sf_comparative_wages.R

library(tidycensus)
library(tigris)
library(tidyverse)
library(sf)
library(srvyr)
options(tigris_use_cache = TRUE)

sf_area <- metro_divisions() %>%
  filter(GEOID %in% c("4186042034", "4186041884"))

sf <- pumas(year = 2019, cb = TRUE, state = "CA") %>%
  st_point_on_surface() %>%
  st_filter(sf_area) %>%
  pull(PUMACE10)

sf_wage_occ <- get_pums(
  variables = c("WAGP", "OCCP", "ESR"),
  state = "CA",
  puma = sf,
  survey = "acs1",
  year = 2021,
  recode = TRUE,
  variables_filter = list(
    COW = 1:8
  ),
  rep_weights = "person"
) %>%
  to_survey()

sf_compare <- sf_wage_occ %>%
  mutate(comparison = case_when(
    OCCP == "6230" ~ "Carpenters",
    OCCP == "1021" ~ "Software developers",
    TRUE ~ "Other"
  ))%>%
  filter(comparison != "Other")

sf_compare %>%
  group_by(comparison, ESR_label) %>%
  summarize(average = survey_mean(WAGP, na.rm = TRUE))

This was great - I just modified this for a research project. For others looking for occupation codes, you can find them here: https://www.eeoc.gov/federal-sector/management-directive/eeoc-federal-sector-occupation-cross-classification-table