bpsr allows users to get Statistics Indonesia’s (BPS) datasets and other resources from R by wrapping up its application programming interface (API).
Installation
You can install the development version of bpsr like so:
# install.packages("devtools")
devtools::install_github("dzulfiqarfr/bpsr")Usage
You need to register yourself on BPS API’s website first to be able to use bpsr.
All functions from bpsr have a bps_ prefix, so we can benefit from an automatic code completion.
Most functions to get resources take a string of a resource identifier (ID) as the first argument. So we typically need to look it up first in the table of the resource of interest. Functions to get such tables have a noun-y name. For example, use bps_dataset() to request the dataset table.
table <- bps_dataset(lang = "eng")
table
#> # A tibble: 10 × 9
#> dataset_id title subject_id subject def notes vertical_var_group_id unit
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <lgl>
#> 1 70 Popula… 2 Commun… <NA> <… 1 NA
#> 2 111 Percen… 2 Commun… <NA> Sour… 1 NA
#> 3 391 Percen… 2 Commun… <NA> Sour… 1 NA
#> 4 392 Active… 2 Commun… <NA> Sour… 1 NA
#> 5 393 Percen… 2 Commun… <NA> Sour… 1 NA
#> 6 395 Percen… 2 Commun… <NA> Sumb… 1 NA
#> 7 396 Percen… 2 Commun… <NA> <… 1 NA
#> 8 398 Percen… 2 Commun… <NA> <… 1 NA
#> 9 402 Percen… 2 Commun… <NA> Sour… 152 NA
#> 10 403 Househ… 2 Commun… <NA> Sour… 1 NA
#> # ℹ 1 more variable: graph <int>Functions to request datasets have bps_get_ prefix. Use bps_get_dataset() to request a dataset.
data <- bps_get_dataset(table$dataset_id[[1]], lang = "eng")
data
#> # A tibble: 558 × 5
#> vertical_var derived_var year period var
#> <chr> <chr> <int> <chr> <dbl>
#> 1 INDONESIA Male 2014 Annual 18.8
#> 2 INDONESIA Male 2015 Annual 23.7
#> 3 INDONESIA Male 2016 Annual 27.2
#> 4 INDONESIA Male 2017 Annual 34.5
#> 5 INDONESIA Male 2018 Annual 42.3
#> 6 INDONESIA Male 2019 Annual 50.5
#> 7 INDONESIA Male 2020 Annual 56.6
#> 8 INDONESIA Male 2021 Annual 65.0
#> 9 INDONESIA Female 2014 Annual 15.4
#> 10 INDONESIA Female 2015 Annual 20.2
#> # ℹ 548 more rows
#> # ℹ Read the metadata with `bps_metadata()`Under the hood, bpsr parses the JSON returned by the API into a tibble. More importantly, the package arranges the dataset in a tidy structure, which is well suited for analysis.
Use bps_get_trade_hs_*() to request trade datasets. These trade functions take the Harmonized System (HS) chapter or code as the first argument. To look up the HS chapter or code, use bps_hs_code().
hs <- bps_hs_code()
export <- bps_get_trade_hs_chapter(
hs$hs_chapter[[1]],
"export",
year = 2021,
freq = "annual"
)
export
#> # A tibble: 67 × 7
#> hs_chapter description port partner year net_weight export
#> <chr> <chr> <chr> <chr> <int> <dbl> <dbl>
#> 1 01 Binatang hidup ATAPUPU EAST T… 2021 15036. 1.98e5
#> 2 01 Binatang hidup KUALA NAMU INTERNA… JAPAN 2021 5 1.03e2
#> 3 01 Binatang hidup KUALA NAMU INTERNA… SINGAP… 2021 14950 1.77e4
#> 4 01 Binatang hidup KUALA TANJUNG MALAYS… 2021 23601 6.51e4
#> 5 01 Binatang hidup NGURAH RAI (U) EAST T… 2021 1932. 1.13e4
#> 6 01 Binatang hidup NGURAH RAI (U) HONG K… 2021 240 2.05e4
#> 7 01 Binatang hidup SEKUPANG SINGAP… 2021 25167463. 5.59e7
#> 8 01 Binatang hidup SOEKARNO-HATTA (U) AFGHAN… 2021 128 5.50e4
#> 9 01 Binatang hidup SOEKARNO-HATTA (U) AUSTRIA 2021 60 2.07e2
#> 10 01 Binatang hidup SOEKARNO-HATTA (U) BAHRAIN 2021 127 8.40e3
#> # ℹ 57 more rows
#> # ℹ Read the metadata with `bps_metadata()`Some resources are available for download. We can use the download functions to get those resources. This is especially useful to get a dataset stored in a spreadsheet, which comes in a Microsoft Excel file format with the “.xls” extension.
spreadsheet <- bps_dataset_spreadsheet(lang = "eng")
bps_download_spreadsheet(spreadsheet$dataset_id[[1]], "spreadsheet.xls")Aside from the common usage, bpsr also provides other features, including:
- requesting multiple datasets at once using
bps_get_datasets(); and
- downloading multiple resources at once using
bps_download().
Learn more in the Getting Started vignette (vignette("bpsr")).