SAVE Subcommand (FACTOR command)

SAVE allows you to save factor scores from any rotated or unrotated extraction as new variables in the active dataset. You can use any of the three methods for computing the factor scores.

  • Only one SAVE subcommand is executed. If you specify multiple SAVE subcommands, only the last is executed.
  • SAVE must follow the last ROTATION subcommand.
  • If no ROTATION subcommand is specified after the last EXTRACTION subcommand, SAVE must follow the last EXTRACTION subcommand and no rotation is used.
  • If neither ROTATION nor EXTRACTION is specified, SAVE must follow the last ANALYSIS subcommand and the default extraction and rotation are used to compute the factor scores.
  • SAVE subcommands before any explicitly specified ANALYSIS, EXTRACTION, or ROTATION subcommands are ignored.
  • You cannot use the SAVE subcommand if you are replacing the active dataset with matrix materials. (See the topic Matrix Output (FACTOR command) for more information. )
  • The new variables are added to the end of the active dataset.

Keywords to specify the method of computing factor scores are:

REG. Regression method. This is the default.

BART. Bartlett method.

AR. Anderson-Rubin method.

DEFAULT. The same as REG.

  • After one of the above keywords, specify in parentheses the number of scores to save and a rootname to use in naming the variables.
  • You can specify either an integer or the keyword ALL. The maximum number of scores you can specify is the number of factors in the solution.
  • FACTOR forms variable names by appending sequential numbers to the rootname you specify. The rootname must begin with a letter and conform to the rules for variable names. For information on variable naming rules, see Variable Names .
  • If you do not specify a rootname, FACTOR forms unique variable names using the formula FACn_m, where m increments to create a new rootname and n increments to create a unique variable name. For example, FAC1_1, FAC2_1, FAC3_1, and so on will be generated for the first set of saved scores and FAC1_2, FAC2_2, FAC3_2, and so on for the second set.
  • FACTOR automatically generates variable labels for the new variables. Each label contains information about the method of computing the factor score, its sequential number, and the sequential number of the analysis.

Example

FACTOR VARIABLES=V1 TO V12
  /CRITERIA FACTORS(4)
  /ROTATION
  /SAVE REG (4,PCOMP).
  • Since there is no EXTRACTION subcommand before the ROTATION subcommand, the default principal components extraction is performed.
  • The CRITERIA subcommand specifies that four principal components should be extracted.
  • The ROTATION subcommand requests the default varimax rotation for the principal components.
  • The SAVE subcommand calculates scores using the regression method. Four scores will be added to the file: PCOMP1, PCOMP2, PCOMP3, and PCOMP4.