us_phds.Rmd

---
title: "US PhDs"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

```{r}
library(tidyverse)
theme_set(theme_light())

# Major field of study
major_fields_raw <- readxl::read_xlsx("~/Downloads/data_tables/sed17-sr-tab012.xlsx",
                  skip = 3)

major_fields <- major_fields_raw %>%
  rename(field = `Field of study`) %>%
  gather(key, value, -field) %>%
  mutate(year = as.numeric(ifelse(str_detect(key, "X__"), NA, key)),
         type = ifelse(!str_detect(value, "Number|Percent"), NA, value),
         value = as.numeric(value)) %>%
  fill(year, type) %>%
  select(-key) %>%
  filter(!is.na(value)) %>%
  spread(type, value)
```

```{r}
fine_fields <- readxl::read_xlsx("~/Downloads/data_tables/sed17-sr-tab013.xlsx",
                  skip = 3) %>%
  rename(field = 1) %>%
  gather(year, number, -field) %>%
  mutate(year = as.numeric(year),
         number = as.numeric(number)) %>%
  filter(!is.na(number))

fine_fields %>%
  filter(field %in% sample(unique(field), 6)) %>%
  ggplot(aes(year, number, color = field)) +
  geom_line()
```

```{r}
# get the broad field names and the major field names

sex <- c("All", "Male", "Female", "All doctorate recipientsa", "All fieldsa")

broad_fields <- readxl::read_xlsx("~/Downloads/data_tables/sed17-sr-tab014.xlsx", skip = 4) %>%
  rename(field = 1) %>%
  filter(!field %in% sex) %>%
  mutate(field = fct_recode(field,
                            "Life sciences" = "Life sciencesb",
                            "Other" = "Otherc")) %>%
  pull(field) %>%
  as.character()
```


```{r}
recipients_year_field_sex <- readxl::read_xlsx("~/Downloads/data_tables/sed17-sr-tab015.xlsx", skip = 3) %>%
  rename(field = 1) %>%
  select(-contains("change")) %>%
  mutate(field = as.character(fct_recode(field, "All" = "All doctorate recipientsa",
                                         "Other" = "Otherb")),
         sex = if_else(field %in% sex, field, NA_character_),
         broad_field = ifelse(field %in% broad_fields, field, NA)) %>%
  fill(sex, broad_field) %>%
  gather(year, number, -sex, -broad_field, -field) %>%
  mutate(year = as.numeric(year),
         number = as.numeric(number)) %>%
  filter(!field %in% sex) %>%
  filter(!is.na(number))

recipients_year_field_sex %>%
  filter(sex != "All",
         broad_field == "Mathematics and computer sciences") %>%
  ggplot(aes(year, number, color = sex)) +
  geom_line() +
  expand_limits(y = 0) +
  facet_wrap(~ field)

recipients_year_field_sex %>%
  spread(sex, number) %>%
  mutate(pct_male = Male / All) %>%
  filter(broad_field == "Engineering") %>%
  mutate(field = fct_reorder(field, -pct_male)) %>%
  ggplot(aes(year, pct_male, color = field)) +
  geom_line() +
  scale_y_continuous(labels = scales::percent_format()) +
  labs(x = "Year",
       y = "% of PhD recipients reporting as male",
       color = "Major field",
       title = "Breakdown by sex over time within Engineering major fields")

recipients_year_field_sex %>%
  spread(sex, number) %>%
  mutate(pct_male = Male / All) %>%
  filter(broad_field == "Humanities and arts") %>%
  mutate(field = fct_reorder(field, -pct_male)) %>%
  ggplot(aes(year, pct_male, color = field)) +
  geom_line() +
  scale_y_continuous(labels = scales::percent_format()) +
  labs(x = "Year",
       y = "% of PhD recipients reporting as male",
       color = "Major field",
       title = "Breakdown by sex over time within Humanities & Arts major fields")

recipients_year_field_sex %>%
  spread(sex, number) %>%
  mutate(pct_male = Male / All) %>%
  filter(broad_field == "Education") %>%
  mutate(field = fct_reorder(field, -pct_male)) %>%
  ggplot(aes(year, pct_male, color = field)) +
  geom_line() +
  scale_y_continuous(labels = scales::percent_format()) +
  labs(x = "Year",
       y = "% of PhD recipients reporting as male",
       color = "Major field",
       title = "Breakdown by sex over time within Education major fields")
```

Three levels:

* Broad field (Life sciences)
* Major field (Biological and biomedical sciences)
* Subfield (Computational biology)