Decision Making in an Automated Environment: The Effects of Anonymity and Proximity with a Group Decision Support System

Avoidance of Groupthink

Groupthink is a problem thatplagues decision-making groups. This article reviews antecedent and procedural conditions leading to groupthink and examines productive meeting processes that can prevent groupthink. The article then explores the possible role that group support systems (GSS) may play in the development of these productive meeting processes. Group support systems are an advanced information technology that provide electronic support to groups involved in teamwork. This technology possesses certain inherent resources or structuralfeatures. This article proposes aframeworkfor the study of the effects of GSS on groupthink. It examines the effects of GSS structuralfeatures on antecedent and procedural conditions predisposing groupthink. Existing GSS research relevant to groupthink-related variables is reviewed. The review of GSS structural features and of existing GSS research indicates that GSS use might indeed be an appropriate method of preventing groupthink. Finally, the article proposes a methodologyfor the empirical study of the impact of GSS use on groupthink.

Inspiring Group Creativity

A laboratory study examined effects of group members'perceptions of behavioral components of transformational and transactional leadership on group creativity underanonymous and identified electronic brainstorming conditions. Results indicated that perceptions of transactional goal setting and inspirational leadership were each positively related to group creativity. Intellectual stimulation and individualized consideration were negatively related to group creativity. Except for the effect of intellectual stimulation on group creativity, these effects were stronger under anonymous versus identified conditions.

New Contexts for Relational Communication in Groups

This article discusses the importance of relational communication in groups. Connections are made with potentially fruitful theoretical concepts, while findings from related group research are used to discuss new and interesting directions with regard to relational communication in three pivotal group contexts. Each section provides a rationale for why these group contexts should be studied and suggests propositions to guide future research. The discussion highlights the many facets of the relational side of members’ intentions in groups: cooperation, connection, autonomy, similarity, flexibility/rigidity, cohesion/withdrawal, harmony/conflict, stereotyping, and stigmatization. Concerning group process analysis, the authors revisit the family, as one’s first group experience, from a relational communication perspective. Next, the influence of heterogeneity of group membership and intercultural diversity is discussed with regard to relational dynamics. Finally, the effects of computer-mediated communication (CMC) on relational communication in groups are investigated.

The Effects of Variations in Gdss Capabilities on Decision-Making Processes in Groups

Group Decision Support Systems (GDSSs) provide mechanisms for reducing equivocality and facilitating consensus on interpretations by decision-making groups. Whereas much previous research compares groups with GDSS support to nonsupported groups, less research has been conducted which varies the type ofsupportprovided in a GDSS. This study addresses two questions: What are the impacts of variations in the level of sophistication of a GDSS on group decision-making processes? What is the significance of these process impacts on group outcomes? Two different GDSS designs that delivered varying levels of decision-making support were employed in the study. The conceptual model identifies six specific process impacts. The results indicate that variations in GDSS capabilities have an impact on elements of the group decision-making process, such as the ability to develop procedural insight or ideational connection and use formal evaluation. Further, the process impacts were found to be significantly related to the group outcomes. 7he ideas presented in the article are expected to offer a foundation for developing finer-grained theories of GDSS effects on group decision making.

Physical Proximity Effects on Computer-Mediated Group Idea Generation

An experimental study measured the idea generation performance of groups offour and eight members in both proximate and distributed conditions using the same synchronous computer-mediated communication system. Proximate groups worked with all members in the same room. Distributed groups worked with all groups in separate, distributed offices. In terns of the total number of unique, total quality, and number of high-quality ideas generated, distributed groups outperformed proximate groups. A group size by proximity interaction was predicted to result in a higher rate of performance gainsfor increased group size within distributed groups, yet no interaction wasfound. The implications ofthese results for researchers and organizations contemplating the adoption of computermediated support technologies is discussed.

Group Size and Anonymity Effects on Computer-Mediated Idea Generation

Some for,ns of computer-mediated communication enable groups to engage simultaneously in multiple electronic conversations. One interesting aspect of this technology, in addition to this parallelism, is that the channel is configurable to support forms of group interaction that were previously difficult, if not impossible. This article reports a laboratory experiment that examined the effects of group size (3 and 9 members) and group member anonymity on the performance of groups using a computer-mediated idea-generation system. Although group members in all conditions made, on average, the same number of comments, larger groups generated significantly more ideas (and higher-quality ideas as rated by judges) than did smaller groups. Anzonymity had no effect on ideational performance. Members of small-idenitified groups made the fewest critical remarks, were the most satisfied, and rated themselves more effective than group members from the other experimental conditions.

Some Liberating Effects of Anonymous Electronic Brainstorming

The high productivity of electronic brainstorming groups has been attributed to several factors, one of which is anonymity. The authors assessed the role of anonymity by randomly assigning brainstorming groups to one offour conditions: anonymous electronic, nonanonymous electronic, verbal, and nominal. The authors found some evidence that more controversial ideas were produced by members of anonymous electronic groups than by members of the other groups. The authors also found clear evidence that anonymous brainstorming groups produced more nonredundant ideas than did nonanonymous brainstorming groups.

Theoretical and Methodological Issues in Group Support Systems Research

The amount of research on group support systems (GSS) is growing quickly. One component of GSS, anonymous interaction, has received a great deal of attention recently. The quantitative and qualitative research thus far on anonymous GSS interaction suggests that the effects of GSS anonymity on group processes and outcomes are positive and/or neutral. In this article, the authors explore the potentialfor negative and/or dysfunctional consequences of GSS anonymity and discuss the relevant implications and research questions to be asked and answered.

Effects of Transformational Leadership and Anonymity on Idea Generation in Computer-Mediated Groups

A laboratory experiment was conducted to evaluate the effects of high and low levels of transformational leadership style and anonymity on 36 undergraduate student work groups using a Group Decision Support System (GDSS) to perform an idea generation task. Transformational leadership was manipulated through confederate leaders who displayed leadership behaviors and entered comments into the GDSS according to scripts. Results of analysis of variance indicated that groups working under high transformational leadership generated more original solutions, supportive remarks, solution clarifications, and questions about solutions and reported higher levels of perceived performance, extra effort, and satisfaction with the leader than groups working under low transformational leadership. Results also indicated that anonymous groups generated more critical comments and less problem clarifications than identified groups. Implications of these results for organizational researchers and managers are discussed.

Effects of Participative and Directive Leadership in Electronic Groups

A laboratory experiment was conducted to assess the effects of participative and directive leadership on participation, performance, and satisfaction of 24 undergraduate student work groups that interacted electronically via a Group Decision Support System (GDSS) to performa creativity task. Participative and directive leadership were manipulated through confederate leaders who entered scripted comments into the GDSS. Performance was measured in terms of quality and uniqueness of solutions.Results of partial least squares analysis indicated that perceptions of both leader participativeness and directiveness were positively related to levels of participation. Participation in turn was positively related to performance but negatively related to satisfaction. Problem structure moderated all these relationships except for the relationship between participation and performance. Practical and research implications, study contributions, and limitations are discussed.

Impact of Anonymity on Information Sharing through Internal Psychological Processes

Many web users post their opinions and information without revealing their identities (i.e., demographics, social standing, and/or expertise); anonymity has become a common form of information exchange in online communities such as social media sites and Internet forums. In the meantime, negative social consequences have resulted from anonymous activities in online communities. Existing literature reports largely mixed results about the impact of online anonymity on online community outcomes. In an effort to clarify these mixed results, this study proposes a new operational definition of online anonymity that consists of three dimensions: unlinkability, unobservability, and pseudonymity. The data used in this study was collected from 238 online community users through a web-based survey conducted in South Korea. The results show that unlinkability, unobservability and pseudonymity positively influence self-discrepancy, which in turn positively influences the quality and quantity of information sharing.

When Cultures Clash Electronically

This research examines the extent to which the email medium exacerbates the aggressiveness of opening offers made by negotiators from two distinct cultures. Hypotheses derived from negotiation, communication, and culture research predict that Hong Kong Chinese negotiators using email would exhibit a reactance effect and consequently engage in more aggressive opening offers and claim higher distributive outcomes than similar negotiators in the United States. Study 1 examines intercultural email negotiations and results indicate that Hong Kong Chinese negotiators made more aggressive opening offers and attained higher distributive outcomes than their U.S. counterparts. Study 2 results replicate Study 1 findings in an intracultural negotiation setting and also show favorable outcomes for Hong Kong email negotiators when compared to both Hong Kong and U.S. face-to-face negotiators. Overall, the findings suggest that Hong Kong Chinese and U.S. negotiators vary substantially in how they negotiate via email and face to face, which results in differences in distributive outcomes.

Effects of Anonymity and Social Comparison of Rewards on Computer-Mediated Group Brainstorming

Anonymity in computer-mediated communication (CMC) is valued in organizations because it can facilitate participation in discussions, especially of sensitive issues. Concerns over maintaining the motivation to participate in such discussions, because of the inability to reward people for their contributions, have led to the development of techniques that allow rewards to be allocated without identifying their distributors and recipients. It was hypothesized, however, that anonymous rewards are not equivalent to identified rewards in the ability to motivate people. Results from a laboratory experiment using a group electronic brainstorming task supported this hypothesis. Task effort was higher when reward recipients were identified than when they were anonymous; comparing rewards further increased the effects of identification. Identification combined with comparison increased task effort slightly more for recipients of the middle level than the top or bottom levels of rewards. Negligible effects were found for idea quality. Implications are discussed for anonymity, comparison targets, task structure, and the intersection of technical and social factors in organizations.

Anonymity and Counter-Normative Arguments in Computer-Mediated Discussions

A laboratory study examines (a) the effects of source and participant anonymity on the incidence of counter-normative arguments during computer-mediated discussion of an issue that has a normative position, and (b) how the incidence of counter-normative arguments, in turn, affects members' agreement with the normative position. In source-anonymous groups, members know their group's composition but they do not know who has provided what input. In participant-anonymous groups, members do not know their group's composition or the source of any input. When initial opinions among group members differ, source anonymity leads to a lower incidence of counter-normative arguments compared with the identified or participant-anonymity conditions. In groups with a lower incidence of counter-normative arguments, there is greater agreement with the issue's normative position, and members also deviate less from each other after the discussion. Theoretical and practical implications of the results are presented.

Perceived Proximity in Virtual Work: Explaining the Paradox of Far-but-Close

One's colleagues can be situated in close physical proximity, yet seem quite distant. Conversely, one's colleagues can be quite far away in objective terms, yet seem quite close. In this paper, we explore this paradoxical phenomenon of feeling close to geographically distant colleagues and propose a model of perceived proximity (a dyadic and asymmetric construct which reflects one person's perception of how close or how far another person is). The model shows how communication and social identification processes, as well as certain individual and socio-organizational factors, affect feelings of proximity. The aim is to broaden organizational studies' theoretical understandings of proximity to include the subjective perception of it. By shifting the focus from objective to perceived proximity, we believe that scholars can resolve many conflicting findings regarding dispersed work. By understanding what leads to perceived proximity, we also believe that managers can achieve many of the benefits of co-location without actually having employees work in one place.

Social processes and group decision making: anonymity in group decision support systems

Various social processes in decision-making groups are considered detrimental to the quality of decisions. It is often assumed that removing the ability for groups to exert strong social influence on its members improves group decisions. Group Decision Support Systems (GDSSs) are increasingly used to remedy the social faults of the decision-making process in groups. In these systems, anonymity is seen as a tool to reduce the impact of the group over its members, and therefore as the key to improved group performance. This meta-analytic review examines the assumption that anonymity in GDSSs is beneficial for group decision-making on a range of performance indicators. In 6 meta-analyses of 12 independent investigations there is no support for this hypothesis. The only reliable effect of anonymity was to lead to more contributions, especially more critical ones. An alternative model is presented to account for the findings. This model argues that performance in decision-making groups depends on the social context and relevant social norms as well as on system characteristics such as anonymity. It is concluded that the integration of anonymity into phases of group decision support does not guarantee improved performance.

A test of task-technology fit theory for group support systems

Group support systems (GSS) provide both promise and puzzlement. Experimental studies of different systems over the years have resulted in conflicting findings --- sometimes enhancing group performance, at other times having no effect, and occasionally even resulting in worse performance for GSS-supported groups than for traditional groups. Researchers have speculated that the mixed results are due to a poor fit of the GSS with the group's task. A recent model of task-technology fit has provided a theoretical perspective from which to test this issue. In this paper, a theory of task-technology fit is tested by applying it to a selected set of published GSS experiments. Key constructs in the theory are operationalized via coding instruments, and the application of the coding scheme provides support for the theory.

The role of group support technology in developing the HEALNet research agenda

HEALNet, a Canadian Network of Centres of Excellence Program, has undergone a building exercise to focus and consolidate its research program for the 1998 to 2002 funding period. A number of challenges were identified while trying to develop this research agenda within a short time frame. These included how to bring together a diverse group of researchers, ensuring that everyone's voice was heard, and ensuring that the collective interests were placed ahead of individual interests. Furthermore, as three separate workshops were being held, a method for quickly capturing the workshop output and preparing reports was required. A Group Support System (GSS) was used to facilitate this work and to help meet these challenges. This report describes how the GSS was used. While the GSS supported brainstorming activities, it was less useful in assisting the group in developing the details of research programs. It did facilitate the timely production of reports from the workshops. Such systems have a role to play in supporting group work at the idea generation stage, but further work is required in order for them to be useful in the development of detailed research programs.