MISSING Subcommand (CSDESCRIPTIVES command)

MISSING specifies how missing values are handled.

• All design variables must have valid data. Cases with invalid data for any design variable are deleted from the analysis.

The SCOPE keyword specifies which cases are used in the analyses. This specification is applied to analysis variables but not design variables.

ANALYSIS. Each statistic is based on all valid data for the analysis variables used in computing the statistic. Ratios are computed by using all cases with valid data for both of the specified variables. Statistics for different variables may be based on different sample sizes. This setting is the default.

LISTWISE. Only cases with valid data for all analysis variables are used in computing any statistics. Statistics for different variables are always based on the same sample size.

The CLASSMISSING keyword specifies whether user-missing values are treated as valid. This specification is applied only to categorical design variables (strata, cluster, and subpopulation variables).

EXCLUDE. Exclude user-missing values among the strata, cluster, and subpopulation variables. This setting is the default.

INCLUDE. Include user-missing values among the strata, cluster, and subpopulation variables. Treat user-missing values for these variables as valid data.