PROC SURVEYFREQ uses the degrees of freedom of the variance estimator to obtain the t distribution’s percentile for confidence limits for proportions, totals, and other statistics. The procedure also uses the degrees of freedom to compute the F statistics in the Rao-Scott and Wald chi-square tests.
PROC SURVEYFREQ computes the degrees of freedom based on the variance estimation method and the sample design. Alternatively, you can use the DF= option in the TABLES statement to specify the degrees of freedom.
For Taylor series variance estimation, PROC SURVEYFREQ calculates the degrees of freedom (df) as the number of clusters minus the number of strata. If there are no clusters, then df is the number of observations minus the number of strata. If the design is not stratified, then df is the number of clusters minus one. These numbers are based on the observations included in the analysis of the individual table request. These numbers do not count observations that are excluded from the table because of missing values. For more information, see the section Missing Values. If you specify the MISSING option, missing values are treated as valid nonmissing levels and are included when computing degrees of freedom. If you specify the NOMCAR option for Taylor series variance estimation, observations with missing values of the TABLES variables are included when computing degrees of freedom.
If you use a REPWEIGHTS statement to provide replicate weights, the degrees of freedom equals the number of replicates, which is the number of REPWEIGHTS variables that you provide. Alternatively, you can use the DF= option in the REPWEIGHTS or the TABLES statement to specify the degrees of freedom.
For BRR variance estimation (when you do not use a REPWEIGHTS statement), PROC SURVEYFREQ calculates the degrees of freedom as the number of strata. By default, PROC SURVEYFREQ computes the number of strata by using all valid observations in the data set. If you specify the DFADJ method-option, PROC SURVEYFREQ computes the number of strata as the number of nonmissing strata for the current table request. This excludes any empty strata that occur when observations are excluded because of missing values of a TABLES variable. For more information, see the section Missing Values.
For bootstrap and jackknife variance estimation (when you do not use a REPWEIGHTS statement), PROC SURVEYFREQ calculates the degrees of freedom (df) as the number of clusters minus the number of strata. If there are no clusters, then df is the number of observations minus the number of strata. If the design is not stratified, then df is the number of clusters minus one. By default, PROC SURVEYFREQ computes the stratum and cluster counts by using all valid observations in the data set. If you specify the DFADJ method-option, PROC SURVEYFREQ computes the stratum and cluster counts by using only nonmissing observations for the current table request. This excludes any empty strata or clusters that occur when observations are excluded because of missing values of a TABLES variable. For more information, see the section Missing Values.