The STRATA statement names variables that form the strata in the standardization. The combinations of categories of STRATA variables define the strata in the
population.
The STRATA variables are one or more variables in all input data sets. These variables can be either character or numeric.
The formatted values of the STRATA variables determine the levels. Thus, you can use formats to group values into levels.
See the FORMAT procedure in the
Base SAS Procedures Guide and the FORMAT statement and SAS formats in
SAS Language Reference: Dictionary for more information.
When the STRATA statement is not specified or the statement is specified without variables, all observations in a data set
are treated as though they are from a single stratum.
You can specify the following options in the STRATA statement after a slash (/):
-
EFFECT
-
displays a table of the stratum-specific effect estimates and associated confidence limits. This option applies only when
METHOD=DIRECT with two study populations and when METHOD=MH, where two study populations are required. If the EFFECT=DIFF
option in the PROC STDRATE statement is specified, the stratum-specific rate or risk difference effects are displayed. Otherwise,
the stratum-specific rate or risk ratio effects are displayed.
-
MISSING
-
treats missing values as a valid (nonmissing) category for all STRATA variables. When PROC STDRATE determines levels of a
STRATA variable, an observation with missing values for that STRATA variable is excluded, unless the MISSING option is specified.
-
ORDER=DATA | FORMATTED | INTERNAL
-
specifies the order in which the values of the categorical variables are to be displayed. You can specify the following values
for the ORDER= option:
- DATA
-
sorts by the order in which the values appear in the input data set.
- FORMATTED
-
sorts by their external formatted values.
- INTERNAL
-
sorts by the unformatted values, which yields the same order that the SORT procedure does.
By default, ORDER=INTERNAL. For ORDER=FORMATTED and ORDER=INTERNAL, the sort order is machine-dependent.
-
STATS <( CL=LOGNORMAL | NONE | NORMAL | POISSON )>
-
displays tables for stratum-specific statistics such as stratum-specific rates and risks. You can specify the following values
of the CL= suboption to request confidence limits for the rate or risk estimate in each stratum:
-
LOGNORMAL
-
requests confidence limits based on a lognormal approximation.
-
NONE
-
suppresses confidence limits.
-
NORMAL
-
requests confidence limits based on a normal approximation and also displays the standard error for the rate estimate in each
stratum.
-
POISSON
-
requests confidence limits based on a Poisson distribution for stratum-specific rates. This values applies only when STAT=RATE
in the PROC STDRATE statement.
The default is CL=NORMAL.
-
SMR <( CL=LOGNORMAL | NONE | NORMAL | POISSON )>
-
displays tables for stratum-specific SMR estimates. This option applies only when METHOD=INDIRECT is specified in the PROC
STDRATE statement. You can specify the following values of the CL= suboption to request confidence limits for the SMR estimate
in each stratum:
-
LOGNORMAL
-
requests confidence limits based on a lognormal approximation.
-
NONE
-
suppresses confidence limits.
-
NORMAL
-
requests confidence limits based on a normal approximation and also displays the standard error for the SMR estimate in each
stratum.
-
POISSON
-
requests confidence limits based on a Poisson distribution for stratum-specific SMR estimates. This values applies only when
STAT=RATE in the PROC STDRATE statement.
The default is CL=NORMAL.
