CensusProtection

How to use the open-source versions of the recommended census protection methods

This repository explains how to use tools and methods recommended for census 2021 protection. Part 1 is related to targeted record swapping while part 2 focuses on the cell-key method. The methods can be used independently: it is either possible to compute unperturbed hypercubes with swapped microdata (part 2 can be ignored), perturbed hypercubes with original microdata (part 1 can be skipped), or perturbed hypercubes with swapped microdata (part 1 then part 2).

The instructions below contain references to the respective Github-repositories with the updated versions of the tools recommended for the protection of census. The activities related with development of tools and methods for the census protection were co-funded by Eurostat.

Part 1: Targeted record swapping

To apply record swapping to original microdata it is possible to use either the R package sdcMicro (paragraph 1.1) or MuARGUS (paragraph 1.2).

Both implementation use the same underlying new C++ code which is optimized to be incredibly fast. This new implementation of the targeted record swapping method is intended to be very close to the SAS codes tested in the previous SGA based on the ONS codes. More information regarding the small differences can be found in the vignettes vignette("recordSwapping") from sdcMicro package.

1.1 R-package for targeted record swapping

Targeted record swapping is part of the package github.com/sdcTools/sdcMicro, function recordSwap(). The vignette for targeted record swapping is callable via vignette("recordSwapping").

1.2 MuARGUS

The latest release of MuARGUS that includes the implementation of record swapping is available at github.com/sdcTools/muargus/releases where also a quick reference documentation with explanations and screenshots can be downloaded.

Part 2: Cell-key method

The cell-key method can be applied either with the R package cellKey (paragraph 2.2) or tauArgus (paragraph 2.3). Both implementation use the ptable package (paragraph 2.1).

2.1 R-package ptable

The package is developed at github.com/sdcTools/ptable where also installation instructions and example usage can be found. The package includes a vignette and a graphical user interface and it can draw plots that describe the perturbations that will be applied to the hypercubes based on a specific pTable with the cellKey package, tauArgus or SAS macros.

2.2 R-package cellKey

The package is developed at github.com/sdcTools/cellKey where also installation instructions and example usage can be found. The package includes also a detailed vignette that helps users getting started with the functionality of the package.

Instructions for installing R-packages

install.packages(c("data.table","devtools","R.utils","sdcTable","sdcMicro","sdcHierarchies","remotes"))

# install ptable
devtools::install_github("sdcTools/ptable", dependencies=c("Depends","Imports"), force=TRUE, build_opts="--build-vignettes")
# install cellKey
remotes::install_github(repo = "sdcTools/cellKey", dependencies = TRUE, build_opts = "--no-resave-data", force = TRUE)

2.3 TauARGUS

The latest release of TauArgus that includes the implementation of the cellKey method is available at github.com/sdcTools/tauargus/releases where also a quick reference documentation with explanations and screenshots can be downloaded.