STANDARDIZE Subcommand (PROXIMITIES command)
Use STANDARDIZE to standardize data values for either cases or variables before
computing proximities. One of two options can be specified to control
the direction of standardization:
VARIABLE. Standardize the values for each variable. This setting is the default.
CASE. Standardize the values within each case.
Several standardization methods are available.
These methods allow you to equalize selected properties of the values.
All methods can be used with either VARIABLE or CASE. Only one standardization
method can be specified.
- If
STANDARDIZEis omitted, proximities are computed by using the original values (keywordNONE). - If
STANDARDIZEis used without specifications, proximities are computed by using Z scores (keywordZ). -
STANDARDIZEcannot be used with binary measures.
NONE. Do not standardize. Proximities are computed by using the original values. This setting
is the default if STANDARDIZE is omitted.
Z. Standardize
values to Z scores, with a mean
of 0 and a standard deviation of 1. PROXIMITIES subtracts the mean value for the variable
or case from each value that is being standardized, and then PROXIMITIES divides by the standard deviation.
If the standard deviation is 0, PROXIMITIES sets all values for the case or variable to 0. This setting is
the default if STANDARDIZE is
used without specifications.
RANGE. Standardize
values to have a range of 1. PROXIMITIES divides each value that is being standardized
by the range of values for the variable or case. If the range is 0,
PROXIMITIES leaves all values
unchanged.
RESCALE. Standardize
values to have a range from 0 to 1. From each value that
is being standardized, PROXIMITIES subtracts the minimum value and then divides by the range for the
variable or case. If a range is 0, PROXIMITIES sets all values for the case or variable to 0.50.
MAX. Standardize
values to a maximum magnitude of 1. PROXIMITIES divides each value that is
being standardized by the maximum value for the variable or case.
If the maximum of the values is 0, PROXIMITIES divides each value by the absolute magnitude of the smallest value
and adds 1.
MEAN. Standardize
values to a mean of 1. PROXIMITIES divides each value that is being standardized by the mean of the
values for the variable or case. If the mean is 0, PROXIMITIES adds 1 to all values for the
case or variable to produce a mean of 1.
SD. Standardize
values to unit standard deviation. PROXIMITIES divides each value that is being standardized
by the standard deviation of the values for the variable or case. PROXIMITIES does not change the values
if their standard deviation is 0.
Example
PROXIMITIES A B C
/STANDARDIZE=CASE RANGE.
- Within each case, values are standardized to have ranges of 1.