Over the last decades system dynamicists have experimented with approaches to achieve more involvement of their clients in the model building process. As a result the number of reports in the literature on the use of system dynamics as an organizational intervention tool has increased dramatically. From the literature we have identified 107 cases that provide details on the modeling process and the assessment of results. The cases show a wide variety in the way the interventions are reported and assessed. From a research point of view this is clearly an undesirable state of affairs. This article reports on a meta analysis of findings of these studies and tries to provide an overview of outcome studies on system dynamics interventions. It attempts to draw some preliminary conclusions on the effectiveness of system dynamics modeling interventions, to clarify differences in definitions of outcomes, and to provide guidelines for more standardized assessments and reports. Rather than remaining in the stage of single case descriptions, the latter will enable the accumulation of research results in the future, a prerequisite for institutional learning within the system dynamics community.

Group model building (GMB) is a participatory method for involving stakeholders in the process of developing system dynamics models. GMB has historically consisted of undocumented structured small-group exercises. This paper describes an effort to document GMB scripts called Scriptapedia, and how documented GMB scripts can be used to design more effective GMB sessions that address cultural and ideological barriers to collaboration. A case study of a project to develop a coordinated community response to domestic violence is used to illustrate the use of scripts for planning collaboration. The paper concludes with a discussion of potential limitations of scripts and implications for future research.

to refer to a bundle of techniques used to construct system dynamics models working directly with client groups on key strategic decisions. We use facilitated face-to-face meetings to elicit model structure and to engage client teams directly in the process of model conceptualization, formulation, analysis, and decision making."Group Model Building" (GMB) as we use the term has a cosy, narrow, some might say parochial ring to it. In this brief title, we fail to mention that the models our groups build are always system dynamics models. Furthermore as we discuss below, GMB has much in common with six other named techniques for gaining client involvement in strategic problem finding and problem resolving activities, all using system dynamics models, none of which use the term Group Model Building. We also see our work as sharing many common concerns with the viewpoints expressed in JORS' previous symposium on Problem Structuring Methods. Eden and Ackermann (2006) list four similarities between PSMs: (1) use of a model as a transitional object, (2) increasing the overall productivity of group process, (3) attention to facilitation and effective group process, and (4) appreciation of the significance of facilitation skills. If we were to use Eden and Ackermann's (2006) four points of similarity as a loose definition of PSMs we would have to conclude that GMB is another PSM.We are probably an example of Westcombe, Franco, and Shaw's (2006) fifth challenge "to design a new PSM that explicitly integrates with a certain approach to hard modelling" while trying to avoid working "simply as a bolt-on to hard modelling".

Client involvement in modeling is expected to change mental models and thereby foster implementation of conclusions. Leading authors have pointed out that a lack of knowledge on the crucial elements of modeling interventions hinders accumulation of research results. There is no clear evidence for the effectiveness of group model building, and a conceptual model linking elements of the modeling process to goals is missing. We propose an integrative conceptual model, drawing on theories of persuasion (mental model change) and the infl uence of beliefs, attitudes, subjective norms and perceived behavioral control on actions. Our study builds on standard operationalizations and a body of research in social psychology. Data from seven group model-building cases indicate that group model building changes attitudes, subjective norms and intentions. Conclusions for group model-building practice are that participants often do not recognize mental model changes and that strengthening of participants' feeling of control needs special attention.

No abstract