matt-saenz/cpsR
Load CPS microdata into R using the Census API
cpsR
Overview
Load Current Population Survey
(CPS) microdata
into R using the Census Bureau Data API, including basic monthly
CPS
and CPS
ASEC
microdata.
Note: This product uses the Census Bureau Data API but is not endorsed
or certified by the Census Bureau.
For a Python version of this package, check out
PyCPS.
Installation
To install cpsR, run the following code:
install.packages("cpsR")To install the development version of cpsR, run the following code:
# install.packages("devtools")
devtools::install_github("matt-saenz/cpsR")Census API key
In order to use cpsR functions, you must supply a Census API
key in one of two ways:
- Using the
keyargument (manually) - Using environment variable
CENSUS_API_KEY(automatically)
Using environment variable (or env var, for short) CENSUS_API_KEY is
strongly recommended for two reasons:
- Saves you from having to copy-paste your key around
- Allows you to avoid including your key in scripts
It is important to avoid including your key in scripts if you plan to
share your code with others (like in the example below)
since you should keep your key secret.
You can set up env var CENSUS_API_KEY in two steps:
First, open your .Renviron file. You can do so by running:
# install.packages("usethis")
usethis::edit_r_environ()Second, add your Census API key to your .Renviron file like so:
CENSUS_API_KEY='your_key_here'
This enables cpsR functions to automatically look up your key by
running:
Sys.getenv("CENSUS_API_KEY")Example
library(cpsR)
library(dplyr)
library(purrr)
# Simple use of the basic monthly CPS
sep21 <- get_basic(
year = 2021,
month = 9,
vars = c("prpertyp", "prtage", "pemlr", "pwcmpwgt")
)
sep21
#> # A tibble: 103,858 × 4
#> prpertyp prtage pemlr pwcmpwgt
#> <int> <int> <int> <dbl>
#> 1 2 80 5 1361.
#> 2 2 85 5 1411.
#> 3 2 80 5 4619.
#> 4 2 80 5 4587.
#> 5 2 42 1 3677.
#> 6 2 42 1 3645.
#> 7 1 9 -1 0
#> 8 2 41 1 3652.
#> 9 2 32 7 4117.
#> 10 2 67 1 2479.
#> # ℹ 103,848 more rows
sep21 %>%
filter(prpertyp == 2 & prtage >= 16) %>%
summarize(
pop16plus = sum(pwcmpwgt),
employed = sum(pwcmpwgt[pemlr %in% 1:2])
) %>%
mutate(epop_ratio = employed / pop16plus)
#> # A tibble: 1 × 3
#> pop16plus employed epop_ratio
#> <dbl> <dbl> <dbl>
#> 1 261765646. 154025931. 0.588
# Pulling multiple years of CPS ASEC microdata
asec <- map_dfr(2020:2021, get_asec, vars = c("h_year", "marsupwt"))
count(asec, h_year, wt = marsupwt)
#> # A tibble: 2 × 2
#> h_year n
#> <int> <dbl>
#> 1 2020 325268182.
#> 2 2021 326195440.