In KTH-Library/institutions: Global Research Identifier Database (GRID)

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(institutions)
library(dplyr)

institutions_download()

A static dataset from October 2020 from the Research Organization Registry Community has been bundled in this R package.

It provides GRID identifiers for research organizations and can therefore be used to link to the data otherwise provided through functions in this R package.

Some examples of usage follow:

# get a GRID identifier for a specific institution
id_kth_grid <- 
  institutions_search("Royal Institute of Technology")$grid_id

# get the corresponding ROR identifier
id_kth_ror <- 
  ror$ror_orgs %>%
  filter(GRID == id_kth_grid) %>%
  pull(id)

# what ROR core info can we get for this identifier?
ror$ror_orgs %>% filter(id == id_kth_ror) %>% glimpse()

# what other identifiers are associated with this org?
ror$ror_ids %>% filter(id == id_kth_ror)

# ROR child table entries for labels, aliases, acronyms, links and types?

# what aliases?
ror$ror_aliases %>% filter(id == id_kth_ror)

# what acronyms?
ror$ror_acronyms %>% filter(id == id_kth_ror)

# what links?
ror$ror_links %>% filter(id == id_kth_ror)

# what types?
ror$ror_types %>% filter(id == id_kth_ror)

Some summaries

The full dataset is available and can be summarized.

Using "child tables" for summaries

A few examples:

# what top five types of orgs are represented in this data?
ror$ror_types %>% count(types) %>% arrange(desc(n)) %>% head(5)

# top 5 companies with many aliases
ror$ror_aliases %>% count(aliases) %>% arrange(desc(n)) %>% head(5)

# why does IBM have so many acronyms? seemingly because it is a multi-national corp.
ror$ror_aliases %>% filter(aliases == "International Business Machines Corporation") %>% 
  inner_join(ror$ror_orgs) %>%  pull(name) %>% paste0(collapse = ", ")

# what is the most frequently used acronym and which orgs use this abbreviation for its name?
ror$ror_acronyms %>% count(acronyms) %>% arrange(desc(n))
ror$ror_acronyms %>% filter(acronyms == "CCC") %>% inner_join(ror$ror_orgs) %>% pull(name) %>% paste0(collapse = ", ")

# some orgs have more than one link, seems to be duplicates?
ror$ror_links %>% group_by(id) %>% count(links) %>% filter(n > 1) %>% ungroup()
ror$ror_links %>% filter(id == "https://ror.org/01eea1w69")

External identifiers

extids <- ror$ror_ids

# which external identifiers are most frequently provided?
# one given identifier can have multiple other identifiers esp FundRef
extids %>%
  group_by(name) %>%
  count() %>%
  arrange(desc(n))

# which WikiData identifiers does a set of orgs have?
ids_wikidata <- 
  extids %>% filter(
  id %in% c("https://ror.org/052gg0110", "https://ror.org/013meh722"),
  name == "Wikidata") %>%
  distinct() %>%
  pull("value")

knitr::kable(escape = FALSE, tibble(url = sprintf("<a href='https://www.wikidata.org/wiki/%s'>%s</a>", ids_wikidata, ids_wikidata)))