Using Textual Data in System Dynamics Model Conceptualization

Abstract

Abstract:Qualitative data is an important source of information for system dynamics modeling. It can potentially support any stage of the modeling process, yet it is mainly used in the early steps such as problem identification and model conceptualization. Existing approaches that outline a systematic use of qualitative data in model conceptualization are often not adopted for reasons of time constraints resulting from an abundance of data. In this paper, we introduce an approach that synthesizes the strengths of existing methods. This alternative approach (i) is focused on causal relationships starting from the initial steps of coding; (ii) generates a generalized and simplified causal map without recording individual relationships so that time consumption can be reduced; and (iii) maintains the links from the final causal map to the data sources by using software. We demonstrate an application of this approach in a study about integrated decision making in the housing sector of the UK.

Assessing the efficacy of group model building workshops in an applied setting through purposive text analysis

“…The results also show how GMB produces higher quality models than can be obtained by eliciting individual mental models in isolation. The insights gleaned from this work have also demonstrated a beneficial use of PTA for generating CLDs of collective and shared mental models from multiple individual interview transcripts (Kim, 2009b;Kim and Andersen, 2012;Turner et al, 2013;Eker and Zimmermann, 2016). Overall, we have demonstrated that substantial changes in alignment can be achieved through a low-resource GMB workshop that can be implemented in a substantially shorter time frame (SI 7) than much of the existing work in the community-based system dynamics literature (e.g.…”

“…The PTA process entailed coding transcribed interviews for discrete statements of causality that describe a direct, non-ambiguous relationship between a cause factor and an effect factor, and the implied or stated polarity of the relationship. To ensure the rigor of this process, the full text of the causal statement was recorded in a codebook and cataloged according to five deductive (pre-determined) codes as specified in Table 2 (Kim and Andersen, 2012;Turner et al, 2013;Eker and Zimmermann, 2016). These five codes represented areas of the text that identify the causal link, the cause factor, the effect factor, and the polarity and dynamics (i.e.…”

Research Design and Major Issues in Developing Dynamic Theories by Secondary Analysis of Qualitative Data

“…Researchers also experienced the limitations of the qualitative dataset in terms of extracting enough modeling information for a simulation model. The Grounded Theory process needs a significant amount of data [14] and effort [12] to generate enough information for modeling, as reported in the literature.…”

“…System Dynamics researchers have noted the lack of detailed protocols on mixed methods in the modeling process in the last decade [7] despite the existence of successful publications developed with the mixed methodology [8]. Since then, some studies have begun to address this need [9][10][11][12][13][14][15][16][17][18][19]. This article adds to this emerging line of work by describing new insights into the use of a mixed-method research design and combining Grounded Theory and System Dynamics modeling.…”

Systems Thinking to Understand National Well-Being from a Human Capital Perspective

“…Existing formal coding methods of deriving causal maps from textual data [31][32][33][34] have a common framework of five core phases: (i) open coding to identify the themes in the data, (ii) the identification of individual causal relationships, (iii) visualizing these relationships in word-and-arrow diagrams, (iv) generalizing and simplifying these diagrams (axial coding) and (v) recording the links between the final causal map and the data source explicitly in a data source reference table.…”