strategy

The use of System Dynamics within strategy courses aims to teach students dynamic theories to explain firm performance. This paper presents an analysis of the development of students' strategic systems thinking skills after taking a course on strategic development processes. Students that achieved the best performance employed graphs representing the performance of key variables, described how the strategy worked using verbal explanations of feedback processes, and validated the results obtained to detect poor or unreasonable strategies. However, a large group of students did not follow System Dynamics practices and performed poorly in their strategies. We extract two lessons: more emphasis on the areas shown by the best students, which are along the modeling for learning practices, are needed to improve students' strategic systems thinking skills, and it is important to adopt multidimensional tools to evaluate systems thinking skills given the diversity in learning preferences. Copyright © 2012 System Dynamics Society.

This paper deals with the representation of the conceptual structure of mental models of dynamic systems (MMDS). Heretofore, this structure has not been consistently defined. Consequently, studies about MMDS continue to use different conceptual structures to measure mental models. Even such properties as feedback loops and delays, which lie at the core of dynamic systems, are often not considered. This situation leads to incompatible findings and stagnating research. We review the literature about mental models in the field of system dynamics. In addition, we refer to dynamic systems theory as the mathematical basis for system dynamics to complement and validate our conceptual structure. One may conclude that most of the existing mental model studies measure only parts of the structure that we propose. The paper's contribution is to elaborate the conceptual structure of an MMDS and to use this structure to operationally enhance the definition of an MMDS. Copyright © 2012 System Dynamics Society

Controlled experiments, long used in psychology and economics, are increasingly being used to develop theories in the areas of interest to management science, including decision making and systems. In this paper we review laboratory experiments relating to system dynamics. We discuss how these experiments help to (1) determine causes for subjects' poor performance in dynamic environments and (2) model such behaviour in system dynamics models. Finally, we present the main insights from the works reviewed and propose some ideas for a research agenda. Copyright © 2012 System Dynamics Society

The misuse of Personal Protective Equipment results in health risk among smallholders in developing countries, and education is often proposed to promote safer practices. However, evidence point to limited effects of education. This paper presents a System Dynamics model which allows the identification of risk-minimizing policies for behavioural change. The model is based on the IAC framework and survey data. It represents farmers' decision-making from an agent-oriented standpoint. The most successful intervention strategy was the one which intervened in the long term, targeted key stocks in the systems and was diversified. However, the results suggest that, under these conditions, no policy is able to trigger a self sustaining behavioural change. Two implementation approaches were suggested by experts. One, based on constant social control, corresponds to a change of the current model's parameters. The other, based on participation, would lead farmers to new thinking, i.e. changes in their decision-making structure. Copyright © 2011 System Dynamics Society

Although it has been suggested that people use the wrong cognitive procedures in solving stock and flow (SF) problems, we know little of what these mental procedures are. We present two experiments aimed at demonstrating the influence of analogical reasoning on SF failure. Results of Experiment 1 show that SF failure decreases when people are asked to compare problems that share behavioral similarity (common relations). However, the benefit of behavioral similarity depends on the surface similarity (common superficial object attributes). Results from Experiment 2 demonstrate that when the behavioral characteristics of the problems are unknown by the participants, the process of comparing two problems with behavioral similarity improves responses to a subsequent SF problem, regardless of the surface similarity between the problems. Surface similarity helps only when the behavioral similarity between the problems is already known. Implications for training and education in system dynamics are discussed. Copyright © 2011 System Dynamics Society