<b>"Effects of computer-mediated communication on group and individual behavior."<p> </b>

Computer mediated communication as a form of communication has increased dramatically over the last decade. The Internet alone as of January 1993 connects over 1,300,000 hosts in nearly all major countries. Private local and wide area networks formed by the corporate world collectively are estimated to be at least as large as the Internet itself. By the year 2000 these "internetworks will link several hundred million computers together, and the total number of users with access to the global electronic information matrix will exceed 500 million (1)". The economic, political and social ramifications of this level of connectivity are themselves a dynamic that can only be studied as they evolve. The means of this change are not only an improvement in the efficiency of information exchange as a result of technical innovations but also the changes in individual, group and organizational behavior as a result of utilization of these technologies (2).

The application of email as a information channel has contributed heavily to the creation of the current electronic society. Email as a form of human communication is used to transmit messages that are formal, informal, personal, intimate and anonymous but is limited to the capabilities of the utilized computer software. These limitations caused by the inability to use anything other than basic text effect both the user and the message (3). This paper is a discussion of the impact of the use of email at both an individual and group level.

In order to understand the changes occurring through email it is helpful to first define communication. Figure A provides a typical illustration of the communication model (4).

Figure A

A Model of Communication Process

Sender Channel Receiver

____________________ ____________ ______________________

| _________ _________ | | ___________ | | __________ __________ |

--[[cedilla]]|| MEANING | | ENCODE | |--[[cedilla]]| | MESSAGE | |--[[cedilla]] | | DECODE | | MEANING | |--

| | | | | | | |

| __________ |

| | NOISE | |

| |

| |

----------------------------------------FEEDBACK---------------------------------------

A message is created and encoded into a particular medium by the sender, transmitted though a channel and/or medium and then decoded (and hopefully understood) by the receiver who responds with feedback. This process is summarized by Porter and Roberts as "an attempt to share meaning via transmission of messages from senders to receivers." (5) Other factors that influence the communication process are social context and conventions, the relationship between the sender and receiver, the topic of communication, and the personal and physical characteristics of the participants (6). It important to realize that channels of communication consist of more than verbal, written and electronic mediums but also include nonverbal behaviors such as use of language, timing, tone of voice, gestures, facial expressions.

Email by its nature eliminates many of the interpersonal aspects of the various forms of spoken and written communication. Though similar to a letter in form, email is a very different channel of communication since none of the physical and social qualities exist such as paper quality, letterhead, handwriting, choice of font, format and position or status of the sender. While individuals who use email often create headers, trailers and crude graphics, the personalizations do not present the same information as a touchable physical item that is read in its original form as created by the sender. Email is not a physical item but an electronic transmission that may have its type font and format altered by the sending and receiving machines while the content is easily edited, forwarded and erased by the receiver (7). Most of the contextual cues that are a part of written, telephone and face-to-face communication are not available to the receiver, creating a new channel of communication whose rules, etiq

The use of email has a number of implications for the user. It provides more access to a wide variety of individuals and groups, irrespective to local geography. It can provide a medium for conversations irrespective of time and place, access to discussions of a particular topic, collaborative research, and expansion of social and work related relationships. A frequent difficulty faced by users is learning to effectively filter the flood of incoming information. Additionally there is the problem of associated with learning to use a particular computer system and frustrations caused by little direct feedback. Another consequence perhaps positive or negative is the generation of increased perceived need for information (8).

Personal experience suggests that information-seeking behavior is altered by the use of email. Responding to an individual's query for information on people's behavior on the Internet, I was able to establish communication with a doctoral student at the University of Massachusetts who is studying the effects of computer-mediated communication on groups and organizations. He forwarded me his current bibliography that became the basis of research for this paper. (He will receive a copy of this paper to add to his collection of information.) I realized that the my usual methods of locating information were augmented and streamlined by a few timely connections on the Internet.

Finholt and Sproull have organized the variables that influence group behavior in a conventional, i.e., face-to-face setting, under the three classes of group attributes, group processes and organizational consequences. They also hypothesized how these variables are effected by communication in electronic groups. Figure B illustrates the structure they composed which provides a useful framework for discussion of the effects of email on group behavior (9).

Figure B

Comparing Conventional Group Behavior with That of Electronic Groups.

Conventional Group Behavior in Organizations

Group Attributes [[cedilla]] Group Processes [[cedilla]] Org. Consequences

(1) Physical settings (1) Interaction: Synchronous (1) Participation

(2) Member characteristics (2) Influence: Multiple social cues (2) Performance

(3) Membership criteria (3) Identity maintenance: (3) Learning

(4) Task type Multiple physical & social cues

Hypothesized Electronic Group Behavior in Organizations

Group Attributes [[cedilla]] Group Processes [[cedilla]] Org. Consequences

(1) No shared physical setting (1) Interaction: Asynchronous (1) Participation:

Larger groups,

dispersed groups,

re. equality across

members

(2) "Invisible" members (2) Influence attempts: (2) Performance:

Few social cues Reduced process

loss, increased

information access

(3) Membership criteria (3) Identity maintenance: (3) Learning:

Through text primarily New information

buffers, rapid

feedback, rapid

diffusion

(4) Topic

______________________________________________________________________

The first category "Group Attributes" raised questions about the traditional definition of a group with respect to physical setting, membership and the purpose of the group. Email obviously lacks of a definite physical setting. Meetings and discussion are held with all the participants sending and receiving messages over a wide range of time. Chair and table arrangement, room aesthetics, style of furniture and the number of individuals present at the meeting are physical components of a meeting location that also influence behavior and interaction (10). These elements are absent from the virtual context of email that often creates a much more informal setting for participants.

Membership characteristics effect group interaction through the context created by individual members' physical appearance, social status, gender, task expertise and hierarchical position. These characteristics also can influence how the group perceives itself, how it is perceived by others and how an individual's input is to be valued. Most users of lists are "invisible" or "lurkers" who digest and probably use information at their discretion but who do not contribute to the group dynamic though active participation or personal presence as in a physical meeting. In an electronic group it is often impossible to determine the exact roster of members (though this may be available to group moderators) and even if participants personally know each other, they are still less aware of personal characteristics (11).

Membership criterion for any group determines the level of formality for interaction. More restrictive requirements lend themselves to more formal groups will voluntary membership lends itself to more informal groups (12). Within the Internet obtaining membership to a discussion group may be no more complex than having an email account and applying to a central host, while application to an individual may be necessary for inclusion in others. Although there is no little information available on this topic it seems likely that the more restrictive discussion groups will communicate more formally than those in "open" groups. One benefit of electronic groups, even those more strictly controlled, is that exclusion of individuals based upon factors such as age, sex, ethic origin, sexual orientation, physical handicaps, etc., is more difficult (13). (The SYSTERS discussion group on the Internet is an example of a group the specifically limits its membership to women. Questions about maintaining such restri

Finholt and Sproull's final group attribute of purpose shows that conventional groups are organize around a task while electronic groups form to address a topic. Personal experience suggests that this is true generally but there are exceptions. For example, the Operation SOS reflector list was created to facilitate communication for coordinating group activities and keep individuals informed of events.

Continuing with Finholt and Sproull's group behavior structure we move the to variables associated with the "Group Process," specifically interaction and influence. This area currently is under much discussion by researchers, observers and participants alike, since many of the topics in this area lead to the most volatile results of electronic communication.

Email communication provides a great contrast to conventional physical meetings where one or a few individuals usually dominate a meeting due to hierarchical status, aggressive speakers and conversation norms necessary since only once person can speak at a time (14). Asynchronous communication dramatically changes this group dynamic by allowing individuals to send complete thoughts simultaneously without fear of interruption and allowing more diversity in opinions and information available to the group (15).

Influence exerted by participants in a conventional meeting is done in a variety of ways including hierarchical status, seating position, physical or verbal charisma and personality, speaking style, gestures, manner and appearance. All these factors contribute to the contextual setting of a meeting and give the sender and received clues about appropriateness of content, message channel and timing of delivery. In the context of email most, if not all of these factors, are eliminated. Individuals who are intimidated in physical meetings are more likely to participate in an email group discussion. This is an important effect since higher status or more vocal individuals tend to dominate face-to-face meetings even if they are not an expert on the topic of discussion (16).

The absence of social cues in email communication appears to cause deindividuation of individuals -- feelings of anonymity and reduction in self-awareness and self-regulation (17). Deindividuation causes individuals to become self-centered and be less concerned with how they are presenting themselves via the content and style of messages. Email users often ignore or forget that the real audience is composed of individual people and not a computer. The effects of these phenomenon lead to "more extreme, more impulsive and less socially differentiated" behavior (18) such as verbal abuse and criticism, i.e., "flaming". Socialization and regulation of "email etiquette" is difficult there are no widely distributed guidelines. Most groups rely on the discretion and self-moderating comments of its participants. Two other examples of uninhibited behavior are the willingness to communicate bad news or negative information and the flouting of social conventions, such as sending personal messages while at w

At this point it is instructive to deviate from the structure provided by Finholt and Sproull to discuss the primary communication roles, gatekeeper, liaison and participant/isolate, identified in organization communication research. Any individual in an organization will fill one or more of these roles in an information network. A person's place in the network can determine attitude and behavior with respect to job satisfaction and relationships to peers, subordinates and superiors (20).

Gatekeepers control the flow of information to either subordinates or superiors and within peer groups. Within the realm of email the closest role to this is a discussion list moderator who may view all messages before posting to the entire group. Since anyone on an email discussion list can post any message the role of gatekeeper becomes much more distributed. In the case of organizations, an increased flow of direct information from lower level to higher level employees, due to the equalization effect of email, has the potential to impact how these employee relations and organization attitudes (21).

The liaison role is related to gatekeeper but the emphasis is keeping various groups informed of each others' activities. In conventional settings, individuals are generally assigned such roles in various forms such as having persons from one committee sit on another committee. Interoffice memos and telephone conversations are specific channels than can be used to fulfill this role. In the email environment the liaison role, like that of the gatekeeper role, is diffused to an unlimited number of people.. Participants in multiple discussion groups will often forward messages to other lists or the originator of a message will be the liaison by sending the same message to multiple lists. Again the potential of email to effect the functioning of organizational communication networks is tremendous, but still relatively unknown.

Participants/isolates are essentially those who are inside or outside any kind of communication loop. Those in an information network in a general setting are belong to various groups and will receive and pass on information. Isolates are those who do not participate either voluntarily or involuntarily in established information networks. Comparing this to computer mediated communication provides some interesting contrasts. Isolates are generally those who do not have access to email or the knowledge of how to use it. This situation can be intentional or beyond an individual's control, such as no access to a modem. Participants are those who do have access to email, but as discussed earlier it is difficult to identify all or most participants. Large numbers of individuals receive access to a large amount of information and make use of it without participating in the information loop. It is interesting that while individuals have more access to information and other email users, there is also th

Another role that effects group behavior is that of the leader. Certainly in conventional meetings, individuals will interact within societal norms and follow a structure established by the recognized leader or facilitator. However in electronic discussion groups there is usually no identified leader(s). While there are individuals who "talk" more frequently, it is difficult to continuously monopolize a medium where communication does not occur conversationally. The lack of social context cues a more distributed level of participation that is not as dependent on higher/lower status paradigms of face-to-face situations (22).

Consequences of changing group attributes and processes discussed earlier and the shifts in communication roles are a differentiation of behaviors. Returning to the structure from Figure B, these "Organizational Consequences" are classified under the areas of participation, performance and learning.

The nature of participation electronic groups is more evenly distributed among members in group primarily due to the equalization of status because of the lack of social context cues and the uninhibited behavior of deindividuation. Higher levels of participation in electronic groups give rise to larger memberships than in a more conventional setting. This increase in numbers leads to more connections between individuals and leads to the creations of more groups as the topics of discussion expand (23). A potential negative for organizations with email is that while it improves information access and interaction of group members, it also provides a vehicle to communicate and magnify discontent (24).

One question is how these various effects change the actual decision making process or performance of groups. Porter and Roberts (25) found that face-to-face communication is more effective to create a change in opinion, while comprehension of information is higher in written or textual forms of exchange. While there is no data to support this, it would seem reasonable to assume that this would also be true of email since it is very effective at transmitting text, but is devoid of almost all the nonverbal cues that help lead to influence opinions. During face-to-face meetings is a group will come to consensus usually on the first decision proposal suggested and use predominately non-confrontational behavior In electronic groups it appears that lower status individuals are just as likely to make the first decision proposal as higher status persons. Additionally, it appears that electronic groups use more time to come to consensus since there is no clear leader and more opinions and concerns are

groups using computer-mediated communication to reach consensus, as compared with groups using face-to-face discussions, experience some inefficiencies in communicating, participate more equally within the e group, are more uninhibited, and reach decisions which deviate further from initial individual preferences. These findings are consistent with our ideas about the potential effects of a communication medium which increases focus of attention on messages, which reduces the ability to communicate social and personal feedback, and which depersonalizes the situation (27).

The potential for learning in electronic groups is increased by the speed and efficiency of requests and feedback. A query to a discussion group reaches a large number of individuals. When responses are also sent to the group the diffusion of information is expanded. Liaisons between groups amplify this activity by posting information to other groups. Certainly this type of activity can lead to a high volume of messages and information overload if not moderated properly, but it does provide a large arena for the discussion of questions, ideas and opinions.

The lines of communication within an organization, top-down, bottom-up and among peer groups, are altered by the more direct communication of email the behavioral changes described previously. This is expected to decentralize organizational communication, through more access to everyone, which will lead to more democratic processes, better decision making and an increase in productivity. This hypothesis is in need of further field research, but current work suggests that the effects of computer-mediated communication on organizational power structures are a complex problem since the same people react and behave differently when using different communication channels (28).

The sociological and organizational impacts of email are expanding with the growth of computer networks in all over the world. While the potential for increased information access and communication is apparent, the effects of this shift in the group behaviors in the electronic mode of communication are a new and growing field of study with respect to scholarly research (29).

1. Thomas F. Mandel. "Surfing the Wild Internet." Scan No. 2109, SRI International: Business Intelligence Program, 1993.

2. Tom Finholt and Lee Sproull. "Electronic Groups at Work." Organization Science, 1(1):41-64, 1990.

3. Howard Rosenbaum, and Gregory B. Newby. "An Emerging Form of Human Communication: Computer Networking." Proceedings of the 53rd ASIS Annual Meeting, 27:300-325 1990.

4. Charles A. O'Reilly, III and Louis R. Pondy. "Organizational Communication." in Organizational Behavior, ed. by Steven Kerr (Columbus, OH: Grid Publishing, 1979), pp. 119-150.

5. Lyman W. Porter and Karlene H. Roberts. "Communication in Organizations," in Handbook of Industrial and Organizational Psychology, ed. by Marvin D. Dunnette (Chicago: Rand McNally, 1976), pp.1553-1589.

6. Lee Sproull and Sara Kiesler. "Reducing Social Context Cues: Electronic mail in Organizational Communication." Management Science, 32(11):1492-1512, 1986.

7. Sara Kiesler. "The Hidden Messages in Computer Networks." Harvard Business Review, 64:46-48+, 1986.

8. Elaine B. Kerr and Starr Roxanne Hiltz. Computer-Mediated Communication Systems: Status and Evaluation, Human Communication Research Series, ed. by Peter R. Monge (New York: Academic Press, 1982), pp. 97-111.

9. Finholt and Sproull, p. 43,

10. Porter and Roberts, p. 1565.

11. Finholt and Sproull, p. 43.

12. Finholt and Sproull, p. 43-44.

13. Kerr and Hiltz, p. 115-116.

14. Jane Siegel, "Group Processes in Computer-Mediated Communication." Organizational Behavior and Human Decision Processes, 37:157-187, 1986.

15. Sara Kiesler and Lee Sproull. "Group Decision Making and Communication Technology." Organizational Behavior and Human Decision Processes, 52:96-123, 1992.

16. Jane Siegel, et al. "Group Processes in Computer-Mediated Communication." Organizational Behavior and Human Decision Processes, 37:157-187, 1986.

17. Sara Kiesler, et al. "Social Psychological Aspects of Computer-Mediated Communication." American Psychologist, 39(10):1123-1134, 1984.

18. Kiesler and Sproull, p. 103.

19. Sproull and Kiesler, pp. 1507-1508.

20. Peter R. Monge, et al. "The Determinants of Communication and Communication Structure in Large Organizations: A Review of Research." Organizational Science: Communication Yearbook 2, pp. 312-331.

O'Reilly and Pondy, pp. 119-150.

21. Finholt and Sproull, p. 60.

22. Kiesler, et al, p. 1129.

23. Kiesler and Sproull, p. 116-117.

24. Finholt and Sproull, p. 60.

25. Porter and Roberts, p. 1563.

26. Kiesler and Sproull, pp. 109-113.

27. Siegel, et al, p. 175.

28. Paul Attewell and James Rule. "Computing and Organizations: What We Know and What We Don't Know." Communications of the ACM, 27(12):1184-1191, 1984.

Siegel, et al, p. 184.

29. Prior to completing to this paper I received an announcement of an electronic research group that has been studying computer-mediated commication for the last eigth months. The message was in invitation for individuals to join the group and take part in the data collection and analysis phases. Individuals to contact were: Sheizaf Rafaeli, Hebrew University of Jerusalem, sheizafr@shum.cc.huji.ac.il; Fay Sudweeks, University of Sydney, fays@archsci.arch.su.edu.au

1

Bibliography

Attewell, Paul and Rule, James . "Computing and Organizations: What We Know and What We Don't Know." Communications of the ACM, 27(12):1184-1191, 1984.

Finholt, Tom and Lee Sproull. "Electronic Groups at Work." Organization Science, 1(1):41-64, 1990.

Kerr, Elaine B. and Starr Roxanne Hiltz. Computer-Mediated Communication Systems: Status and Evaluation, Human Communication Research Series, ed. by Peter R. Monge, New York: Academic Press, 1982.

Kiesler, Sara, et al. "Social Psychological Aspects of Computer-Mediated Communication." American Psychologist, 39(10):1123-1134, 1984.

Kiesler, Sara and Lee Sproull. "Group Decision Making and Communication Technology." Organizational Behavior and Human Decision Processes, 52:96-123, 1992.

Kiesler, Sara. "The Hidden Messages in Computer Networks." Harvard Business Review, 64:46-48+, 1986.

Mandel, Thomas F. "Surfing the Wild Internet." Scan No. 2109, SRI International: Business Intelligence Program, 1993. (Contact author at mandel@netcom.com for futher copies.)

Monge, Peter R., et al. "The Determinants of Communication and Communication Structure in Large Organizations: A Reivew of Research." Ogranizational Science: Communication Yearbook 2, pp. 312-331.

O'Reilly, III, Charles A. and Louis R. Pondy. "Organizational Communication." in Organizational Behavior, ed. by Steven Kerr (Columbus, OH: Grid Publishing, 1979), pp. 119-150.

Porter, Lyman W. and Karlene H. Roberts. "Communication in Organizations," in Handbook of Industrial and Organizational Psychology, ed. by Marvin D. Dunnette (Chicago: Rand McNally, 1976), pp.1553-1589.

Rosenbaum, Howard and Gregory B. Newby. "An Emerging Form of Human Communication: Computer Networking." Proceedings of the 53rd ASIS Annual Meeting, 27:300-325, 1990.

Siegel, Jane, et al. "Group Processes in Computer-Mediated Communication." Organizational Behavior and Human Decision Processes, 37:157-187, 1986.

Sproull, Lee and Sara Kiesler. "Reducing Social Context Cues: Electronic mail in Organizational Communication." Management Science, 32(11):1492-1512, 1986.

Slack, Jennifer Daryl. "Surveying the Impacts of Communication Technologies," in Progress in Communication Science: Volume V, ed. by Brenda Dervin and Melvin J. Voigt, (Norwood, New Jersey: Ablex, 1984), pp. 73-109.