Talk:What is 6 sigma?


 * if it is called 6 sigma, why does it cover +/- 4.5 standard deviations..why not call it 9 sigma?
 * The 1.5 sigmas added to the name Six Sigma are arbitrary and they are called "1.5 sigma shift" (SBTI Black Belt material, ca 1998).
 * y-axis on the six sigma figure for product tolerance +/- 0.05 is not correct. It is unlikely that the maximum value would be 1.  this axis is misleading.
 * Note that in the introduction of analysis methods, minitab is introduced. Actually excell does these same calculations, as does Mathematica.
 * Where did the figures for the control charts come from? Are these new or taken from another source?  Is that source open-source?
 * For example 2 there is a small problem in that the flow rates appear to be sampled frequently enought that they are not independent measurements. The statistical criteria for control charts assume that measurements are independent.  As a thought experiment, assume that the technician had measured the data 100x more frequently.  By measuring more frequently the technician would call the process out of control more frequently.
 * For example 2 also, it seems that the first part of the data is a better indication that the system is out of statistical control.. why focus on the second half of the data when the first looks worse?
 * How come 1 standard deviation has 68% of a sample within its range, but with 6 sigma, 1 standard deviation has 70% outside it's range?


 * The formula for Standard Deviation is incorrect. It shows 1/N under the radical and is the "uncorrected sample standard deviation (Sn)) and generally not used. and it should be 1/(N-1) which is the "Corrected Sample Standard Deviation". The text describes it correctly, however the graphic is wrong.|Standard Deviation--Navywings 14:49, 26 July 2013 (EDT)
 * The percentage for 3.4 DPMO in the table is wrong. (1-3.4)/1000000 = .9999966 (hence the term "five 9's).|Six Sigma for Dummie Cheat Sheet--Navywings 14:49, 26 July 2013 (EDT)