All models are wrong…
I am a regular reader of the AOK-Forum list hosted on Yahoo! A recent post had links to a series of fascinating diagrams. One of them had a list of features that one should consider in constructing a model and thus help us to make sense of the world. George Box, the industrial statistician, is credited with the quote ‘all models are wrong, some are useful’. The characteristics of a good model contained in the link should help us to make our models more useful.
About Mark Schenk
Mark works globally with senior leadership teams to improve their ability to communicate clearly and memorably. He has been a Director of Anecdote since 2004 and helped the company grow into one of the world’s leading business storytelling consultancies. Connect with Mark on:
Comments
Comments are closed.
List of features of models
Idiagram has published an excellent list of features that all conceptual models should share. To quote: Broadly speaking we use the term ‘model’ to refer to any structured knowledge that accurately reflects and enables us to make sense of the…
I am not sure whether it was George Box or W. Edwards Deming who should be qouted for “all models are wrong, some models are useful” as I have seen both qouted for this in various articles. Who does it belong to?
I thought it was Deming at first but did some searching and found that Box used it as a heading in a book chapter in 1979. Here is the citation:
Box, G.E.P., Robustness in the strategy of scientific model building, in Robustness in Statistics, R.L. Launer and G.N. Wilkinson, Editors. 1979, Academic Press: New York.
a great model on models above guys, now i wil have to go back and reflect on the one’s i have drawn!
I am confident this quote is Box’s. See artilces by George Box, if you are interested. Also see some of my thoughts on Deming’s ideas.
Regardless of who the quote belongs to, it is a classical example of making sense out of nonsense.