Title: Re: Stats, Models, Rumsfeld
At 5:52 -0800 3/18/04, Phillip Good wrote:
As I suspected you are unable to offer a single justification for your actions just as Rumsfield was unable to offer a single justification for his...

I cannot help but to send my $0.02 worth in defense of the Secretaty of Defense, Donald Rumsfeld, who, in my opinion, did a rather excellent job in putting what us human factors folks have long been trying to grasp in a layman's language (I also think that the Plain English Campaign's premier Foot In Mouth trophy he received was utterly undeserved!).  It is the unknown unknowns that James Reason ("Human Error," 1990) called "latent errors" and that are the most insidious form of human error, often taking operators of complex systems completely unawares and resulting in large-scale catastrophes.  What "known unknowns" and "unknown unknowns" have to do with statistics was, again IMHO, eloquently and succinctly captured by George Box in his often quoted statement.  When we model reality, it is of utmost importance to at least have some qualitative sense of by how much and to what direction the models' outputs differ from the true state of the world.  To uncritically take models as representative of reality, that is, to forget or ignore the "known unknowns" and "unknown unknowns" is a dangerous folly.

Esa
-- 
------------------------------------
Esa M. Rantanen, Ph.D.
Assistant Professor
University of Illinois at Urbana-Champaign
Institute of Aviation, Aviation Human Factors Division
Willard Airport-One Airport Road, Q5, MC-394
Savoy, IL  61874
Tel. 217-244-8657 (AHFD)
Tel. 217-244-7397 (Psych.)
Tel. 217-373-8276 (Home)
Fax 217-244-8647
e-mail: [EMAIL PROTECTED]
url: http://www.humanfactors.uiuc.edu
------------------------------------



Reply via email to