Current Population Survey (CPS)

The principal labor force survey, providing income, poverty, and health insurance coverage estimates.

One table with one row per sampled household, a second table with one row per family within each sampled household, and a third table with one row per individual within each of those families.
A complex sample designed to generalize to the civilian non-institutional population of the US.
Released annually since 1998, linkable to the Basic Monthly releases.
Administered jointly by the US Census Bureau and the Bureau of Labor Statistics.

Please skim before you begin:

Current Population Survey 2023 Annual Social and Economic (ASEC) Supplement
Wikipedia Entry
This human-composed haiku or a bouquet of artificial intelligence-generated limericks

# jobs robbed by robot
# luddite rebellion looming
# blue, due to red pill

Download, Import, Preparation

Download and unzip the 2023 file:

library(httr)

tf <- tempfile()

this_url <- "https://www2.census.gov/programs-surveys/cps/datasets/2023/march/asecpub23sas.zip"

GET( this_url , write_disk( tf ) , progress() )

unzipped_files <- unzip( tf , exdir = tempdir() )

Import all four files:

library(haven)

four_tbl <- lapply( unzipped_files , read_sas )

four_df <- lapply( four_tbl , data.frame )

four_df <- lapply( four_df , function( w ){ names( w ) <- tolower( names( w ) ) ; w } )

household_df <- four_df[[ grep( 'hhpub' , basename( unzipped_files ) ) ]]
family_df <- four_df[[ grep( 'ffpub' , basename( unzipped_files ) ) ]]
person_df <- four_df[[ grep( 'pppub' , basename( unzipped_files ) ) ]]
repwgts_df <- four_df[[ grep( 'repwgt' , basename( unzipped_files ) ) ]]

Divide weights:

household_df[ , 'hsup_wgt' ] <- household_df[ , 'hsup_wgt' ] / 100
family_df[ , 'fsup_wgt' ] <- family_df[ , 'fsup_wgt' ] / 100
for ( j in c( 'marsupwt' , 'a_ernlwt' , 'a_fnlwgt' ) ) person_df[ , j ] <- person_df[ , j ] / 100

Merge these four files:

names( family_df )[ names( family_df ) == 'fh_seq' ] <- 'h_seq'
names( person_df )[ names( person_df ) == 'ph_seq' ] <- 'h_seq'
names( person_df )[ names( person_df ) == 'phf_seq' ] <- 'ffpos'

hh_fm_df <- merge( household_df , family_df )
hh_fm_pr_df <- merge( hh_fm_df , person_df )
cps_df <- merge( hh_fm_pr_df , repwgts_df )

stopifnot( nrow( cps_df ) == nrow( person_df ) )

Save locally

Save the object at any point:

# cps_fn <- file.path( path.expand( "~" ) , "CPS" , "this_file.rds" )
# saveRDS( cps_df , file = cps_fn , compress = FALSE )

Load the same object:

# cps_df <- readRDS( cps_fn )

Survey Design Definition

Construct a complex sample survey design:

library(survey)
    
cps_design <-
    svrepdesign(
        weights = ~ marsupwt ,
        repweights = "pwwgt[1-9]" ,
        type = "Fay" ,
        rho = ( 1 - 1 / sqrt( 4 ) ) ,
        data = cps_df ,
        combined.weights = TRUE ,
        mse = TRUE
    )

Variable Recoding

Add new columns to the data set:

cps_design <- 
    update( 
        cps_design , 

        one = 1 ,

        a_maritl = 
            factor( 
                a_maritl , 
                labels = 
                    c( 
                        "married - civilian spouse present" ,
                        "married - AF spouse present" ,
                        "married - spouse absent" ,
                        "widowed" ,
                        "divorced" , 
                        "separated" , 
                        "never married"
                    )
            ) ,
            
        state_name =
            factor(
                gestfips ,
                levels = 
                    c(1L, 2L, 4L, 5L, 6L, 8L, 9L, 10L, 
                    11L, 12L, 13L, 15L, 16L, 17L, 18L, 
                    19L, 20L, 21L, 22L, 23L, 24L, 25L, 
                    26L, 27L, 28L, 29L, 30L, 31L, 32L, 
                    33L, 34L, 35L, 36L, 37L, 38L, 39L, 
                    40L, 41L, 42L, 44L, 45L, 46L, 47L, 
                    48L, 49L, 50L, 51L, 53L, 54L, 55L, 
                    56L) ,
                labels =
                    c("Alabama", "Alaska", "Arizona", "Arkansas", "California", 
                    "Colorado", "Connecticut", "Delaware", "District of Columbia", 
                    "Florida", "Georgia", "Hawaii", "Idaho", "Illinois", "Indiana", 
                    "Iowa", "Kansas", "Kentucky", "Louisiana", "Maine", "Maryland", 
                    "Massachusetts", "Michigan", "Minnesota", "Mississippi", "Missouri", 
                    "Montana", "Nebraska", "Nevada", "New Hampshire", "New Jersey", 
                    "New Mexico", "New York", "North Carolina", "North Dakota", "Ohio", 
                    "Oklahoma", "Oregon", "Pennsylvania", "Rhode Island", "South Carolina", 
                    "South Dakota", "Tennessee", "Texas", "Utah", "Vermont", "Virginia", 
                    "Washington", "West Virginia", "Wisconsin", "Wyoming")
            ) ,

        male = as.numeric( a_sex == 1 )
    )

Analysis Examples with the `survey` library

Unweighted Counts

Count the unweighted number of records in the survey sample, overall and by groups:

sum( weights( cps_design , "sampling" ) != 0 )

svyby( ~ one , ~ state_name , cps_design , unwtd.count )

Weighted Counts

Count the weighted size of the generalizable population, overall and by groups:

svytotal( ~ one , cps_design )

svyby( ~ one , ~ state_name , cps_design , svytotal )

Descriptive Statistics

Calculate the mean (average) of a linear variable, overall and by groups:

svymean( ~ ptotval , cps_design )

svyby( ~ ptotval , ~ state_name , cps_design , svymean )

Calculate the distribution of a categorical variable, overall and by groups:

svymean( ~ a_maritl , cps_design )

svyby( ~ a_maritl , ~ state_name , cps_design , svymean )

Calculate the sum of a linear variable, overall and by groups:

svytotal( ~ ptotval , cps_design )

svyby( ~ ptotval , ~ state_name , cps_design , svytotal )

Calculate the weighted sum of a categorical variable, overall and by groups:

svytotal( ~ a_maritl , cps_design )

svyby( ~ a_maritl , ~ state_name , cps_design , svytotal )

Calculate the median (50th percentile) of a linear variable, overall and by groups:

svyquantile( ~ ptotval , cps_design , 0.5 )

svyby( 
    ~ ptotval , 
    ~ state_name , 
    cps_design , 
    svyquantile , 
    0.5 ,
    ci = TRUE 
)

Estimate a ratio:

svyratio( 
    numerator = ~ moop , 
    denominator = ~ ptotval , 
    cps_design 
)

Subsetting

Restrict the survey design to persons aged 18-64:

sub_cps_design <- subset( cps_design , a_age %in% 18:64 )

Calculate the mean (average) of this subset:

svymean( ~ ptotval , sub_cps_design )

Measures of Uncertainty

Extract the coefficient, standard error, confidence interval, and coefficient of variation from any descriptive statistics function result, overall and by groups:

this_result <- svymean( ~ ptotval , cps_design )

coef( this_result )
SE( this_result )
confint( this_result )
cv( this_result )

grouped_result <-
    svyby( 
        ~ ptotval , 
        ~ state_name , 
        cps_design , 
        svymean 
    )
    
coef( grouped_result )
SE( grouped_result )
confint( grouped_result )
cv( grouped_result )

Calculate the degrees of freedom of any survey design object:

degf( cps_design )

Calculate the complex sample survey-adjusted variance of any statistic:

svyvar( ~ ptotval , cps_design )

Include the complex sample design effect in the result for a specific statistic:

# SRS without replacement
svymean( ~ ptotval , cps_design , deff = TRUE )

# SRS with replacement
svymean( ~ ptotval , cps_design , deff = "replace" )

Compute confidence intervals for proportions using methods that may be more accurate near 0 and 1. See ?svyciprop for alternatives:

svyciprop( ~ male , cps_design ,
    method = "likelihood" )

Regression Models and Tests of Association

Perform a design-based t-test:

svyttest( ptotval ~ male , cps_design )

Perform a chi-squared test of association for survey data:

svychisq( 
    ~ male + a_maritl , 
    cps_design 
)

Perform a survey-weighted generalized linear model:

glm_result <- 
    svyglm( 
        ptotval ~ male + a_maritl , 
        cps_design 
    )

summary( glm_result )

Replication Example

This example matches the count and share of individuals with health insurance in Table H-01:

count_covered <- svytotal( ~ as.numeric( cov == 1 ) , cps_design )

stopifnot( round( coef( count_covered ) , -5 ) == 304000000 )

stopifnot(
    round( coef( count_covered ) - confint( count_covered , level = 0.9 )[1] , -3 ) == 746000
)

share_covered <- svymean( ~ as.numeric( cov == 1 ) , subset( cps_design , cov > 0 ) )

stopifnot( round( coef( share_covered ) , 3 ) == 0.921 )

stopifnot(
    round( coef( share_covered ) - confint( share_covered , level = 0.9 )[1] , 3 ) == 0.002
)

Poverty and Inequality Estimation with `convey`

The R convey library estimates measures of income concentration, poverty, inequality, and wellbeing. This textbook details the available features. As a starting point for CPS users, this code calculates the gini coefficient on complex sample survey data:

library(convey)
cps_design <- convey_prep( cps_design )

cps_household_design <- subset( cps_design , a_exprrp %in% 1:2 )

svygini( ~ htotval , cps_household_design )

Analysis Examples with `srvyr`

The R srvyr library calculates summary statistics from survey data, such as the mean, total or quantile using dplyr-like syntax. srvyr allows for the use of many verbs, such as summarize, group_by, and mutate, the convenience of pipe-able functions, the tidyverse style of non-standard evaluation and more consistent return types than the survey package. This vignette details the available features. As a starting point for CPS users, this code replicates previously-presented examples:

library(srvyr)
cps_srvyr_design <- as_survey( cps_design )