Six Sigma–Metrics Reveal Secrets

Why are metrics important in Six Sigma?   According to the authors of the book Six Sigma:  The Breakthrough Management Strategy Revolutionizing the World’s Top Corporations, Ikel Harry, Ph.D., and Richard Schroder, metrics establish the difference between perception, intuition, and reality.

Let’s say there is some problem that is causing defects, a problem that is detected by an inspection process.   That outcome is referred to as “Y”.   Now let’s say that engineers get together and brainstorm to find out what the potential cause is of that visible problem, and let’s call that potential cause “X”.   In order to change the outcome “Y”, you have to change the potential cause “X”.

The Six Sigma process focuses on the potential cause “X”.   Let’s say the engineers mentioned in the paragraph above change the process so as to change the potential cause “X”, and then they see if the outcome “Y” occurs.   If “Y” still occurs, then despite their intuition that “X” caused “Y” to occur, they have found through the Six Sigma process that “X” did not indeed cause “Y” as they believed.   It separates the intuition they had about where the problem was and the reality of where it wasn’t.   That’s obviously not the end of the story, because now the engineers have to go back to the proverbial blackboard and figure out a different culprit for the hidden cause of the visible outcome “Y”.

This means that, once the hidden cause “X” is found, changed and then proved via statistics that it also changes the outcome “Y”, then this means that engineers have found a correlation between the two.   Wow, that means they are done, right?

Not necessarily!   It is possible that “X” and “Y” are BOTH caused by some other hidden variable, which we shall call “Z”.  Let me tell you a story to illustrate.   I remember in an Introduction to Psychology course back in college that the teacher was trying to explain the principle that “correlation is not causation”.   He showed a graph which was taken by a census of a lot of French rural towns where the X-axis had the number of storks cited in a given year in the town, and the Y-axis had the number of babies born in that same town during that given year.   The graph showed an almost perfect correlation between the two:   as the number of storks cited in the town increased, so did the number of babies reported being born!

Someone might look at these two variables and to speculate that somehow the storks are causing the birth of the babies, as told in folk legend and the Warner Brothers Looney Tunes cartoon “The Apes of Wrath”, where a drunken stork mistakenly delivers Bugs Bunny to an expectant couple of gorillas (with hilarious consequences).

Well, I puzzled over the problem for a few seconds and them remembered a factoid I learned in French class, that in rural France, storks often make their nest in the chimneys of houses.   And so if there are more houses, there are more storks.  And also if there are more houses, there are more families, and therefore more babies.   So the “stork” variable and the “baby” variable were both independent of each other, but dependent on the variable of “houses”, and that’s why they seemed to go up together.   Not because one was causing the other, but that both were being caused by something else in the same way.

I mentioned my hypothesis to the teacher, and he said that not only was I right, but I was the only was in all of his three classes to get the answer right.   But it was only because I understood the concept he was trying to get at, AND I had a piece of information that the students didn’t possess, mainly the correlation between the storks (a variable mentioned in the problem) and the houses (which was not mentioned in the problem).   Knowing this made me solve the problem and uncover the secret “cause” of the higher number of babies:   not their conveyance via stork, but their parents’ accommodation with additional houses.

This shows that metrics, if properly understood and used, can reveal secrets that may remain hidden otherwise.   And that is only one reason for their adoption by companies as tools for measurement and change.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: