation analysis. Pearson corre-
lation is calculated for the variables measured by interval or
ratio scales. The simple correlation among all the research
variables is shown in Table 5. Although several variables
showed significant correlations, their tolerance values
ranged from 0.621 to 0.957, indicating that multicollinearity
is not a likely threat to the parameter estimates (Hair,
Anderson, Tatham, & Black, 1995). The means, standard
deviations, reliabilities, and rwgðjÞ values of the variables are
together provided in Table 5.
Table 6 shows the results of the multiple regression
analyses, testing the hypotheses. The results indicate that the
first regression model is significant at p , 0:05 level ðF-
value ¼ 3:896Þ and other two regression models are
significant at p , 0:01 level (F-value ¼ 8:590 and 5.193).
Also, the predictors of each model explain 11, 21 and 18%
of the total variance, respectively.
Hypotheses 1 and 2 examine the relationships between
community knowledge sharing activity and community
outcomes. Because community size is expected to affect
the two community outcome variables, but not a variable
we are interested in, we treated it as a control variable.
In terms of hypotheses 1 and 2, community knowledge
sharing activity was significantly related to both commu-
nity participation and community promotion (b ¼ 0:368;
p , 0:01; b ¼ 0:441; p , 0:01). Thus, hypotheses 1 and
2 were supported. In hypotheses 3 and 4 with a control
variable of community age, considering the time horizon
issue for testing the relationships, loyalty to the virtual
community provider (VCP) was associated not with
community participation but with community promotion
(b ¼ 0:228; p , 0:1). Consequently, hypothesis 3 was
not supported, while hypothesis 4 was supported.
Based on the statistically supported hypotheses, the
overall paths of virtual community dynamics are graphically
shown in Fig. 2. The two thick lines in Fig. 2 present the
significance at p , 0:01 level, and the thin line indicates
the significance at p , 0:1 level. In particular, Fig. 2 shows
Table 5
Correlation analysis between the research variables ðn ¼ 77Þ
Variables Cronbach’s alpha rwgðjÞ Mean SD 1 2 3 4 5
1. Knowledge sharing activity 0.660 N/A 7.094 1.669
2. Community participation 0.856 0.877 3.866 0.416 0.324***
3. Community promotion 0.784 0.797 3.774 0.476 0.446*** 0.622***
4. Loyalty to the virtual community
provider
0.841 0.909 3.393 0.441 0.049 0.313*** 0.300***
5. Community size N/A N/A 1124 (Persons) 1921 0.480*** 0.093 0.261** 20.046
6. Community age N/A N/A 11 (Months) 6.67 0.189 0.024 0.180 20.231** 0.285**
**: p , 0:05; ***: p , 0:01; N/A: not applicable.
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166162
the dynamics of a virtual community based on knowledge
sharing activity from an e-business perspective.
6. Discussions and implications
Testing the hypotheses, we examined whether commu-
nity knowledge sharing activity, measured objectively, is
related to perceptions of virtual community performance,
and whether the perceptions of community performance are
associated with loyalty toward the virtual community
provider. On the link between community activity and
community outcomes, community knowledge sharing
activity predicted both community participation and com-
munity promotion. We interpret this to mean that a simple
measure such as the number of knowledge posting activity
(or knowledge viewing activity) is an accurate indicator of
positive perceptions of community membership and loyalty
to the community portal.
We also found a significant link between the
community outcome and the virtual community provider
outcome (i.e. the link between community promotion and
loyalty to the virtual community provider). This implies
that effectively managed communities may ultimately
have potential for economic gains of their community
portal (Rothaermel & Sugiyama, 2001). For example,
avatar selling or B2C commercial exchange in the portal
might be increased as the communities are activated.
Since the relationship between the community outcome
and loyalty to the virtual community provider was found
significant in this study, several elaborate ways to link
them need to be explored.
Lastly, it is noteworthy to understand the role of
community promotion in virtual community dynamics.
It was considered as the indicator of loyalty toward
Table 6
Results of hypotheses tests (from H1 to H4)
Model R2 F b Results
(1) Participation (PART)
PART ¼ ACT þ N þ errors 0.109 3.896**
ACT 0.368*** H1 was supported
N (control variable) 20.112 N/A
(2) Promotion (PRO)
PRO ¼ ACT þ N þ errors 0.212 8.590***
ACT 0.441*** H2 was supported
N (control variable) 0.037 N/A
(3) Loyalty (LOY)
LOY ¼ PART þ PRO þ T þ errors 0.180 5.193***
PART 0.172 H3 was not supported
PRO 0.228* H4 was supported
T (control variable) 20.277** N/A
ACT: community knowledge sharing activity; N : community size; T : community age. ***: p , 0:01; **: p , 0:05; *: p , 0:1:
Fig. 2. Validated model for virtual community dynamics.
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166 163
the community portal as well as a significant outcome of
community knowledge sharing activity, possibly mediating
the relationship between the level of community knowledge
sharing activity and loyalty toward the portal. In addition,
community promotion may play a critical role in reaching a
critical mass (Markus, 1990) of community members within
a short period, and so bring the benefit of network
externalities (Yannelis, 2001) to the community.
In short, the most immediate beneficiaries of the findings
of this study may be the managers of virtual community
service portals who are concerned about vitalizing their
virtual communities (commercial or non-commercial) from
an e-business perspective. They can derive specific calls for
action from the study’s findings:
1. Use simple and objective measures (such as knowledge
posting and viewing activities) as good proxies for
evaluating the state of health of a virtual community.
2. Develop and elaborate the link between virtual commu-
nities and the portal through activities such as avatar
selling or B2C transactions (between shopping malls the
portal operates and community members).
7. Conclusion
In this study, we examined how virtual community
activity for knowledge sharing explains virtual community
performance and whether such community performance is
related to loyalty toward the community service portal. We
developed the measure of the level of community knowl-
edge sharing activity, using posting and viewing activities,
and found that it is significantly related to positive
perceptions of virtual community members such as
community participation and community promotion. This
implies that community portal managers may use simple
measures such as knowledge posting and viewing activities
as good surrogates for evaluating actual virtual community
stimulation. However, these interpretations are not based
on specific causalities. Possibly a large number of knowl-
edge postings might not be the cause of community
loyalty, but be an eventual result of community loyalty. For
this, in this study, we note we just validated that the level
of community knowledge sharing activity is closely related
to perceptual community loyalty (i.e. community partici-
pation and promotion) and proposed that it may be a proper
proxy for the state of health of a virtual community. As
virtual communities become mature, a longitudinal study
for the causal relationships between them needs to be
conducted.
A major finding of this study is that the level of loyalty
toward the community service provider is associated with
the level of the community outcome such as community
promotion. We suggest to virtual community providers that
they should trace the elaborate links between their
communities and the virtual community providers
themselves, which would help create profitable Internet
business models. Linking community knowledge sharing
activity to commercial activity is critical to virtual
community providers who expect that many future B2C
commercial transactions will be conducted via on-line
communities (Bressler & Grantham, 2000).
This study is expected to be useful to researchers who
are interested in knowledge management through virtual
communities, virtual teams, or CoPs, and profitable
Internet business models by virtual communities. Also,
practitioners such as virtual community service providers
may benefit from this research both by realizing the
potential business value of virtual communities and by
finding simple and objective measures of community
knowledge sharing activity (i.e. knowledge posting and
viewing activities) as good proxies for the state of health of
a virtual community.
There are several limitations in this study. Since the data
was collected in a single website as well as a single country
(South Korea), the general applicability of the findings is
limited. More extensive data collection is needed for
greater generalizability. Secondly, there were relatively
high correlations between some variables. All of them may
be causal outcomes of certain affecting variables (e.g.
community leadership, off-line activities, or usefulness of
content) though we did not consider them in the research
model of the study. Future research to find which key
virtual community characteristics influence community
outcomes from a knowledge management perspective is
required. Thirdly, in the sampling procedure, we suspect
that the samples may be biased since all the sample
communities voluntarily participated in the survey. Next,
since we simply measured the variable of loyalty to the
virtual community provider instead of ultimate outcomes
such as real commercial transactions or profits, the direct
relationship between the level of community activity and
its financial contribution to the portal was not explored in
the study. For future research, investigating the relationship
between the level of community knowledge creating/shar-
ing activities and its financial contribution to the commu-
nity service portal, would be valuable. Finally, since this
study focused on non-profit virtual communities inside a
community portal, a study on profit communities such as
eBay.com or brand communities (McWilliam, 2000) is
needed.
Appendix A
Table A1
Appendix B
Table B1
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166164
References
Abbott, A. (1988). The system of professions: An essay on the division of
expert labor. Chicago, IL: University of Chicago Press.
Alavi, M., & Leidner, D. E. (2001). Review: Knowledge management and
knowledge management systems: Conceptual foundations and research
issues. MIS Quarterly, 25(1), 107–136.
Axelrod, R. (1984). The evolution of cooperation. New York: Basic Books.
Balasubramanian, S., & Mahajan, V. (2001). The economic leverage of the
virtual community. International Journal of Electronic Commerce, 5(3),
103–138.
Barnard, C. I. (1938). The functions of executive. MA: Harvard University Press.
Bateman, T. S., & Organ, D. W. (1983). Job satisfaction and the good
soldier: The relationship between affect and citizenship. Academy of
Management Journal, 26(4), 587–595.
Bauer, C., & Scharl, A. (2000). Quantitative evaluation of web site content
and structure. Internet Research: Electronic Networking Applications
and Policy, 10(1), 31–43.
Bhattacherjee, A. (2001). Understanding information systems continuance:
An expectation–confirmation model. MIS Quarterly, 25(3), 351–370.
Bressler, S. E., & Grantham, C. E. (2000). Communities of commerce. New
York: McGraw-Hill.
Brown, J. S., & Duguid, P. (1991). Organizational learning and communities
of practice. Organization Science, 2(1), 40–57.
Brown, J. S., & Duguid, P. (2000). The social life of information. Boston:
HBS Press.
Bucy, E. P., Lang, A., Potter, R. F., & Grabe, M. E. (1999). Formal features
of cyberspace: Relationships between Web page complexity and site
traffic. Proceedings of 49th Annual Conference of the International
Communication Association (ICA-99), San Francisco.
Butler, B. S. (2001). Membership size, community activity, and sustain-
ability: A resource-based model of online social structures. Information
Systems Research, 12(4), 346–362.
Choi, B. G., & Lee, H. S. (2002). Knowledge management strategy and its
link to knowledge creation process. Expert Systems with Applications,
23(3), 173–187.
Espinosa, J. A., Cummings, J. N., Wilson, J. M., & Pearce, B. M. (2003).
Team boundary issues across multiple global firms. Journal of
Management Information Systems, 19(4), 157–190.
Fernback, J. (1999). There is a there there: Notes toward a definition of
cyber-community. In S. G. Jones (Ed.), Doing internet research:
Critical issues and methods for examine the net (pp. 203–220).
Thousand Oaks, CA: Sage Publications.
Fernback, J., & Thompson, B. (1995). Virtual communities: Abort, retry,
failure? Computer Mediated Communication and the American Collec-
tivity, Available: http://www.ewll.com/user/hlr/texts/Vccivil.html/
Finholt, T., & Sproull, L. (1990). Electronic groups at work. Organization
Science, 1, 41–64.
Godwin, M. (1994). Nine principals for making virtual communities work.
Wired, June(2.06), 72–73.
Gusfield, J. (1975). The community: A critical response. New York: Harper
Colophon.
Hagel, J., III, & Armstrong, A. M. (1997). Net gain: Expanding markets
through virtual communities. Boston: HBS Press.
Hair, J. F., Anderson, R., Tatham, R., & Black, W. (1995). Multivariate
data analysis with readings (4th ed.). Englewood Cliffs, NJ:
Prentice-Hall.
Hansen, H. R. (1996). Klare sicht am Info-Highway: Geschafte via Internet
and Co. Vienna: Orac.
Hare, A. P. (1976). Handbook of small group research (2nd ed.). New
York: Free Press.
Hater, J. J., & Bass, B. M. (1988). Superiors’ evaluations and subordinates’
perceptions of transformational and transactional leadership. Journal of
Applied Psychology, 73(4), 695–702.
Heller, K. (1989). Return to community. American Journal of Community
Psychology, 17(1), 1–15.
Table A1
Measures of the research variables
Variables Items
Community I take an active part in our virtual community
participation I do my best to stimulate our virtual community
I often provide useful information/contents for our
virtual community members
I eagerly reply to postings by the help-seeker of
our virtual community
I take care about our virtual community members
I often help our virtual community members who
seek support from other members
Community
promotion
I invite my close acquaintances to join our
virtual community
I often talk to people about benefits of our virtual
community
I often introduce my peers or friends to our virtual
community
Loyalty to the
virtual community
I recommend to my acquaintances that they enroll in
freechal.com
provider (VCP) I often talk about benefits of freechal.com
I often talk to my peers in my company or school
about freechal.com
I even give freechal.com ideas/suggestions on
planning and operations
I will visit freechal.com continuously, even if my
virtual community vanishes
Table B1
Factor structure of measures of the variables
Scale items Rotated component
Factor 1 Factor 2 Factor 3 Factor 4
Knowledge posting activity 0.246 0.132 0.287 0.808
Knowledge viewing activity 0.088 20.054 0.191 0.845
Community promotion1 0.164 0.161 0.761 0.172
Community promotion2 0.308 20.080 0.811 0.195
Community promotion3 0.265 0.163 0.625 0.227
Community participation1 0.690 0.274 0.303 0.205
Community participation2 0.767 0.151 0.255 0.047
Community participarion3 0.703 0.021 0.397 20.015
Community participation4 0.725 0.086 0.308 20.062
Community participation5 0.890 20.017 0.070 0.191
Community participation6 0.868 0.093 20.042 0.207
Loyalty1 0.122 0.780 0.330 0.048
Loyalty2 0.085 0.895 0.120 0.069
Loyalty3 0.039 0.848 0.168 0.116
Loylaty4 0.173 0.766 20.025 20. 246
Loyalty5 20.126 0.755 20.181 0.062
Eigenvalue 5.744 3.147 1.610 0.980
Percentage of variance
explained
35.902 19.668 10.064 6.124
Cumulative percentage 35.902 55.570 65.635 71.759
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166 165
Hiltz, S. R., & Wellman, B. (1997). Asynchronous learning networks as a
virtual classroom. Communications of ACM, 40(9), 44–49.
James, L. R., Femaree, R. G., & Wolf, G. (1984). Estimating within-group
interrater reliability with and without bias. Journal of Applied
Psychology, 69(1), 85–98.
Jones, Q. (1997). Virtual communities, virtual settlements and cyber-
archaeology: A theoretical outline. Journal of Computer-Mediated
Communications, 3(3) Available: http://www.ascusc.org/jcmc/vol3/
issue3/jones.html/
Jordan, J., & Jones, P. (1997). Assessing your company’s knowledge
management style. Long Range Planning, 30(3), 392–398.
Kim, A. J. (2000). Community building on the web. Berkeley: Peachpit Press.
Kim, Y.-G., Yu, S.-H., & Lee, J.-H. (2003). Knowledge strategy planning:
Methodology and case. Expert Systems with Applications, 24(3), 295–307.
Kimberly, J. R., & Evanisko, M. J. (1981). Organizational innovation: The
influence of individual, organizational, and contextual factors on
hospital adoption of technological and administrative innovations.
Academy of Management Journal, 24(4), 689–713.
King, S. (1994). Analysis of electronic support groups for recovering
addicts. Interpersonal Computer Technology, 2(3), 47–56.
Kollock, P. (1998). Design principles for online communities. PC Update,
15(5), 58–60.
Kraemer, K. L., & Dedrick, J. (2002). Strategic use of the Internet and
e-commerce: Cisco systems. Journal of Strategic Information System,
11(1), 5–29.
Larsen, K. R. T., & Mclnerney, C. R. (2002). Preparing to work in the
virtual organization. Information & Management, 39(6), 445–456.
Lee, J.-H., & Kim, Y.-G. (2001). A stage model of organizational
knowledge management: A latent content analysis. Expert Systems with
Applications, 20(3), 299–311.
Liebowitz, J. (1999). Building organizational intelligence. London: CRC Press.
Lu, H., & Lin, J. C. (2002). Predicting customer behavior in the market-
space: A study of Rayport and Sviokla’s framework. Information and
Management, 40(1), 1–10.
Magretta, J. (1998). The power of virtual integration: An interview with Dell
Computer’s Michael Dell. Harvard Business Review, 76(2), 72–85.
Markus, M. L. (1990). Toward a critical mass theory of interactive media.
In J. Fulk, & C. Steinfield (Eds.), Organizations and communication
technology (pp. 194–218). Newbury Park, CA: Sage Publications.
McWilliam, G. (2000). Building stronger brands through online commu-
nities. Sloan Management Review, 41(3), 43–54.
Miles, G., Miles, R. E., Perrone, V., & Edvinsson, L. (1998). Some
conceptual and research barriers to the utilization of knowledge.
California Management Review, 40(3), 281–288.
Miranda, S. M., & Saunders, C. S. (2003). The social construction of
meaning: An alternative perspective on information sharing. Infor-
mation System Research, 14(1), 87–106.
Moorman, R. H. (1991). Relationship between organizational justice and
organizational citizenship behaviors: Do fairness perceptions influence
employee citizenship? Journal of Applied Psychology, 76(6), 845–855.
Morrison, E. W. (1994). Role definition and organizational citizenship
behavior: The importance of the employee’s perspective. Academy of
Management Journal, 37(6), 1543–1567.
Mosteller, F., & Tukey, J. W. (1977). Data analysis and regression: A
second course in statistics. Pearson Education POD.
Nonaka, I. (1994). A dynamic theory of organization knowledge creation.
Organization Science, 5(1), 14–37.
Nonaka, I., & Konno, N. (1998). The concept of ‘Ba’ Building a foundation
for knowledge creation. California Management Review, 40(3), 40–54.
Nunally, J. C., Jr. (1978). Psychometric theory. New York: McGraw-Hill.
Obst, P., Zinkiewicz, L., & Smith, S. G. (2002). Sense of community in
science fiction fandom. Part 1. Understanding sense of community in an
international community of Interest. Journal of Community Psychology,
30(1), 87–103.
Organ, D. W. (1988). Organizational citizenship behavior: The good
soldier syndrome. Lexington, MA: Lexington Books.
Ostrom, E. (1990). Governing the commons: The evolution of institutions
for collective action. New York: Cambridge University Press.
Pasmore, W. A. (1995). Social science transformed: The socio-technical
perspective. Human Relations, 48(1), 1–21.
Patel, N. V. (2002). Emergent forms of IT governance to support global e-
business models. Journal of Information Technology Theory and
Application, 4(2), 33–48.
Preece, J. (2000). Online communities: Designing usability, supporting
sociability. New York: Wiley.
Reinartz, W., & Kumar, V. (2002). The mismanagement of customer
loyalty. Harvard Business Review, 80(7), 86–94.
Rice, R. E., & Love, G. (1987). Electronic emotion: Socioemotional content
in a computer-mediated communication network. Communication
Research, 14(1), 85–108.
Romm, C., & Clarke, R. (1995). A preliminary draft for a comprehensive
model. Proceedings of the Sixth Australasian Conference on Infor-
mation Systems, pp. 57–70.
Rothaermel, F. T., & Sugiyama, S. (2001). Virtual Internet communities
and commercial success: Individual and community-level theory
grounded in the atypical case of TimeZone.com. Journal of Manage-
ment, 27(3), 297–312.
Schneider, B., & Bowen, E. (1985). Employee and customer perceptions of
service in banks: Replication and extension. Journal of Applied
Psychology, 70(3), 423–433.
Schubert, P., & Selz, D. (1999). Web assessment—measuring the
effectiveness of electronic commerce sites going beyond traditional
marketing paradigms. Proceedings of the 32nd Hawaii International
Conference on Systems Sciences, Hawaii.
Scott, J. E. (2000). Facilitating interorganizational learning with information
technology. Journal of Management Information Systems, 17(2), 81–114.
Sinclair, J. T. (2001). eBay: The smart way (2nd ed.). New York: AMACOM.
Srinivasan, S. S., Anderson, R., & Kishore, P. (2002). Customer loyalty in
e-commerce: An exploration of its antecedents and consequences.
Journal of Retailing, 78(1), 41–50.
Swan, J., Newell, S., & Robertson, M. (2000). Limits of IT-driven
knowledge management for interactive innovation processes: Towards
a community-based approach. Proceedings of the 33rd Hawaii
International Conference on System Sciences, Hawaii.
Tan, M., & Teo, T. S. H. (2000). Factors influencing the adoption of
Internet banking. Journal of AIS, 1(5) Available: http://jais.isworld.org/
articles/1-5/.
Wegner, E. C., & Synder, W. M. (2000). Communities of practice: The
organizational frontier. Harvard Business Review, 78(1), 139–145.
Weinreich, F. (1997). Establishing a point of view toward virtual
communities. CMC Magazine, 4(2) Available: http://www.december.
com/cmc/mag/1997/feb/wein.html
Wellman, B., & Gulia, M. (1999). Virtual communities as communities:
Net surfers don’t ride alone. In M. A. Smith, & P. Kollock (Eds.),
Communities in cyberspace (pp. 167–194). London: Routledge.
Whitaker, V. M., & Parker, C. M. (2000). The factors enabling and
inhibiting the development of agricultural Internet virtual communities:
An Australian case study. White Paper.
Wiig, K. M., Hoog, R., & Spek, R. (1997). Supporting knowledge
management: A selection of methods and techniques. Expert Systems
with Applications, 13(1), 15–27.
Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and
organizational commitment as predictors of organizational citizenship
and in-role behaviors. Journal of Management, 17(3), 601–617.
William, R. L., & Cothrel, J. (2000). Four smart ways to run online
communities. Sloan Management Review, 41(4), 81–91.
Yannelis, D. (2001). On the simple welfare economics of network
externality. International Journal of Social Economics, 28(4), 344–353.
Yoo, W. S., Suh, K. S., & Lee, M. B. (2001). Exploring the factors
enhancing member participation in virtual communities. Proceedings of
the Fifth Pacific Asia Conference on Information Systems, Seoul.
Zeithaml, V. A., Berry, L. L., & Parasuraman, A. (1996). The behavioral
consequences of service quality. Journal of Marketing, 60(2), 31–46.
J. Koh, Y.-G. Kim / Expert Systems with Applications 26 (2004) 155–166166
