Information systems (IS) process modeling using the technique of Data Flow Diagramming (viz., Systems Analysis) can be defined as a complex task for IS designers. This study draws from the domains of educational psychology and organizational behavior in examining the training of novices in conceptual process modeling. Specifically, an experiment was conducted to determine what effects cooperative, team based participation has on self-efficacy and learning outcomes in dataflow diagramming (DFD) tasks. Results showed novice learners of DFDs performed better when working in cooperative teams rather than learning alone. For those learning in cooperative teams, neither team conflict nor team cohesion had any effect on DFD skill acquisition.

Information systems (IS) process modeling can be considered an ill-structured, complex task that requires higher-order thinking skills. A simple system may require a simple, easy to create process model. However, as systems become more multi-faceted, the complexity of creating process models increases (Millet, 1999). It is difficult for those with minimal training to comprehend the complexity of sophisticated process models

To facilitate a more rapid progression through the stages of IS design, studies have focused on the similarities and differences between expert and novice designers in constructing conceptual or logical models of systems, the improvement of conceptual design through feedback, and the usage of heuristics in the modeling process Despite the advances that have been made, there continues to be a call for additional research on the pedagogy of training novices effectively and efficiently (Brown and Klein, 2003). This study draws from the domains of educational psychology and organizational behavior in examining the training of novices in Data Flow Diagramming (i.e.. IS process modeling). Specifically, an experiment was conducted to determine what effects cooperative, team based participation had on self-efficacy and learning outcomes in systems analysis tasks.

Data flow diagramming (DFD) is one technique used to document process models and is often a part of Systems ; Analysis and Design (SA&D) curriculums. In one survey, over half of the respondents (53%) devoted 5-10% of class time to teaching DFDs, and 73% believed it was definitely important to teach DFDs in SA&D courses (Tastle and Russell, 2003). In this research, we examine the factors that may affect the ability of students to learn how to create DFDs.

This paper builds on a cumulative research tradition by extending the research model of Ryan, Bordoloi, and Harrison (2000) that investigated the influence of learning mode (cooperative versus individual learning) and selfefficacy on skill acquisition. Ryan et al., (2000) found that self-efficacy was significantly and positively related to skill acquisition. However, contrary to what was expected, performance for subjects in cooperative teams was not significantly different from subjects working individually. Because of this unexpected result, we were motivated to build on this research in two ways. First, we wanted to see whether team-based cooperative learning is beneficial for other complex tasks in information systems development. Thus, we examine the creation of DFDs rather than ERDs to see if cooperative learning teams lead to higher level performance as theory predicts. secondly, we include new research questions as suggested by the work of Ryan et al. (2000) that further examine the relationship between cooperative learning teams and performance outcomes. Specifically, we investigate various aspects of intra-group processes that might impact performance by cooperative learning groups as well as the impact of motivation throughout the learning process. By examining these issues, we hope to be able to provide recommendations to instructors on techniques that may enhance process modeling skill acquisition.

2. LITERATURE REVIEW, HYPOTHESES AND RESEARCH MODEL

As recommended for multidisciplinary cumulative research, we selected constructs based on existing MIS studies and prior work in related disciplines (Bandura, 1986; Lee and Bobko, 1994) to guide this empirical investigation. Based on Ryan et al.'s (2000) call for future research to examine the influence that intra-group processes might have on achievement, we include the intra-group process constructs of conflict resolution and team cohesion. In the following few subsections, we discuss these constructs and the related literature, the hypotheses, and our proposed research model.

2.1 Learning Mode

Cooperative learning has been defined by Cohen (1994) as:

"Students working together in a group small enough that everyone can participate on a collective task that has been clearly assigned. Moreover, students are expected to carry out their task without direct and immediate supervision of the instructor".