Work seamlessly with World Bank World Development Indicators (WDI) • wbwdi

wbwdi is an R package to access and analyze the World Bank’s World Development Indicators (WDI) using the corresponding API. WDI provides more than 24,000 country or region-level indicators for various contexts. wbwdi enables users to download, process and work with WDI series across multiple geographies and time periods.

The package is designed to work seamlessly with International Debt Statistics (IDS) provided through the wbids package and shares its syntax with its sibling Python library wbwdi. It follows the principles of the econdataverse.

This package is a product of Christoph Scheuch and not sponsored by or affiliated with the World Bank in any way, except for the use of the World Bank WDI API.

Installation

You can install the development version of wbwdi like this:

pak::pak("tidy-intelligence/r-wbwdi")

Usage

The main function wdi_get() provides an interface to download multiple WDI series for multiple geographies and specific date ranges.

wdi_get(
  geographies = c("MX", "CA", "US"), 
  indicators = c("NY.GDP.PCAP.KD", "SP.POP.TOTL"),
  start_year = 2020, end_year = 2024
)

You can also download these indicators for all geographies and available dates:

wdi_get(
  geographies = "all", 
  indicators = c("NY.GDP.PCAP.KD", "SP.POP.TOTL")
)

Some indicators are also available on a monthly basis, e.g.:

wdi_get(
  geographies = "AUT", 
  indicators = "DPANUSSPB",         
  start_year = 2012, end_year = 2015, 
  frequency = "month"
)

Similarly, there are also some indicators available on a quarterly frequency, e.g.:

wdi_get(
  geographies = "NGA", 
  indicators =  "DT.DOD.DECT.CD.TL.US",
  start_year = 2012, end_year = 2015, 
  frequency = "quarter"
)

You can get a list of all indicators supported by the WDI API via:

wdi_get_indicators()

You can get a list of all supported geographies via:

wdi_get_geographies()

You can also get the list of supported indicators and geographies in another language, but note that not everything seems to be translated into other languages:

wdi_get_indicators(language = "es")
wdi_get_geographies(language = "zh")

Check out the following function for a list of supported languages:

wdi_get_languages()

In addition, you can list supported regions, sources, topics and lending types, respectively:

wdi_get_regions()
wdi_get_sources()
wdi_get_topics()
wdi_get_lending_types()

If you want to search for specific keywords among indicators or other data sources, you can use the RStudio or Positron data explorer. Alternatively, this package comes with a helper function:

indicators <- wdi_get_indicators()
wdi_search(
  indicators,
  keywords = c("inequality", "gender"),
  columns = c("indicator_name")
)

Relation to Existing R Packages

The most important differences to existing packages are that wbwdi is designed to (i) have a narrow focus on World Bank Development Indicators, (ii) have a consistent interface with other R packages (e.g., wbids), (iii) have an MIT license, and (iv) have a shared interface with Python libraries (e.g., wbwdi). wbwdi also refrains from using cached data because this approach frequently leads to problems for users due to outdated caches and it uses httr2 to manage API requests and parse responses.

More specifically, the differences of existing CRAN releases (apart from interface design) are:

WDI: uses cached data by default, does not allow downloading meta data from the WDI API (e.g., languages, sources, topics), has a GPL-3 license, and does neither use httr nor httr2 for requests.
worldbank: does not have a narrow focus because it includes the Poverty and Inequality Platform and Finances One API.
wbstats: uses cached data by default, does not use httr2.