Review and summary of "Selling system dynamics to (other) social scientists", by Nelson P. Repenning, System Dynamics Review Vol 19 No. 4 (Winter 2003) 303-327,
Published on-line by Wiley InterScience. Accessed 8/13/05 (subscription may be required)
Author: "Nelson P Repenning is the J. Spencer Standish Associate Professor of Management at the MIT Sloan School of Management. His work focuses on understanding the factors that contribute to the successful implementation, execution, an design of business processes."
Quotes:
"Can one do research that both meets the standards of good system dynamics practice and is acceptable to other social scientists? ... practitioners resonated with this line of work almost immediately. Academics, however, were not similarly receptive."
I [initially] interpreted these reactions as evidence of irreconcilable differences - similar to those outlined by Meadows (1980) - between systems dynamics and other social science disciplines."
"Changing the labels to match the existing literature was a critical step in gaining acceptance for my model."
"The referees were, quite correctly, critical of a paper that proposed to link these concepts but failed to cite any of the previous attempts to define and understand them."
"I now believe my difficulties were rooted in my failure to build the reader's intuition. ... A claim that model is trivial is a s much a statement about the reader's understanding as it is about the underlying model.... People often do not recognize gaps in their intuition... Any discomfort that readers do experinece from not fully understanding a model is likely to manifest itself in indirect sways.
Suggesting that a model is "too complicated' that the "Insights are trivial", an that more analysis is needed are all likely responses.
"The answer is, I now believe, straightforward, if sobering: standard modes of presenting SD models, while potentially effective for SD audience, are ineffective when presenting to non-SD audiences, even those weth technical backgrounds. ...the popularity of modeling and estimation methods that build on the assumptions of equilibrium and mono-causality suggest that the social scenties will find it more, not less, difficult to develop intuition from a system model.
...I now believe that many of the difficulties I experienced in selling my work were, to a large degree, self inflicted....
I have found it easier to s ell my work to those scholars whose primary interest lies in understanding real world phenomenon.
...my sense is ... that many of the errors will onnot change without significant intervention.
"The consequence of using mainstream economics as a referent (in the 60's and 70's) is that movers of the field found little common ground with the rest of social science world (s they defined it). The early focus on economic appears to have fueled a vicious cycle of increasing isolation from the rest of social science that persists to this day."
"As a dedicated student of system dynamics, to me the conclusion that we are at least party responsible for the situation we now face seems inescapable.... The result is a community that, today, is largely isolated from mainstream social science. . The use of new modeling methods is on the rise in other parts of social science but this growth has not included system dynamics. For example in organization theory, the field with which I am most familiar, papers in so-called "complex adaptive systems " models and agent-based representations far out-nmber system dynamics models, although a strong case can be made that the latter is more appropriate to the task at hand.
The system dynamicist must sell her client on the fundamental premise of her enterprise; the forecaster faces no such barrier.
Gaining such sanction, however requires that scholars who use systems dynamics get and keep jobs at management and other professional schools. Developing research using SD that is of interest to the larger social science community is ... central to achieving the mission ...
No comments:
Post a Comment