Module Reference

This section contains details of the data control modules that are supplied with SuperSERVER.

Perturbed Continuous RSE — Used in conjunction with other modules to update RSE for weighted datasets to take into account the effect of perturbation. This module adds the ability to perturb the RSE and variance cubes for continuous variables.

Perturbed Estimates — Adjusts the sum of a measure so that the ratio of perturbed sum and perturbed mean is equal to the perturbed contributor count.

Perturbed Mean Variance — Used in conjunction with other modules to update RSE for sums and means to take into account the effect of continuous perturbation. This module adds the ability to perturb the population mean values in the variance cube. In order to use the module, you must have the FREQ cube enabled and have a VTable available to lookup the correction value.

Perturbed Population Estimate Variance — Used in conjunction with other modules to update RSE for sums to take into account the effect of perturbation. This module adds the ability to perturb the population estimate values in the variance cube. In order to use the module, you must have the FREQ cube enabled and have a ZTable available to lookup the correction value.

RSE Annotation — Annotates RSE values.

RSE Calculation — Calculates Relative Standard Error.

RSE Poor Table Check — Uses the RSE cube to check table reliability against a configured threshold.

RSE Suppression — Suppresses RSE values over a given threshold.

Example Configuration — This section contains some examples of complete method configuration for some typical scenarios.

Module Requirements

Many of the data control modules are designed to work together. The following table lists the combinations of modules that are required for common scenarios:

	Count Perturbation	Count and Continuous Perturbation	Weighted Dataset with Count Perturbation	Weighted Dataset with Count and Continuous Perturbation	Weighted Dataset (No Perturbation)
Average Cell Weight			Required	Required
Continuous Perturbation		Required		Required
Output Scaling	Optional	Optional	Optional	Optional	Optional
Perturbation	Required	Required	Required	Required
Perturbed Continuous RSE			Required if dataset has measures	Required
Perturbed Count RSE			Required	Required
Perturbed Estimates	Required if dataset has measures	Required	Required if dataset has measures	Required
Perturbed Mean Variance				Required
Perturbed Population Estimate Variance			Required if dataset has measures	Required
RSE Annotation			Optional	Optional	Optional
RSE Calculation			Runs Automatically	Runs Automatically	Runs Automatically
RSE Poor Table Check			Optional	Optional	Optional
RSE Suppression			Optional	Optional	Optional
Sparsity Check	Optional	Optional	Optional	Optional	Optional

Refer to the Example Configuration section for some examples of these scenarios.

Using Multiple Data Control Modules

Where multiple data control modules are required, they generally need to be run in a specific sequence, so that the results from one module can be used by another. To manage this, the Data Control API uses a priority system, which allows you to configure the sequence in which the modules are executed. The priority is a numeric value; modules with a lower value will be executed first.

When you require multiple data control modules to work together, you can choose either to create just one method and add all the individual module plugins that you need to that method, or you can create multiple methods and apply them individually to the dataset.

If you are using one method, set the priority of each module when you add it to the method: method <method_id> adddcplugin <plugin_id> <plugin_filename> <priority>

If you are using multiple methods, set the priority of each method when you apply it to the dataset: cat <dataset_id> addmethod <method_id> <priority>

You can change the priority of a module or method if necessary, refer to the method comand and the cat command for more details.

Refer to the individual module reference pages for details on their specific priority requirements.

Logging

By default, the Data Control modules will log limited information about the affect of each module. To increase the amount of logging output for specific modules, do the following:

Open the SuperSERVER logging properties file in a text editor. If you installed to the default location this file is C:\ProgramData\STR\SuperSERVER SA\log4j.scsa.xml.

Add the following section before the closing </log4j:configuration> tag:

XML

<logger name="AUDIT_DataServer.Dcapi" additivity="false">
    <level value ="DEBUG" />
    <appender-ref ref="MainLog" />
    <appender-ref ref="CONSOLE"/>       
</logger>

For each specific module that you want to log additional information, add the following property to your method:
CODE
```
method <method_id> <plugin_id> addproperty EnableLogCube "true"
```
By default the log output will be written to the SuperSERVER log directory. If you wish to change the location of the log file for an individual module, add the following property:
CODE
```
method <method_id> <plugin_id> addproperty LogCubePath <path>
```
For example:
CODE
```
method MyPerturbationMethod PerturbedMeanVariance addproperty EnableLogCube "true"
method MyPerturbationMethod PerturbedMeanVariance addproperty LogCubePath "C:\\dcapi\\logs"
```