- Can the "factor-tree" methodology for qualitative causal social-science modeling be usefully extended to uncertainty-sensitive computational modeling?
- If so, can it be done so as to promote reviewability, reusability, and composability?
This report builds on earlier RAND research (e.g., Understanding and Influencing Public Support for Insurgency and Terrorism, 2012) that reviewed and integrated social science relevant to terrorism and insurgency. That research used qualitative conceptual causal models called "factor trees" to identify the factors that contribute to various aspects of terrorism or insurgency at a slice in time and how the factors relate to each other qualitatively.
This report goes beyond the conceptual and qualitative by specifying a prototype uncertainty-sensitive computational model for one of the factor trees from the earlier research, one that describes public support for terrorism and insurgency. The authors first detail their approach to designing such a model, emphasizing the challenges they encountered in assigning mathematical meaning to the factor tree's numerous factors and subfactors, identifying suitable "building block" combining algorithms, and the uncertainty in their values and the relationships among them. They then describe how they implemented the model in a high-level visual-programming environment, show how the model can be used for exploratory analysis under uncertainty, and discuss their initial experience with it.
Methodologically, the work illustrates a new approach to causal, uncertainty-and-context-sensitive, social-science modeling. It also illustrates how such models can be reviewable, reusable, and potentially composable.
Qualitative 'Factor Trees' Can Structure Social-Science Causal Knowledge Systemically and Be the Basis for Uncertainty-Sensitive Computational Modeling That Reflects Limitations of Knowledge
- Challenges to building a computational model based on a factor tree involved (1) defining the factors and their values; (2) defining how to reflect relationships among factors within the tree and the varied significance of influences among factors; (3) dealing with uncertainty about factor values and combining rules, and showing results of exploratory analysis across uncertainties; and (4) implementing the model in a computer program in which substantive content is transparent, comprehensible, and as language-independent as possible.
- The work found approaches to overcome these challenges and demonstrated that the extension of the "factor tree" methodology to uncertainty-sensitive computational modeling is possible and that the result can be reviewed, understood, and either used or adapted by other researchers.
- The model and documentation should be distributed to the wider community for evaluation and experiments in model composition.
- The approach used should be applied more generally as part of pulling together, substantively reviewing, encapsulating, and exploiting "modules" of social-science knowledge relevant to counterterrorism and other subjects.
Table of Contents
Specifying the Model
Implementation in a High-Level Language
Looking Ahead to Exploratory Analysis Under Uncertainty
Using the Model for Knowledge Elicitation, Discussion, and Diagnosis
Primer on Factor Trees (a reprint)
Verification and Validation
Eliciting Factor Values
Mathematics for "And" and "Or" Relationships
The research described in this report was prepared for the Office of the Secretary of Defense (OSD). The research was conducted within the RAND National Defense Research Institute, a federally funded research and development center sponsored by OSD, the Joint Staff, the Unified Combatant Commands, the Navy, the Marine Corps, the defense agencies, and the defense Intelligence Community.
This report is part of the RAND Corporation technical report series. RAND technical reports may include research findings on a specific topic that is limited in scope or intended for a narrow audience; present discussions of the methodology employed in research; provide literature reviews, survey instruments, modeling exercises, guidelines for practitioners and research professionals, and supporting documentation; or deliver preliminary findings. All RAND reports undergo rigorous peer review to ensure that they meet high standards for research quality and objectivity.
Permission is given to duplicate this electronic document for personal use only, as long as it is unaltered and complete. Copies may not be duplicated for commercial purposes. Unauthorized posting of RAND PDFs to a non-RAND Web site is prohibited. RAND PDFs are protected under copyright law. For information on reprint and linking permissions, please visit the RAND Permissions page.
The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.