Sense of Virtual Community: Conceptual Framework
and Empirical Validation
Joon Koh* and Young-Gul Kim**
KAIST Graduate School of Management
E-mail: email@example.com*, firstname.lastname@example.org**
In this article, we introduce the “sense of virtual community” construct and explore the factors that
are expected to affect it. We also examine the moderating effect of the virtual community origin. By
analyzing 172 responses, we found that the sense of virtual community is affected by the virtual
community characteristics such as (1) leaders’ enthusiasm, (2) perceived similarity, (3) off-line activities,
and (4) playfulness. From the moderated regression analysis, we also found that the virtual community
origin moderates the relationship between virtual community characteristics and sense of virtual
As the internet coverage broadens rapidly, the virtual community, has become a topic of interest to
IT professionals and management researchers [e.g., Jones, 2000]. Despite the explosive growth of virtual
communities on the internet, limited empirical research has been conducted to study the issues related to
the psychological states of the virtual community members.
The objective of this study is to enhance the existing knowledge about virtual community by
introducing a new construct, sense of virtual community, and empirically validating the effects of virtual
community characteristics on the sense of virtual community. We draw from the relevant literature and
focus on developing a conceptual foundation for understanding virtual community. More specifically, this
study intends to answer the following questions:
• What is the sense of virtual community? Is there any unique property differentiating it from that of
a traditional community?
• What are the key factors affecting the sense of virtual community at the individual level?
• Does the origin of virtual community moderate the relationship between virtual community
characteristics and sense of virtual community?
2. Sense of Virtual Community, and Virtual Community Origin
Computer-mediated community environments need commitment of their members just like the
traditional environment, based on the fact that we understand virtual community to be “a community
extended via emerging technologies.” Among the four elements (membership, influence, integration, and
emotional connection) that McMillan & Chavis  regarded as the components of sense of
community, membership and influence are considered as common perception factors in both the virtual
and traditional communities. Reflecting the unique characteristics of the virtual space, we introduce a new
dimension of immersion, using an expanded concept of flow [Csikszentimihalyi, 1975; Hoffman & Novak,
1996]. Thus, sense of virtual community is treated as having three dimensions: (1) membership−people
experience feelings of belonging to their virtual community, (2) influence−people influence other
members or their community, and (3) immersion−people feel the state of flow during virtual community
navigation. We define sense of virtual community as the psychological state of perceiving three
dimensions of membership, influence, and immersion, formed through the activities of virtual community.
With respect to the virtual community origin, virtual communities can be dichotomized as online
originated and offline originated. We are interested in examining the moderating role of the virtual
community origin on the proposed relationships between virtual community characteristics and sense of
3. Research Model and Hypotheses
We concentrate our efforts on developing a conceptual foundation for understanding a virtual
community, introducing sense of virtual community, its determinants, and the moderating variable. The
model for this research is shown in Fig. 1.
Leaders’ enthusiasm helps members feel that the virtual community is activated, and it also fosters
them to pay attention to the virtual community [Kim, 2000]. Some practitioners even argue that, with only
one passionate leader, a virtual community can be activated and its members get to participate actively in
the community activities. Even though leadership necessary for virtual communities may differ from
traditional leadership, leaders’ enthusiasm is expected to influence community members to feel greater
membership toward the community.
Hypothesis 1: There is a positive relationship between leaders’ enthusiasm and membership.
Fig. 1. Research model
H8 H9 H10
Exclusion of non-members is a characteristic of the virtual space as is in the real world, and is
related to a concept of perceived boundaries in the real life [McMillan & Chavis, 1986]. In-group
members are distinguished from out-group members through scarcity of resources. A group based on
similar interests tends to exclude strangers, which keeps the community homogenous. We can take the
closed user group (CUG) as an example. Since most of the virtual communities have been formed, based
on similar interests and themes, similarity perception is important for reinforcing membership within a
virtual community [Sarason, 1974].
Hypothesis 2: There is a positive relationship between similarity and membership.
Offline meetings can play a role in complementing the low social presence inherent of most
computer-mediated environments [Lombard & Ditton, 1997]. Kiesler et al.  argued that balancing
online with offline activities is critical in sustaining a virtual community, implying that use of diverse
communication channels may be effective in the computer-mediated environment. While Walther 
argues that on-line interactions are as sociable and intimate as in-person interactions over time, we
believe that off-line meetings will help facilitate the virtual community activism and lead to higher sense
of virtual community.
Hypothesis 3: There is a positive relationship between offline activities and membership.
Hypothesis 4: There is a positive relationship between offline activities and influence.
Hypothesis 5: There is a positive relationship between offline activities and immersion.
We also propose that playfulness is a useful construct for understanding individuals’ evaluation and
affection of virtual communities. Previous research has revealed that attitudinal outcomes such as
emotion, pleasure, and satisfaction result from the playfulness experience [Csikszentimihalyi, 1975].
Playfulness will affect flow−immersion by human-machine interaction [Griffiths, 1998]. Furthermore,
playfulness lets members perceive influence on other members or on their community because it contains
interactions among members or between members and their community. Consequently, when a virtual
community provides entertainment value for its members in the online context, it would be more likely to
be activated [Moon & Kim, 2001].
Hypothesis 6: There is a positive relationship between playfulness and influence.
Hypothesis 7: There is a positive relationship between playfulness and immersion.
As mentioned above, virtual communities are classified into two types. Online originated virtual
communities are launched based on the common interests and themes communicated via computer-
mediated communications. Weak ties are mostly observed in the initial stage [Wellman & Gulia, 1999].
For example, membership and influence are likely to be low in the early stage of the online originated
virtual community. On the other hand, social relationship in an offline originated virtual community tends
to be strong even at the beginning of the online-communication due to prior offline interactions
[Blumstein & Kollock, 1988]. Therefore, higher levels of leaders’ efforts and offline activities are
necessary for increasing membership and influence in the case of online originated virtual communities
than in the case of offline originated virtual communities. Besides, online originated virtual communities
are probably relatively homogenous in their interests and attitudes while being heterogeneous in terms of
the participants’ age, gender, social class, ethnicity, and other aspects of their demography [Wellman &
Gulia, 1999]. The homogenous interests of online originated virtual community members may foster a
relatively high level of empathetic understanding and mutual support [Marsden, 1983]. Hence, perceived
similarity with the interests and values will affect membership more strongly in the case of the online
originated virtual community than in the case of the offline originated virtual community. Additionally,
immersion is generated by the online oriented communications [Young, 1996] that the online originated
virtual community depends on. It is unlikely that the offline originated virtual community will incur high
level of immersion of members because it mainly depends on the cognitive mechanisms already formed
by the prior face-to-face communication. Thus, affecting factors seem to influence immersion more
strongly in the case of the online originated virtual community than in the case of the offline originated
On the basis of the above discussion, we derived the following hypotheses to validate the moderating
effect of the virtual community origin on the basic relationships between virtual community
characteristics and sense of virtual community.
Hypothesis 8: The virtual community origin moderates the relationship between virtual community
characteristics and membership.
Hypothesis 9: The virtual community origin moderates the relationship between virtual community
characteristics and influence.
Hypothesis 10: The virtual community origin moderates the relationship between virtual community
characteristics and immersion.
4. Methods and Results
We developed the instruments for our variables based on the relevant literature and the results of the
prior interviews with the system operators (sysops) of the four representative virtual communities in
Korea. With the questionnaire modified through the pre-testing, we selected 44 virtual communities that
have been recognized as the popular virtual community sites in Korea and distributed 220 questionnaires,
assigning 5 questionnaires to each community. From the 220 questionnaires that were distributed, 172
questionnaires were received and used for analysis. The response rate was 78 percent. In terms of
gender and age of the respondents, 47 percent of them were male, and 71 percent were less than 26 years
old. In addition, 92 respondents were enrolled in the on-line originated virtual community, and 80
individuals were members of the off-line originated virtual communities.
We conducted the factor analysis, the reliability test, the multiple regression analyses, the moderated
regression analyses [Atuahene-Gima & Li, 2000] and others. First, all instruments demonstrated high
convergent and discriminant validity and their Cronbach’s alpha values were all satisfactory, ranged from
0.742 to 0.930. Conducting the multiple regression analyses, virtual community characteristics affecting
the sense of virtual community were detected at the individual level (See Table 1). Membership was
significantly affected by (1) leaders’ enthusiasm, (2) perceived similarity, and (3) offline activities.
Offline activities had the strongest impact on membership and influence in the virtual community. It is
consistent with the results of the prior studies that strong ties cannot be sustained without physical cues
[e.g., Beniger, 1987]. However, the variable of offline activities does not seem to be a significant factor
affecting immersion. We interpret this to mean that offline activities incur two effects: one is fostering
members to commit to the community and the other is preventing them from becoming fully immersed in
the community online activity. We presume that the two contradicting effects of offline activities on
immersion might cancel each other and suggest that further research will be necessary for developing a
relevant contingency theory. Immersion was influenced only by playfulness, while perceived influence
was significantly affected by offline activities and playfulness. The result implies that a virtual community
should find and meet the needs of its members, supporting Kim’s  observations that successful
communities evolve to keep pace with the changing needs of their members. It also suggests that virtual
communities should encourage their members to meet each other in the offline setting, so that their
members may perceive high influence.
Six irrelevant questionnaires returned were discarded
Table 1: Results of hypotheses tests (From H1 to H7)
(1) Membership (MEM)
MEM = E + S + O + errors 0.592 80.753***
E 0.219*** H1 was supported.
S 0.255*** H2 was supported.
O 0.475*** H3 was supported.
(2) Influence (INF)
INF = O + P + errors 0.478 77.307***
O 0.631*** H4 was supported.
P 0.129** H6 was supported.
(3) Immersion (IMM)
IMM = O + P + errors 0.159 15.802***
O - 0.007 H5 was rejected.
P 0.402*** H7 was supported.
(Note) E, Leaders’ enthusiasm; S, Similarity; O, Offline activities; P, Playfulness.
Another important finding of this study is that the impacts of virtual community characteristics on
the sense of virtual community are contingent on the virtual community origin (See Table 2 for models of
moderated regression analyses and Table 3 for results of it). Both leaders’ enthusiasm and similarity
affected membership more strongly in the case of an online originated virtual community than in the case
of an offline originated one (Hypothesis 8 is supported). Also, offline activities affected influence more
strongly in the online originated virtual community than in the offline originated one, which supports
Hypothesis 9. The results imply that in the case of an online originated virtual community, practitioners
should concentrate more on empowering active leaders, enhancing perceived similarity with other
members, fostering interactions between members and holding various events in an offline setting.
However, in another moderated regression analysis in terms of immersion, no significant increased R
found (Hypothesis 10 is not supported).
Table 2: Models for moderated regression analyses
(Note) MEM, Membership; INF, Influence; IMM, Immersion; E, Leaders’ enthusiasm; S, Similarity; O,
Offline activities; P, Playfulness; CO, Community Origin.
Models for moderated regression analyses Related Hypothesis
MEM = E + S + O + CO + E × CO + S × CO + O × CO + errors
INF = O + P + CO + O × CO + P × CO + errors
IMM = O + P + CO + O × CO + P × CO + errors
Table 3: Results of moderated regression analyses: Moderating effects of the community origin (CO) on
sense of virtual community
Standardized Regression coefficient (β)
Enthusiasm 0.279*** N/A N/A
Similarity 0.347*** N/A N/A
Offline activities 0.503*** 0.838*** 0.142
Playfulness N/A 0.166** 0.344***
Community Origin (CO) 0.786*** 0.387 - 0.442
0.600 0.480 0.248
CO × Enthusiasm
- 0.323* N/A N/A
CO × Similarity
- 0.368* N/A N/A
CO × Offline activities
- 0.121 - 0.510*** - 0.123
CO × Playfulness
N/A - 0.076 0.230
0.034 0.041 0.003
F Change 5.097 7.121 0.289
Significance of F change 0.002 0.001 0.749
0.634 0.521 0.251
0.618 0.507 0.228
F-value 40.372*** 36.139*** 10.968***
d/f 7/163 5/166 5/164
*p<0.1; ** p<0.05; ***p<0.01
In this study, we proposed and validated that the sense of virtual community construct is composed of
three dimensions: membership, influence, and immersion. Besides, we found that the sense of virtual
community is affected by (1) leaders’ enthusiasm, (2) perceived similarity, (3) offline activities, and (4)
playfulness. We also validated that the effect of virtual community characteristics (leaders’ effort,
similarity, offline activities and playfulness) on sense of virtual community is contingent on the virtual
For the researchers on virtual community, we suggest to explore objective and behavior-oriented
outcome variables such as frequency of visit, duration time, or organizational citizenship behaviors for
future research. We suspect that the sense of virtual community may mediate the links between
independent variables and such outcome behaviors. To be able to trace such link over time, a longitudinal
study may need to be conducted that observes the same set of virtual communities over time.
Since the data was collected only in Korea and the target communities were chosen with a
convenient sampling method, the general applicability of the findings is limited. Nevertheless, we believe
that establishing the sense of virtual community construct provides a valuable perspective for in-depth
understanding of virtual community and its members’ behaviors.
Atuahene-Gima, K. and Li, H., arketing’s Influence Tactics in New Product Development: A
Study of High Technology Firms in China,_ Journal of Product Innovation Management, Vol. 17, 2000,
Beniger, J., “Personalization of Mass Media and the Growth of Pseudo-community,”
Communication Research, Vol. 14, 1987, pp. 352-371.
Bernard, J., The Sociology of Community, Scott, Foresman and Company, 1973.
Blumstein, P. and Kollock, P., “Personal Relationships,” Annual Review of Sociology, Vol. 14, 1988,
Csikszentmihalyi, M., Beyond Boredom and Anxiety, The Jossey-Bass Publishers, San Francisco,
Griffiths, M., “Internet Addiction: Does It Really Exist?” in Psychology and the Internet:
Intrapersonal, Interpersonal, and Transpersonal Implications, J. Gackenbach (ed.), Academic Press, New
York, 1998, pp. 61-75.
Hoffman, D.L. and Novak, T.P., “Marketing in Hypermedia Computer-mediated Environments:
Conceptual Foundations,” Journal of Marketing, Vol. 60, No. 3, 1996, New York, pp. 50-73.
Jones, Q., “Time to Split, Virtually: Expanding Virtual Publics into Vibrant Virtual Metropolises,” in
Proceedings of the 33rd HICSS, Hawaii, January 2000.
Kiesler, S., Siegel, J. and McGuire, T.W., “Social Psychological Aspects of Computer-mediated
Communication,” American Psychologist, Vol. 39, No. 10, 1984, pp. 1123-1134.
Kim, A.J., Community Building on the Web, Peachpit Press, Berkeley, 2000.
Lombard, M. and Ditton, T., “At the Heart of It All: The Concept of Presence,” Journal of Computer
Mediated Communications, Vol. 3, No. 2, 1997. [Online]. Available:
Marsden, P., “Restricted Access in Networks and Models of Power,” American Journal of Sociology,
Vol. 88, 1983, pp. 686-717.
McMillan, D.W. and Chavis, D.M., “Sense of Community: A Definition and Theory,” Journal of
Community Psychology, Vol. 14, 1986, pp. 6-23.
Moon, J.W. and Kim, Y.G., “Extending the TAM for World-Wide-Web Context,” Information &
Management, Vol. 38, 2001, pp. 217-230.
Sarason, S.B., The Psychological Sense of Community: Prospects for a Community Psychology,
Jossey-Bass, San Francisco, 1974.
Walther, J.B., “Relational Aspects of Computer-mediated Communication: Experimental
Observations over Time,” Organization Science, Vol. 6, No. 2, 1995, pp. 186-203.
Wellman, B. and Gulia, M., “Virtual Communities as Communities: Net Surfers Don’t Ride Alone,”
in Communities in Cyberspace, M. A. Smith and P. Kollock (eds.), Routledge, London, 1999, pp. 167-194.
Young, K.S., “Psychology of Computer Use: XL. Addictive Use of the Internet: A Case that Breaks
the Stereotype,” Psychological Reports, Vol. 79, No. 3, 1996, pp. 899-902.