polisci-journal-oa.R

# Setup and Functions -----------------------------------------------------

library(httr2)
library(jsonlite)
library(dplyr)
library(ggplot2)
journals = c("AJPS", "APSR", "BJPS", "CPS", "EJIR", "EJPR", "IO", "ISQ", "JOP", "JCR")
issns = c("1540-5907", "0003-0554", "1469-2112", "0010-4140", "1354-0661", "1475-6765", "0020-8183", "1468-2478", "1468-2508", "1552-8766")
years = c("2020", "2021", "2022", "2023")

# Returns a list with totalArticles and oaArticles from CrossRef API
# Input is ISSN and Year, both as a string
getOAArticles = function(issn, year) {
  dateString = paste0('from-pub-date:', year, '-01-01,until-pub-date:', year, '-12-31')
  
  # Construct API Request
  journalData = httr2::request("https://api.crossref.org/journals/") |> 
    req_url_path_append(issn) |> req_url_path_append("works") |>
    req_url_query(`filter`=dateString, `facet`='license:*') |>
    req_perform() |>resp_body_json()
  
  
  totalArticles = journalData$message$`total-results`
  
  licenseFacet = journalData$message$facets$license$values
  # sum up all creativecommons license
  oaArticles = sum(unlist(licenseFacet[grepl("creativecommons.org", names(licenseFacet))]))
  return (list("totalArticles" = totalArticles, "oaArticles" = oaArticles))
}



# Data retrieval ----------------------------------------------------------


data = data.frame(matrix(ncol =5, nrow = 0))
colnames(data) = c("journal",  "issn", "year", "totalArticles", "oaArticles" )

for (i in 1:length(years)) {
  year = years[i]
  for (j in 1:length(issns)) {
    journal = journals[j]
    issn = issns[j]
    articles = getOAArticles(issn, year)
    totalArticles = articles$totalArticles
    oaArticles = articles$oaArticles
    data = rbind(data, data.frame("journal" = journal, "issn" = issn, "year" = year, "totalArticles" = totalArticles, "oaArticles" = oaArticles))
    Sys.sleep(2)
    print(paste0("Done with ", journal, ", ", year))
  }
}


# Clean and plot ----------------------------------------------------------


# Drop JOP, which has no crossref data
data = data |> filter(journal != "JOP")

data$oaPercentage = (data$oaArticles / data$totalArticles) * 100

# ggplot code with chatGPT assistance
# 2023 bar graph
data |> filter(year == "2023") |> ggplot(aes(x = journal, y = oaPercentage, fill = journal)) +
  geom_bar(stat = "identity", width = 0.7) +
  labs(title = "Share of OA Articles by Journal in 2023",
       x = "Journal",
       y = "Percentage of OA Articles (%)") +
  theme_minimal(base_size = 15) +  # Clean theme with larger base font size
  theme(
    axis.text.x = element_text(angle = 0, vjust = 0.5, hjust = 0.5),
    axis.text.y = element_text(size = 12),
    plot.title = element_text(hjust = 0.5, size = 18, face = "bold"),
    legend.position = "none"  # Remove legend for journal names
  ) +
  scale_fill_brewer(palette = "Set3")  # Add nice color palette for the bars

# All years
ggplot(data, aes(x = journal, y = oaPercentage, fill = as.factor(year))) +
  geom_bar(stat = "identity", position = position_dodge(width = 0.7), width = 0.7) +  # Group bars by journal
  geom_text(aes(label = year), 
            position = position_dodge(width = 0.7),  hjust = 1.5, size = 3.5, angle = 90) +  # Add labels for years
  labs(title = "Share of OA Articles by Journal (2020-2023)",
       x = "Journal",
       y = "Percentage of OA Articles (%)",
       fill = "Year") +
  theme_minimal(base_size = 15) +  # Clean theme with larger base font size
  theme(
    axis.text.x = element_text(angle = 45, hjust = 1, size = 12),  # Tilt journal names for better readability
    axis.text.y = element_text(size = 12),
    legend.position = "none",
    plot.title = element_text(hjust = 0.5, size = 18, face = "bold"),
  ) +
  scale_fill_brewer(palette = "Set3")  # Consistent colors for the years