Nobel Prize

Author

Renee Watson

Nobel Prize

Approach

For this assignment, I plan to use the Nobel Prize organization’s public APIs to explore patterns in Nobel Prize awards and laureates. I will retrieve the data in JSON format using R and the jsonlite package. Since the API data contains nested information, I will use purrr and dplyr from the tidyverse to organize the records into tidy data frames. I will create separate tables for prize records, laureate information, and individual laureate awards, then join the tables when needed for comparisons. I plan to investigate four questions: 1. Which Nobel Prize category has awarded the most prizes? 2. Which decade had the highest number of individual Nobel laureate awards? 3. Which Nobel Prize category has the highest percentage of female laureates? 4. Which Nobel Prize category has the highest median age among recipients?

Some challenges I anticipate include working with nested JSON fields, accounting for laureates who have received more than one award, and distinguishing individual laureates from organizations. I will also need to check for missing birth dates when estimating recipient’s ages. Because the age calculation uses birth year and award year, the ages will be approximate. For the decade comparison, I will note that the 2020s are not yet complete. I will use summary tables to present the findings and explain what each result means.

Data sources: Nobel Prize API, prize records

          Nobel Prize API, laureate records
          
          Nobel Prize Developer Zone

Reference: Nobel Prize Outreach. (n.d.). Nobel Prize Developer Zone (API version 2.1). https://www.nobelprize.org/about/developer-zone-2/

Codebase

library(tidyverse)
library(jsonlite)
library(knitr)

The API returns data in pages. The function below retrieves every page, so the analysis is not limited to the first 25 or 1,000 records. simplifyVector = FALSE preserves the nested JSON structure while the data are being read.

base_url <- "https://api.nobelprize.org/2.1"

get_all_records <- function(endpoint, field, page_size = 1000) {
  all_records <- list()
  offset <- 0

  repeat {
    url <- paste0(
      base_url, "/", endpoint,
      "?limit=", page_size, "&offset=", offset
    )

    response <- jsonlite::fromJSON(url, simplifyVector = FALSE)
    page <- response[[field]]

    if (is.null(page) || length(page) == 0) break

    all_records <- c(all_records, page)

    if (length(page) < page_size) break
    offset <- offset + page_size
  }

  all_records
}

prizes_json <- get_all_records("nobelPrizes", "nobelPrizes")
laureates_json <- get_all_records("laureates", "laureates")

cat("Prize records retrieved:", length(prizes_json), "\n")
Prize records retrieved: 688 
cat("Laureate records retrieved:", length(laureates_json), "\n")
Laureate records retrieved: 1028 

Some JSON properties are missing for certain records, particularly for organizations. This helper converts a missing value into NA.

get_text <- function(x) {
  if (is.null(x) || length(x) == 0) {
    NA_character_
  } else {
    as.character(x[[1]])
  }
}


prizes <- purrr::map_dfr(prizes_json, function(x) {
  tibble(
    award_year = as.integer(x$awardYear),
    category = get_text(x$category$en),
    recipients = length(x$laureates)
  )
})


laureates <- purrr::map_dfr(laureates_json, function(x) {
  tibble(
    laureate_id = as.character(x$id),
    name = coalesce(
      get_text(x$knownName$en),
      get_text(x$orgName$en),
      get_text(x$fullName$en)
    ),
    gender = get_text(x$gender),
    birth_date = get_text(x$birth$date)
  )
})

laureate_awards <- purrr::map_dfr(laureates_json, function(x) {
  purrr::map_dfr(x$nobelPrizes, function(p) {
    tibble(
      laureate_id = as.character(x$id),
      award_year = as.integer(p$awardYear),
      category = get_text(p$category$en)
    )
  })
}) %>%
  distinct(laureate_id, award_year, category)

individual_awards <- laureate_awards %>%
  left_join(laureates, by = "laureate_id") %>%
  filter(gender %in% c("male", "female"))

cat("Individual laureate-award records:", nrow(individual_awards), "\n")
Individual laureate-award records: 1005 

Question 1: Which Nobel Prize category has awarded the most prizes?

This question compares the number of awarded prizes in each category. I use the nobelPrizes data and exclude records for years when no prize recipient was named.

category_counts <- prizes %>%
  filter(recipients > 0, !is.na(category)) %>%
  count(category, name = "prizes_awarded", sort = TRUE)

knitr::kable(
  category_counts,
  col.names = c("Nobel Prize category", "Prizes awarded"),
  caption = "Number of awarded prizes by category"
)
Number of awarded prizes by category
Nobel Prize category Prizes awarded
Physics 120
Literature 119
Chemistry 118
Physiology or Medicine 117
Peace 107
Economic Sciences 58

The results show that Physics has the highest number of Nobel Prizes awarded, with 120 prizes, followed by Literature with 119 and Chemistry with 118. Physiology or Medicine has 117 prizes, while Peace has 107. Economic Sciences has the fewest, with 58 prizes.

Question 2: Which decade had the highest number of individual Nobel laureate awards?

For this question, I use the laureate-award data and group the award years into decades. I count each individual award once, including laureates who may have received more than one prize. Organizations are excluded from this comparison.

decade_counts <- individual_awards %>%
  filter(!is.na(award_year)) %>%
  mutate(decade = (award_year %/% 10) * 10) %>%
  count(decade, name = "individual_awards") %>%
  arrange(desc(individual_awards)) %>%
  mutate(decade = paste0(decade, "s"))

knitr::kable(
  decade_counts,
  col.names = c("Decade", "Individual laureate awards"),
  caption = "Individual Nobel laureate awards by decade"
)
Individual Nobel laureate awards by decade
Decade Individual laureate awards
2000s 119
2010s 117
1970s 103
1990s 101
1980s 94
2020s 82
1960s 75
1950s 71
1900s 56
1930s 55
1920s 54
1940s 40
1910s 38

The results show that the 2000s had the highest number of individual Nobel laureate awards, with 119, followed by the 2010s with 117. The 1970s ranked third with 103 awards. The 1910s had the fewest, with 38 awards.

Question 3: Which Nobel Prize category has the highest percentage of female laureates?

I compare gender and prize category using the individual laureate records. I count each person only once per category, even if that person received more than one prize in that category. Organizations are excluded because they are not individual men or women.

gender_by_category <- individual_awards %>%
  distinct(laureate_id, category, gender) %>%
  group_by(category) %>%
  summarise(
    women = sum(gender == "female"),
    total_individuals = n(),
    female_percent = round(100 * women / total_individuals, 1),
    .groups = "drop"
  ) %>%
  arrange(desc(female_percent))

knitr::kable(
  gender_by_category,
  col.names = c("Category", "Female laureates", "All individual laureates", "Female (%)"),
  caption = "Percentage of female Nobel laureates by category"
)
Percentage of female Nobel laureates by category
Category Female laureates All individual laureates Female (%)
Peace 21 113 18.6
Literature 19 123 15.4
Physiology or Medicine 14 235 6.0
Chemistry 8 200 4.0
Economic Sciences 3 101 3.0
Physics 5 230 2.2

The results show that Peace has the highest percentage of female Nobel laureates, with 18.6% (21 out of 113), followed by Literature at 15.4% (19 out of 123). Physiology or Medicine has 6.0%, Chemistry has 4.0%, and Economic Sciences has 3.0%. Physics has the lowest percentage of female laureates at 2.2% (5 out of 230).

Question 4: Which Nobel Prize category has the highest median age among recipients?

This question requires combining information from different fields. I join each person’s birth date to their Nobel Prize award year using laureate_id. I then subtract the birth year from the award year and compare the median estimated age across categories. The ages are estimates based on birth year and award year rather than exact dates, and recipients with missing birth information are excluded from the analysis.

ages_at_award <- individual_awards %>%
  mutate(
    birth_year = as.integer(substr(birth_date, 1, 4)),
    estimated_age = award_year - birth_year
  ) %>%
  filter(
    !is.na(estimated_age),
    between(estimated_age, 0, 120)
  )

age_by_category <- ages_at_award %>%
  group_by(category) %>%
  summarise(
    laureate_awards = n(),
    median_age = round(median(estimated_age), 1),
    mean_age = round(mean(estimated_age), 1),
    .groups = "drop"
  ) %>%
  arrange(desc(median_age))

knitr::kable(
  age_by_category,
  col.names = c("Category", "Awards with birth year", "Median age", "Mean age"),
  caption = "Estimated age at Nobel Prize award by category"
)
Estimated age at Nobel Prize award by category
Category Awards with birth year Median age Mean age
Economic Sciences 101 67 67.2
Literature 123 67 65.0
Peace 113 62 61.0
Chemistry 202 58 59.5
Physiology or Medicine 235 58 59.0
Physics 231 56 57.8

The results show that Economic Sciences and Literature have the highest median age at the time of the award, both at approximately 67 years. Peace follows with a median age of 62 years, while Chemistry and Physiology or Medicine both have a median age of 58 years. Physics has the lowest median age at 56 years.

Conclusion

This analysis used JSON data from the Nobel Prize APIs to examine patterns in Nobel Prize awards and recipients. The results showed that Physics had the most prizes awarded, while the 2000s had the highest number of individual laureate awards. Peace had the highest percentage of female laureates. Economic Sciences and Literature had the highest median age among recipients. Overall, the findings demonstrate how JSON data can be transformed and analyzed in R to explore historical patterns and differences across Nobel Prize categories.