Abstract for: The Qualitative Systems Exploration Model (QSEM): A framework to support the structural analysis of Causal Loop Diagrams
Interpreting and analysing Causal Loop Diagrams (CLD) within Participatory System Dynamics can be challenging when justifying and prioritising key system structures. Traditionally, selections made during model interpretation can lack transparency, making it difficult to trace decisions to expert input and consensus. We introduce the Qualitative Systems Exploration Model (QSEM), a semi-quantitative framework designed to bring greater clarity and rigor to CLD analysis by integrating structured steps for quantification and prioritisation. QSEM integrates with Group Model Building (GMB) methods and contains 3 core phases: (1) System Factor Classification – categorises factors in the CLD based on local impact and/or control potential; (2) Loops of Interest – prioritises impact and/or control-based feedback structures for exploration; and (3) Archetype Identification – maps identified structures to known system archetypes, supporting insight generation for intervention and decision-making. QSEM draws from adjacency matrix methods, scenario-based techniques, and systems thinking. Applying QSEM in a commissioned government project identified the most influential factors and feedback mechanisms within the CLD. The framework highlighted high‐impact/control loops and archetypal structures, while real‐time adjustments to QSEM’s weighting coefficients and phase‐two feedback‐based loop metrics supported dynamic quantitative exploration within a qualitative paradigm, helping to prioritise system components. This guided consensus on strategic areas for change, enhanced systemic learning, and deepened participants’ understanding of system dynamics. The application of QSEM supported model interpretation and transparency in decision-making. The approach facilitated participant engagement, making it easier to identify key factors and feedback mechanisms. Further refinements, such as exploring dedicated software solutions, enhancing data visualisation techniques, better integrating with GMB Scripts, and improving archetype elucidation processes, could enhance QSEM’s scalability and broader applicability. Future work should focus on testing the framework across diverse system models and domains. Creation and/or verification of complex spreadsheet formulas for metric calculations.