Loading...
Please wait, while we are loading the content...
Similar Documents
Team information sharing 1 RUNNING HEAD : Team Information Sharing Will They Share ? Team Problem-Solving in Computer-Mediated Environments
| Content Provider | Semantic Scholar |
|---|---|
| Author | Ferzandi, Lori A. Skattebo, Amie L. Terrell, Ivanna S. Bains, Priya |
| Abstract | Despite the purported appeal of teamwork, teams do not always reap the benefits of their collective knowledge. Unique information is rarely ascribed the same level of importance as commonly-held information. This poster will discuss the shortcomings of computer-mediated group problem-solving, along with the implications that such findings have for the future creation and implementation of virtual teams. The current investigation extended co-located Information Sampling theory research by Stasser and Titus (1987) to the computer-mediated cooperative team domain. Data from thirty-seven three-person teams indicate that, despite its criticality, information not shared equally by all team members is frequently discounted, and often ignored during team discussions. Team information sharing 3 Introduction Given the increasing reliance on inter-organizational and global cooperation, many organizations have begun experimenting with virtual or distributed teams (Lipnack & Stamps, 1997; Mohrman, 1999; Townsend, DeMarie, & Hendrickson, 1998). While Olson and Olson (2003) have suggested that tools developed to aid collaboration at a distance are becoming increasingly widespread, they have simultaneously lamented the fact that “there are still many research issues about how to design such systems and what effects they have on the individuals, groups, and organizations that use them” (p.584). Recognizing that researchers have faced challenges in attempting to gather group process information in the distributed domain, researchers like Grudin (1994) argue that it is nonetheless important to determine whether phenomena commonly associated with co-located teams operate similarly in distributed team environments. The current research seeks to address this issue. Specifically, the goal of the current research was to examine the effect of shared knowledge profiles within distributed teams. In order to address this issue, the current study gathered problem-solving information from distributed groups whose only means of collaboration were through the use of computer-mediated chat rooms. Transcripts from these chat room conversations were collected and analyzed to offer insight into the process features of distributed groups whose members had varying degrees of shared and novel information. The results of this study will begin to address Grudin’s (1994) concerns regarding the need for both theoretical and empirical extensions of research related to co-located teamwork phenomena. Theoretical Background Decision-making groups appear to have an advantage over individual decision-makers, as groups are able to collate and compare the information and resources contained within the minds Team information sharing 4 of their group members. This is sometimes referred to as partial mission overlap, networking, or the combination of core competencies (e.g., Bultje & van Wijk, 1998). Despite this apparent wealth of knowledge, many groups are found to reach decisions that are not supported by their collective knowledge. Specifically, research involving face-to-face group communication by Stasser and Titus (1987) suggests, that when critical pieces of information are held by individual group members they are not always factored into the group’s final decision. As the research presented below will demonstrate, the commonly-held belief that “two heads are better than one” may not always be accurate. According to the common knowledge effect, final task decisions are largely influenced by information that is common to most (if not all) members of the team (Gigone & Hastie, 1993, 1996). Using this common information, team members formulate opinions that are likely to bias their final judgments (in the direction of these initial opinions). They are often unwilling to accept unique or previously unshared information as valid not only because it would require additional effort to incorporate this information into the existing base of information, but also because it may actually run counter to the information already gathered (Larson, Christensen, Abbott, & Franz, 1996; Stewart & Stasser, 1995). This unique information may in fact be deemed detrimental to the perceived progress that the team feels it has made (Stasser, 1999). Social validation seems to play an equally important role in understanding why individuals may or may not be motivated to present unique information to their teammates (Stasser, 1999). For example, Hinsz (1990) found that teams would only seriously consider the validity of unique information when (a) one member could be absolutely certain of the accuracy of that information and/or (b) the information could be correctly recalled by another team member. These results would seem to suggest that individuals who experience a sense of Team information sharing 5 rejection after presenting unique information to the team would be much less likely to present additional unique information throughout the remainder of the problem-solving discussion. Ultimately, the likelihood that a unique piece of information will be discussed by a team after initial mention by one member has been determined to be a function of collective information sampling (Stasser & Titus, 1987). The collective information sampling (CIS) model states that the probability of item discussion is largely influenced by (a) the number of team members who can potentially recall this unique piece of information after it is first mentioned and (b) the likelihood that an individual will mention this unique piece of information to the group in the first place. According to this model, the proportion of unique (unshared) information mentioned to the group (compared to the overall level of discussion) decreases as the number of members comprising a team increases (Stasser, Taylor, & Hanna, 1989). The use of computer-mediated communication in the current study adds a unique dimension to the teamwork research conducted by Stasser and colleagues in an effort to determine whether distributed teams fall prey to the same common knowledge effects (Gigone & Hastie, 1993, 1996) and/or information sampling tendencies (Stasser & Titus, 1987) that have been cited in co-located team research. Given research by Walther (1996) which suggests that computer mediated communication tends to filter out social cues that otherwise aid communication in FtF settings, we would expect that team problem-solving effectiveness and efficiency in a computer-mediated setting would be negatively impacted by the uneven distribution of information across team members. In an attempt to examine in greater detail, the frequent inability of teams to take unique information into account when solving tasks, the current study utilized the technique of hidden profiles (Stasser & Stewart, 1992; Hollingshead, 1996). In a hidden profile setting, critical Team information sharing 6 information (that ultimately leads to the problem solution) is unequally distributed among team members, such that each member has only part of the information needed to accurately solve the problem. Team members who problem-solve using only the information they receive up front, will more often than not, generate incorrect or sub-optimal solutions. Accordingly, teams in which problem-critical information is shared unequally among members (i.e., hidden-profile teams) are likely to face challenges in uncovering and recognizing the relevance of such information, despite its importance in obtaining the correct solution. The failure to recognize unique information becomes even more troublesome when teams are not advised of this unequal distribution of information. As a result, we hypothesized that hiddenprofile teams in the current study (who were initially unaware of this unequal sharing of information) would be less effective in generating the correct solution, than teams in which all problem-critical information was shared equally (i.e., full-profile teams). While effectiveness has typically been measured using outcome measures such as solution accuracy, we felt that it was equally important to evaluate the content of each team’s discussion prior to making overall performance ratings. For example, a team that consistently focuses on relevant and critical problem space information throughout the majority of their discussion, yet fails to obtain the correct solution due to a math miscalculation, should not be completely discounted and overlooked. As demonstrated by Stasser (1992), discussions that are focused on key information are correlated most highly with the correct solution. Thus, it was important that the transcripts in the current study be analyzed according to the focus of each team’s discussion, rather than simply the accuracy of their proposed final solution. In order for teams to generate the correct solution in the current study, all nine pieces of critical information must be (a) brought to the attention of the team, (b) validated by other team Team information sharing 7 members, and (c) factored into the final solution of the problem. Accordingly, hypotheses 1 and 2 are direct tests of whether or not teams mentioned critical pieces of information in the first place (i.e., step “a”). Because each member of a full-profile team was given all nine pieces of critical information up front, there was a greater likelihood that each piece of information would be brought up by at least one of team member. In contrast, there was less of a chance that critical information would be brought into a hidden profile team’s discussion since only one or two members were privy to this information at the start of the problem-solving task. Hypothesis 1: Full-profile teams will mention more problem-critical pieces of information throughout the duration of their discussion than hidden-profile teams. Along the same lines, an increase in the number of critical pieces of information mentioned would have a tendency to reduce the amount of non-critical information that coul |
| File Format | PDF HTM / HTML |
| DOI | 10.1037/e518632013-325 |
| Alternate Webpage(s) | http://www.personal.psu.edu/users/a/l/als383/Docs/SIOP2004TeamSharing.pdf |
| Alternate Webpage(s) | https://doi.org/10.1037/e518632013-325 |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
