Summary of survey software: Methods
This is a summary of the information included under the heading
"Methods used for variance estimation" for each of the
software packages described on these pages. Select the appropriate
title for more information on any package.
Mainly through Taylor Series methods. Methods based on replication
(Jackknife, Balanced repeated, Bootstrap) are available for some analyses.
Bascula uses Balanced Repeated Replication (BRR) and Taylor series methods for
variance estimation.
No information on this topic for this package
Taylor linearisation.
Taylor linearized deviation approach.
Jackknife or design-based formulae for the Generalized Regression
(GREG) estimator with Taylor linearization for non-linear parameters.
Jackknife and/or Taylor series approaches. Multiple imputation for missing
data.
Taylor linearization.
- Taylor series linearization
- Replication weighting
Taylor expansion.
Taylor-series linearization is used in the survey analysis commands.
There are also commands for jackknife and bootstrap variance estimation;
although these are not specifically oriented to survey data they will
accomodate survey features like clustering and stratification.
The Taylor series linearization method (GEE for regression models) is
used combined with variance estimation formulas specific to the sample
design. The user does not need to develop special replicate weights
since the sample design can be specified directly to the program.
Jackknife and Balanced Repeated Replication (BRR) variance estimation is
also supported.
Replication methods, primarily the jackknife. Half-sample replication
and random group replication are also available, and the weights associated
with these schemes may be modified by the user.
Balanced repeated replication (including the Fay method), jackknife
(several variants) and other replication methods specified by users
through the development of replicate weights (e.g., bootstrap).
Return to main page for survey software