Subject: Is this a causal loop?

Posted by John Heinbokel on 2/27/2004
In Reply To:Is this a causal loop? Posted by Della Robertson on 1/27/2004

Message:

Hi Again, Della,

Steve's comments reminded me that I HAD meant to address your second question, but lost the thread as I was responding initially.

One of the nicest descriptions of tying CLDs to Stock-Flow structures (I too assume that's what you mean by "the model." I do recognize, however, that a loop is a perfectly valid 'model' in that it's a simplified depiction of the real system of interest just as is a Behavior Over Time Graph or as is a system of differential equations. Quick, stop me before I digress again!) is presented in Andy Ford's book Modeling the Environment (Island Press; his Chapters 7 and 15, in particular). One of his points, that has been pounded into me multiple times by George Richardson as well, is to start with the stocks. George is clear that every proper CLD will contain at least one stock. Figure out what that is, extract it, put it in the center of your page, and then carefully begin identifying and attaching the flow(s) that add or subtract stuff from that stock. Where then does a stock directly influence a flow (sketch in that 'connector')? Where is there a converter (or auxilliary) that belongs in a feedback loop that connects a stock to a flow? And so on until the CLD is translated into a stock flow structure that can be simulated precisely by the computer. BTW, Ford spends some valuable time discussing translating in the other direction - going from a Stock-Flow structure to a CLD. All good stuff and worth practising. IMHO it's only when we're comfortable with that sort of back and forth translation that we can really be comfortable and effective guiding the kids through these minefields.

john heinbokel