Subject: Systems Language - closed loops

Posted by Eric Stiens on 2/5/2010
In Reply To:Systems Language - closed loops Posted by Niall Palfreyman on 2/5/2010

Message:

> Here it is no longer a question of two separate agents influencing each other, but rather a process influences a stock, which in return dictates (not influences) the process. There is something about this SD way of viewing the loop which my students find difficult to absorb, and it seems to them a needless abstraction.

Ah, the first order loop. The exponential decay loop can be even harder to get across at first because the causality is moving opposite the direction of the outflow.

I have found that disaggregating than reaggregating (for example, splitting the rabbit stock into rabbit babies and rabbit adults, with the amount of rabbit adults affecting the birth inflow, and then just reaggregating into one rabbit stock) can be helpful , as well as switching back and forth between a S/F view and a CL view. Also continuing to make the point that the flows are actual units over time (though abstracting this into a percentage growth rate etc seems to be a problem for older students, not younger students)

Another point I make is that "nothing changes without a cause" -- so in this case if you can show that 10 rabbits are being born every year at year 1 and 20 rabbits are being born every year at year 7 the question that follows is -- what changed the amount of rabbits being born?

I think you are probably right that to your students it seems so drop-dead obvious that why even note it (duh! more rabbits are alive to have more babies!) and it might well be awhile until noting it carries any deep-level systemic insights with it, rather than just being a formal exercise ("I do this because I'm supposed to") -- but really hammering the point early that nothing can change without a cause helps later on when aspiring model-builders start doing strange things because their implicit understanding of the system is still much greater than their skill at externalizing and making explicit