CRAN status Grand total pipeline status

BFS

The BFS package allows to search and download public data from the Swiss Federal Statistical Office (BFS) in a dynamic and reproducible way.

Installation

# Install the released version from CRAN
install.packages("BFS")

To get a bug fix, or use a feature from the development version, you can install BFS from GitHub.

# install from Github
devtools::install_github("lgnbhl/BFS")

Usage

library(BFS)

Download metadata

To search and download data from the Swiss Federal Statistical Office, you first need to retrieve information about the available public datasets. The function bfs_get_metadata() returns a data frame/tibble containing the titles, publication dates, observation periods, data sources, website urls and download urls of all the BFS datasets available in a given language. You can get the metadata in German (“de”, by default), French (“fr”), Italian (“it”) or English (“en”). Note that Italian and English metadata give access to less datasets.

meta_en <- bfs_get_metadata(language = "en")

head(meta_en)
## # A tibble: 6 x 6
##   title         observation_peri… published  source   url_bfs        url_px     
##   <chr>         <chr>             <chr>      <chr>    <chr>          <chr>      
## 1 University o… 1997-2019         25.03.2020 Federal… https://www.b… https://ww…
## 2 University o… 1997-2019         25.03.2020 Federal… https://www.b… https://ww…
## 3 University s… 1980-2019         25.03.2020 Federal… https://www.b… https://ww…
## 4 University s… 1990-2019         25.03.2020 Federal… https://www.b… https://ww…
## 5 Criminal off… 2009-2019         23.03.2020 Federal… https://www.b… https://ww…
## 6 Defendants r… 2009-2019         23.03.2020 Federal… https://www.b… https://ww…

Search for data

To search for a specific dataset title in the BFS metadata, you can use the bfs_search() function. This function leverages the R base function grepl() but calls the data argument first to allow the use of the pipe operator %>%.

library(magrittr)

meta_en_uni <- bfs_get_metadata("en") %>%
  bfs_search("university students")

print(meta_en_uni)
## # A tibble: 2 x 6
##   title         observation_peri… published  source   url_bfs        url_px     
##   <chr>         <chr>             <chr>      <chr>    <chr>          <chr>      
## 1 University s… 1980-2019         25.03.2020 Federal… https://www.b… https://ww…
## 2 University s… 1990-2019         25.03.2020 Federal… https://www.b… https://ww…

Download dataset

To download a BFS dataset, add the related url link from the url_px column of the downloaded metadata as an argument to the bfs_get_dataset() function. You can choose the language (German, French, Italian or English if any) in which the dataset is downloaded with the language argument.

df_uni <- bfs_get_dataset(url_px = meta_en_uni$url_px[1], language = "en")

print(df_uni)
## # A tibble: 16,800 x 5
##    level_of_study                           gender isced_field      year   value
##    <fct>                                    <fct>  <fct>            <fct>  <dbl>
##  1 First university degree or diploma       Male   Education scien… 2019/…    46
##  2 Bachelor                                 Male   Education scien… 2019/…   149
##  3 Master                                   Male   Education scien… 2019/…   131
##  4 Doctorate                                Male   Education scien… 2019/…   120
##  5 Further education, advanced studies and… Male   Education scien… 2019/…    14
##  6 First university degree or diploma       Female Education scien… 2019/…    62
##  7 Bachelor                                 Female Education scien… 2019/…   696
##  8 Master                                   Female Education scien… 2019/…   540
##  9 Doctorate                                Female Education scien… 2019/…   313
## 10 Further education, advanced studies and… Female Education scien… 2019/…    24
## # … with 16,790 more rows

You can access additional information about the downloaded dataset using the R base attributes() function.

attributes(df_uni) %>%
  str()
## List of 14
##  $ names        : chr [1:5] "level_of_study" "gender" "isced_field" "year" ...
##  $ row.names    : int [1:16800] 1 2 3 4 5 6 7 8 9 10 ...
##  $ download_date: Date[1:1], format: "2020-03-25"
##  $ contact      : chr "Section Educational Processes, e-mail  <a href=mailto:[email protected]>[email protected]</a>"
##  $ description  : chr "University students by year, ISCED field, gender and level of study"
##  $ last_update  : chr "20200325 08:30"
##  $ link         : chr "https://www.bfs.admin.ch/asset/en/px-x-1502040100_131"
##  $ note         : chr "<B>Meta information</B>#Data as of: 25.03.2020#Survey: Students and graduates in higher education institutions "| __truncated__
##  $ subject_area : chr "15 - Education and science"
##  $ survey       : chr "Students and degrees of higher education institutions (SHIS-studex)"
##  $ title        : chr "University students by Year, ISCED field, Gender and Level of study"
##  $ source       : chr "FSO - Students and degrees of higher education institutions - © FSO"
##  $ units        : chr "Person"
##  $ class        : chr [1:3] "tbl_df" "tbl" "data.frame"

In case the function fails to download the dataset, you can have a look at its related BFS webpage using the url_bfs link.

browseURL(meta_en_uni$url_bfs[1]) # open webpage

Sometimes the PC-Axis file of the dataset doesn’t exist. You should then use the “STAT-TAB - interactive table” service provided by BFS to download manually the dataset.

Data caching

Data caching is handled using the pins R package. To open the folder containing all the BFS datasets you already downloaded, you can use the bfs_open_dir() function.

bfs_open_dir()

If a dataset has already been downloaded during the day, the functions bfs_get_metadata() and bfs_get_dataset() retrieve the already downloaded datasets from your local pins caching folder. Caching speeds up code and reduces BFS server requests. However, if a given dataset has not been downloaded during the day, the functions download it again to be sure that you have the lastest data available. You can also force the download using the force argument.

Other information

A blog article showing a concret example about how to use the BFS package.

Alternative R package: pxweb.

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