MISSING Subcommand (FACTOR command)

MISSING controls the treatment of cases with missing values.

• If MISSING is omitted or included without specifications, listwise deletion is in effect.
• MISSING must precede the ANALYSIS subcommand.
• The LISTWISE, PAIRWISE, and MEANSUB keywords are alternatives, but any one of them can be used with INCLUDE.

LISTWISE. Delete cases with missing values listwise. Only cases with nonmissing values for all variables named on the VARIABLES subcommand are used. Cases are deleted even if they have missing values only for variables listed on VARIABLES and have valid values for all variables listed on ANALYSIS. Alias DEFAULT.

PAIRWISE. Delete cases with missing values pairwise. All cases with nonmissing values for each pair of variables correlated are used to compute that correlation, regardless of whether the cases have missing values for any other variable.

MEANSUB. Replace missing values with the variable mean. All cases are used after the substitution is made. If INCLUDE is also specified, user-missing values are included in the computation of the means, and means are substituted only for the system-missing value. If SELECT is in effect, only the values of selected cases are used in calculating the means used to replace missing values for selected cases in analysis and for all cases in computing factor scores.

INCLUDE. Include user-missing values. Cases with user-missing values are treated as valid.