Overview (CSGLM command)

CSGLM performs linear regression analysis, as well as analysis of variance and covariance, for samples that are drawn by complex sampling methods. The procedure estimates variances by taking into account the sample design that is used to select the sample, including equal probability and probability proportional to size (PPS) methods, and with replacement (WR) and without replacement (WOR) sampling procedures. Optionally, CSGLM performs analyses for a subpopulation.

Basic Specification

  • The basic specification is a variable list (identifying the dependent variable, the factors, if any, and the covariates, if any) and a PLAN subcommand with the name of a complex sample analysis plan file, which may be generated by the CSPLAN procedure.
  • The default model includes the intercept term, main effects for any factors, and any covariates.
  • The basic specification displays summary information about the sample design, R-square and root mean square error for the model, regression coefficient estimates and t tests, and Wald F tests for all model effects. Additional subcommands must be used for other results.

Operations

  • CSGLM computes linear model estimates for sampling designs that are supported by the CSPLAN and CSSELECT procedures.
  • The input dataset must contain the variables to be analyzed and variables that are related to the sampling design.
  • The complex sample analysis plan file provides an analysis plan based on the sampling design.
  • By default, CSGLM uses a model that includes the intercept term, main effects for any factors, and any covariates.
  • Other effects, including interaction and nested effects, may be specified by using the MODEL subcommand.
  • The default output for the specified model is summary information about the sample design, R-square and root mean square error, regression coefficient estimates and t tests, and Wald F tests for all effects.
  • WEIGHT and SPLIT FILE settings are ignored by the CSGLM procedure.

Syntax Rules

  • The dependent variable and PLAN subcommand are required. All other variables and subcommands are optional.
  • Multiple CUSTOM and EMMEANS subcommands may be specified; each subcommand is treated independently. All other subcommands may be specified only once.
  • The EMMEANS subcommand may be specified without options. All other subcommands must be specified with options.
  • Each keyword may be specified only once within a subcommand.
  • Subcommand names and keywords must be spelled in full.
  • Equals signs (=) that are shown in the syntax chart are required.
  • Subcommands may be specified in any order.
  • The dependent variable and covariates must be numeric, but factors and the subpopulation variable can be numeric or string variables.
  • Across the dependent, factor, and covariate variable lists, a variable may be specified only once.
  • Plan file and subpopulation variables may not be specified on the variable list.
  • Minimum syntax is a dependent variable and the PLAN subcommand. This specification fits an intercept-only model.

Limitations

  • WEIGHT and SPLIT FILE settings are ignored with a warning by the CSGLM procedure.