Overview (WEIGHT command)

WEIGHT gives cases different weights (by simulated replication) for statistical analysis. WEIGHT can be used to weight a sample up to population size for reporting purposes or to replicate an example from a table or other aggregated data. With WEIGHT, you can arithmetically alter the sample size or its distribution.

To apply weights resulting from your sampling design, see Sampling and Testing.

Basic Specification

The basic specification is keyword BY followed by the name of the weight variable. Cases are weighted according to the values of the specified variable.

Syntax Rules

  • Only one numeric variable can be specified. The variable can be a precoded weighting variable, or it can be computed with the transformation language.
  • WEIGHT cannot be placed within a FILE TYPE—END FILE TYPE or INPUT PROGRAM—END INPUT PROGRAM structure. It can be placed nearly anywhere following these commands in a transformation program. See the topic Commands and Program States for more information.

Operations

  • Unlike most transformations, WEIGHT takes effect as soon as it is encountered in the command sequence. Thus, special attention should be paid to its position among commands. See the topic Command Order for more information.
  • Weighting is permanent during a session unless it is preceded by a TEMPORARY command, changed by another WEIGHT command, or turned off with the WEIGHT OFF specification.
  • Each WEIGHT command overrides the previous one.
  • WEIGHT uses the value of the specified variable to arithmetically replicate cases for subsequent procedures. Cases are not physically replicated.
  • Weight values do not need to be integer, and some procedures, such as FREQUENCIES, CROSSTABS, and CTABLES, will use fractional values on the WEIGHT variable. However, most procedures treat the WEIGHT variable as a replication weight and will simply round fractional weights to the nearest integer. Some procedures ignore the WEIGHT variable completely, and this limitation is noted in the procedure-specific documentation. Procedures in the Complex Samples add-on module can use inverse sampling weights specified on the CSPLAN command.
  • Cases with missing or nonpositive values for the weighting variable are treated as having a weight of 0 and are thus invisible to statistical procedures. They are not used in calculations even where unweighted counts are specified. These cases do remain in the file, however, and are included in case listings and saved when the file is saved.
  • A file saved when weighting is in effect maintains the weighting.
  • If the weighted number of cases exceeds the sample size, tests of significance are inflated; if it is smaller, they are deflated.