By George P. Richardson and David F. Andersen
2. David F. Andersen (biography)
What can you do if you are in a group that is trying to deal with problems that are developing over time, where:
- root causes of the dynamics aren’t clear;
- different stakeholders have different perceptions;
- past solutions haven’t worked;
- solutions must take into account how the system will respond; and,
- implementing change will require aligning powerful stakeholders around policies that they agree have the highest likelihood of long-term success?
The fields of systems thinking and system dynamics modelling bring four important patterns of thought to such a group decision and negotiation:
- thinking dynamically;
- thinking in stocks and flows;
- thinking in feedback loops; and,
- thinking endogenously.
Thinking dynamically
This refers to thinking about problems as they have developed over time and will play out in the future. The principal tool to facilitate dynamic thinking is graphs over time. Sketching graphs over time helps groups move from a focus on separate dramatic events to a focus on the persistent, often almost continuous, pressures giving rise to the discrete events we see.
Thinking in stocks and flows
Stocks are accumulations and flows are their rates of change. Therefore, thinking in stocks and flows focuses on populations, physical stocks, inventories, backlogs, and other accumulating characteristics central to the problem, and on the production capacities, resources, and distinctive competencies available to deal with the problem.
Stocks change gradually over the time frame of interest, growing or declining as inflows compete with outflows. System capacities result not from quick changes, but from sustained investment. System policies must work through flows to change key stocks over time.
Thinking in feedback loops
This focuses on circular causality, the likely extended ramifying effects of actions taken by actors in the system. Feedback loops are a source of policy resistance. Exposing reinforcing and balancing feedback loops, active or latent in system structure, gives planners the opportunity to avoid the natural tendencies of complex systems to compensate for or counteract well-intentioned policy initiatives.
Thinking endogenously
This is the most powerful aspect of systems thinking. Thinking endogenously refers to the effort to see the ‘system as cause’, to extend the boundary we naturally place around our thinking about a problem to the point that root causes are seen not as independent forces from outside but linked in circular causal loops with internal forces over which we might have some control. ‘Systems thinking’ drives many apparently diverse schools of thought, but at the core of them all is the mental effort to uncover endogenous sources of system behaviour.
Combining these patterns of thought
These patterns of thought underpin a group model building process using system mapping and modelling that integrates data, other empirical insights, and mental models into strategy and policy processes.
Strategic policy making begins with the pre-existing mental models and policy stories brought into the room by stakeholders affected by the problem and by stakeholders charged with dealing with the problem.
Strategic policy consensus and direction emerge from a process that combines social facilitation with technical modelling and analysis. The method blends dialogue with data. It begins with an emergent discussion and ends with an analytic framework that moves from ‘what is’ baseline knowledge to informed ‘what if’ insights about future policy directions.
The modelling processes themselves are not described here, but can be found in Richardson and Andersen (2010).
There are a number of process features related to building these models that contribute to their appeal for stakeholders:
- Engagement. Stakeholders are in the room as the model is evolving, and their own expertise and insights drive all aspect of the analysis.
- Mental models. The model building process uses the language and concepts that stakeholders bring to the room with them, making explicit the assumptions and causal mental models stakeholders use to make decisions.
- Complexity. The resulting nonlinear simulation models lead to insights about how system structure influences system behaviour, revealing understandable but initially counterintuitive tendencies like policy resistance or ‘worse before better’ behaviour.
- Alignment. The modelling process benefits from diverse, sometimes competing points of view as stakeholders have a chance to wrestle with causal assumptions in a group context. Often these discussions realign thinking and are among the most valuable portions of the overall group modelling effort.
- Refutability. The resulting formal model yields testable propositions, enabling stakeholders to see how well their implicit theories match available data about overall system performance.
- Empowerment. Using the model, stakeholders can see how actions under their control can change the future of the system.
Group modelling merges stakeholders’ causal and structural thinking with the available data, drawing upon expert judgment to fill in the gaps concerning possible futures. The resulting simulation models provide powerful tools for strategy and policy development.
Do you have experience to share with group model building using system mapping and modelling? Do the features outlined above resonate with your experience? Do you have other tips to share?
To find out more:
Richardson, G. P. and Andersen, D. F. (2010). Systems thinking, mapping, and modeling for group decision and negotiation. In, C. Eden and D. N. Kilgour (eds.), The handbook for group decision and negotiation. Springer: Dordrecht, The Netherlands: 313-324.
Biography: George P. Richardson PhD is Emeritus Professor of Public Administration and Policy in the Rockefeller College of Public Affairs and Policy at the University at Albany, State University of New York, USA and affiliated Professor of Informatics in the College of Computing and Information. He is the author of ‘Introduction to System Dynamics Modeling with DYNAMO’ (1981) and ‘Feedback Thought in Social Science and Systems Theory’ (1991, 1999), both of which were honored with the System Dynamics Society’s Jay W. Forrester Award. He founded the System Dynamics Review and later served for seven years as its Executive Editor. He has been honored with awards for Excellence in Teaching (2003) and Excellence in Academic Service (2010). In 2011, the System Dynamics Society recognized him with its award for Outstanding Service for his contributions to the Society and the field.
Biography: David F. Andersen PhD is the O’Leary Distinguished Service Professor, Emeritus, of Public Administration, Public Policy, and Information Science at the Rockefeller College, University at Albany, State University of New York, USA. His work centres on applying system dynamics, systems thinking, and information technology approaches to problems in the public, not-for-profit, and private sectors. He has served as a technical consultant to organizations at the federal, state, and local levels. He is co-author of ‘Introduction to Computer Simulation: The System Dynamics Modeling Approach’ (winner of the Forrester Award in 1983) and ‘Government Information Management’.
The recent posts on knowledge and ignorance https://i2insights.org/2019/09/24/knowledge-and-unknown-unknowns/ were useful to my learning of my place in system change discussion. Especially helpful in thinking of knowledge over time and context.
I’m trying to learn about systems and governance and as a front line mental health worker who has an overall sense, perhaps “magical thinking” – as they say in the game, of the systems in play in mental health. The focus on gaps and needs that are woven in “the science” of system change are the drivers, yet as your article lays out, the 4 patterns need to be engaged with.
I’m trying to get around to engaging in my small, bottom up, service level way with no legitimacy or place to speak of policy and strategy other than involvement with a small reference/advisory group which has aspirations and goals of system improvement. For sure I have no place to address:
“implementing change will require aligning powerful stakeholders around policies that they agree have the highest likelihood of long-term success?”
Yet it is useful, from my context to deploy the “magical thinking” of system policy change even though the system itself is nowhere near such an approach and even though the systems stakeholders pretend outwardly they are.
Thanks for listening to the above, had to lay out my own context.
What I wanted to say on my own learning context is about the organizational context, the organization as a system –as a vital ingredient to address for any of this to be able to shift. Not that this is news to anyone working in implementation science. I stumbled, (I will spare the story) on Barry Oshry’s “Seeing Systems: Unlocking the mysteries of organizational life,” and this video https://www.youtube.com/watch?v=VYpO-pK7IFA where he lays out: top, bottom and middle worlds and how structure shapes our consciousness, our separateness and how we adapt, our feelings and emotions within the system.
His approach is a critical ingredient to talk of system change and helps us address: “system blindness,” “change the pattern they are in… we have to see what is missing in the system…. our integrating with one another.”
You lay out the dynamic very well, and the challenge of thinking “flow” in programs, care systems – while aiming for mental wellness within: programs vs. population health. I’m going to try out your approach from the context of my little program and let you know how it goes. I posted about your article, though with a focus on engaging the front line as I find there is way too much disconnection with the implementors of care.
https://socialhealthpracticeottawa.wordpress.com/2019/08/03/embracing-the-idea-we-are-working-within-a-system-article-from-integration-and-implementation-insights-helps-us-to-move-to-long-term-action-within-our-systems/
I think you are looking in the right place, the gap or disconnection (it usually is one) between thinking about the ‘system’ and thinking about the individuals in it whom we care a lot about. What you (and we) would be looking for are ways front line people, doing their best, implement care that does not (in the aggregate) produce longer term system-wide results no one wants. We often find that feedback loops in these complex systems, which we sometimes don’t even see, gain strength and work to defeat our favorite plans. An analogy is ‘enabling behavior’ on the part of those caring deeply about an alcoholic.
Your Ottawa web site is exciting to see.
Just wanted to say thank you both; helpful to a front line mental health worker, trying to work in advancing… practices. You put so deliberately put substance to the idea of systems, it rings true of the level of breadth involved along with the aim, as you state “the long term” instead of “quick changes.”
Love to hear more about how thinking any of the four patterns of thinking we’re describing here can be of practical help for dealing with population-wide (systemic) mental health problems. Such a difficult and important area to work in. We all can imagine a sequence of stocks of people from the group not having any apparent difficulties, to a (growing?) group of people showing signs of mental health problems, to people whose lives are seriously damaged by such problems, eventually to people who have to be in some form of (institutional?) care. And I suspect that most of the money spent on systemic mental health problems are toward the more extreme end of that sequence of stocks, when probably our greatest leverage would be toward the beginning of that tragic slide.