Feeds:
Posts
Comments

Archive for November, 2010

Some of us never get enough punishment which is why I decided to start a PhD Leadership degree at the age of 40.   Needless to say it is an adjustment.  One of the things I vowed to myself before starting was that I would not loose touch with the outside world while in the program.  Hence, this weekend I decided to try and catch up on current events only to discover that all universes end up colliding with each other.

Here is a link to a fascinating Charlie Rose interview of Siddhartha Mukherjee regarding his new book The Emperor of all Maladies: A Biography of Cancer

Interview: http://www.charlierose.com/guest/view/7009

Book: http://books.simonandschuster.com/Emperor-of-All-Maladies/Siddhartha-Mukherjee/9781439181713

In Statistics class, we are reading book Dirty Rotten Strategies: How We Trick Ourselves and Others Into Solving the Wrong Problems Precisely.  The book goes beyond Type I and Type II errors into what the authors describe as Type 3 and 4 errors.
Here’s how the authors explain types of errors:
llll
Type I: wrongly conclude that new drug is better than old one (aka reject a true null hypothesis)
Type II: wrongly conclude that old drug is better than new one (aka accept a wrong null hypothesis)
Type III: we unintentionally fool and trick ourselves into solving the wrong problem precisely, but we don’t necessarily force our definitions onto others
Type IV: we intentionally force others into solving the wrong problems precisely, to their detriment and for our gain and benefit.
llll
The definitions do not imply malice, just things to be careful about (although authors would probably disagree with me on this hence book title).
llll
While listening to the Mukerjee interview, he describes the evolution of thought with respect to radical mastectomies and I saw it as a clear example of a Type IV error.  According to Mukherjee, proponents of the radical mastectomy procedure  inflicted damage on thousands of womens bodies until patients started demanding empirical evidence which showed there was no difference in benefit (i.e. nothing gained) between regular mastectomy, radical mastectomy and lumpectomy plus radiation.  Hence, those who underwent the procedure had been intentionally convinced to solve the wrong problem precisely (i.e. instead of focusing on curing cancer the focus became the removal of the breast and surrounding tissue in a very precise manner).
llll
In addition, Mukherjee points out that the term “radical” in and of itself caused group think (i.e. what patient when trying to fight for his/her life would refuse “radical” treatment?). This group think enabled the solving of the wrong problem to continue until a select few asked for what the other book we are reading The Black Swan: The Impact of the Highly Improbable refers to as silent evidence (i.e. data proving there was no benefit) to be revealed.
llll
It would be hard to say that doctors practicing the radical mastectomy approach were maliciously trying to mutilate women.  What they failed to do was verify their assumptions were correct before inflicting permanent damage.  To put the group think into philosophical terms, they were blinded by their own semiotics (cultural signs, in this case medical culture, such as nomenclature and conventions, aka biases, that reinforced their belief the radical mastectomy approach was correct, i.e. caused detection errors).
llll
For details on semiotics you can view:
lll
llll
51:31min Yale University lecture on Semiotics and Structuralism http://www.youtube.com/watch?v=VsMfaIOsT3M&feature=related
llll
All this begs the question: how do we avoid such egregious errors, especially as we practice personal leadership (leading through our own behavior and beliefs)….????
lll
I’m still letting this sink in; old noodles & new material take a while to blend together…
lll
Meanwhile, any thoughts or comments on the subject of how to avoid such errors are most welcome!
Advertisements

Read Full Post »