Weightings - SuperCROSS

SuperCROSS can be used to tabulate against sample surveys where each record in the sample survey has one or more associated weights. Population estimates are calculated by summing these weights, and estimates of means are also computed with these weights.

Survey design, planning, and implementation is outside the scope of SuperSTAR. Once the survey is complete and the weights have been reassessed on the basis of non-response, etc, the source data can be prepared so that each unit record has its weight associated with it.

The actual weight values need to be computed outside the SuperSTAR system.

For details about configuring weighting, see Weightings - Data Control.

Replicate Weights

The weightings functionality in SuperCROSS supports the replicate weight method of calculating standard errors of survey data results.

Replicate weights are a series of variables that contain the information necessary for correctly computing (via the replicate weight method) the standard errors of point estimates when analyzing survey data.

The replicate weights are derived from the survey design and processing. Each record in the survey is assigned a number of replicate weights, from 30 to up to several thousand times depending on the actual methodology, and the required accuracy of the variance estimate.

The weights are stored as columns in the source data, and are loaded into a SuperSTAR database through SuperCHANNEL.