SDC-Report_v1

Input Data

The data set consists of 102 observations

Information on selected important (key) variables

  • Categorical key variable(s): DISTRICT | URB_RUR | S5A_ITEM_NAME
  • Continuous key variable(s): Qty_produced | Qty_consumed
  • Weight variable: HH_Final_weight
  • householdID: HOUSEHOLD_ID
  • strataVariable(s): not defined

Modifications

  • Modifications on categorical key variables: FALSE
  • Modifications on continuous key variables: TRUE
  • Modifications using PRAM: FALSE
  • Local suppressions: TRUE

Disclosure risk:

Frequency Analysis for Categorical Key Variables

Number of observations violating

  • 2-Anonymity: 0 (original dataset: 3)
  • 3-Anonymity: 0 (original dataset: 12)

Percentage of observations violating

  • 2-Anonymity: 0.000% (original dataset: 2.941%)
  • 3-Anonymity: 0.000% (original dataset: 11.765%)

Disclosure Risk for Categorical Variables

Expected Percentage of Reidentifications:

  • modified data: 0.048% (~ 0.049 observations)
  • original data: 0.176% (~ 0.179 observations)

10 combinations of categories with highest risks

DISTRICT URB_RUR S5A_ITEM_NAME risk fk Fk hier_risk
KAILAHUN NA Tilapia fish 0.001 5 1198.667 0.0031220
KAILAHUN NA Cat fish 0.001 5 1198.667 0.0031220
KAILAHUN NA Others specify 0.001 5 1198.667 0.0031220
TONKOLILI Rural Tilapia fish 0.001 6 1297.697 0.0027690
TONKOLILI Rural Cat fish 0.001 6 1297.697 0.0027690
TONKOLILI Rural Others specify 0.001 6 1297.697 0.0027690
TONKOLILI NA Tilapia fish 0.001 6 1297.697 0.0027690
TONKOLILI NA Cat fish 0.001 6 1297.697 0.0027690
TONKOLILI NA Others specify 0.001 6 1297.697 0.0027690
KAMBIA Rural Tilapia fish 0.001 8 1503.902 0.0022763

Disclosure Risk Continuous Scaled Variables

The (distance-based) disclosure risk for continous key variables is between 0.000% and 98.039% in the modified data.

In the original data, the risk is assumed to be approximately 100.000%.

Hierarchical risk

  • modified data: 0.148 (0.145%)
  • original data: 0.532 (0.522%)

Data Utility

Frequencies Categorical Key Variables:

Variable: DISTRICT

Categories Original data Modified data
KAILAHUN 6 3
KENEMA 3 NA
KONO 9 9
BOMBALI 27 27
FALABA 0 NA
KOINADUGU 6 6
TONKOLILI 6 6
KAMBIA 12 12
KARENE 0 NA
PORTLOKO 9 9
BO 18 18
BONTHE 3 NA
MOYAMBA 0 NA
PUJEHUN 0 NA
WESTERN RURAL 0 NA
WESTERN URBAN 0 NA
NA 3 12

Variable: URB_RUR

Categories Original data Modified data
Rural 93 93
Urban 3 NA
NA 6 9

Variable: S5A_ITEM_NAME

Categories Original data Modified data
Cat fish 34 34
Others specify 34 34
Tilapia fish 34 34

Local Suppressions

The table below shows for each categorical key variable the number (1st row) and the percentages (2nd row) of suppressed cells.

DISTRICT URB_RUR S5A_ITEM_NAME
Number of Suppression 9 3 0
Percentage 8.824 2.941 0.000

Data Utility of Continuous Scaled Key Variables

Univariate summary of variable Qty_produced

Original Modified Difference
Min. 0.86000 0.9 -0.040000
1st Qu. 15.36000 15.4 -0.040000
Median 30.72000 30.7 0.020000
Mean 73.99828 68.4 5.598276
3rd Qu. 61.44000 61.4 0.040000
Max. 430.00000 286.7 143.300000

Univariate summary of variable Qty_consumed

Original Modified Difference
Min. 0.2150 0.2 0.0150
1st Qu. 10.2400 10.2 0.0400
Median 25.8000 25.8 0.0000
Mean 150.2112 49.4 100.8112
3rd Qu. 61.4400 61.4 0.0400
Max. 3072.0000 227.3 2844.7000

Information Loss Criteria

  • Criteria IL1: 158.590%
  • Difference of Eigenvalues in modified data: 80.505% (0.00% in original data)

Boxplot of Differences

R-Code

Session-Info

About the R-Version

  • Version: R version 4.5.2 (2025-10-31 ucrt)
  • Platform: x86_64-w64-mingw32

Locales

LC_COLLATE=English_United States.utf8 | LC_CTYPE=English_United States.utf8 | LC_MONETARY=English_United States.utf8 | LC_NUMERIC=C | LC_TIME=English_United States.utf8

Attached base packages

stats | graphics | grDevices | utils | datasets | methods | base

Other attached packages

labelled (2.16.0) | questionr (0.8.1) | haven (2.5.5) | readxl (1.4.5) | tidyr (1.3.1) | dplyr (1.1.4) | agrisvyr (0.2.0) | sdcMicro (5.7.9)

Packages loaded via Namespace (but not attached)

tidyselect (1.2.1) | viridisLite (0.4.2) | farver (2.1.2) | R.utils (2.13.0) | S7 (0.2.1) | fastmap (1.2.0) | promises (1.5.0) | digest (0.6.38) | mime (0.13) | lifecycle (1.0.4) | cluster (2.1.8.1) | magrittr (2.0.4) | compiler (4.5.2) | rlang (1.1.6) | sass (0.4.10) | tools (4.5.2) | utf8 (1.2.6) | yaml (2.3.10) | data.table (1.17.8) | knitr (1.50) | askpass (1.2.1) | htmlwidgets (1.6.4) | plyr (1.8.9) | xml2 (1.4.1) | RColorBrewer (1.1-3) | miniUI (0.1.2) | withr (3.0.2) | purrr (1.2.0) | R.oo (1.27.1) | grid (4.5.2) | xtable (1.8-4) | data.tree (1.2.0) | ggplot2 (4.0.1) | scales (1.4.0) | MASS (7.3-65) | cli (3.6.5) | rmarkdown (2.30) | crayon (1.5.3) | generics (0.1.4) | otel (0.2.0) | rstudioapi (0.17.1) | robustbase (0.99-6) | tzdb (0.5.0) | cachem (1.1.0) | stringr (1.6.0) | rhandsontable (0.3.8) | cellranger (1.1.0) | vctrs (0.6.5) | jsonlite (2.0.0) | carData (3.0-5) | hms (1.1.4) | systemfonts (1.3.1) | jquerylib (0.1.4) | shinyBS (0.61.1) | glue (1.8.0) | DEoptimR (1.1-4) | DT (0.34.0) | stringi (1.8.7) | gtable (0.3.6) | later (1.4.4) | tibble (3.3.0) | pillar (1.11.1) | htmltools (0.5.8.1) | openssl (2.3.4) | R6 (2.6.1) | textshaping (1.0.4) | evaluate (1.0.5) | shiny (1.11.1) | kableExtra (1.4.0) | readr (2.1.6) | highr (0.11) | R.methodsS3 (1.8.2) | openxlsx (4.2.8.1) | renv (1.1.5) | httpuv (1.6.16) | bslib (0.9.0) | Rcpp (1.1.0) | zip (2.3.3) | svglite (2.2.2) | xfun (0.54) | fs (1.6.6) | forcats (1.0.1) | usethis (3.2.1) | getPass (0.2-4) | prettydoc (0.4.1) | pkgconfig (2.0.3)

Disclaimer

R-Package sdcMicro is developed and maintained by Statistics Austria (www.statistik.at).

Please use the issue-tracker on github to report any issues:


This report was generated on Wed, 08/04/2026 at 23:32:18.