Query Open Alex for UoA, get OA proportions, plot

.gitignore

@@ -0,0 +1,4 @@
+.Rproj.user
+.Rhistory
+.RData
+.Ruserdata

oa-nz.Rproj

@@ -0,0 +1,13 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX

open-alex-query.Rmd

@@ -0,0 +1,58 @@
+---
+title: "open-alex-query"
+author: "Tom Saunders"
+date: "2023-12-20"
+output: html_document
+---
+
+```{r}
+library(jsonlite)
+library(dplyr)
+library(ggplot2)
+
+options(stringsAsFactors = FALSE)
+```
+
+```{r}
+# Set base url to OpenAlex API with 'works' endpoint
+base_url <- "https://api.openalex.org/works"
+
+# Choose relevant institution based on ROR
+ror <- "institutions.ror:https://ror.org/03b94tp07"
+
+# Choose filters
+my_filters <- paste(c(
+                ror, 
+                "is_paratext:false",
+                "type:article|book|book-chapter",
+                "from_publication_date:2020-01-01",
+                "to_publication_date:2022-12-31",
+                "is_retracted:false"), 
+                collapse = ",")
+
+# Select grouping
+group <- "&group_by=oa_status"
+
+# Create url
+url <- paste0(base_url, "?", "filter=", my_filters, group)
+
+# Make request and parse output
+json <- (fromJSON(url)) 
+
+# Convert to data frame
+oa_status <- as.data.frame(json$group_by)
+
+# Calculate percentages for each oa type
+oa_status <- oa_status |> 
+  mutate(
+    pc = (count/sum(count)*100),
+  )
+
+# Plot
+ggplot(oa_status, aes(reorder(key, -pc), pc)) +
+  geom_col() +
+  theme_classic() +
+  xlab("") +
+  ylab("%")
+```
+