All entries for Saturday 25 October 2008
October 25, 2008
Yesterday during Lila's presentation there was a rather heated debate by Francisco on whether the word "imperial" was accurate to describe Deming's idea of leadership.
After reading the chapter on "Some New Principles of Training and Leadership" from Out of The Crisis, I thought I could reflect a bit on what his idea of leadership is about. The following is an extract from page 248
"The aim of leadership should be to improve the performance of man and machine, to improve quality, to increase output, and simultaneously to bring pride of workmanship to people. Put in a negative way, the aim of leadership is not merely to find and record failures of men, but to remove the causes of failure: to help people to do a better job with less effort." (Deming, 1982; p.g. 248)
If we take that to be Deming's idea of leadership, then it seems clear that there isn't anything remotely suggests Deming's ideal leadership style is imperial, commanding, and one above all e.t.c
Further, Deming lists three things which he considers to be the responsibility of a leader and they are 1) remove obstacles 2) improve the system and 3) achieve consistency.
What does this mean?
Firstly, leaders should identify what causes the worker to perform at sub-optimal level (not in statistical control) and try to remove it. Then through improving the system, the leader creates an environment where workers can fully perform to the best of capability (therefore in statistical control) and generate greater satisfaction. Lastly , leaders should find ways to reduce differences in performances between people and/or occurence.
From the above explanation we see the idea of statistical control come up quite often. Why is that? Isn't statistical control something is is used in ensuring quality of products is consistent? I too, found this puzzling at first but I soon realised Deming has moved a step beyond using SPC for quality management of product and applied it in all sort of management practice including the management of people.
Using an example of golf student, he says , "before the lessons (the golfer's score) were obviously not in a state of control: There are points outside the control limits. Then came lessons. His scores thereupon showed a state of statistical control with the desired result... Here, lessons changed the system."
His points, I think, is to use statiscal methods to understand the variation in worker's performance, and find ways to test for the effect of any improvement measures. Such as lessons to improve performance.
This tool also useful in identifying which worker is worthy of of management attention. An unlearned worker in a state of chaos has potential to improve and whilst a learned worker performing sub-optimally yet is in the state of control must be removed from the system.The following quote should illustrate this idea.
"Anyone, when he has brought his work to a state of statiscal control, whether he was trained well or badly, is in a rut" (Deming, 1982; p.g. 249)
Note: Rut= a usual or fixed practice ; especially : a monotonous routine
"He has completed his learning of that particular job. It is not economical to try to provide further training of the same kind. He may nevertheless, with good training, learn very well some other kind of job." (Deming, 1982; p.g. 249)
Is this method suitable for application in the management of people? I for one hold a more reserved view. One of the concern I have is how it go about establishing the upper and lower control limits because these are crucial in deciding whether a process is in state of statistical control. Furthermore, whether such limits when constructed from a small sample size such as in the author's examples, accurately reflect the control limits when derived from a much larger sample. As I understand it, there is usually a minimum required sample size for a conclusion drawn from statistical tool to be reliable. This point wasn't really discussed in the book. Unless the examples in the book are for illustrative purposes only rather than the orthodox way to carry out this procedure.