JCMC 9 (2) January 2004
Collab-U  CMC Play  E-Commerce Symposium Net Law  InfoSpaces Usenet
  NetStudy VEs VOs O-Journ HigherEd Conversation Cyberspace Web Commerce
Vol. 6 No. 1  Vol. 6 No. 2  Vol. 6 No. 3  Vol. 6 No. 4  Vol. 7 No. 1  Vol. 7 No. 2  Vol. 7 No. 3
 Vol. 7 No. 4  Vol. 8 No. 1  Vol. 8 No. 2   Vol. 8 No. 3  Vol. 8 No. 4   Vol. 9 No. 1

Virtual Behavior Settings: An Application of Behavior Setting Theories to Virtual Communities

Anita Blanchard
University of North Carolina, Charlotte


Abstract

Virtual communities are a new social phenomenon in computer-mediated communication (CMC). Within these communities, a sense of place is emerging that may provide a key to understanding them. This paper proposes that virtual communities can be better understood as operating within an emerging environmental form: virtual behavior settings, a concept informed by Barker's (1968, 1978a) and Wicker's (1987, 1992) theories of (face-to-face) behavior settings, altered by considerations of the distinctive qualities of time, place, and objects in CMC. Virtual behavior settings are examined in terms of the emergence and maintenance of setting programs, their participants, and their operation within physical behavior settings.


Introduction

Over the past 40 years, computer-mediated communication (CMC) has grown in use as a new mode of communication (Hiltz & Turoff, 1993). More recently, researchers, popular media, and participants alike have noticed that communities of on-going, relatively stable groups of computer users have formed and are maintaining themselves over CMC (S. Jones, 1995; Q. Jones, 1997; Rheingold, 1993b; Rothaermal & Sugiyama, 2001). These virtual communities have developed in a variety of areas of people's lives including education, work, and leisure, and they exist over nearly every form of computer communication. The increasing number of people involved in these communities and also the potential for these new communities to change our lives warrants further investigation into how and why virtual communities exist.

As an emerging phenomenon, researchers are only beginning to understand virtual communities. Previous research has focused on how CMC characteristics affect a variety of aspects of individual and group communication including impression formation, developing interpersonal relationships and group decision making (Hancock & Dunham, 2001; Hollingshead & McGrath, 1995; Ramirez et al, 2002; Walther, 1992). Most research specific to virtual communities has been primarily anecdotal and lacks a theoretical basis (Wellman & Gulia, 1999), although there is a growing body of research developing and applying theory to virtual communities (Baym, 1995, 1997; Q. Jones, 1997; Utz, 2000). Nevertheless, much of the research focuses on describing virtual communities and how they may change society, work, or interpersonal relations instead of developing theory about virtual communities.

I propose that virtual communities can be understood as operating within emerging environmental forms called virtual behavior settings. Virtual behavior settings are created through the shared interactions of members and a developing sense of space or "place" in CMC. Although virtual behavior settings are not completely analogous to face-to-face (FtF) behavior settings, the behavior setting concept as developed by Barker and his colleagues (1978a) and further advanced by Wicker (1987, 1992) can clarify virtual behavior settings and their potential for understanding virtual communities.

This paper first provides an overview of computer-mediated communication and virtual communities. Then FtF behavior settings theories are introduced. Next, the conceptual framework of virtual behavior settings is developed, drawing on both the theories of FtF behavior settings and research on CMC. The final section presents the implications of using the virtual behavior settings framework to examine virtual communities.


Overview of Virtual Communities and CMC

Although most people can recognize a community when they see one, arriving at a precise definition of community is difficult (Mann, 1978; Q. Jones, 1997). Community can refer to a group of people who share a physical location (e.g., a neighborhood) or it can refer to a group of people who share a common interest or characteristic (e.g., the scientific community) (Mann, 1978; Schuler, 1996). In addition, these types of community are not mutually exclusive; communities of interest can develop within physically based communities.

A feature that is common to both of these conceptions of community is the members' sense of community. Psychological sense of community is defined as the members' feeling of shared emotional attachment, belonging, influence, and the integration and fulfillment of needs that makes the community different from simply a group of individuals (McMillian & Chavis, 1986).

Sense of community appears to exist in the virtual communities discussed in previous research. Shared emotional attachment is apparent in Turkle's (1995) description of role-playing virtual communities in which members develop emotional relationships with each other including a multitude of friendships, virtual marriages and even divorces. Rheingold (1993a, 1993b) also discusses the attachment members feel for each other in the WELL. Some virtual communities clearly fulfill specific and unique needs for their members. Persons affected by cancer have listservs to augment their FtF support (Turner, Grube, & Meyers, 2001) while wristwatch enthusiasts use a multipurpose website to share knowledge, experience and products with each other (Rathaermel & Sugiyama, 2001). Other social and organizational virtual communities provide members opportunities to influence others by exchanging help, information, and support on topics ranging from parenting to soap operas to software problems to careers (Baym, 1995; Rheingold, 1993a, 1993b; Finholt & Sproull, 1990; Pickering & King, 1995). Based on these and similar examples, the social form called virtual communities does appear to have a psychological sense of community for their members.

Virtual communities are relatively stable groups of people who interact primarily over CMC and who have developed a sense of community. These members interact with each other through the computer, but may have additional modes of communication such as FtF interaction, the telephone, or postal mail. Membership in virtual communities is primarily voluntary. That is, most participants seek out virtual communities with whom they share interests (Wellman & Gulia, 1999) and stay with these communities based on their own interest in maintaining membership. Occasionally in organizations, employees are required to be members of particular virtual communities (Finholt & Sproull, 1990; Sproull & Keisler, 1991), but their functioning as virtual communities may not be as strong as with completely voluntary virtual communities. Some virtual communities consist primarily of discussions which vary from informal chats among participants to more formal discussions of specific topics and issues. In other virtual communities, participants engage in on-going, interactive role-playing games.

Because virtual communities exist over computers, information about communicating via computers should be informative for virtual communities. The following section will describe some of the characteristics of CMC and virtual communities and provide relevant research in these areas.

Characteristics of CMC and Virtual Communities

Table 1 provides an overview of the some of the main forms of text-based CMC on which virtual communities exist and a selection of their characteristics, including the timing of the messages (i.e., synchronous in which messages appear on others' screens as they are typed or asynchronous in which messages are read at a later point in time), the method of access, and examples of virtual communities. Although researchers do not currently believe that CMC characteristics determine virtual community behavior, these characteristics are believed to have some influence on behavior as compared to FtF interactions (Baym, 1995; Walther, 1992; Wellman & Gulia, 1999). In particular, the timing of the CMC (asynchronous or synchronous), the reduction of social cues (e.g., smiles, tone of voice, and other physical information), and the software capabilities (e.g., use of objects in MOOs) influence virtual community behavior (see Culnan & Markus, 1987 and Walther, 1992 for more detail).

Virtual Community
Description
Access
Examples
Asynchronous Listservs Group email discussion. Email account. Must initially join group to receive messages. Only members can receive messages from the listserv. EvalTalk: professionals interested in program evaluation
SurvivorShip: Recovery group for victims of childhood abuse
Newsgroups Group discussions similar to listservs except that messages are stored in a central location. Must go to particular newsgroup site through Internet (e.g., Usenet), direct dial in, or specific Internet Service (e.g., AOL). sci.psychology.misc: for discussions about psychology.
rec.arts.tv.soaps: for soap opera fans
Synchronous Chatrooms Real-time discussions loosely based on particular topic or interest group Must belong to service with this capability. Search through "rooms" or "channels" to find topic or group of interest. Beyond Planet Earth: for people interested in astronomy
30+: for people over 30 years old.
MOOs Real time discussions and activities loosely based around particular topic. There is movement around a MOO, and participants interact with "objects." Telnet to specific MOO. Can maneuver through rooms or areas. MediaMOO: MOO for professionals interested in media research. Set up as a series of office buildings.
LambdaMOO: Well-known social MOO set up as a house.

Table 1. Characteristics of various types of virtual community technology. Note: The names for these communities sometimes vary. Listservs are also known as majordomos; both names refer to the software that runs the technology. Newsgroups are also known as bulletin boards. Chatrooms developed from Internet Relay Channels (IRC). MOOs are similar to other virtual communities known as MUDs, MUSHes, and MUCKs.

Researchers have additionally focused on participants' characteristics, interactions between participants, interactions between participants and the technology (e.g., equipment or software), the rational choices people make in using the technology, and contextual influences on communication practices such as the physical and social environments in which CMC takes place (Baym, 1995; Hollingshead & McGrath, 1995; Markus, 1994; Walther, 1992; Wellman & Gulia, 1999). Researchers are, therefore, examining a broad range of influences on CMC behavior.

Although the research in CMC that is applicable to virtual communities is expanding, research specific to virtual communities is lacking. There are more anecdotal accounts of experiences in virtual communities than there are careful examinations (Wellman & Gulia, 1999). Additionally, the careful examinations tend to address only pieces of people's experiences in virtual communities such as attempts to maintain norms in the group (Kollock & Smith, 1994; McLaughlin, Osborne, & Smith, 1995; Smith, McLaughlin, & Osborne, 1998), identification of virtual communities (Q. Jones, 1997; Liu, 1999), identification of participant types (Myers, 1987), experimentation with different on-line personas (Turkle, 1995), impression formation and management (Douglas & McGarthy, 2002; Hancock & Dunham, 2001; Walther, Shorack & Tidwell, 2001), the development of on-line friendships (Utz, 2000) and the movement of relationships from CMC to other media (Parks & Floyd, 1996). Few researchers attempt to examine virtual communities as a whole.

Notable exceptions are Baym (1995) and Q. Jones (1997). Baym (1995) enumerates a range of CMC characteristics and group use of CMC that affects the emerging social dynamics observed in virtual communities. In particular, she examines how external contexts (e.g., the participant's immediate environment), temporal structure (e.g., limited or on-going interactions), system infrastructure (e.g., characteristics of the CMC), group purposes (e.g., task or social), and characteristics of the group and individual members affect social outcomes such as relationships among members and norms of the group. She argues that these characteristics do not determine the social outcomes, but are influential in the dynamic interaction among the community members.

Q. Jones (1997) takes an archaeological perspective. He argues that like an archaeologist in a village, we can understand how the virtual community functions by examining the characteristics of the place it inhabits and the traces of life left by its occupants. Further, he believes that virtual communities need to have a minimal level of interactive, public communication, a variety of communicators, and a minimal level of sustained membership to be considered virtual communities.

Baym's (1995, 1997) and Q. Jones' (1997) contributions are important in understanding virtual communities. Baym links several important contextual factors to the observed social outcomes in the virtual community. Jones attempts to decouple the virtual community from the place in which it occurs, noting the importance of the place and objective characteristics where the virtual community occurs. In fact, Jones' work notes what may be key to understanding virtual communities: the emergence of a new environmental form. The next section will examine the sense of place developing in virtual communities and their emerging virtual environment.

Development of a Sense of Place

One of the more surprising and perhaps more important developments of on-going CMC is its emergence as a "place" for behaving. Researchers and participants are beginning to refer to CMC as a social or even a conceptual space that people have begun to experience as a place (Beaubien, 1999; Harasim, 1993; S. Jones, 1995; McLaughlin et al, 1995; Rheingold, 1993a).

What contributes to a sense of place? One factor may be the social exchanges that occur in virtual communities (S. Jones, 1995). Rheingold (1993a, p. 64) argues that after a group accumulates a sufficient number of shared social experiences, "[the community] takes on a definite and profound sense of place in people's minds."1 Another factor might be the individual's cognitions of the computer's functioning. Weick (1990, p.14) argues that since much of the work on computers is out of sight, individuals create mental models to help them understand what is going on inside the computer. They use these mental models to understand where they "are" on the computer and may move to different parts of this model when they are interacting with their different virtual communities.

Individuals' mental models may be influenced by the ways they must access their virtual communities and also by the identification of boundaries for a particular virtual community. Methods of access are discussed in Table 1 for the various types of CMC technologies. Participants sign on to listservs similarly to email. Participant sign onto their email account and simply read their email on their local computer system. Receiving communication from the listserv is quite similar to receiving email from a friend, colleague, or family member. The main difference technically is the heading of the email indicating that it is from the listserv group.

Accessing newsgroups, chatrooms, and MOOs, on the other hand, require varying levels of "movement" around the Internet or participants' Internet service provider (e.g., AOL). That is, participants must execute commands to move to a particular part of a computer system or to another system entirely. To access public Usenet newsgroups, participants must connect with a news server (i.e., a central storage location) and join the particular group in which they are interested. To access newsgroups specific to commercial services such as America On-Line or Prodigy, participants use the service's menu or icons. For chatrooms and IRC, members must connect to a publicly available address or use icons and menus to access an Internet service's chatrooms. For MOOs, a participant has to "leave" his or her local computer system via telnet or some other program that allows electronic "movement" to other computer systems to access the MOO. Web pages can contain several of these CMC technologies (e.g., bulletin boards and chatrooms) in one "place" or grouping of web pages.

How may participants' methods of access affect feelings of place? Does a participant in a listserv have the same feeling of being in a place as someone in a MOO? There are several lines of reasoning which could help us understand why people experience a sense of place. If sense of place is weak for listserv participants, but stronger for newsgroups, chat and MOO participants then accessing the virtual community, specifically moving through one's mental model of access may contribute to sense of place. MOOs, newsgroups, and chatrooms require that participants actively access the forum whereas listservs do not. Their mode of access may indicate a dimension of perceived space in CMC: the more active one is in reaching the CMC forum, the more one may feel a sense of place. Therefore, MOOs, newsgroups, and chatrooms may invoke more locational awareness than listservs or other forms of email. If sense of place is similar for listservs compared to other groups, then the shared social interactions within the virtual community may be the key component of the sense of place.

The timing of the CMC communication is another possible contributor to sense of place. If synchronous virtual communities have more sense of place than asynchronous ones, then the speed of interactions may affect members' sense of place. Pargman (2000) argues that synchronous settings allow the development of rituals and conversations between members which increases the feelings of co-presence. This co-presence feeling presumably increases members' conceptions of a persistent virtual place.

The virtual community boundaries may also affect sense of place. Boundaries can indicate whether the person is in or out of the virtual community. One boundary that is often cited is the name of the topic of discussion for the groups (Harasim, 1993; Kollock & Smith, 1994). Thus, within newsgroups or chatrooms, one picks and chooses which virtual community to enter based on the name of the group indicating the topic of discussion. This is the only external distinction between any particular group and any other one on the same system. Even in MOOs, as one wanders through the system of "rooms," the name often indicates the purpose of the room. Participants may expect behavior in a virtual "café" (where people engage in informal chats) to be different from behavior in a virtual "classroom" (where teachers lecture or carry on discussions with students) even if both rooms are on the same system with the same cohort of participants. Listservs are different. The boundary is still the topic or the name of the listserv, but the only thing that separates listserv messages from private email is the header information (e.g., the "from" and "subject") of the email. Thus, listservs' boundaries are more ambiguous.

The emergence of a feeling of place in virtual communities is drawing some attention in virtual community research (S. Jones, 1995; Harasim, 1993). S. Jones (1995) argues that researchers need to understand the connection between social relations and spatial practices, values and beliefs. It seems imperative to examine functioning within this new virtual place. Since virtual communities occur among groups of people interacting in specific virtual locations, it seems that the behavior setting concept should provide a provocative environmental analogy. The following section provides an overview of behavior settings and highlights features that may be relevant in virtual communities.


Overview of the Theory of Behavior Settings

Behavior settings theories grew out of Roger Barker and his associates' (1978b) observations that the individual behavior of children was better explained by their current environment at the time of the observation than by their individual characteristics. He identified these environments as behavior settings and began to examine their characteristics and their influence on people's behavior. Behavior setting researchers then went on to investigate the behavior settings of adults and also to examine the role of cognitions in these settings (see Wicker, 1987).

Behavior settings are small-scale social systems, bounded by time and place, composed of people and physical objects (Wicker, 1992). Time and place boundaries are important for identifying when and where the behavior setting exists, but people and objects are primary components. People are the most malleable part of the behavior setting and are essential to its existence (Barker, 1978b). Without at least one human occupant, the setting does not exist. Nevertheless, the interactions among the people and with the physical objects make a behavior setting what it is.

A key feature of this interaction is the behavior-environment "synomorphy," or the fit between the behavior of the people and the characteristics and arrangements of the physical objects in the setting (Wicker, 1987). That is, people and objects interact in a behavior setting. People arrange objects in the setting (e.g., chairs, desks, computers, etc.). However, people are also constrained by these objects. For example, arranging classroom chairs in a circle may facilitate discussions, but may make lecturing more difficult.

Human-object interactions in behavior settings are not random, but rather they create an orderly pattern of behavior, a prescribed sequence of interactions between the people and the objects in the behavior setting known as the "setting program" (Wicker, 1979). The program is a key characteristic of behavior settings. It is the behavior setting's raison d'être and although it may be subject to modification, drastic changes in the ability of a setting to maintain its program threaten the survival of the setting. For example, in a café, employees use coffee machines and cash registers as they serve coffee to customers, who sit at tables to enjoy the coffee. In this example, each group of participants is necessary for the survival of the setting and do not normally perform the others' behavior.

Behavior settings also serve participants in ways other than their programs. Participants are not totally engaged in carrying out and maintaining setting programs (Wicker,1987) even though that is a major responsibility for most. People may see opportunities to achieve satisfactions which may be only indirectly related to the program, such as self-expression, developing social relationships and carrying out side programs. These are subsidiary to the behavior setting program, but exist nonetheless. However, occupants experience obligations to maintain the setting program (Barker, 1978b). The setting program must be maintained in order for the behavior setting to survive. These opportunities and obligations felt by behavior setting participants work together to keep the program functioning and thus keeping the behavior setting functioning.

Not all participants feel the same level of obligation or opportunity within the setting (Wicker, 1987). Some types of participants are more heavily involved in maintaining the program whereas others are more passive participants (Barker & Schoggen, 1978; Schoggen, 1989). Table 2 lists the names and definitions of the types of participants in behavior settings along with their level of influence. If one considers power in terms of how active the participant is in maintaining the program, Table 2 also provides a distribution of activity level by participants.

Level Name Definition
6 Single Leader Doesn't share power or leadership authority at any one occurrence. May have helpers at level 4.
5 Joint Leaders Share power or leadership authority.
4 Active Functionary Have power over part of the setting, but do not lead the setting
3 Member/Customer Have great potential power, but little immediate power.
2 Audience/Guest Little power, but welcome
1 On-looker Tolerated, but not welcome. No power.
0 Potential Inhabitants
Table 2. Types of behavior setting participants and level of influence.

Participants at levels 4-6 are called the "operatives" of the behavior setting. They are the necessary inhabitants of the setting. They have the most power in the setting, but they also experience the most coercion by the setting. Like the employees of a café, these participants are most strongly affected by the mechanisms which keep the setting operating.

The participants at the lower level of influence are also important, but contribute in an indirect way. These more passive inhabitants assist in the functioning of the behavior settings, like the customers in a café. Interestingly, although most of the participants in each of these levels are considered in positive or neutral terms, those at level 1, the on-lookers, are considered in negative terms. For example, people rarely stare through the windows of a café to observe customers drinking coffee and chatting. These types may be tolerated although not welcome in the behavior setting (Schoggen, 1989).

Behavior settings are self-regulating entities. According to Barker (1968), participants perceive and respond to events in the behavior setting they occupy. If a participant interprets an event as a threat to his or her opportunities for satisfaction, then the participant will act to correct this event or remove it. Wicker (1987, 1992) recast Barker's explanation of the self-regulation of settings in terms of Weick's (1979) process of sensemaking. That is, Wicker argues that behavior settings are social constructions whose structure can be understood in terms of sensemaking and negotiating among participants. Thus, behavior setting programs reflect the continuous organizing and negotiating activities of participants. Individuals develop cause maps (i.e., beliefs about why things happen) as to the functioning and operations of the behavior setting; these cause maps vary in degree of overlap with the cause maps of other behavior setting participants. As expressed by Wicker (1992), the benefits of the sensemaking model over Barker's model is that individuals are differentiated, and cause maps help us understand the direction of cognitions and actions of setting occupants as individuals and group members.

Although behavior settings are self-regulating, they are subjected to influences originating outside of the setting. Behavior settings are open-ended systems that depend on events, conditions, and entities beyond their boundaries (Wicker, 1987). These influences may originate within other settings of a large organization or of a larger community and may include demographic, cultural, political and economic influences.

This section reviewed some of the features of behavior settings that may be useful in examining virtual communities. In the next section, these features will be discussed in terms of behavior and cognitions in virtual communities.


Virtual Behavior Settings

Given the knowledge we have about behavior settings and the observations that virtual communities are developing over CMC, it seems reasonable to propose that a new form of behavior setting, a virtual behavior setting, is emerging. Although virtual behavior settings are conceptualized somewhat differently than physical behavior settings, they can provide insight into people's experiences of community in CMC. That is, observations and theories about people in virtual behavior settings can be informed by observations and theories about people in FtF behavior settings.

Virtual behavior settings are naturally-occurring, computer-accessible social spaces in which groups of people participate in on-going exchanges of communication. They are naturally occurring in that they have not been induced by an experimenter, but have simply emerged through CMC practices. They are accessible only through computer communication and thus cannot be identified or characterized by the typical physical attributes used in identifying traditional behavior settings.

The lack of physical objects in virtual behavior settings as compared to physical behavior settings, which will be discussed later, increases the dependence of virtual behavior settings on electronic "objects." Objects in virtual behavior settings are primarily the messages members exchange. This has two major effects. First, for communication to occur, there must be at least two people in the virtual behavior setting, although not necessarily at the same time. Second, the process of communication exchange is very important regardless of whether its content is information, support, sensemaking, coercion, or a request for help. Participants put objects of communication into the behavior setting that they then perceive and negotiate, and thereby, socially construct the virtual behavior setting (Weick, 1979). However, quite differently than physical behavior settings, these communication objects are also manipulated as members cut, paste, save, exchange and alter all or part of the text communication. Therefore, virtual behavior settings are socially constructed entities in which members manipulate their communication objects.

Two of the biggest differences between FtF and virtual behavior settings are how the concepts of time and place are dealt with. Time and place boundaries are fundamental attributes of physical behavior settings. However, CMC seriously weakens these boundaries (Garton & Wellman, 1996). The next two sections will consider the concepts of time and place within virtual behavior settings.

Time

How does the concept of time apply to virtual behavior settings? In a sense, virtual behavior settings are "open all night" and have 24-hour access. Because anyone can access a virtual behavior setting from anywhere in the world with a computer, theoretically, people can be participating in it at all times. Therefore, it may never "close" or cease to exist.

The effect on the virtual behavior setting differs depending on whether the communication is synchronous or asynchronous. Synchronous virtual behavior settings are the most like 24-hour cafés. Participants can log on or access the settings and expect to communicate with others. However, in synchronous virtual behavior settings, there are greater needs for others to be present in order to enact the setting program. Conceptually, in a FtF 24 hour café, a lone customer can enter and, even devoid of employees, make a cup of coffee, sit at a table, and drink the coffee, thereby enacting the setting program. However, if a lone participant enters a synchronous virtual behavior setting such as a chatroom without any objects or people, it is essentially impossible to enact any setting program. Some synchronous virtual communities such as MOOs have the capacity for solitary behavior such as creating objects or engaging in asynchronous communication. In these virtual behavior settings, only one person may be present to enact the setting program. Otherwise, a synchronous virtual behavior setting exists with just one person present, but it cannot function.

For asynchronous virtual behavior settings, the concept of time is more individually focused. Other members may or may not be "in" (i.e., accessing the communication objects) an asynchronous virtual behavior setting at the same time. The behavior setting's existence depends on the individual's presence: it exists only from an individual's perspective when the individual is reading the messages. The person can see evidence (in the form of text messages) of other peoples' presence, but the person does not have any indication of the simultaneous presence of other people. For listservs and newsgroups and other asynchronous forms of communication, the time dimension focus changes from "when people are present" to "when I am present."

Therefore, in comparing FtF and virtual behavior settings on the concept of time, there are some similarities and differences. Synchronous virtual behavior settings may be more fragile because they depend on people participating at the same time; people in asynchronous settings (similar to FtF settings) need only to see objects such as messages from past participants in order to enact the setting program. However, in asynchronous settings, individuals' conceptions of time are decoupled from each other. For the participants, the general consequence may be the ability to participate at their convenience. But for the theorist, the consequence is that it is more difficult to distinguish when the virtual behavior setting is in existence.

Place

The differences between FtF behavior settings and virtual behavior settings in regards to place have to do with the absence of a physical place despite the presence of an emerging virtual "place." As discussed previously, sense of place in virtual behavior settings comes from the social interactions between participants and their mental maps of the virtual community. This seriously distorts the concept of place as used in FtF behavior settings. Is there any "place" that is similar to FtF behavior settings? Actually, perceptual cues do exist in a virtual behavior setting. Messages flow across the screen in synchronous settings. They are organized in particular ways in asynchronous settings. Participants can be aware (or not) of other people's presence in the setting. One can even identify (i.e., point to) a virtual behavior setting in terms of its electronic address or name.

Although virtual behavior setting participants (like FtF setting participants) share a common location in the electronic address, their perceptions of place (unlike FtF setting participants) may be quite different. In asynchronous settings, users' access software (e.g., Google groups vs. Microsoft Outlook) may organize messages differently making it more or less difficult to follow conversations. Different synchronous software can also offer various options to make following conversations more or less difficult. Therefore, in virtual behavior settings, participants can be literally in the same location at the same time and see quite different "places." Again, this decoupling of the individuals' concept of place from the group's is quite unlike FtF behavior settings. However, in this instance, individuals' unique concepts of place cause them to have easier or more difficult experiences than others in enacting the setting program, due mostly to their access software.

Objects

Theories of behavior settings specify more than simply time and place; a key component of these theories is the interaction between the people and physical objects. The interactions between people and objects create a synomorphic relationship (i.e., the person-object fit). People manipulate and are constrained by objects in the setting. Therefore, synomorphic relationships should also exist in virtual behavior settings between people and their electronic objects.

The most obvious synomorphic relationship exists between the person and the computer. However, this relationship is on the boundary of the virtual behavior setting. A person must use a computer to access a virtual behavior setting, so characteristics of the computer and its software such as ease of use, speed, and access capability have important impacts on individuals involved in CMC and virtual communities (Baym, 1995).

The physical computer on the edge of the virtual behavior setting is important, but it is not the only object with which the participants interact. As discussed previously, the text that is exchanged can be considered an important (electronic) object, and its limitations in conveying social cues and social presence have been examined extensively (Sproull & Keisler, 1986; Hancock & Dunham, 2001).

This limitation in conveying cues considers only part of the synomorphic relationship between people and objects in the virtual behavior settings. According to behavior setting theory, we should expect people to shape and constrain the text to meet their needs. One may argue that CMC users have developed ways to add back in social cues in their text communications such as emoticons, special new vocabulary to denote tone of message (e.g., IMHO for "in my humble opinion"), using signatures that contain personal information or simply typing in a behavior or emotion (e.g., "I'm laughing out loud"). That is, these developments are participants' manipulations of the text objects' capacity to achieve a higher level of social interaction.

MOOs have built into their programming language the capacity for participants to describe the behaviors that they want to do. For example, MOO participant Sally can type "emote waves fondly to others in the room" and all the other participants will see "Sally waves fondly to others in the room" appear on their computer screens. Although MOOs are not as popular a CMC medium as others, their ability to allow for more expression encourages experiences close to FtF interactions (Pargon, 2002). Thus, although text-only communication constrains expression in all CMC, people manipulate their text to meet their goals of communicating more effectively.

Postmes et al. (2000) report that the use of this paralanguage varies between different online groups; not all online groups use emoticons or other paralanguage in the same way. Some virtual behavior settings participants use their paralanguage objects to be more expressive whereas others do not. Is this an example of personal preference or an example of a synomorphic norm within a settting? Postmes et al. rerport that members change their use of paralanguage as they move between different groups. This can be interpreted as an indication of members' awareness of switching virtual settings and virtual behavior setting programs.

One CMC environment, the MOO, has the capacity for a greater diversity of objects. MOOs have special objects that users manipulate. These electronic objects are usually text descriptions of such "things" as clothing, food, drink, books, and furniture. Virtual communities with more advanced graphical capabilities can actually "show" the objects in the setting. People interact with these electronic objects; they pick the objects up, carry them, keep, discard, wear and generally "use" them. Interestingly, these electronic objects are important in people's interactions with each other. Bruckman and Resnick (1996) report that at the inauguration of a new MOO, participants spent more time interacting with the objects and discussing their interactions than simply conversing with each other.

Setting Program

The essential feature of FtF behavior settings is the setting program; programs are important in virtual behavior settings, also. Programs in virtual behavior settings primarily consist of communication and the exchange of messages because the range of behavior in virtual behavior settings is restricted to these elements. It must be noted that the exchange of messages and information is also important in FtF setting programs. However, in those settings, there is also likely to be movement or exchange of physical objects, too.

Often virtual communities are divided into task-oriented and socially-oriented communities. That is, they are characterized by whether the primary purpose of their communications is to accomplish some work goal or to provide emotional and other social support for the participants. This division has been questioned (Baym, 1995; S. Jones, 1995). Critics of this type of division argue that both types of communication occur in all groups. That is, the task at hand may be important, but exchange of social support is important, too (Baym, 1995). Baym's study of a newsgroup based on soap operas (considered quite social as compared to an electronic work group) characterized the group's main functions as "collaborative interpretation and distributing information" (p. 147). Selwyn's (2000) study of a professional online teacher's group described them as primarily exchanging information about teaching children with special needs, but was also noted that the group was used for empathic exchange. S. Jones (1995) argues that dividing groups into relationship-oriented or task-focused groups shows insight into the functions of the group but does not address the connections between the users.

Conceptualizing virtual communities' functioning in terms of a program instead of a purpose allows for these multiple purposes to exist simultaneously and to contribute to the survival of the virtual community. The main setting program is analogous to the primary function (professional or social) of the group. However, individuals may have subsidiary programs or may not be fully engaged in the setting's main program (Wicker, 1987). This would help explain why communities that are primarily work-related or involve mostly information exchange also provide social exchange and support to the participants. Individuals find opportunities to fulfill their personal needs in the virtual behavior setting within the obligations of maintaining the virtual setting program.

Setting Program Maintenance

Another key feature of behavior settings is that they are self-regulating. That is, there are forces acting on the participants to maintain and carry out the setting's program. This concept has been recognized as extremely important in virtual groups and has been addressed in studies of group norms and of the importance of cooperation and coordination for survival of the group (Kollock & Smith, 1994; McLaughlin et al, 1995; Postmes, Spears & Lea, 2000; Sassenberg, 2002). The key issues here are what causes participants to attempt to correct other participants and how can they do it? As in FtF behavior settings, the participants' cognitions about the virtual behavior setting's program, how the setting should operate, and their role as participants should be important in what causes them to try to correct others. The processes of sensemaking and negotiating that occur in a socially constructed entity continue as participants establish, maintain and modify the programs. Additionally, leaders and other committed members of the group are more likely to engage in sanctioning (Sassenberg, 2002).

What actions can be taken to enforce or maintain the program? In FtF behavior settings, participants can attempt to get others to follow the program by correcting their behavior or by eliminating the deviant members from the setting (Barker, 1968). However, sanctioning is quite difficult in a virtual group (Kollock & Smith, 1994). Sanctioning a member who is violating the program or behaving in ways that disturb other participants' activities often consists simply of electronic messages telling the offending party to stop or change his or her behavior. McLaughlin and her colleagues (1995; Smith et al., 1998) found that 15% of all messages from a randomly selected set of newsgroups were what they classified as "conduct correcting." These messages can be highly emotional and are sometimes characterized as flames. However, without the option of physical force, they may be perceived as no more than empty threats.

Not all participants resist changing their behavior to follow the setting program. It may be that the salience of a person's identity as an individual or as a member of the virtual setting affects willingness to change his or her behavior (Spears & Lea, 1992). That is, when personal identity is more salient, the individual behaves as an individual interacting with other individuals and when group identity is more salient, intragroup and intergroup psychological processes are more relevant explanations of behavior. It follows that sanctioning and influence by the group to modify behavior and conform to a program may be more influential if the person identifies as a member the community instead of as an individual.

Indeed, Sassenberg (2002) found that Internet groups in which members have a stronger group identity are more likely to adhere to the group norms than groups with lower group identity. Specifically, Sassenberg found that when members of virtual groups identified with the group as a whole, its goals, and its purposes (in other words, its setting program), these group members were more likely to adhere to that program. But what about those members who do not adhere to the setting program even after they have been informed of their violations?

Ignoring the offending person is one of the few useful sanctions available (Kollock & Smith, 1994). However, for this strategy to be effective every current participant, as well as each new member who joins the group, must agree to do this. Therefore, it is likely that this method of sanctioning is not very effective at stopping a behavior.

However, ignoring an offending person may be quite beneficial to the person who is being offended. If negotiations or sanctions fail, then by simply ignoring or not reading messages from certain people or about certain topics, the individual can eliminate most of his or her direct involvement with others whose perceptions of the setting program do not overlap with his or her own. This supposes that there are others who are also participating in the group who share the person's program expectations. Thus, to an outside observer, there may exist several competing main setting programs within the virtual community, but to the individual participant, there is only one. One may expect that the more that participants' cognitions and expectations differ, the larger the number of competing, higher order programs there will be.

The final step to preserve a virtual setting program is to moderate the virtual behavior settings and to restrict membership (Kollock & Smith, 1994). When asynchronous virtual behavior settings become moderated, one person becomes responsible for screening or compiling the messages and sending them to the rest of the group. Often, a computer program assists the moderator in this job. Thus, messages about certain topics or from certain people are not distributed for the rest of the community to read. This type of moderation is not possible in synchronous communication.

Both synchronous and asynchronous virtual behavior settings can restrict membership. In asynchronous virtual behavior settings (mainly listservs), moderators also make membership decisions as to who is allowed to join. In synchronous virtual behavior settings, an offending person's username and access privileges are deleted by one of the technical operators of the MOO (Turkle, 1995). This rarely happens, and is reserved for instances when the individual has done a particularly antisocial act such as a virtual rape.

Although moderated groups have not been examined extensively, it appears that the movement from an unmoderated virtual community to a moderated one is not undertaken lightly and does not proceed quickly. It also appears that there are a number of groups that restrict membership from the beginning. This initial restriction can reduce the problem of members having different cognitions about the setting program. However, ignoring other people may be more likely than higher levels of sanctioning since it is relatively simple.

Participants

Participants in virtual behavior settings can be characterized by activity level, ranging from leaders to lurkers. Table 3 lists a proposed typology. Leaders fall into three main types: those who have skills to run the virtual behavior setting such as technical leaders and those who are leaders through their communication such as information and social leaders (Myers, 1987; Turkle, 1995). A technical leader may or may not also be a information or social leader. These types of leaders may be indicative of the types of power or control in virtual behavior settings. Technical leaders have the ability to restrict membership or otherwise affect the operating of the virtual behavior settings. Information leaders have expert knowledge relevant to the group. Social leadership may simply depend more on level of activity. If a participant is actively communicating, he or she may be a leader because others are more likely to read his or her communication. Some participants may ignore this highly active person, but unless all participants do, the person may be considered a leader by virtue of the number of his or her communications.

Category Participant Type Definition
Leader Technical Leader Sees that program that supports CMC form runs. Also known as moderator (asynchronous) and wizard (synchronous).
Information Leader Major provider of expertise and knowledge about a topic
Social Leader Major provider of social support among members
Participant Participant Contributes along a continuous range of communication
Lurker Private Communicator Communicates to other participants through private email
Never Communicates Never communicates to other participants or to community
Table 3. Proposed types of virtual behavior setting participant and activity level.


This type of leader is somewhat different than the leaders of FtF behavior settings as defined by Barker and Schoggen (1979). The single leader in FtF behavior settings may be comparable to the technical leader who has power over admitting members. However, the power of technical leaders over the virtual setting's program may be substantially less than single leaders in FtF settings. The joint leaders are similar to the communication leaders in virtual behavior setting in that they share power or leadership in the setting. However, information and social leaders may have more power over influencing the program. In addition, there may be more of them since the amount of power they possess is generally determined by how active they want to be and also in the ways other participants respond to them.

The less active and sometimes entirely silent participants in virtual behavior settings are known as 'lurkers' and are often regarded in a negative light (see Kollock & Smith, 1994). There may be two types of lurkers: those who never communicate to the community, but do to individual members, and those who never communicate at all. Despite their invisibility, virtual community members are highly aware of their presence, often referring to them during group discussions (Weinberg, 2001).

Lurkers are most easily described as on-lookers in Barker and Schoggen's (1978) typology, but could also be considered as members or part of the audience. That lurkers can move to the level of participants by their own volition makes them different from the FtF members and guests.

The ability to lurk in a virtual community may depend on the type of CMC. Participants are generally present in MOOs and chatrooms to engage in real-time communication. However, participants can also wander through rooms and channels listening in on others' public conversations without contributing.2 These silent participants are not entirely invisible to the other participants. Many synchronous communication technologies and programs allow users to display a list of the people in the room. Thus, more active participants can call out salutations to the less active ones or otherwise try to draw these lurkers into the conversation.

Newsgroups and listservs offer the most "invisibility" to the low activity participants. In these venues, participants are never quite sure who their fellow participants are or even how many there are. Lurkers may be more common in these types of groups. This is considered a problem in small groups because responsibility for the viability of the groups is reduced to an even smaller number of participants (Kollock & Smith, 1994). Therefore, survival of the group may depend on a relatively few people conducting a dialogue while others watch or if members only respond to a few topics from the total range of topics.

What leads to a person's willingness to communicate in the setting? Often, it is assumed that personality characteristics play a role in willingness to communicate. However, virtual community participants may experiment with different parts of their personality in their communications with other members (Turkle, 1995). Someone may by quiet in their FtF interactions and quite talkative in the CMC interactions or vice versa.

Another possibility may be the individual's identification with the virtual community. Wicker (1987) argues that in settings that overlap with a person's central life interests, people are more likely to be concerned about and involved in maintaining and enforcing the setting program. Thus, communication leaders may be those for whom the virtual community is important in their lives. However, membership in these virtual communities is primarily voluntary. Therefore, it is possible for the virtual community to fall within the central life interests of the other participants and even lurkers. The difference may be that their lower level of participation meshes with their own expectations about their behavior in the setting. Thus, it is possible that a low contributor who strongly identifies with the group may attempt to keep a program in place when it is threatened, but never feel the need to contribute to a discussion.

Virtual Behavior Setting within FtF Behavior Setting

Many researchers have approached the study of CMC and virtual communities as an isolated phenomenon; few of them consider the context in which CMC occurs and its effect on participants' behavior. Notable exceptions include Baym (1995), who cites the user's immediate situation (e.g., work, school or other organization) as an influence on CMC and on virtual communities and Markus (1994), who observes that users sometimes interrupt FtF communication to attend and respond to CMC. Also, Wellman and Gulia (1999) point out that much analysis treats the interactions through CMC as isolated social events without considering them in relation to other parts of people's social lives.

Participants cannot ever be fully removed from their physical space while engaging in their community in virtual space. When people enter a virtual setting, they are still at home, work, or some other physical place. Although the participants might consider themselves as primarily in the virtual behavior setting, they are still tied to their immediate, physical setting; there should be some impact of FtF behavior settings on virtual behavior settings.

One problem with examining the impact of a person's FtF behavior settings on virtual behavior settings is the lack of information about where people use their computers to access their virtual communities. It is generally assumed that the primary places of use tend to be at home, at work or at school although it is not clear at which location there is the most usage. For this section, the impact of use at work and home will be examined.

A sizable proportion of people access their work-related and non-work related virtual communities while they are working. Baym's (1995) study of a recreational newsgroup devoted to soap operas found that most of the communication took place on Monday through Friday during office hours. As more employees have more access, organizations are very concerned about their email and Internet use (Shaw, 2002).

Several factors are important to consider about how physical behavior settings at work affect virtual behavior settings. First, employees must either have Internet access at work or there must be access to intraorganizational CMC and virtual communities. Second, time pressures, workload levels, and employee autonomy among other work-related issues can affect how much attention and energy the employees can give the virtual community while at work. Third, employees can access both work-related virtual communities and those unrelated to work. If the employee accesses non-work related virtual communities, the employee's self-presentation to colleagues and superiors as a hard worker may cause the employee to access these virtual communities in a clandestine manner. In actuality, it may be relatively simple for an employee to appear to be diligently working while he or she is "elsewhere" communicating with others. If the communities the employee accesses are work-related or professionally enhancing, then the pressure to appear to be working should be much less.

In terms of behavior settings, the employee engaged in CMC is actually straddling the two settings and may be acutely aware of both of them simultaneously. The program obligations and opportunities of each setting affect the other. That is, the program obligations of the FtF behavior setting at work affect the employee's ability to communicate and participate in the virtual behavior setting, and meeting the obligations and opportunities of the virtual behavior setting could limit the amount of time and attention the employee gives to his or her work and FtF colleagues.

Accessing virtual communities from home is likely to be different than accessing them from work. At home, the person may be using his or her personal account (e.g., America On-line) not accessible from work. This person is also likely to be paying for this access, which he or she would not directly pay for at work (Baym, 1995). Again, people can access both work and non-work related virtual communities. However there may be less need for clandestine access of either type of virtual community at home except when pressure from the family or other relationships exists. The physical behavior setting at home may make fewer demands for attention and obligations and people may more easily cross the boundaries between the virtual and the physical behavior settings. However, this scenario assumes that childcare and other forms of family or social interaction do not demand attention from the virtual behavior setting participant.


Implications of the Virtual Behavior Settings Framework

This paper proposes using virtual behavior settings as a way of conceptualizing how virtual communities function. This final section will identify some key implications of using the virtual behavior settings framework and how it can help us develop theory for virtual communities.



Figure 1 shows the proposed virtual behavior settings model. This framework highlights how people and objects affect each other and the setting program. The solid lines indicate what is objective in the setting and can be seen by others. The dotted lines show what is subjectively perceived by each participant (i.e., time and place) and may be individual to the participant. The grouping of the participants ranges from the most central members (i.e., leaders) to the most peripheral (i.e., lurkers). Each member's perceptions of the group and its setting program overlap to various extents with other members. The central members' perceptions are more likely to agree than members in the periphery. The notion of time in this model is that members enact the program which subsequently constrains how the members act. Researchers have assumed that asynchronous and synchronous virtual communities are essentially the same. However, member behavior and community functioning may differ in synchronous and asynchronous virtual communities due to both the demands of the community's setting program, the necessity for member co-presence, and the possibility of lurking.

The most important component of the virtual behavior settings framework is the setting program. It is the set of behaviors in which members engage that keep the virtual community functioning. Within each virtual behavior setting, members are constrained by what they can do with the objects within the setting and the virtual place in which it occurs. For example, participants in MOO settings can have more complex setting programs than members of chatroom settings. Additionally, members are constrained in their behavior by the generally accepted setting program. They can certainly enact subprograms that meet their own needs as long as they do not seriously disrupt the main setting program.

In order for a virtual community to survive, the setting program must be easy to determine and enact. Developers of new virtual communities must create a clear program of activities and ensure that at least some members perform these activities. For established communities, new members arriving must be able to determine what the program is and how they can participate in it. Both of these requirements may be easier for asynchronous virtual communities because asynchronous virtual communities depend less on the co-presence of members. That is, a few active members can enact the program at their convenience, keeping the community viable. Additionally, asynchronous virtual community members can observe behavior for a longer period of time before they are required to act, if they are required to act at all.

The virtual behavior setting framework allows for a variety of member types. Leaders and participants, as described in Table 3, are essential for a virtual community to survive. However, lurkers have an important role, too. They should be considered more positively as potentially active members or even an audience for the active members; they do not need to be regarded negatively as freeloaders. Virtual communities are voluntary groups. There is likely to be turnover in membership. Having a significant base of potentially active members increases the probability of at least some members' moving from lurker to participant as currently active members leave the group. Again, this may be easier for asynchronous communities than synchronous communities because lurking is easier and more acceptable in asynchronous virtual communities.

This framework also implies that FtF behavior settings will have to accommodate people participating in both FtF and virtual settings. Work organizations, in particular, must modify acceptable program behaviors to allow employees to participate in their work-related virtual communities. For example, employees must be able to participate in their virtual settings without being sanctioned in their physical settings. They must also have enough freedom from engaging in the physical setting to appropriately engage in the virtual setting. Telecommuters will need to learn to negotiate the boundaries between their FtF home and virtual work settings. Additionally, researchers of both FtF behavior settings and virtual communities must become better at acknowledging and incorporating the context in which the behavior setting occurs. Virtual community researchers must acknowledge the users' FtF settings and their demands while FtF behavior setting researchers must acknowledge members' simultaneous participation in both.

Finally, the virtual behavior setting framework focuses on the group members' behavior and their personal cognitions of it but not necessarily their feelings about the setting. Thus, this framework could be used to understand or develop both virtual communities and other types of virtual entities. Organizations could conceptualize online help desks and other online employee groups as virtual behavior settings. Additionally, health care workers could use the virtual behavior setting concept to set up virtual counseling or other therapeutic groups to help hard-to-reach or stigmatized populations (e.g., for HIV or teen pregnancy counseling). As long as these settings have stable members who are clear about what setting program to enact and an easy way for new members to learn how to enact the program, they do not need to be virtual "communities" to be active virtual behavior settings.

This paper has proposed a new way of conceptualizing how the place in which virtual communities exist interacts with the membership of the community to help us understand how virtual communities function. The paper has focused primarily on text-based virtual communities. New forms of virtual community which could include video and audio capabilities will have to address these same issues. As a conceptual framework, the virtual behavior settings concept provides a starting point for developing theory to understand all forms of virtual communities.


Footnotes

1. This is particularly interesting when compared to television, which has also been attributed to the separation of physical and social place (Meyrowitz, 1985). Viewers develop shared social experiences by watching the same popular shows. However, watching television is a passive activity that allows one-way communication whereas CMC involves more engaging activities such reading and accessing information and allows for multiple modes of communication.

2. These participants can also engage in private communications such as "whispers" or "instant messages" that only those two people can see.


References

Barker, R. G. (1968). Ecological psychology: Concepts and methods for studying the environment of human behavior. Stanford, CA: Stanford University Press.

Barker, R. G. (Ed.). (1978a). Habitats, environments, and human behavior: Studies in eco-behavioral science from the Midwest Psychological Field Station. Stanford, CA: Stanford University Press.

Barker, R. G. (1978b). Theory of Behavior Settings. In R. G. Barker (Ed.), Habitats, environments, and human behavior: Studies in eco-behavioral science from the Midwest Psychological Field Station (pp. 213-228). Stanford, CA: Stanford University Press.

Barker, R. G., & Schoggen, P. (1978). Measures of habitat and behavior output. In R. G. Barker (Ed.), Habitats, environments, and human behavior: Studies in eco-behavioral science from the Midwest Psychological Field Station (pp. 229-244). Stanford, CA: Stanford University Press.

Baym, N. (1995). The emergence of community in computer mediated communication. In S. G. Jones (Ed.), Cybersociety: Computer mediated communication and community (pp. 35-68). Thousand Oaks, CA: Sage.

Baym, N. (1997). Interpreting soap operas and creating community: Inside an electronic fan culture. In S. Keisler (Ed.), Culture of the Internet (pp. 103-142). Manhaw, NJ: Lawrence Erlbaum Associates.

Beaubien, M. P. (1999). Playing at community: Multi-user dungeons and social interaction in cyberspace. In L. Strate, R. Jacobsen, & S. Gibson (Eds.), Communication and cyberspace: social interaction in electronic environments (pp. 179-188). Bridgehampton, NY: Hampton Press.

Culnan, M. J., & Markus, M. L. (1987). Information technologies. In F. M. Jablin, L. L. Putnam, K. H. Roberts, & L. W. Porter (Eds.), Handbook of organizational communication: An interdisciplinary perspective (pp. 420-443). Newbury Park, CA: Sage.

Douglas, K. M., & McGarty, C. (2002). Internet identifiability and beyond: A model of the effects of identifiability on communicative behavior. Group Dynamics: Theory, Research, and Practice, 6, 17-26.

Finholt, T., & Sproull, L. S. (1990). Electronic groups at work. Organization Science, 1, 41-64.

Garton, L. & Wellman, B. (1996). Social impacts of electronic mail in organizations: A review of the research literature. Communication Yearbook, 18, 434-453.

Hancock, J. T., & Dunham, P. J. (2001). Impression formation in computer-mediated communication revisited. Communication Research, 28, 325-347.

Harasim, L. M. (1993). Networlds: Networks as social space. In L. M. Harasim (Ed.) Global networks: Computers and international communication (pp. 15-34). Cambridge, MA: The MIT Press.

Hiltz, S. R., & Turoff, M. (1993). The network nation: Human communication via computer (2nd Edition). Cambridge, MA: The MIT Press.

Hollingshead, A. B., & McGrath, J. E. (1995). Computer-assisted groups: A critical review of empirical research. In R. Guzzo & E. Salas (Eds.), Team effectiveness and decision making in organizations (pp. 46-78). San Francisco, CA: Jossey-Bass Publications.

Jones, S. G. (1995). Understanding community in the information age. In S. G. Jones (Ed.), Cybersociety: Computer mediated communication and community (pp. 1-34). Thousand Oaks, CA: Sage.

Jones, Q. (1997). Virtual communities, virtual settlements, and cyber-archaeology: A theoretical outline. Journal of Computer-Mediated Communication, 3. Retrieved March 23, 2003 from http://jcmc.indiana.edu/vol3/issue3/jones.html.

Kollock, P. & Smith, M. (1994). Managing the virtual commons: Cooperation and conflict in computer communities. Retrieved January 20, 2004 from http://www.sscnet.ucla.edu/soc/csoc/vcommons.htm.

Liu, G. Z. (1999). Virtual community presence in Internet relay chatting. Journal of Computer-Mediated Communication, 5. Retrieved March 23, 2003 from http://jcmc.indiana.edu/vol5/issue1/liu.html.

Mann, P. A. (1978). Community psychology: Concepts and applications. New York: Free Press.

McMillian, D. W. & Chavis, D. M. (1986). Sense of community: A definition and theory. Journal of Community Psychology, 14, 6-23.

Markus, M. L. (1994). Finding a happy medium: Explaining the negative effects of electronic communication on social life at work. ACM Transactions of Information Systems, 12, 119-149.

McLaughlin, M. L., Osborne, K. K., & Smith, C. B. (1995). Standards of conduct on Usenet. In S. G. Jones (Ed.), Cybersociety: Computer mediated communication and community (pp. 90-111). Thousand Oaks, CA: Sage.

Meyrowitz, J. (1985). No sense of place: The impact of electronic media on social behavior. New York: Oxford University Press.

Myers, D. (1987). "Anonymity is part of the magic": Individual manipulation of computer mediated contexts. Qualitative Sociology, 10, 251-266.

Pargman, D. (2000). Code begets community: On social and technical aspects of managing a virtual community. PhD dissertation, Linkoping Universitet, Sweden. Retrieved January 20, 2004 from http://esplanaden.lysator.liu.se/svmud/pargman/.

Parks, M., & Floyd, K. (1996). Making friends in cyberspace. Journal of Computer Mediated Communication, 1. Retrieved March 23, 2003 from http://cwis.usc.edu/dept/annenberg/vol1/issue4/parks.html.

Pickering, J. M., & King, J. L. (1995). Hardwiring weak ties: Interorganizational computer-mediated communication, occupational communities, and organizational change. Organization Science, 6, 479-504.

Postmes, T., Spears, R., & Lea, M. (2000). The formation of group norms in computer-mediated communication. Human Communication Research, 26, 341-371.

Ramirez, A., Walther, J. B., Burgoon, J. K., & Sunnafrank, M. (2002). Information-seeking strategies, uncertainty, and computer-mediated communication: Toward a conceptual model. Human Computer Research, 28, 213-228.

Rheingold, H. (1993a). A slice of life in my virtual community. In L. M. Harasim (Ed.), Global networks: Computers and international communication (pp. 57-80). Cambridge, MA: The MIT Press.

Rheingold, H. (1993b). The virtual community: Homesteading on the electronic frontier. Reading, MA: Addison-Wesley.

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 Management, 27, 297-312.

Sassenberg, K. (2002). Common bond and common identity groups on the Internet: Attachment and normative behavior in on-topic and off-topic chats. Group Dynamics: Theory, Research, and Practice, 6, 27-37.

Schoggen, P. (1989). Behavior settings: A revision and extension of Roger G. Barker's ecological psychology. Stanford, CA: Stanford University Press.

Schuler, D. (1996). New community networks: Wired for change. New York: ACM Press.

Selwyn, N. (2000). Creating a "connected" community? Teachers' use of an electronic discussion group. Teachers College Record, 102, 750-778.

Shaw, R. (2002, Oct. 22). Net use at work spikes-and so do worries. USA Today. Retrieved March 23, 2003 from http://www.usatoday.com/tech/webguide/internetlife/2002-10-21-net-work_x.htm.

Smith, C. B., McLaughlin, M. L., & Osborne, K. K. (1998). From terminal ineptitude to virtual sociopathy: How conduct is regulated on Usenet. In F. Sudweeks, M. McLaughlin, & S. Rafaeli (Eds.), Network and netplay: Virtual groups on the Internet (pp. 95-112). Menlo Park, CA: AAAI Press.

Spears, R., & Lea, M. (1992). Social influence and the influence of the 'social' in computer-mediated communication. In M. Lea (Ed.), Contexts of computer-mediated communication (pp. 36-64). London: Harvester Wheatsheaf.

Sproull, L., & Keisler, S. (1986). Reducing social context cues: Electronic mail in organizational communication. Management Science, 17, 1492-1512.

Sproull, L. & Keisler, S. (1991). Increasing personal connections. In L. Sproull & S. Keisler (Eds.), Connections: New ways of working in networked organizations (pp. 417-429). Cambridge, MA: The MIT Press.

Turkle, S. (1995). Life on screen: Identity in the age of the internet. New York: Simon & Schuster.

Turner, J. W., Gruber, J. A., Meyers, J. (2001). Developing an optimal match within online communities: An exploration of CMC support communities and traditional support. Journal of communication, June, 231-251.

Utz, S. (2000). Social information processing in MUDs: The development of friendships in virtual worlds. The Journal of On-line Behavior, 1. Retrieved March 23, 2003 from http://www.behavior.net/JOB/v1n1/utz.html.

Walther, J. B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19, 52-90.

Walther, J. B., Slovacek, C. L., & Tidwell, L. C. (2001). Is a picture worth a thousand words? Photographic images in long-term and short-term computer-mediated communication. Communication Research, 28, 105-134.

Weick, K. E., (1979). The social psychology of organizing (2nd Edition). New York: Random House.

Weick, K. E. (1990). Technology as equivoque: Sensemaking in new technologies. In P. S. Goodman & L. S. Sproull (Eds.), Technology and organizations (pp. 1-44). San Francisco: Jossey-Bass.

Weinberg, H. (2001). Group process and group phenomena on the Internet. International Journal of Group Psychotherapy, 51, 361-378.

Wellman, B. & Gulia, M. (1999). Net surfers don't ride alone: virtual communities as communities, In P. Kollock & M. Smith (Eds.), Communities in cyberspace (pp. 331-367). Berkley, CA: University of California Press.

Wicker, A. W., (1979). An introduction to ecological psychology. Monterey, CA: Brooks Cole Publishing.

Wicker, A. W. (1987). Behavior settings reconsidered: Temporal stages, resources, internal dynamics, context. In D. Stokols & I. Altman (Eds.), Handbook of environmental psychology (pp. 613-653). New York: John Wiley & Sons.

Wicker, A. W. (1992). Making sense of environments. In W. B. Walsh, K. H. Clark, & R. H. Price (Eds.), Person-environment psychology: Models and perspectives (pp. 158-191). Hillsdale, NJ: Lawrence Erlbaum Associates.


About the Author

Anita Blanchard, Ph.D. is Assistant Professor of organizational psychology at the University of North Carolina, Charlotte. Her current area of research interest is how virtual communities maintain themselves and are experienced as "communities."
Address: Dept of Psychology, UNC Charlotte, 9201 University City Blvd., Charlotte, NC 28223. Tel (704) 687-4847, Fax (704) 687-3096.

©Copyright 2004 Journal of Computer-Mediated Communication