Both the internet and bookstores are replete with advice concerning how virtual communities should be designed. However, most of the material focuses on specific social practices, design elements, and tools that a web developers have in their arsenal for building environments that encourage participation in online communities. Although this information is undoubtedly valuable to those attempting to create an online community, it is neccesary that designers develop a more complete understanding of the social psychological processes that underlie online community development in order to most effectively choose software conducive to the communities purposes and to establish productive patterns of interaction in their community. While a complete analysis of common advice for creating virtual communities is beyond the scope of this paper, it is important to explore the social psychological phenomena which provide a foundation for some of the most prevalent practices in building, maintaining, and evaluating online communities.
Building Online Community
When building a virtual community it is one must draw a distinction between the community itself and the space it inhabits. Traditionally, communities were defined as tightly knit networks of interpersonal ties that were primarily confined to a bounded geographic area (Wellman, 2001). The development of more effective transportation and, subsequently, telecommunications has led progressively to communities of networked individualism. These communities are not tied to particular physical places but, rather, are grounded in the network of personal relationships between individuals who may or may not be geographically collocated. Conceiving of community as a social network is useful when determining a starting point for building a virtual community because it implies that the community is distinct from the space it inhabits. That is, communities grow organically through interpersonal relationships; they are not built. Therefore, an emphasis on site design can be misplaced when creating an online space for community if the designer does not consider the ways in which users desire to interact with one another and the motivations for which they use the site.
There are a number of key considerations when creating an online community. First, it is important to assess community members’ needs and the tasks which users would like to accomplish in the community through interacting with the anticipated community members. This analysis suggests not only the type of software but design elements as well (Preece, 2000). For example, while there are many motivations for community members to join an online community, one extremely common one is information seeking (Ridings & Gefen, 2004). Indeed, one element of the early online community the WELL that fascinated Rheingold (1993) was the community’s ability to provide him quickly with the information that he needed. Therefore, it follows that if interviews and focus groups of potential community members reveal that they desire to share information, software selection and site design can be chosen so as to assist them by augmenting search capabilities, providing organization for user knowledge contributions, and providing links to external content that is relevant to community members (Ridings & Gefen, 2004). Additionally, under these conditions software, such as a bulletin board system, might be chosen that would allow information to persist on the site overtime as opposed to chat which allows only fleeting discussions.
Software selection should not occur in isolation but proceed in parallel with social planning. For example, a community’s whose function is to provide information to its members may want to establish guidelines concerning the correct way to post a request for a particular fact. Providing well thought out rules and expectations based upon the community’s needs, expressed using consistent language and tone, can be especially important to establishing a community’s culture. Social Identity Deindividuation theory posits that the absence of nonverbal cues when communicating via CMC forces participants to rely on contextual cues in the text to determine appropriate behavior and group norms (Walther & Parks, 2002). Furthermore, participants often over attribute similarity as a result of minimal cues thus creating social attraction to the group. Powazek (2002) notes that the “tone of the content you give your users is replicated and amplified a thousand times in the response it generates” (pg. 20). This is because the developer, in both content and design, often provides some of the most visible social cues to the participants about the community that inhabits the site. Therefore, establishing a friendly and helpful tone from the beginning likely can reduce flaming and hostile interaction on the site.
Another example of social planning rooted in the community’s needs is the determinination of whether it is desirable to set barriers to community participation (Preece, 2000). The appropriate level of barriers that community members must overcome in order to participate should be based upon the nature of the community. For example, a group that provides social support for geographically isolated homosexual teens might want to have a relatively high barrier to participation in order to discourage offensive comments and harassment. In contrast, a site that focuses on mountain biking may find only a very low barrier is neccesary. In all cases, however, care must be taken that barriers remain somewhat permeable and are not set so high as to discourage new participants. Often a high percentage of online community participants are lurkers who rarely contribute to discussion (Preece, 2000). While some may view these members as undesirable, benefiting from the community without giving back, they actually play an important role in the development of the community. Movement between levels of involvement in the community often occurs as interests shift and new topics emerge leading core members to take on less central roles while peripheral members become more involved in the communitu’s life (Wenger, McDermott, & Snyder, 2002).
After prototypes for a website are designed they should be tested by the community to provide feedback regarding usability and the extent to which the system meets the community’s needs. Participant design, an iterative process of testing and redesigning that involves the end user, should be used to refine the system and correct any existing problems. Technological design shapes not only human behavior but also the social context in which it is embedded (Harrison & Zappen, 2003). Moreover, this relationship is reflexive with human behavior and culture in turn shaping technological design. Therefore, when engaging in design, it is important to consider the social consequences of one’s choices. Specifically, Harrison and Zappen (2003) argue that web developers must engage phronetic research, research that is value laden, which asks not only about the desirability of where design choices might be leading us but also whose values are represented within the design.
Finally, a developer should invite engaging and diverse participants to help build a critical mass for the community. Diversity of online communities has historically been a problem because they are primarily communities of interest in which individuals seek others similar to themselves (Baym, 1998). This lack of diversity in online communities presents a particular problem for fledgling communities because research shows that homogenous groups of similar others may run out of things to talk about (Ling, Beenen, Ludford, Wang, Chang, Li, Cosley, Frankowski, Terveen, Rashid, Resnick, & Kraut, 2005). Furthermore, individuals are more likely to make contributions to a community when they believe that their contribution will be unique. One solution is to actively recruit diverse individuals that will enrich the community with varying viewpoints. Another potential solution would be to actively engage participants by presenting specific topics that stretch the scope of the community. This could provide participants with the opportunity to discover realms of opinion in which they might differ.
In short, when building a community it is important to first assess the community’s needs and desires and then carefully select software and develop social mechanisms conducive to productive interaction. Furthermore, given the relationship between technological design and human behavior, the consequences of such choices must be considered along with the values that they reflect. Finally, a diverse community, with the unique contributions each member can provide, results in greater participation.
Maintaining Online Community
Creating an online community, however, goes beyond simply designing a space in which community members interact. Numerous problems can arise in online communities necessitating the development of a culture that discourages “flaming” and social mechanisms for the community to police itself. Kiesler, Siegel, and McGuire (1984) define flaming as “the practice of expressing oneself more strongly on the computer than one would in other communication settings” (pg. 1130). They argue that messages online provide minimal cues and are depersonalized resulting in deindividuation, which is the process of losing social-awareness resulting in a lack of evaluating one’s actions. Ultimately, this can lead to uninhibited verbal behavior such as cursing, insulting others, and hostile comments.
Although early findings showed that aggressive, uninhibited behavior was more common in computer-mediated groups than face-to-face groups, there are several things that an online community can do to discourage flaming, uninhibited verbal attacks and abuse. First, flaming may be prevented by encouraging the development of interpersonal relationships based upon in-depth knowledge of other participants. These interpersonal relationships can counteract flaming, which is rooted in deindividuation and the depersonalization of others as a result of reduced social cues. While the reduced social cues of computer-mediated communication (CMC) inhibit relationship development, they do not prevent it. Walther (1992) noted that early CMC experiments provided individuals with artificial time limits that restricted the number of messages computer-mediated participants could exchange compared to face-to-face participants. Subsequent experimentation showed that, given enough time, computer-mediated participants’ levels of impression formation could approach that of face-to-face participants (Walther, 1993). Higher levels of impression development result in individuated participants, reducing the potential for deindividuation and flaming within an online community. The implications are that mature communities, where tightly knit social networks of interpersonal relationships have had time to form, may be more resistant to flaming than communities in the early phases of development.
There may be an optimal community size with respect to using interpersonal relationships for the prevention of flaming. Human beings only maintain relationships with roughly 1,000 individuals with whom they know well enough to converse (Wellman, 2001). Should very large communities find it impossible to promote interpersonal relationships among all members, they might attempt to establish a social identity and associated norms in order to discourage flaming. Although on the surface social identity deindividuation theory seems to conflict with interpersonal perspectives of relationship development online, it is possible for these to operate in tandem. For example, a participant might first relate to another individual on the basis of their shared social identity. However, as time passes and personal, individuating information is exchanged they may begin to relate on an interpersonal level. This progression highlights the need for backend messaging systems within communities that can provide semi-public or private spaces for social interaction (Wenger et al., 2002). In short, a community might find it self particularly strengthened when it creates a social identity that discourages flaming and places mechanisms within the community to allow for the development of interpersonal relationships
While anonymity in online communities can be useful for providing honest, open communication, because some participants will inevitably use it to disrupt the community it can also be a disaster. The solution is to create a system that encourages community participants to behave responsibly by tying their virtual identities to their real world identities. Warrant is the concept of connecting the self-presentation to the self. Online communities may have difficulty establishing warrant because the greater the extent to which a community allows for anonymity, the greater the opportunity is for individuals to self-present unrealistic identities (Walther & Parks, 2002). Communities should encourage their participants to present warranting information to encourage accurate self-presentations and to prompt others to behave in an appropriate manner. Warranting information is information which can be confirmed and is difficult or impossible for the person to whom it refers to manipulate (Walther & Parks, 2002). Examples of information that provide partial warrant are real names, directory information, and access to an individual’s face-to-face network. Therefore, online communities can move toward tying the real world to the virtual world by requiring new community members to create publicly viewable profiles.
Another good policy for online communities is to design them in such a way that the community can police itself. However, this does not mean deputizing certain community members to oversee others. Indeed, creating stiff hierarchies with consolidated power is not only detrimental to community development but it can also be an invitation for the abuse of power. For example, Wikipedia has been critizied for providing a cadre of administrators with so much power that they can effectively determine who has a right to use the site and the manner in which it should be used (Wikimedia, 2006). Community, when rooted in the concept of communitas, is an egalitarian experience in which participants find themselves in a liminal state where power differentials between participants are minimized. Creating a hierarchy serves to hamper expressiveness and impede the emotional connection experienced by participants. Therefore, web developers should carefully consider the mechanisms for self-policing that they put in place. A good strategy is to distribute community power as widely as possible among responsible community members. This policy has been successfully employed by many websites, the prototypical case being Slashdot.net. Slashdot community members are asked to occasionally moderate a maximum of five comments (Powazek, 2002). These moderators are themselves rated by meta-moderators who determine whether the comment moderations were fair. A study of this rotating power system suggests that the system works well, with most moderations being judged as fair, and that “rebel” moderators were uncommon (Lampe & Resnick, 2004). Therefore, a system of distributed power can truly reflect the values of the community.
In short, interpersonal relationships discourage flaming because participants know the person on the other side of the screen as a real person. Furthermore, interpersonal relationships can work in parallel with social identity to prevent hostile interactions online. Ideally, interpersonal ties that create both accountability and understanding should be encouraged to the full extent possible. Beyond that, designers should attempt to create a system with distributed power that encourages equality and egalitarian interaction.
Evaluating Online Community
Although there is a generally accepted lay concept of community, in academic literature the concept is multilayered and, at times, not well understood. I0n a study of the emergence of online community in usenet groups, Baym (1998) proposes a number of dimensions along which a community may be potentially measured. Specifically, Baym posits that community development can be distinguished by linguistic innovation, relationship development, and the existence of group norms. Baym holds that every group develops unique forms of expression. These can include new words, phrases, and acronyms as well as special types of jokes and stories. Furthermore, groups sometimes codify these linguistic innovations in documents that serve to inform and initiate new community members. Codification of knowledge and meanings owned by the group is not unique to online communities. Indeed, “one of the primary tasks of a community of practice is to establish this common baseline and standardize what is well understood so that people can focus their creative energies on the more advanced issues” (Wegner et al., 2002, pg. 11). Thus, the level of community development can be measured by conducting a content analysis of text the community creates for linguistic innovations. Additionally, the analysis of the level of knowledge codification can inform scholars of the stage of community development.
Baym (1998) also notes that community development is tied to the development of interpersonal relationships between its members. Although some early research suggested that creating interpersonal relationships via CMC would be difficult, if not impossible, research since the mid-1990’s has found a wealth of evidence suggesting that relationships not only form in online communities but flourish. Parks and Floyd (1996) found that 60.7 % of participants of a usenet group reported at least one interpersonal relationship. Furthermore, relationships formed online progressed to other media with 35.3% of these participants communicating via the telephone, 28.34% using the postal service, and a 33.3% eventually meeting face-to-face. Therefore, one method of measuring online relationship development is the number of media through which a particular relationship progresses.
However, noting the number of media through which a relationship progresses may not be the best method for defining online relationship development. This approach tacitly assumes that relationship development is tied to movement through progressively richer media, ultimately placing a premium on face-to-face interaction. However, many relationships considered quite intimate by their participants may never progress to another medium. A more inclusive and, possibly, more accurate way of measuring relationship development is to consider the amount and content of self-disclosure. Altman and Taylor (1973) proposed social penetration theory which measured relationship development by the breadth and depth of self-disclosure. Social penetration theory represents information about an individual as a series of concentric spheres. The perimeter of the sphere represents the breadth of self-disclosure and each successive sphere represents the depth of self-disclosure. Breadth of self-disclosure refers to the number of topics addressed during interaction, while depth of self-disclosure refers to a self-disclosures level of intimacy reached by a dyad. Relational development is usually characterized by greater breadth on relatively non-intimate topics with more limited forays into intimate spheres. Relationship development in online communities could be operationalized by the both the depth and breadth of self-disclosures because relationships in online environments are based solely on the verbal exchange of messages..
Finally, community development may be determined by the existence of behavioral norms, tacit rules understood by community members. Users reinforce these norms by providing social sanctions against those who violate them. Often these sanctions involve “shaming people into compliance by drawing attention to their violations” (Baym, 1998, pg. 61). Norms may be assessed by engaging in participant observation within the community and codifying apparent norms. Participant observation would be ideal because norms may operate below the level of consciousness and not be apparent until they are violated. However, participant observation may not be neccesary in those communities in which software allows text to persist in the system; simply analyzing past interactions may be enough to determine group norms.
Site developers should remember that “virtual” communities and the online spaces they inhabit are not synonymous. When creating online spaces for communities, site developers should remember to assess the needs and desires of a community and incorporate this information into the final design. Furthermore, web developers have a moral obligation to reflect on the consequences of their design, considering the values it promotes and whether this will lead the community in a desirable direction. Designers can allow the community some self-determination by designing sites so that the community polices itself, thereby determining appropriate content. They can also promote egalitarian exchange by eliminating power hierarchies that might serve to stifle participation and distributing power broadly among participants. Most important, web developers must remember that community is rooted in the interpersonal bonds created between participants. If design does not nurture the growth of these bonds the result is a bedlam of fighting or, worse, an empty community space.
Works Cited:
(2006, October 6th). Quitting wikipedia and wanted you to know why. Retrieved November 13, 2006, from Wikimedia mailing lists Web site: http://mail.wikimedia.org/pipermail/wikien-l/2006-October/054949.htm
Altman, I., & Taylor, D. (1973). Social penetration: the development of interpersonal
relationships. New York: Holt, Rinehart and Wilson, Inc..
Baym, N. (1998). The emergence of on-line community. In S. Jones (Ed.), Cybersociety 2.0 (pp. 35-68). Thousand Oaks: Sage.
Kiesler, S., Siegel, J., & McGuire. (1984). Social psychological aspects of computer mediated communication. American Psyschologist, 39(10), 1123-1134.
Lampe, C & Resnik, P. (2004). Slash(dot) and burn: distributed moderation in a large online conversation space. Proceedings of the SIGCHI conference on Human factors in computing systems, p.543-550, April 24-29, 2004, Vienna, Austria.
Ling, K., Beenen, G., Ludford, P., Wang, X., Chang, K., Li, X., Cosley, D., Frankowski, D., Terveen, L., Rashid, A. M., Resnick, P., and Kraut, R. (2005). Using social psychology to motivate contributions to online communities. JCMC, 10(4), Retrieved November, 4th 2006, from http://jcmc.indiana.edu/vol10/issue4/ling.html
Parks, M. R., & Floyd, K. (1996). Making friends in cyberspace. Journal of
Communication, 46(1), 80-97.
Powazek, D. (2002). Design for community. Indianapolis, IN: New Riders.
Preece, J. (2000). Online communities: Designing usability, supporting sociability. New York, NY: John Wiley & Sons Ltd.
Rheingold, H. (1993). The virtual community: Homesteading on the electronic frontier. New York, NY: HarperPerrenial.
Ridings, C. & Gefen, D. (2004). Virtual community attraction:Why people hang out online. JCMC 10(1), Retrieved November, 4th 2006, from http://jcmc.indiana.edu/vol10/issue1/ridings_gefen.html
Walther, J.B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19, 52-90.
Walther, J.B. (1993). Impression development in computer-mediated interaction. Human Communication Research, 20, 473-501.
Walther, J. B., & Parks, M. R. (2002). Cues filtered out, cues filtered in: Computer-mediated communication and relationships. In M. L. Knapp & J. A. Daly (Eds.), Handbook of interpersonal communication (3rd ed., pp. 529-563). Thousand Oaks, CA: Sage.
Wellman, B. (2001). Physical place and cyber place: the rise of personalized networks. International Journal of Urban and Regional Research, 25(2), 227-252
Wenger , E., McDermott, R., & Snyder, W. (2002). Cultivating communities of practice. Boston, MA: Harvard Business School Press.
Labels: CMC, Community, design, social psychology
