Introduction:

This is a dataset that I’ve chosen from the Harvard Dataverse containing a list of awarded noble prizes and metadata about the winners.

https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/HYRJDX

Load Libraries: 

library(tidyr)
library(dplyr)
## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
library(ggplot2)
Read in data: 

raw_data_path<-("https://raw.githubusercontent.com/amedina613/Data-607-Project-2/main/complete.csv")

raw_data <- read.csv(raw_data_path, sep = ",")
print(colnames(raw_data))
##  [1] "awardYear"                  "category"                  
##  [3] "categoryFullName"           "sortOrder"                 
##  [5] "portion"                    "prizeAmount"               
##  [7] "prizeAmountAdjusted"        "dateAwarded"               
##  [9] "prizeStatus"                "motivation"                
## [11] "categoryTopMotivation"      "award_link"                
## [13] "id"                         "name"                      
## [15] "knownName"                  "givenName"                 
## [17] "familyName"                 "fullName"                  
## [19] "penName"                    "gender"                    
## [21] "laureate_link"              "birth_date"                
## [23] "birth_city"                 "birth_cityNow"             
## [25] "birth_continent"            "birth_country"             
## [27] "birth_countryNow"           "birth_locationString"      
## [29] "death_date"                 "death_city"                
## [31] "death_cityNow"              "death_continent"           
## [33] "death_country"              "death_countryNow"          
## [35] "death_locationString"       "orgName"                   
## [37] "nativeName"                 "acronym"                   
## [39] "org_founded_date"           "org_founded_city"          
## [41] "org_founded_cityNow"        "org_founded_continent"     
## [43] "org_founded_country"        "org_founded_countryNow"    
## [45] "org_founded_locationString" "ind_or_org"                
## [47] "residence_1"                "residence_2"               
## [49] "affiliation_1"              "affiliation_2"             
## [51] "affiliation_3"              "affiliation_4"

Tidy Data

There are multiple columns with the same information, such as “residence_1”, and “residence_2.” In an effort to tidy data, let’s concatenate the info: 

raw_data <- raw_data %>%
  mutate(residence = coalesce(residence_1, residence_2)) %>%
  select(-c(residence_1, residence_2))
Concatenate affiliation columns into one column to tidy up the data: 

raw_data <- raw_data %>%
  mutate(affiliation = coalesce(affiliation_1, affiliation_2, affiliation_3, affiliation_4)) %>%
  select(-c(affiliation_1, affiliation_2, affiliation_3, affiliation_4))

This is quite a wide data set and these are just some examples of what you can do

Tidy data for analysis

Split up the data into two separate tables: Awards (awardYear, category, prizeAmount) and Recipients (name, birth_date, gender): 

awards <- na.omit(subset(raw_data, select = c("awardYear", "category", "prizeAmount")))

recipients <- na.omit(subset(raw_data, select = c("name", "birth_date", "gender")))
Some of the recipients are organizations. Filter those out for the sake of the gender analysis using dplyr’s filter function 

recipients_without_gender <- recipients %>%
  filter(is.na(gender) | gender == "")

recipients <- recipients %>%
  filter(!is.na(gender), gender != "")

I made a df for the recipients without gender to make sure everything checks out. It does, 950-27 is 923, which is the number of obs. in the filtered final df.

What proportion of the awards are given to women vs men? How has this changed over time?

Create a data frame of gender data: 

gender_counts <- table(recipients$gender)

proportion_women <- gender_counts["female"] / sum(gender_counts)
proportion_men <- gender_counts["male"] / sum(gender_counts)

gender_data <- data.frame(
  Gender = c("Female", "Male"),
  Proportion = c(proportion_women, proportion_men)
)
Plot the data: 

ggplot(gender_data, aes(x = Gender, y = Proportion, fill = Gender)) +
  geom_bar(stat = "identity") +
  labs(title = "Proportion of Awards by Gender",
       x = "Gender",
       y = "Proportion") +
  scale_fill_manual(values = c("Female" = "blue", "Male" = "red")) +
  theme_minimal()

Which field has received the largest prize amount? Which field has received the lowest?

Group by field and calculate the total prize amount for each field: 

field_prize_sum <- awards %>%
  group_by(category) %>%
  summarise(total_prize_amount = sum(prizeAmount, na.rm = TRUE))
Print the min and max prize: 

max_prize_field <- field_prize_sum %>%
  filter(total_prize_amount == max(total_prize_amount))

min_prize_field <- field_prize_sum %>%
  filter(total_prize_amount == min(total_prize_amount))


print("Field with the largest prize amount:")
## [1] "Field with the largest prize amount:"
print(max_prize_field)
## # A tibble: 1 × 2
##   category total_prize_amount
##   <chr>                 <int>
## 1 Physics           725890928
print("Field with the lowest prize amount:")
## [1] "Field with the lowest prize amount:"
print(min_prize_field)
## # A tibble: 1 × 2
##   category   total_prize_amount
##   <chr>                   <int>
## 1 Literature          289282102

How have the number of awards given out each year changed over time?

I attempted to make a bar plot to visualize the number of awards given out each year for each award type, however there was so much information in the plot, it was not decipherable. So I decided to simplify the plot by aggregating by decade.

The floor function was used for this.

awards_by_decade_type <- awards %>%
  group_by(decade = floor(awardYear / 10) * 10, category) %>%
  summarise(num_awards = n(), .groups = 'drop')

ggplot(awards_by_decade_type, aes(x = as.factor(decade), y = num_awards, fill = category)) +
  geom_bar(stat = "identity", position = "dodge") +
  labs(x = "Decade", y = "Number of Awards", fill = "Award Type") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))