BFS

Search and download data from the Swiss Federal Statistical Office

The BFS package allows to search and download public data from the Swiss Federal Statistical Office (BFS stands for Bundesamt für Statistik in German) APIs in a dynamic and reproducible way.

Installation

install.packages("BFS")

You can also install the development version from Github.

devtools::install_github("lgnbhl/BFS")

Usage

library(BFS)

Get the data catalog

Retrieve the list of publicly available datasets from the data catalog in any language (“de”, “fr”, “it” or “en”) by calling bfs_get_catalog_data().

catalog_data_en <- bfs_get_catalog_data(language = "en")

catalog_data_en

## # A tibble: 184 × 7
##    title                language publication_date    number_asset url_bfs url_px
##    <chr>                <chr>    <dttm>                     <dbl> <chr>   <chr> 
##  1 Acknowledgment of p… en       2023-06-22 08:30:00     25945442 https:… https…
##  2 Adoptions by differ… en       2023-06-22 08:30:00     25945406 https:… https…
##  3 Deaths by instituti… en       2023-06-22 08:30:00     25945423 https:… https…
##  4 Deaths by sex, citi… en       2023-06-22 08:30:00     25945436 https:… https…
##  5 Deaths since 1803    en       2023-06-22 08:30:00     25945437 https:… https…
##  6 Dissolved partnersh… en       2023-06-22 08:30:00     25945438 https:… https…
##  7 Divorces by canton,… en       2023-06-22 08:30:00     25945378 https:… https…
##  8 Divorces by duratio… en       2023-06-22 08:30:00     25945381 https:… https…
##  9 Divorces by institu… en       2023-06-22 08:30:00     25945387 https:… https…
## 10 Live births by inst… en       2023-06-22 08:30:00     25945410 https:… https…
## # ℹ 174 more rows
## # ℹ 1 more variable: catalog_date <dttm>

You can search in the data catalog using the following arguments:

language: The language of a BFS catalog, i.e. “de”, “fr”, “it” or “en”.
title: to search in title, subtitle and supertitle.
spatial_division: choose between “Switzerland”, “Cantons”, “Districts”, “Communes”, “Other spatial divisions” or “International”.
prodima: by specific BFS themes using one or multiple prodima numbers.
inquiry: by inquiry number.
institution: by institution.
publishing_year_start: by publishing year start.
publishing_year_end: by publishing year end.
order_nr: by BFS Number (FSO number).

For example, you can search data related to students:

bfs_get_catalog_data(language = "en", title = "students")

## # A tibble: 4 × 7
##   title                 language publication_date    number_asset url_bfs url_px
##   <chr>                 <chr>    <dttm>                     <dbl> <chr>   <chr> 
## 1 University of applie… en       2023-03-28 08:30:00     24367605 https:… https…
## 2 University of applie… en       2023-03-28 08:30:00     24367607 https:… https…
## 3 University students … en       2023-03-28 08:30:00     24367723 https:… https…
## 4 University students … en       2023-03-28 08:30:00     24367729 https:… https…
## # ℹ 1 more variable: catalog_date <dttm>

English (“en”) and Italian (“it”) data catalogs offer a limited list of datasets. For the full list please get the French (“fr”) or German (“de”) data catalogs.

Download data in any language

The function bfs_get_data() allows you to download any dataset from the data catalog using its BFS number (FSO number).

You need first to find the asset number of the dataset.

library(dplyr) #install.packages("dplyr")

asset_number_students <- bfs_get_catalog_data(language = "en", title = "students") |>
  dplyr::filter(title == "University students by year, ISCED field, sex and level of study") |>
  dplyr::pull(number_asset)

asset_number_students

## [1] 24367729

You can then find the BFS number by calling bfs_get_asset_metadata(). This function returns a list containing the metadata of the asset. For the student data, the BFS number is in the orderNR variable.

asset_meta_students <- bfs_get_asset_metadata(number_asset = asset_number_students)

bfs_number_students <- asset_meta_students$shop$orderNr

bfs_number_students

## [1] "px-x-1502040100_131"

You can also manually find the BFS number (FSO number) by opening the related URL official webpage.

url_bfs_students <- bfs_get_catalog_data(language = "en", title = "students") |>
  dplyr::filter(title == "University students by year, ISCED field, sex and level of study") |>
  dplyr::pull(url_bfs)

# open students dataset webpage
browseURL(url_bfs_students)

Finally you can get the data using the number_bfs argument in a given language (“en”, “de”, “fr” or “it”) from the official PXWeb API of the Swiss Federal Statistical Office.

bfs_get_data(number_bfs = bfs_number_students, language = "en")

## # A tibble: 18,060 × 5
##    Year    `ISCED Field`     Sex    `Level of study`       `University students`
##    <chr>   <chr>             <chr>  <chr>                                  <dbl>
##  1 1980/81 Education science Male   First university degr…                   545
##  2 1980/81 Education science Male   Bachelor                                   0
##  3 1980/81 Education science Male   Master                                     0
##  4 1980/81 Education science Male   Doctorate                                 93
##  5 1980/81 Education science Male   Further education, ad…                    13
##  6 1980/81 Education science Female First university degr…                   946
##  7 1980/81 Education science Female Bachelor                                   0
##  8 1980/81 Education science Female Master                                     0
##  9 1980/81 Education science Female Doctorate                                 70
## 10 1980/81 Education science Female Further education, ad…                    52
## # ℹ 18,050 more rows

“Too Many Requests” error message

When running the bfs_get_data() function you may get the following error message (issue #7).

Error in pxweb_advanced_get(url = url, query = query, verbose = verbose) : 
  Too Many Requests (RFC 6585) (HTTP 429).

This could happen because you ran too many times a bfs_get_*() function (API config is here). A solution is to wait a few seconds before running the next bfs_get_*() function. You can add a delay in your R code using the delay argument.

bfs_get_data(
  number_bfs = "px-x-1502040100_131", 
  language = "en", 
  delay = 10
)

If the error message remains, it could be because you are querying a very large BFS dataset. Two workarounds exist: a) download the BFS file using bfs_download_asset() to read it locally or b) query only specific elements of the data to reduce the API call (see next section).

Here an example using the bfs_download_asset() function:

BFS::bfs_download_asset(
  number_bfs = "px-x-1502040100_131", #number_asset also possible
  destfile = "px-x-1502040100_131.px"
)

library(pxR) # install.packages("pxR")
large_dataset <- pxR::read.px(filename = "px-x-1502040100_131.px") |>
  as.data.frame()

Note that reading a PX file using pxR::read.px() gives access only to the German version.

Query specific elements

First you want to get the metadata of your dataset, i.e. the variables (code and text) and dimensions (values and valueTexts). For example:

metadata <- bfs_get_metadata(number_bfs = "px-x-1502040100_131", language = "en")

# tidy metadata
library(dplyr)
library(tidyr) # for unnest_longer

metadata_tidy <- metadata |>
  select(-valueTexts) |>
  unnest_longer(values) |>
  mutate(valueTexts = metadata |>
           select(valueTexts) |>
           unnest_longer(valueTexts) |>
           pull(valueTexts)) |>
  select(code, text, values, valueTexts, everything())

metadata_tidy

## # A tibble: 92 × 7
##    code  text  values valueTexts time  elimination
##    <chr> <chr> <chr>  <chr>      <lgl> <lgl>      
##  1 Jahr  Year  0      1980/81    TRUE  NA         
##  2 Jahr  Year  1      1981/82    TRUE  NA         
##  3 Jahr  Year  2      1982/83    TRUE  NA         
##  4 Jahr  Year  3      1983/84    TRUE  NA         
##  5 Jahr  Year  4      1984/85    TRUE  NA         
##  6 Jahr  Year  5      1985/86    TRUE  NA         
##  7 Jahr  Year  6      1986/87    TRUE  NA         
##  8 Jahr  Year  7      1987/88    TRUE  NA         
##  9 Jahr  Year  8      1988/89    TRUE  NA         
## 10 Jahr  Year  9      1989/90    TRUE  NA         
## # ℹ 82 more rows
## # ℹ 1 more variable: title <chr>

Then you can filter the dimensions you want to query using the text and valueTexts variables and build the query dimension object with the code and values variables.

# select dimensions
dim1 <- metadata_tidy |>
  filter(text == "Year" & valueTexts %in% c("2020/21", "2021/22"))
dim2 <- metadata_tidy |>
  filter(text == "Level of study" & valueTexts %in% c("Master", "Doctorate"))
dim3 <- metadata_tidy |>
  filter(text == "ISCED Field" & valueTexts %in% c("Education science"))
dim4 <- metadata_tidy |>
  filter(text == "Sex") # all valueTexts dimensions

# build dimensions list object
dimensions <- list(
  dim1$values,
  dim2$values,
  dim3$values,
  dim4$values
)

names(dimensions) <- c(
  unique(dim1$code), 
  unique(dim2$code), 
  unique(dim3$code), 
  unique(dim4$code)
)

dimensions

## $Jahr
## [1] "40" "41"
## 
## $Studienstufe
## [1] "2" "3"
## 
## $`ISCED Fach`
## [1] "0"
## 
## $Geschlecht
## [1] "0" "1"

Finally you can query BFS data with specific dimensions.

BFS::bfs_get_data(
  number_bfs = "px-x-1502040100_131",
  language = "en",
  query = dimensions
  )

## # A tibble: 8 × 5
##   Year    `ISCED Field`     Sex    `Level of study` `University students`
##   <chr>   <chr>             <chr>  <chr>                            <dbl>
## 1 2020/21 Education science Male   Master                             151
## 2 2020/21 Education science Male   Doctorate                          121
## 3 2020/21 Education science Female Master                             555
## 4 2020/21 Education science Female Doctorate                          306
## 5 2021/22 Education science Male   Master                             143
## 6 2021/22 Education science Male   Doctorate                          115
## 7 2021/22 Education science Female Master                             599
## 8 2021/22 Education science Female Doctorate                          318

Catalog of tables

A lot of datasets are not accessible through the official PXWeb API. They are listed in the catalog of tables. You can search for specific tables using bfs_get_catalog_tables().

catalog_tables_en_students <- bfs_get_catalog_tables(language = "en", title = "students")

catalog_tables_en_students

## # A tibble: 5 × 7
##   title              language publication_date    number_asset url_bfs url_table
##   <chr>              <chr>    <dttm>                     <dbl> <chr>   <chr>    
## 1 Students at unive… en       2023-04-05 00:00:00     24865589 https:… https://…
## 2 Students at unive… en       2023-04-05 00:00:00     24865590 https:… https://…
## 3 Students at unive… en       2023-03-28 08:30:00     24345362 https:… https://…
## 4 Students at unive… en       2023-03-28 08:30:00     24345374 https:… https://…
## 5 Students at unive… en       2023-03-28 08:30:00     24345366 https:… https://…
## # ℹ 1 more variable: catalog_date <dttm>

Most of the BFS tables are Excel or CSV files. You can download an table with bfs_download_asset() using the number asset.

library(dplyr)

tables_asset_number_students <- catalog_tables_en_students |>
  dplyr::filter(title == "Students at universities and institutes of technology: Basistables") |>
  dplyr::pull(number_asset)

file_path <- BFS::bfs_download_asset(
  number_asset = tables_asset_number_students,
  destfile = "su-e-15.02.04.01.xlsx"
)

Access geodata catalog

Display geo-information catalog of the Swiss Official STAC API using bfs_get_catalog_geodata().

catalog_geodata <- bfs_get_catalog_geodata(include_metadata = TRUE)

catalog_geodata

## # A tibble: 281 × 12
##    collection_id     type  href  title description created updated crs   license
##    <chr>             <chr> <chr> <chr> <chr>       <chr>   <chr>   <chr> <chr>  
##  1 ch.are.agglomera… API   http… Citi… "The list … 2021-1… 2023-0… http… propri…
##  2 ch.are.alpenkonv… API   http… Alpi… "The perim… 2021-1… 2022-0… http… propri…
##  3 ch.are.belastung… API   http… Load… "Passenger… 2021-1… 2022-0… http… propri…
##  4 ch.are.belastung… API   http… Load… "Passenger… 2021-1… 2022-0… http… propri…
##  5 ch.are.belastung… API   http… Load… "Vehicles … 2021-1… 2022-0… http… propri…
##  6 ch.are.belastung… API   http… Load… "Vehicles … 2021-1… 2022-0… http… propri…
##  7 ch.are.erreichba… API   http… Acce… "Accessibi… 2021-1… 2022-0… http… propri…
##  8 ch.are.erreichba… API   http… Acce… "Accessibi… 2021-1… 2022-0… http… propri…
##  9 ch.are.gemeindet… API   http… Typo… "The typol… 2021-1… 2022-0… http… propri…
## 10 ch.are.gueteklas… API   http… Publ… "The publi… 2021-1… 2023-0… http… propri…
## # ℹ 271 more rows
## # ℹ 3 more variables: provider_name <chr>, bbox <list>, inverval <list>

Download geodata

For example you can get information about the dataset “Generalised borders G1 and area with urban character”.

library(dplyr)

geodata_g1 <- catalog_geodata |>
  filter(title == "Generalised borders G1 and area with urban character")
  
geodata_g1

## # A tibble: 1 × 12
##   collection_id      type  href  title description created updated crs   license
##   <chr>              <chr> <chr> <chr> <chr>       <chr>   <chr>   <chr> <chr>  
## 1 ch.bfs.generalisi… API   http… Gene… Administra… 2022-0… 2023-0… http… propri…
## # ℹ 3 more variables: provider_name <chr>, bbox <list>, inverval <list>

Download dataset by collection id with bfs_download_geodata() and unzip file if needed.

# Access Generalised borders G1 and area with urban character
borders_g1_path <- bfs_download_geodata(
  collection_id = "ch.bfs.generalisierte-grenzen_agglomerationen_g1", 
  output_dir = tempdir() #  temporary directory
)

# you may need to unzip the file
unzip(borders_g1_path[4], exdir = "borders_G1")

By default, the files are downloaded in a temporary directory. You can specify the folder where saving the files using the output_dir argument.

Some layers are accessible using WMS (Web Map Service):

library(leaflet)

leaflet() %>% 
  setView(lng = 8, lat = 46.8, zoom = 8) %>%
  addWMSTiles(
    baseUrl = "https://wms.geo.admin.ch/?", 
    layers = "ch.bfs.generalisierte-grenzen_agglomerationen_g2",
    options = WMSTileOptions(format = "image/png", transparent = TRUE),
    attribution = "Generalised borders G1 © 2024 BFS")

Cartographic base maps

You can get cartographic base maps from the ThemaKart project using bfs_get_base_maps(). The list of available geometries in the official documentation.

The default arguments of bfs_get_base_maps() can be change to access specific files:

# default arguments
bfs_get_base_maps(
  geom = NULL,
  category = "gf", # "gf" for total area (i.e. "Gesamtflaeche")
  type = "Poly",
  date = NULL,
  most_recent = TRUE, #get most recent file by default
  format = "shp",
  asset_number = "24025646" #change to get older ThemaKart data
)

A typical base maps ThemaKart file looks like this:

# list of geometry names: https://www.bfs.admin.ch/asset/en/24025645
switzerland_sf <- bfs_get_base_maps(geom = "suis")
communes_sf <- bfs_get_base_maps(geom = "polg", date = "20230101")
districts_sf <- bfs_get_base_maps(geom = "bezk")
cantons_sf <- bfs_get_base_maps(geom = "kant")
cantons_capitals_sf <- bfs_get_base_maps(geom = "stkt", type = "Pnts", category = "kk")
lakes_sf <- bfs_get_base_maps(geom = "seen", category = "11")

library(ggplot2)

ggplot() + 
  geom_sf(data = communes_sf, fill = "snow", color = "grey45") + 
  geom_sf(data = lakes_sf, fill = "lightblue2", color = "black") +
  geom_sf(data = districts_sf, fill = "transparent", color = "grey65") + 
  geom_sf(data = cantons_sf, fill = "transparent", color = "black") +
  geom_sf(data = cantons_capitals_sf, shape = 18, size = 3) +
  theme_minimal() +
  theme(axis.text = element_blank()) +
  labs(caption = "Source: ThemaKart, © BFS")

You can create an interactive map easily with the mapview R package.

library(mapview)

BFS::bfs_get_base_maps(geom = "bezk") |>
  mapview(zcol = "name", legend = FALSE)

Swiss Official Commune Register

The package also contains the official Swiss official commune registers for different administrative levels:

register_gde
register_gde_other
register_bzn
register_kt
register_kt_seeanteile
register_dic

# commune register data
BFS::register_gde

## # A tibble: 2,136 × 8
##    GDEKT GDEBZNR GDENR GDENAME            GDENAMK      GDEBZNA GDEKTNA GDEMUTDAT
##    <chr>   <dbl> <dbl> <chr>              <chr>        <chr>   <chr>   <chr>    
##  1 ZH        101     1 Aeugst am Albis    Aeugst am A… Bezirk… Zürich  1976-11-…
##  2 ZH        101     2 Affoltern am Albis Affoltern a… Bezirk… Zürich  1848-09-…
##  3 ZH        101     3 Bonstetten         Bonstetten   Bezirk… Zürich  1848-09-…
##  4 ZH        101     4 Hausen am Albis    Hausen am A… Bezirk… Zürich  1911-01-…
##  5 ZH        101     5 Hedingen           Hedingen     Bezirk… Zürich  1848-09-…
##  6 ZH        101     6 Kappel am Albis    Kappel am A… Bezirk… Zürich  1911-01-…
##  7 ZH        101     7 Knonau             Knonau       Bezirk… Zürich  1848-09-…
##  8 ZH        101     8 Maschwanden        Maschwanden  Bezirk… Zürich  1848-09-…
##  9 ZH        101     9 Mettmenstetten     Mettmenstet… Bezirk… Zürich  1848-09-…
## 10 ZH        101    10 Obfelden           Obfelden     Bezirk… Zürich  1848-09-…
## # ℹ 2,126 more rows

You can use registers to ease geodata analysis.

library(dplyr)
library(sf)

communes_sf <- bfs_get_base_maps(geom = "polg", date = "20230101")

communes_ge <- communes_sf |>
  inner_join(BFS::register_gde |> 
               filter(GDEKTNA == "Genève"), 
             by = c("id" = "GDENR"))

bbox_ge <- sf::st_bbox(communes_ge)

lake_leman <- bfs_get_base_maps(geom = "seen", category = "11") |>
  filter(name == "Lac Léman")

communes_ge |> 
  ggplot() + 
  geom_sf(data = lake_leman, fill = "lightblue2", color = "grey65") +
  geom_sf(fill = "snow", color = "grey65") + 
  geom_sf_text(aes(label = name), size = 3, check_overlap = T) + 
  # bounding box
  coord_sf(
    xlim = c(bbox_ge$xmin, bbox_ge$xmax),
    ylim = c(bbox_ge$ymin, bbox_ge$ymax)
  ) +
  theme_minimal() +
  theme(axis.text = element_blank()) +
  labs(title = "Communes du canton de Genève",
       x = NULL, y = NULL, 
       caption = "Source: ThemaKart, © BFS")

Main dependencies of the package

Under the hood, this package is using the pxweb package to query the Swiss Federal Statistical Office PXWEB API. PXWEB is an API structure developed by Statistics Sweden and other national statistical institutions (NSI) to disseminate public statistics in a structured way. To query the Geo Admin STAC API, this package is using the rstac package. STAC is a specification of files and web services used to describe geospatial information assets.

You can clean the column names of the datasets automatically using janitor::clean_names() by adding the argument clean_names = TRUE in the bfs_get_data() function.

Other information

This package is in no way officially related to or endorsed by the Swiss Federal Statistical Office (BFS).

Contribute

Any contribution is strongly appreciated. Feel free to report a bug, ask any question or make a pull request for any remaining issue.

BFS
BFS copied to clipboard

Metadata

BFS

Installation

Usage

Get the data catalog

Download data in any language

“Too Many Requests” error message

Query specific elements

Catalog of tables

Access geodata catalog

Download geodata

Cartographic base maps

Swiss Official Commune Register

Main dependencies of the package

Other information

Contribute

← Metadata

Owner

Metadata

BFS BFS copied to clipboard

Metadata

BFS

Installation

Usage

Get the data catalog

Download data in any language

“Too Many Requests” error message

Query specific elements

Catalog of tables

Access geodata catalog

Download geodata

Cartographic base maps

Swiss Official Commune Register

Main dependencies of the package

Other information

Contribute

← Metadata

Owner

Metadata

BFS
BFS copied to clipboard