Skip to contents

Assess Statistical disclosure risk based on an intrusions scenario

Source: R/kobo_anonymise.R
kobo_anonymise.Rd

When personal data is being collected, performing basic de-identification (i.e. removal of direct identifiers) and assessing risk of re-identification (i.e. using indirect identifiers to re-identify individuals) is a key sep to perform in order to be able to share the data with multiple analyst.

The initial step consist in defining potential intrusion scenario. This suppose to document the anonymise cell for each variable

TypeDescription---------------------------Direct_identifierCan be directly used to identify an individual. E.g. Name, Address, Date of birth, Telephone number, GPS location

Direct identifiers will be automatically removed from the data. The function will perform the measurement of various statistical disclosure risk measurement for the selected quasi_identifier and sensitive_information.

Usage

kobo_anonymise(datalist, dico)

Arguments

datalist

An object of the "datalist" class as defined in kobocruncher

dico

An object of the "kobodico" class format as defined in kobocruncher

Examples

# dico <- kobo_dico( xlsformpath = system.file("sample_xlsform.xlsx", package = "kobocruncher") )
# datalist <- kobo_data(datapath = system.file("data.xlsx", package = "kobocruncher") )
# 
# kobo_anonymise(datalist = datalist,
#                    dico = dico,
#                 indicatoradd = indicatoradd   )