The SURVEYLOGISTIC procedure provides logistic regression analysis for sample survey data. Logistic regression analysis investigates the relationship between discrete responses and a set of explanatory variables. PROC SURVEYLOGISTIC fits linear logistic regression models for discrete response survey data by the method of maximum likelihood and incorporates the sample design into the analysis. The SURVEYLOGISTIC procedure enables you to specify categorical classification variables (also known as CLASS variables) as explanatory variables in the model by using the same syntax for main effects and interactions as in the GLM and LOGISTIC procedures.

The following link functions are available for regression in PROC SURVEYLOGISTIC: the cumulative logit function (CLOGIT), the generalized logit function (GLOGIT), the probit function (PROBIT), and the complementary log-log function (CLOGLOG). The procedure performs maximum likelihood estimation of the regression coefficients with either the Fisher scoring algorithm or the Newton-Raphson algorithm.

PROC SURVEYLOGISTIC also performs logistic regression analysis for domains (subpopulations).

See Chapter 87, The SURVEYLOGISTIC Procedure, for more information.