The complexity of concept mapping for policy analysis

Abstract

Concept mapping is a participatory mixed methodology that enables diverse participant groups to develop shared conceptual frameworks that can be used in a variety of policy contexts to identify or encourage complexity, and the adaptive emergent properties associated with it. The method is consistent with an evolving paradigm of complex adaptive systems thinking and helps groups address complexity in several ways: it is inductive, allowing shared meaning to emerge; it is based on a simple set of rules (operations) that generate complex patterns and results; it engages diverse agents throughout the process through a range of participation channels (synchronous or asynchronous web, face-to-face, etc.); the visual products the concept maps, pattern matches, action plots provide high-level representations of evolving thinking; the results are generative, encouraging shared meaning and organizational learning while preserving individuality and diversity; the maps themselves provide a framework that enables autonomous agents to align action with broader organizational or systems vision. The concept mapping process involves free listing, unstructured sorting and rating of ideas, and a sequence of statistical analyses (multidimensional scaling, hierarchical cluster analysis) that produce maps and other results that the participants then interpret. An example is provided of a web-based project that mapped the practical challenges that need to be addressed to encourage and support effective systems thinking and modeling in public health work. It is suggested that using concept mapping especially in combination with other types of human simulation provides a valuable addition to our methodological tools for studying complex human systems.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 43 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.