A matter of trust: using student designed codes of conduct in face-to-face and virtual group environments.

ABSTRACT

This study looked at 42 face-to-face and 4 virtual groups to determine if student-generated codes of conduct would improve group communication. Codes of conduct are one way to improve the communication process for face-to-face groups. The codes help establish common expectations, shared values, and group procedures, all of which contribute to group trust. However, improved communication does not necessarily improve group outcomes. In addition, it is unclear as to whether virtual groups can even develop a code of conduct; much less use it to improve group processes.

Keywords: Group processes, codes of conduct, group communication, group trust, virtual groups, face-to-face group work

1. INTRODUCTION

With the advent of the internet and changes in corporate structure, more and more companies are relying on functional teams, both face-to-face and virtual, in the workplace. This trend is reflected in the classroom, as many business schools now incorporate group and team projects into the curriculum. Yet, for all its benefits, group work can also present some challenges, both in the workplace and the classroom, that may result in the overall group experience not being as positive and rewarding as intended. There may be personality conflicts between group members, which interfere with group communication and performance. There may be differences in the goals that individual team members' hold, especially as the workplace becomes more diverse. Further, there may be significant differences in ability and motivation. Members need to feel comfortable communicating their ideas and opinions and moving towards identification and completion of common goals. Small group research suggests that there must be consensus on task allocation and a sense of confidence that each member will do their fair share in order for teams to be effective (Barker, Abrams, Tiyaamornwong, Seibold, Duggan, Park, & Sebastian, 2000). One of the goals of group work in the business classroom should be to teach students how to handle these conflicts so they will be prepared for future workplace teams.

We see an opportunity for codes of conduct to be used to improve the work processes and communication within student work groups. Thus, this study combines previous research in the area of group work communication and codes of conduct. Specifically, it seeks to assess the usefulness of group-designed codes of conduct for enhancing students' communication and group processes within a team. We will be evaluating codes of conducts generated by the groups themselves to determine if these self-designed rules improve the communication process needed for productive groups. In addition, we will compare the effectiveness of codes of conduct for face-to-face groups and those in a virtual environment to determine if they are equally effective.

2. BACKGROUND

Specific social skills required in an effectively functioning group, be it virtual, face-to-face or a combination of the two, include "the ability to control progress through the tasks, the skills to manage competition and conflict, the ability to modify and use different viewpoints as well as the willingness to give mutual support (Cohen, 1994, p.5)." Thus for group learning to be successful, a number of communicative conditions associated with the project, the group, and the individual must be met. The two common issues with unsuccessful teams, both virtual and face-to-face, is lack of trust and poor communication (Jarvenpaa & Leidner, 1998; Van derVeen, Collis, & Jones, 2003). Trust can be divided into cognitive trust and social trust. Cognitive trust is developed through the exchange of reliable information, the communication of processes, and the development of shared knowledge through group processes. Social trust is developed through the exchange of personal information, shared group vision, and creation of member responsibilities and accountability (Kanawattanachai & Yoo, 2002).

Despite the strong evidence that there is little difference in group success between face-to-face and strictly virtual groups, some researchers did find evidence that virtual groups may have more difficulty in creating a cohesive group (Jarvenpaa & Leidner, 1998; Kayworth & Leidner, 2001/2002; Van der Veen, Collis, & Jones, 2003). Virtual groups are unique from traditional face-to-face groups in that group members are asked to work with others that they may never or will never meet, except in virtual space. In a traditional classroom, a learning community is created (for the most part) at each class session. There is obvious instructor presence, non-verbal social cues, and group roles that establish a student's identity (albeit, accurately or inaccurately) (Hoag, Jayakar, & Erickson, 2003; McConnell, 1997). Therefore, virtual groups may focus more on the development of group trust, using communication to exchange information and ensuring the reliability of that information. Face-to-face groups, on the other hand may focus more on creating social trust through group small talk and developing member relationships (Kanawattanachai & Yoo, 2002).

Research demonstrates that a sense of community and connectedness can be developed in exclusively virtual environments. However, research has also suggested that more time is needed to develop virtual communities (Henri, 1995; Hoag, Jayakar, & Erickson, 2003; McConnell, 1997; Monteith & Smith, 2001; Oren, Miodusser, & Nachmias, 2002; Rovai, 2002). The sense of community depends on the amount of trust between group members, individual identity, community support and group facilitation, and active virtual participation, both in quantity and quality. McMahen and Dawson (1995) also identify factors such as common interests, specific tasks to be completed (as opposed to well-defined general goals or outcomes), a sense of obligation to the community or team, and reliable technology as important to creating a sense of virtual group identity. In addition, technological problems can affect trust and communication, thus affecting group success (Kanawattanachai & Yoo, 2002; Yonkers, Segneri, & Ramos, 2003).

Group composition alone does not appear to be a predictor of communication competence, with one exception. When two sub-groups are collaborating on a virtual activity, members tend to have allegiance to their local group, where there is face-to-face interaction (Van der Veen, Collis, & Jones, 2003; Yonkers, Segneri, & Ramos, 2003). This is also true with groups that are composed of equal numbers of two recognizable groups. This can be explained with the "outsider within" phenomenon described by diversity and cross-cultural researchers (Ayoko, Hartel, & Callan, 2002; Jehn, Northcraft, & Neale, 1999; McConnell, 1997). When there is a clear majority and minority with similar values within a working group, the minority tends to accommodate their communication patterns and compromise their values for that of the majority (Barker, Abrams, Tiyaamornwong, Seibold, Duggan, Park, & Sebastian, 2000). A truly heterogeneous group with a variety of values may take longer to develop trust, but can function well. In fact, intercultural workgroup research indicates that after a period of time, heterogeneous groups function as well, and in some cases better, than homogenous groups (Ayoko, Hartel, & Callan, 2002; Barker, et al, 2000; Jarvenpaa & Leidner, 1998; Watson, Kumar, & Michaelson, 1993) . However, when there is no clear majority between the two groups, communication can breakdown. Traditionally underrepresented group members may refuse to compromise their values when they have others to support them. As a result, the team becomes divided, with lack of trust between members that have different values, and greater trust between members with the same values (Jehn, Northcraft, & Neale, 1999).

The work of Fisscher, Nijhof, & Steensma (2003) contends that responsible behavior is based on social interaction and expectations rather than normative philosophy. Farrell, Cobbin & Farrell's (2002) findings indicate that a shared ethical vision within a workgroup resulted in the enforcement of the codes of conduct. When employees or upper management did not share the same vision of ethics, codes of conduct were nothing more than something to get around. Jehn, Northcraft, & Neale, (1999) found that groups with diverse values were less productive than information diverse groups (different skills, knowledge, experience, access to information) and demographically diverse groups (race, language, gender, age). It is important, therefore, that codes of conduct are based on shared values. These shared values, social interaction, and group expectations would then be the basis to develop group trust.

Jarvenpaa and Leidner (1998) identified a number of communicative factors that improved or destroyed trust in virtual environments. Groups with a high level of initial group trust had a high level of social communication, the ability to cope with technical and task uncertainty, and individual intuition. However, in order for groups to sustain trust, there needed to be predictable communication, substantial responses, positive leadership (not appointed), flexibility, and, most importantly transition from social to procedural to task focus. Groups that focused too much on the procedures or social interaction, created unreasonable group exceptions, and lacked follow-through lacked trust and were less successful in completing their assignment. These same factors can be applied to face-to-face groups and/or teams (Parker, Wall, & Cordery, 2001).

Our earlier research (Buff & Yonkers, 2004) suggested that group designed codes of conduct effectively reinforced group processes. Anecdotally, they appeared to be useful for improving group interaction and performance. In addition, we identified the following categories that were included in all student generated codes of conduct: Ideals (aspirations): social expectations, communication, meetings, values, goals; Rules: meetings decision-making, notification, work; Recourse: consequences. While the content differed from group to group, each team included these factors into their codes of conduct. These factors could be correlated with those that were identified by Jarvenpaa and Leidner (1998), with Ideals related to creating initial trust, Rules related to sustaining trust, and Recourse being a way to deal with behavior that results in lack of trust.

3. THE STUDY

The current research was conducted over a two-semester time frame. The research was conducted with two upper-level consumer behavior classes, one upper-level retailing class, four lower-level (generally freshmen) business communications classes, and four upper-level international marketing classes, three domestic and one abroad. In each of these classes, there was a major group project assigned. In most classes, these projects included a written report and oral presentation. There were differences in how students were assigned to groups. In the retailing class, students were able to self select their teammates. In the other classes, students were assigned at random to a group for completion of a major course project. Still in others, students were assigned to groups based on interest, schedule, or experience. One of the first tasks that each group performed was designing a group code of ethics. This was generally completed during one of the groups' first meetings. A short discussion of ethics and group work usually preceded the development of the group codes of conduct.

The goals of student-generated codes of conduct were twofold. They included:

(1) To improve group interaction and functioning by allowing students to define group expectations and sanctions themselves,

(2) To make students responsible for their own group processes and outcomes.

Student groups designed a code of conduct that group members would abide by. Members signed the code to verify their acceptance of the expectations and sanctions. In some cases, instructors reviewed the codes and required groups to clarify with more details. However, instructors did not have any influence on the content, resulting in the codes being totally student generated. As this was completed fairly early in the semester, groups had no significant experience of working together before completing the code of conduct.

We had two hypotheses:

1. Group communication, discourse management, and processes would be augmented by the codes of conduct;

2. The virtual group communication would improve using codes of conduct and be at the same level as the face-to-face groups.

A total of 42 face-to-face groups' codes of conduct were collected over a period of two semesters. Four of the groups were located in France. In addition to the 42 groups, four additional virtual groups' codes of conduct (created out of 8 of the 42 groups) were assigned. Of the four, only two virtual groups created codes of conduct, with doubt as to the final version of one of those.

We used a two-tier method in analyzing codes of conduct. First we gathered information about the codes, student outcomes, and group processes in order to determine how the codes of conduct affected the communication process in completing tasks. We collected information using three different techniques. First, student feedback in the form of end of course and end of project questionnaires or discussions were elicited. Students were asked to comment on the group process, the usefulness of the codes of conduct, and the fairness of grading. The last measure was used to evaluate if students felt they had a role in the process. The second method was qualitative, based on teacher observations of group work and an analysis of the quality of work by groups with and without codes of conduct. Finally, the instructors looked at the quality of group work for those semesters under study and compared it with previous semesters. They did this by looking at final project outputs such as group presentations or reports. They also used a variety of methods to analyze individual contributions to the final project including reports, meeting minutes, statements of responsibilities, and project coherence.

In addition to gathering basic statistics about the groups, including group satisfaction measurements and group grades, we used a variety of qualitative analysis methods to analyze the data. These methods included content analysis, ethnography, and case analysis. In our initial design, we had hoped to have 10 virtual groups from which to draw data. However, due to scheduling changes, only 4 were created. With an n=4, therefore, we decided to develop a case for the most successful virtual group. We used Merriam's (1998) approach to interpretive educational case study, creating a multifaceted examination of a phenomenon. In this case, the phenomenon was the group interaction using codes of conduct. We wanted to understand the factors that affected this group's interaction that allowed them to be successful in completing their assignment. While this type of analysis is limited in its ability to generalize factors, it would allow us to understand what contributed to this group's success, thus identifying future areas for broader research. The case we developed can be found in Appendix A.

4. FINDINGS

Student feedback indicated that the majority of the students in teams with completed codes of conduct were satisfied with the group process when using codes of conduct. Many of the successful groups, however, commented that they did not need to use the code of conduct. This could be because they had similar values to begin with or that the development of the code of conduct created common expectations. We could not determine which was the case based on the student feedback collected. During the two semesters of the study, no group needed instructor intervention to resolve conflicts. When there were group problems, the code of conduct was used by the group itself to help in resolving conflicts. The use of the codes in conflict resolution included reminding group members of their responsibilities, creating an environment for constructive conflict, or making group members accountable for their actions. In some cases, groups indicated that while they used the code of conduct because all members had approved it, the code itself was flawed and should have been revised to be more useful.

A review of the content in each code of conduct revealed some consistencies. Groups identified the expectations for how work would be assigned, how communication and decision making was expected to occur, expectations regarding meeting attendance and level of preparedness, expectations for task performance, and clear guidelines for recourse and enforcement. This was true for the French groups as well. However, the specific content and emphasis of each of these factors varied depending on class, group composition, and culture.

Evaluation of the final group outputs found that the quality of work was no different than the previous semesters when codes of conduct were not used. In addition, the codes of conduct did not guarantee equal quality and quantity of individual member contributions. Some of the final products did not represent a group effort as much as a collection of individual contributions. Therefore, our research supports the first hypothesis in the context of face-to-face groups that group communication, discourse management, and processes would be augmented by the codes of conduct. However, this does not mean that the group product will be improved.

In reviewing the cases, we also found that codes of conduct, when used properly, helped the virtual group to plan the processes needed to complete tasks. The codes also created expectations for communication. When those expectations were not met, the group would need to react in some way. However, in our virtual group, they did not use the recourse developed by the group. The code was simply a way to identify communication breakdowns and group process expectations. We were not able to make any conclusions about the second hypothesis due to the small sample size. It is apparent that our research question was, in fact, premature. Before determining if virtual group communication would improve by using codes of conduct we should have investigated if virtual groups would be able to create codes of conduct in the first place. Our preliminary findings indicate that this would be difficult. In analyzing the case of the one successfully developed virtual code of conduct, we discovered that the code of conduct would be difficult to enforce. We did conclude that virtual groups need more time than face-to-face groups to develop codes of conduct. There is also an indication that groups that are homogenous, have successfully implemented their own local codes of conduct, and use the code as a framework for task planning (rather than a guideline for group norms), may be able to develop a virtual code of conduct. However, it is not clear if this would be a "code of conduct" or a group planning and communication document.

5. DISCUSSION

There are a number of factors that explain why codes of conduct might improve the group communication process in face-to-face groups. Our study would seem to support the idea that shared values, social interaction, and group expectations would be the basis to develop group trust (Farrell, Cobbin & Farrell's, 2002). Group trust, especially social trust, in turn, is the basis for effective conflict resolution and the creation of a safe environment for group interaction. In analyzing information from our study, it appears that face-to-face groups, like the virtual groups Jarvenpaa and Leidner (1998) studied, need predictable communication, substantial responses, positive leadership (not appointed), flexibility, and, most importantly transition from social to procedural to task focus. In our analysis of all the groups, groups that were unable to overcome conflict or who performed below class standards tended to lack one or more of these attributes, resulting in group dysfunction and lack of trust. The code of conduct did help in determining how to address these issues. It also appears that successful face-to-face groups must also go through the transition from social to procedural to task focus (Jarvenpaa & Leidner, 1998). In some cases the codes of conducts focused too much on social or procedural guidelines, so group members were unable to complete tasks. For example, one group's code had well defined group roles, with each member assigned specific tasks. However, the group was so focused on each person's individual responsibilities that the group as a whole suffered, with little coordination of individual tasks. Group communication was limited. As a result, they were unable to successfully complete the group assignment.

Successful communication did not guarantee successful group outcomes. Likewise, successful group communication does not necessarily mean that there will be free exchange of information and equal decision-making power (Dennis, 1996; Jehn, Northcraft, & Neale, 1999). As a result, cognitive trust is not developed either. It would appear that cognitive trust, as well as social trust, is necessary for successful group outcomes (Kanawattanachai & Yoo, 2002). The cognitive trust in a face-to-face group might take a different form than virtual groups. A group may not know the most effective way to use individual members resources or disregard ideas because they do not fit the group norm. As a result, a group that communicates effectively still may not optimize its outputs.

One area for future research would be to determine if the process of creating the code of conduct helped to develop shared values, social interaction, and group expectations or if the successful groups came into the process with the same values and communication styles, thus making the creation process easier. Since our research did not focus on group composition, it would be useful in the future to include information on individual values and determination of groups (i.e. self-selection vs. group composition based on ability, culture, and/or experience). Related to group composition would be the question of what factors affect the code of conduct content. One observation we had in analyzing the codes of content was that groups with a majority of undergraduate (freshmen, sophomores) tended to have a much more rigid code of conduct. In looking at the French and American groups in the case, there was a difference in the tone of the codes. More research should be conducted on the factors that affect the development and implementation of the codes of conduct. For example, is it better to have groups develop a code of conduct after they have some social contact? What process should be developed and under what conditions should codes of conduct be modified? Are codes of conduct more affective if they are general or specific? Our analysis of the virtual groups suggests that homogenous groups and groups that have had previous interaction might have less trouble developing a code of conduct. Both of the subgroups of the virtual group had had common coursework, and in some cases, classes. This could indicate shared values developed over time (Jarvenpaa & Leidner, 1998; Jehn, Northcraft, & Neale, 1999). Comparing successful and unsuccessful groups, these research questions may help in answering what the differences were creating and implementing their codes of conduct.

The results of the international virtual groups have implications for future research. In the successful case of the virtual teams, the group arrived at a project outcome before they could agree on the code of conduct. It appeared that the problem with the code of conduct for the international virtual groups were twofold. First, groups tended to behave as Van der Veen, Collis, & Jones (2003) reported in their study. Students had greater allegiance to their face-to-face group than the virtual group. One reason is that it was very difficult to enforce virtually the consequences to a breech in the code. For example, one virtual group required communication on a weekly basis, but did not exchange anything until the last week of the project. In a face-to-face group, social and non-verbal, in addition to evaluative, consequences can be used to enforce the code. This is difficult in a virtual environment. Therefore, for codes of conduct to work, there needs to be more research on relevant consequences in a virtual environment and how to enforce them. Secondly, the code of conduct development process was more time consuming than faceto-face in a virtual and international environment. A solution to this would be Van der Veen, Collis, & Jones' (2003) suggestion that real-time and alternative technologies (video conferencing) be used initially to help establish social communication and trust. Also, there needs to be built into any virtual activity a long time frame in order to overcome difficulties in communication and technology.

Despite the limits to using the code of conduct, it does appear that they provide a planning mechanism for group processes and provide group expectations. As long as the code of conduct is perceived as a guideline rather than a static structure, it may be able to help groups go from socializing to processes to tasks. However, the question still remains, can virtual groups develop their own codes of conducts, and is it worth the time and effort? Is there a more effective mechanism to help virtual codes improve informational and social trust, thereby improving communication and group processes?

6. CONCLUSION

Codes of conduct are one way to improve the communication process for face-to-face groups. The codes help establish common expectations, shared values, and group procedures, all of which contribute to group trust. However, improved communication does not necessarily improve group outcomes. In addition, it is unclear as to whether virtual groups can even develop a code of conduct; much less use it to improve group processes. More research needs to be conducted on the process needed to create virtual group codes of conduct. In addition, there needs to be further studies on the affects of group composition, including culture, maturity and experience on the development of codes of conduct. The use of self-generated group codes of conduct has many implications outside of the classroom. While this research focused on the classroom, this is an idea that should be tested in the workplace, to determine if workplace teams could improve group communication and outcomes.

APPENDIX A: VIRTUAL GROUP CASE

This group was comprised of two groups: one at a four-year liberal arts college in the US; the other at an American school of business in France. The American group had four members assigned to the group based on their schedule. With the exception of gender, this group was homogenous including: seniors, commuters, and marketing/management majors. Some of the group members had classes together, but none of the group members seemed to interact socially out of school. Less information was available about the composition of the French group. The group consisted of five members in the capstone class of a four-year management program. The program was multi-national, taught in English. Therefore, we could conclude that the students had studied together for four years, had a good command of English, and were used to multi-national interactions.

The American group was given the code of conduct exercise the second week of class, and subsequently worked successfully on group activities in class. The French group was given the code of conduct exercise about three weeks after classes began. Because of the differences in school schedules, the American group worked together about four weeks longer than the French group when they were given the joint assignment. Both groups were given an ethics case with four questions to help guide their analysis. Each group was asked to prepare their own answer. The instructors then exchanged a profile, the code of conduct, and the analysis of the ethics case from each group.

The French group's code of conduct was more professional and procedural than the American's. The French code included the number of absences, due dates, assignment of responsibilities, and specific consequences for lack of performance. The French code also outlined specific performance outcomes that were expected for each member. The American code, on the other hand, was vague, only outlining group expectations on contributions, interaction, and expectations on the quality of work. The consequences were approached in a humorous manner. The first consequence of missing a deadline, for example, was to bring a candy bar for each of the group members. They did use this consequence in one instance, which was enough for the group to apply pressure, albeit humorously, to maintain the group responsibilities. The joint code of conduct the two groups created for their virtual work included communication expectations (each group would post at least once a week), deadlines, and procedural requirements including how the group would make their decisions. During the final feedback presentation, the American group commented that they did not, in fact, follow the code of conduct. Deadlines were continually missed, and there was little interaction before the final analysis was completed. The major barriers to their accomplishing the analysis were time constraints (they only had four weeks to complete the project), communication barriers including the level of understanding, language, and style; differences in business practices; and inconsistency in technology (system crashes, incompatible programs which made down loading difficult).

In fact, while this was the most successful virtual group, the American group was not sure that the final project they submitted to the French group was actually agreed upon by the virtual group or if it was accepted because of the time constraints. Since they never heard anything in reply to the contrary, the American group assumed it was the final product. While each of the local groups seemed to work well together, there did not seem to be the same sense of cohesion with the virtual group. The American group handled conflict differently in the face-to-face than in the virtual environment. They indicated that in the face-to-face environment, there could be enforceable rules that were not possible in the on-line environment. For example, when the French group missed posting on a weekly basis, the American group could only send more e-mails reminding the French group of their responsibility. The American group also handled conflict with humor, which they did not feel comfortable doing virtually or with another cultural context. On the other hand, neither group seemed tied to the procedures they had developed in the code of conduct, but rather focused on accomplishing the task. In the end, both groups were able to negotiate a final project by modifying their original analyses, although the virtual group did not feel that the code of conduct helped the process.

REFERENCES

Ayoko, O., Hartel, C., and Callan, V., "Resolving the Puzzle of Productive and Destructive Conflict in Culturally Heterogeneous Workgroups: A Communication Accommodation Theory Approach", International Journal of Conflict Management, Vol. 13 (2), 2002, 165-196.

Barker, V., Abrams, J., Tiyaamornwong, V., Seibold, D., Duggan, A., Park, H., & Sebastian, M., "New Contexts for Relational Communication in Groups", Small Group Research, Vol. 31 (4), 2000, 470-503.

V. Buff, Cheryl and Yonkers, Virginia, "Using Student Generated Codes of Ethics in the Classroom to Reinforce Business Ethics Education", Manuscript submitted for publication, 2004.

Cohen, E., "Restructuring the Classroom: Conditions for Productive Small Groups", Review of Educational Research, Vol. 64(1), 1994, 1-35.

Dennis, Alan, "Information Exchange and Use in Group Decision Making: You Can Lead a Group to Information, But You Can't Make It Think", MIS Quarterly, Vol. 20 (4), 1996, 433-458.

Farrell, Brian, Cobbin, Deirdre, and Farrell, Helen, "Can Codes of Ethics Really Produce Consistent Behaviours?", Journal of Managerial Psychology, Vol. 17 (6), 2002, 468-491.

Fisscher, O., Nijhof, A. and Steensma, H., "Dynamics in Responsible Behavior in Search of Mechanisms for Coping with Responsibility", Journal of Business Ethics, Vol. 44(2/3), 2003, 209-223.

Henri, France, "Distance Learning and Computer-Mediated Communication: Interactive, Quasi-Interactive or Monologue?", in O'Malley, C (Ed) Computer supported collaborative learning, 145-161, Springer Verlag, Berlin, Germany, 1995.

Hoag, A., Jayakar, K., & Erickson, K., "The Role of Trust in Virtual and Interpersonal Environments: Implication for Team Learning and Case Method Pedagogies", Journalism & Mass Communication Educator, Vol. 57 (4), 2003, 370.

Jarvenpaa, Sirkka and Leidner, Dorothy, "Communication and trust in global virtual teams", Journal of Computer Mediated Communication, Vol. 3 (4), 1998, www.ascusc.org.

Jehn, Karen, Northcraft, Gregory, and Neale, Margaret, "Why Differences Make a Difference: A Field Study of Diversity, Conflict, and Performance in Workgroups", Administrative Science Quarterly, Vol. 44 (4), 1999, 741-764.

Kanawattanachai, Prasert, and Yoo, Youngjin, "Dynamic Nature of Trust in Virtual Teams", The Journal of Strategic Information Systems, Vol. 11 ,(3-4), 2002, 187-213.

Kayworth, Timothy, and Leidner, Dorothy, "Leadership Effectiveness in Global Virtual Teams", Journal of Management Information Systems. Vol. 18 (3), 2001-2001, 7-41.

Kidwell, Linda A., "Student Honor Codes as a Tool for Teaching Professional Ethics", Journal of Business Ethics, Vol. 29, 2001,45-49.

McConnell, D., "Interaction Patterns of Mixed Sex Groups in Educational Computer Conferences; Part 1--Empirical Findings", Gender & Education, Vol. 9 (3), 1997, 345-364.

McMahen, Chris and Dawson, A.J., "The Design and Implementation of Environmental Computer-Mediated Communication (CMC) Projects", Journal of Research on Computing in Education Vol. 27 (3), 1995, 318-336.

Merriam, Sharan, Qualitative Research and Case Study Application in Education, Jossey-Bass San Francisco, CA, 1998.

Monteith, Moira, and Smith, Jan, "Learning in a Virtual Campus: The Pedagogical Implication of Students' Experiences", Innovations in Education and Teaching International, Vol. 38 (2), 2001, 119-132.

Oren, A., Miodusser, D., & Nachmias, R., "The Development Of Social Climate In Virtual Learning Discussion Groups", International Review of Research in Open and Distance Learning, Vol. 3 (1), 2002, 1-19.

Parker, Sharon, Wall, Toby, & Cordery, John, "Future Work Design Research and Practice: Towards an Elaborated Model of Work Design", Journal of Occupational and Organizational Psychology, Vol. 74 (4), 2001,413-441.

Rovai, Alfred, "Building Sense of Community at a Distance", International Review in Open and Distance Learning, Vol. 3 (1), 2002, 1-12.

Van der Veen, Jan, Collis, Betty, and Jones, Val, "Network Applications for Group-Based Learning: Is More Better?", Interactive Learning Environments, Vol. 11 (2), 2003, 127-146.

Watson, W., Kumar, K., & Michaelson, L., "Cultural Diversity's Impact on Interaction Process and Performance: Comparing Homogeneous and Diverse Task Groups", Academy of Management Journal, Vol. 36 (3), 1993, 590-603.

Yonkers, Virginia, Segneri, Lucinda, & Ramos, Edgar, "Developing International Business Communication Skills Long-distance", Proceedings of Association for Business Communication Spring Conference, www.businesscommunication.org. Presented April 5, 2003, Toronto, Canada.

Virginia Yonkers, University at Albany, Albany, New York, USA

Cheryl L. Buff, Siena College, Loudonville, New York, USA

Virginia Yonkers is a PhD student in Instructional Design and Technology at the University at Albany, with an Master in International Management from the University of Denver. She has taught, worked, and published in the field of International Business and Communication.

Dr. Cheryl Buff earned her Ph.D. at Union College in 1998. She is an Assistant Professor of Marketing and Management at Siena College. Her research interests include marketing ethics, brand strategies, buying behavior, market segmentation, and pedagogy.