ACS Data Users Group

Is there a way to define in R which variables to pull from an ACS data set using API? Below is the code I use to pull the entire table for B01003 for all US zip codes, but this table only has Estimate and MOE, so it's not too large. But for other tables, I need to exclude some variables to make this work.

# Install required packages
if (!require(tidycensus)) install.packages("tidycensus")
if (!require(dplyr)) install.packages("dplyr")

# Load the packages
library(tidycensus)
library(dplyr)

# Set your Census API key
census_api_key("your_api_key", install = TRUE, overwrite = TRUE)

# Define the years you're interested in (2011 to 2022)
years <- 2011:2022

# Function to fetch population data for all U.S. ZCTAs for a specific year
get_population_data_us <- function(year) {
tryCatch({
data <- get_acs(geography = "zip code tabulation area",
year = year,
survey = "acs5",
table = "B01003",
output = "wide") %>%
mutate(Year = year)
return(data)
}, error = function(e) {
message("Error with year ", year, ": ", e$message)
return(NULL)
})
}

# Initialize an empty list to store the data
all_data_us <- list()

# Loop over each year, fetching data for all U.S. ZCTAs
for (year in years) {
data <- get_population_data_us(year)
if (!is.null(data)) {
all_data_us[[as.character(year)]] <- data
}
}

# Combine the data from different years
combined_data_us <- bind_rows(all_data_us)

# Save the combined data to a CSV file on your desktop
write.csv(combined_data_us, "~/Desktop/Population_by_ZIP_All_US_2011_2022.csv", row.names = FALSE)

From the documentation for the `tidycensus` R package (https://walker-data.com/tidycensus/articles/basic-usage.html#working-with-acs-data), it looks like you can pass a `variables` argument to the `get_acs` function:

vt <- get_acs(geography = "county",
variables = c(medincome = "B19013_001"),
state = "VT",
year = 2021)

Rather than using the "tables" argument, you can use the "variables" argument, as an other replier suggested. Here's some similar code I used to pull the total population ACS variable over a series of 5 years for all metro areas.

# Which Years? -----------------------------------------------------
years <- 2017:2022
names(years) <- years

# Get metro area population change --------------------------------
metro_populations <- map_dfr(years,~{
get_acs(
geography = "cbsa",
variables = "B01003_001",
year = .x,
survey = "acs5"
)
}, .id = "year")

If you want more than one variable, you can put them in an object just like the "years" object here. Then, reference that object in the "variables" argument.

you don't even need to do that. all you need is a vector of variable IDs. optionally you can also use a named vector to rename the variable IDs to whatever you want in your output.

Here's a minimal example to try: walker-data.com/.../an-introduction-to-tidycensus.html