Type: Package
Title: Reapportion Data from One Geography to Another
Version: 0.2.0
Maintainer: François Briatte <f.briatte@gmail.com>
Description: A port of the 'spReapportion' package, using Simple Features in order to lose the dependencies to the retired 'maptools' and 'rgeos' packages.
License: GPL-3
LazyData: TRUE
URL: https://github.com/briatte/sfReapportion
BugReports: https://github.com/briatte/sfReapportion/issues
Depends: R (≥ 4.1)
Imports: dplyr, sf, sp (≥ 2.0-0)
Suggests: areal, populR, testthat
RoxygenNote: 7.3.3
Encoding: UTF-8
NeedsCompilation: no
Packaged: 2026-04-20 15:21:33 UTC; fr
Author: François Briatte ORCID iD [aut, cre], Joël Gombin ORCID iD [aut]
Repository: CRAN
Date/Publication: 2026-04-21 20:32:10 UTC

Voter addresses in the 20th arrondissement of Paris

Description

All voter addresses located in the 20th arrondissement of Paris, according to the Répertoire électoral unique (REU). We first extracted polling stations from table-bv-reu.parquet, and then extracted addresses located within them from table-adresses-reu.parquet. Addresses were converted into spatial points of class sf. The weight variable is nb_adresses, although that variable does not count voters per se: it measures the number of voter addresses at that particular location, which is probably the best publicly available proxy at that level of precision.

Usage

Paris20eAddresses

Format

An object of class sf (inherits from grouped_df, tbl_df, tbl, data.frame) with 5820 rows and 3 columns.

Source

INSEE: https://www.data.gouv.fr/datasets/bureaux-de-vote-et-adresses-de-leurs-electeurs

Paris IRIS shapefile, 2013

Description

A shapefile of the IRIS (French census tract) of Paris in 2013.

Usage

ParisIris

Format

An object of class SpatialPolygonsDataFrame with 992 rows and 7 columns.

Source

IGN: https://cartes.gouv.fr/rechercher-une-donnee/dataset/IGNF_CONTOURS-IRIS

Paris polling stations shapefile, 2012

Description

A shapefile of the polling stations in Paris in 2012.

Usage

ParisPollingStations2012

Format

An object of class SpatialPolygonsDataFrame with 867 rows and 5 columns.

Source

Ville de Paris: https://opendata.paris.fr/explore/dataset/zones-de-rattachement-des-bureaux-de-vote-en-2012/table/

Socio-professional breakdown of the population aged 15+ in Paris, 2011

Description

A dataset containing the number of inhabitants by socio-professional category in 2011 in Paris, by IRIS (the French census tract).

Usage

RP_2011_CS8_Paris

Format

An object of class data.frame with 992 rows and 10 columns.

Details

Source

INSEE: https://www.insee.fr/fr/statistiques/2028584

Aggregate Spatial Polygons

Description

A replacement for rgeos::unaryUnion.

Usage

sfAggregate(sp, id)

Arguments

sp

an SpatialPolygonsDataFrame object

id

a grouping vector, as required by aggregate

Value

a SpatialPolygonsDataFrame object with unique and valid feature IDs

Note

Source: https://github.com/r-spatial/sf/issues/2563

Author(s)

Roger Bivand (thanks)

Reapportion data from one geography to another

Description

This function allows to reapportion data from one geography to another, for example in the context of working with different administrative units.

Usage

sfReapportion(
  old_geom,
  new_geom,
  data,
  old_ID,
  new_ID,
  data_ID,
  variables = names(data)[-which(names(data) %in% data_ID)],
  mode = "count",
  weights = NULL,
  weight_matrix = NULL,
  weight_matrix_var = NULL
)

Arguments

old_geom

a 'SpatialPolygonsDataFrame' or 'sf' object representing the initial geometry.

new_geom

a 'SpatialPolygonsDataFrame' or 'sf' object representing the geometry you want to reapportion data to.

data

a 'data.frame' containing the data to reapportion, and an ID allowing to match it to the 'old_geom'.

old_ID

a string, the name of the ID variable in the 'old_geom'.

new_ID

a string, the name of the ID variable in the 'new_geom'.

data_ID

a string, the name of the ID variable in the 'data'.

variables

a character vector, representing the names of the variables in the 'data' set to reapportion. By default, all data variables except for the ID.

mode

either "count" or "proportion". "count" is for absolute values, "proportion" is for, well, proportions (expressed between 0 and 1). If "proportion", a 'weights' variable needs to be provided. Note: "proportion" has only been lightly tested.

weights

(optional, lightly tested only) In case the variables are proportions, the name of the variable containing weights (i.e. the total number of observations per unit in the 'old_geom').

weight_matrix

(optional, lightly tested only) a 'SpatialPointsDataFrame' or 'sf' object indicating the spatial coordinates of the observations (inhabitants, voters, etc.).

weight_matrix_var

(optional, lightly tested only) the name of the (numeric) variable containing the weights in weight_matrix.

Value

a 'data.frame' containing new_ID and the reapportioned variables from data

Note

Inspiration from https://stackoverflow.com/a/17703903 and https://rpubs.com/PaulWilliamson/6577. Original sp version available at https://github.com/joelgombin/spReapportion. All mistakes are mine, obviously. More details appear in the README file.

Author(s)

Joël Gombin (initial sp version), François Briatte (sf port)