Trends in gender neutral names in the US

babynames.R

library(ggplot2)
library(data.table)
library(magrittr)

# From https://www.ssa.gov/oact/babynames/limits.html
file <- "https://www.ssa.gov/oact/babynames/state/namesbystate.zip"
destination <- "/home/elio/Downloads/babies.zip"
download.file(file, destination)

dir <- "/home/elio/Downloads/babies"
dir.create(dir)
utils::unzip(destination, overwrite = TRUE, exdir = dir)

bb_state <- list.files("/home/elio/Downloads/babies/", full.names = TRUE, pattern = "*.TXT") %>% 
  lapply(fread) %>% 
  rbindlist() %>% 
  setnames(c("state", "sex", "year", "name", "n")) %>% 
  dcast(state + year + name ~ sex, value.var = "n", fill = 0) %>% 
  .[, odds := (0.5 + abs(F/(M + F) - 0.5))]

babynames <- babynames::babynames %>% 
  setDT() %>% 
  dcast(year + name ~ sex, value.var = "n", fill = 0) %>% 
  .[, odds := (0.5 + abs(F/(M + F) - 0.5))]

babynames %>% 
  .[, .(ratio = weighted.mean(odds, F + M)), by = .(year)] %>% 
  ggplot(aes(year, ratio)) +
  geom_line() +
  hrbrthemes::theme_ipsum_rc() +
  labs(x = "Year of birth", y = NULL, 
       title = "Probability of correctly deducing the sex of a random person\nby knowing their name",
       subtitle = "More prople are named with more gender-neutral (-ish) names.",
       caption = "source: babynames package -> US SSA")
  


babynames %>% 
  .[, .(ratio = mean(odds)), by = .(year)] %>% 
  ggplot(aes(year, ratio)) +
  geom_line() +
  hrbrthemes::theme_ipsum_rc() +
  labs(x = "Year of birth", y = NULL, 
       title = "Probability of correctly deducing the sex of a random person\nby knowing their name",
       subtitle = "More prople are birthed with more gender-neutral (-ish) names.")


library(geofacet)


bb_state %>% 
  .[year >= 1950] %>% 
  .[, .(ratio = weighted.mean(odds, F + M)), by = .(year, state)] %>% 
  ggplot(aes(year, ratio)) +
  geom_line() +
  scale_x_continuous(breaks = c(1950, 2010)) +
  scale_y_continuous(breaks = c(0.97, 1)) +
  coord_cartesian(ylim = c(0.97, NA)) +
  facet_geo(~ state, label = "code") +
  theme_minimal() +
  labs(x = "Year of birth", y = NULL, 
       caption = "source: US SSA")