|
Refers to the mean of the squares of the
deviations from the mean. It is equivalent
to the Variance and is calculated from:
See Mean Square Deviation
Measures of dispersion measure the spread
of the data. Dispersion is central to process
improvement because a fundamental aim of
Six Sigma is to minimize the amount of variation.
The 'standard deviation' is the most common
measure of process improvement. The Mean
Square Error is a closely related concept
used in, for example, ANOVA.
In 'normal' statistics the standard deviation
is calculated using the 'RMSE' method. The
Range Method is widely used in Statistical
Process Control and Measurement Systems
Analysis.
The difference between the
largest and smallest value in a subgroup.
It can be used to estimate the process standard
deviation, see Range Method.
A method for estimating the standard deviation
used in Statistical Process Control and
Measurement Systems Analysis. It uses the
formula:
where 's' is the process standard deviation,
and  is
the mean of more than 15 (preferably 30)
samples, each containing an equal number
of items. The constant d2 is
obtained from tables. If there are 15 samples,
or less, then the formula is:
Root Mean Square Error, an alternative
term for the standard deviation, calculated
using the formula:
A measure of the dispersion of a data set.
It can be calculated using the Range Method
or the RMSE method.
A measure of the dispersion of a data set.
The variance of a population is estimated
from a sample using the formula:
|
