بررسی تأثیر مدلهای ذهنی تیم بر استفاده از اطلاعات و عملکرد پروژه در توسعه سیستم

Teams in system development must share information about tasks in order to be successful. A traditional view considers the processes in place to take the knowledge resident in a team and effectively share it with other members. Newer perspectives require conditions of shared models and understanding among the team members to effectively employ the dispersed information to develop successful systems. A model of teamwork mental models is presented to explore both perspectives so that the importance of both process and shared models is represented. A sample of system developers provides the data to test the model. Results show that team building activities can promote both aspects, yielding a greater chance for a quality team effort.

Information system (IS) development projects are typically complex, dynamic, and unstructured (Schwalbe, 2007). The implementation of these projects requires the communication of knowledge and expertise from different domains to effectively diagnose the problems and design the solutions (Tesch et al., 2009). For these reasons, teams are an effective structure for organizing IS projects (Tiwana and McLean, 2005). IS project teams are expected to disseminate and utilize information that resides among team members (Faraj and Sproull, 2000). In fact, teams are viewed by IS management as a primary mechanism for leveraging the specialized knowledge of individual team members (Tiwana, 2004). This conventional wisdom is often represented by an input-process-output model (McGrath, 1964), where the collective information in the team is shared via interaction to achieve the desired project outcomes. A team’s ability to share and utilize available information is critical to the success of a project (Mesmer-Magnus and DeChurch, 2009). However, teams do not always apply collective information to the necessary tasks in order to complete the project successfully (He et al., 2007). Recent research indicated there must be a collective cognition in the team to serve as behavioral mediators, a model termed the input-mediator-output model (Mathieu et al., 2008). Team cognition refers to the team mental models (TMMs) collectively held by a group of individuals that enable them to effectively communicate and accomplish tasks by acting as a coordinated unit. These mental models serve as a potential factors in completing tasks and should indicate the likelihood of having an effective team (Cannon-Bowers and Salas, 2001 and Mohammed and Dumville, 2001). Mediational factors that intervene and translate the influence of team inputs to outcomes are not processes but better conceived as cognitive, motivational, or affective states – referred as emergent states (Ilgen et al., 2005). Team collective cognition is one of the most critical emergent states for team performance (Marks et al., 2001). However, the importance of shared mental models will vary under different contingencies, making the study of such models in system development projects more relevant (Kozlowski and Ilgen, 2006). In the literature, researchers have studied the formation and evolution of team cognition in the IS development project context (He et al., 2007). They observed a positive impact between shared task understanding and project performance. From these two independent streams of work, it appears both processes and mediation are important. Empirical analyses in other disciplines indicate that emergent states and processes are often highly correlated (Barrick et al., 2007 and Kiffin-Petersen, 2004). Thus, studies that consider only one feature variable type (either process or emergent state) are susceptible to problems associated with omitted or 3rd variable issues (Mathieu et al., 2008). In other words, the standard input-process-output models and input-mediator-output models should be merged. Does the logic hold such that there is a relationship between the emergent state of shared team mental models and the process of information utilization as shown in prior work in other disciplines (Mesmer-Magnus and DeChurch, 2009). Team members with a better understanding of how to interact should be able to effectively exchange and utilize the information collectively held by the group. If true, would common management mechanisms, such as continual team building activities, promote an improvement to the shared mental models that then improve performance? The purpose of this study is, therefore, to explore the relationship of team cognition in the form of mental models to their potential impact on IS project performance. Specifically, we argue that a teamwork mental model (teammate model of domain information and interaction model of task completion knowledge) will be positively associated with information utilization. Both the shared mental model presence and effective information utilization (mediation and process) will lead to improved project performance. This argues for the importance of both an input-process-output and input-mediator-output model to explain IS project performance. Finally, we argue that common managerial interventions can help form a common understanding among team members. Team building strategies are often adopted by managers during the team formation stage to enhance teamwork and improve the quality of interaction among team members (Lee-Kelley and Blackman, 2005). Continuation of the activities can serve to reinforce the lessons and further improve chances for eventual success (Jones, 2008). Survey data of IS developers from IS service firms located in India was collected to explore the proposed model. The results indicate that activities associated with team building intervention leads to higher shared TMMs, which in turn facilitate better information utilization within the team. IN addition, IS project performance is influenced by the levels of TMMs and information utilization. In other words, information utilization serves as a partial mediator between TMMs and final project performance.

The teamwork MMs and their impact on project performance had not been explored in the context of IS development projects. To diagnose problems and generate solutions, information from different sources or possessed by different team members must be processed and integrated to successfully complete an IS development. Project managers face important issues impacting effective information utilization, especially in the IS outsourcing context. In this study, we explore whether team members with a better understanding of how to find the knowledge needed from among team members will more effectively exchange and utilize the information collectively held by the group. To help determine effective means of promoting the needed mental models, we explore whether team building activities have a positive effect on developing teamwork MMs among team members. Formally, the study includes the relationship between the teamwork mental model and information utilization to their eventual impact on IS project performance. The model tested is a mixture of an input-process-output model and an input-mediation-output model. Data collected from 194 IS employees of outsourcing vendors in India illustrate that continual team building activities of communication, problem solving, goal setting and role clarification lead to higher shared teamwork MMs of where information resides among team members. This knowledge of information location might, in turn, facilitate better information utilization by the team members in solving problems, revising goals, learning, and improving performance. Final project performance is positively associated with the levels of teamwork MMs and information utilization, as suggested in the literature. In other words, this exploration indicates that information utilization might improve project performance in the IS development context. The findings of this study have several implications to IS researchers. First, in the TMMs literature, existing empirical documents have treated teamwork process as a black box by indicating an existence of a positive relationship between TMMs and team performance outcomes. This study demonstrates that teamwork MMs is positively associated with team performance through information utilization. Thus, the combined input-process-output/input-mediator-output model suggests mediating mental models will facilitate team processes which, in turn, affect the final team outcomes. Although previous studies address the impact of TMMs in general teams or groups of individuals without a common goal, this study extends the research stream by showing the impact TMMs on IS development teams. We believe that understanding the role of TMMs is meaningful for a project team and the context different than functional department teams that have a long co-working history. Members of a project team must develop their co-working philosophy in a limited time period. In addition, many studies of TMMs apply to loosely delineated groups without a common goal or clear roles and responsibilities – project teams must develop common understandings of each other, the project goals, and how to work together. Secondly, the study shows that management interventions and practices, such as involving all members in the decision making process, might facilitate the forming of teamwork MMs. If so, then the levels of TMMs can be targeted and managed by instilling effective communication practices, setting objectives, and properly defining roles to ensure the success of team processing. Future research can advance our understanding by examining other intervention mechanisms. Finally, this study indicates that the information utilization by the team is affected by the level of common understanding among team members on how to interact with others in finding required information. IS research has consistently documented the importance of information processing on IS project success; however, relatively little research indicates the foundation for the process of effective information utilization in the IS project teams. The results of this study add to the wisdom that IS developers should possess the interpersonal skills and communication skills to effectively interact with users and others (Tesch et al., 2003). Since there is a diversity of TMMs and we employed just one in our model, it is up to future studies to examine whether other types of TMMs have similar effects on teamwork processes or final outcomes. In our model, TMMs are treated as part of an emergent state which has an impact on teamwork processes. Based on this concept, researchers are encouraged to explore the impact of TMMs on different team work processes, such as coordination and communication. Although understanding the consequences of TMMs is vital for IS project management, researchers should also focus on understanding possible antecedents of TMMs to better understand how to form a team that can work in a high quality manner. Furthermore, we only focus on the perceived similarity of members’ mental models, not whether the shared model is the best or most appropriate for the project environment. However, whether members share a correct mental model should be correlated with teamwork processes which determine final performance. For example, Mathieu et al. (2005) study the impact of the quality of the team mental model as determined by an “ideal” or “expert” model. Future research can extend the current study by examining how the quality of the mental model affects IS development teamwork along with the relationship between mental model similarity and teamwork. However, researchers interested in this issue should be cautious; due to the complex and dynamic nature of IS development, it is difficult to define an ideal mental model. For practitioners, this study brings into focus the following suggestions. First, conducting team building activities might facilitate the forming of team mental models. Team building activities allow team members to be familiar with each other and provide a better understanding of how to interact with each others to attain needed information to proceed in the development of a system. These activities should be directed at improving communications, involving all members in problem solving, clarifying roles, and making certain goals are set. Establishment of these features and ongoing reinforcement is especially important for teams with high turnover rates or diverse team members – common for IS projects (Jones, 2008). Secondly, information utilization is a key for problem solving in IS development projects. Managers must encourage and monitor the extent of information utilization within the IS project teams. Without a high degree of information utilization among team members, the odds of success likely lessen. IS managers must foster the information exchange among team members by encouraging an open environment. Finally, in addition to an indirect effect, teamwork MMs are directly associated with final project outcomes. IS managers must not under-estimate the impacts of the teamwork mental models on IS project performance. Teamwork MMs could also affect the effectiveness of other team processes. Unfortunately, few guidelines exist in the current literature. This study is not without limitations. First, we surveyed only one team member to measure teamwork MMs. To understand whether all team members really share a mental model, future research may adopt approaches recommended by Mohammed et al. (2000) to obtain multiple, individual mental models to aggregate into a team mental model. The proposed relationships can then be reassessed through aggregate data. Secondly, common method variance (CMV) may inflate or deflate the dependent-independent relationship when a single information source is adopted. Although Harman’s single-factor test shows that CMV is not a threat to our study, data from multiple viewpoints are encouraged in future studies. Third, the research scope is limited by validating relationships through cross-sectional data since causality cannot be inferred. Future researchers are encouraged to collect multi-wave data to better demonstrate the proposed causal relationships. Finally, a survey design was adopted to test our hypotheses. This approach is useful in understanding the correlation between team building and the consequences. Future research can adopt experimental or quasi-experimental designs to study the difference between team building interventions and the effect of different team building activities.