American Time Use Survey (ATUS)
Sampled individuals write down everything they do for a single twenty-four hour period, in ten minute intervals. Time use data allows for the study of uncompensated work like cooking, chores, childcare.
Many tables with structures described in the user guide, linkable to the Current Population Survey.
A complex survey generalizing to person-hours among civilian non-institutional americans aged 15+.
Released annually since 2003.
Administered by the Bureau of Labor Statistics.
Please skim before you begin:
This human-composed haiku or a bouquet of artificial intelligence-generated limericks
# don't judge me bruno
# eat one hour, sleep the rest
# it's my lazy dayDownload, Import, Preparation
Define a function to download, unzip, and import each comma-separated value dat file:
library(httr)
atus_csv_import <-
function( this_url ){
this_tf <- tempfile()
this_dl <- GET( this_url , user_agent( "email@address.com") )
writeBin( content( this_dl ) , this_tf )
unzipped_files <- unzip( this_tf , exdir = tempdir() )
this_dat <- grep( '\\.dat$' , unzipped_files , value = TRUE )
this_df <- read.csv( this_dat )
file.remove( c( this_tf , unzipped_files ) )
names( this_df ) <- tolower( names( this_df ) )
this_df
}Download and import the activity, respondent, roster, and weights tables:
act_df <- atus_csv_import( "https://www.bls.gov/tus/datafiles/atusact-2021.zip" )
resp_df <- atus_csv_import( "https://www.bls.gov/tus/datafiles/atusresp-2021.zip" )
rost_df <- atus_csv_import( "https://www.bls.gov/tus/datafiles/atusrost-2021.zip" )
wgts_df <- atus_csv_import( "https://www.bls.gov/tus/datafiles/atuswgts-2021.zip" )Specify which variables to keep in each of the data.frame objects:
act_df <- act_df[ c( 'tucaseid' , 'tutier1code' , 'tutier2code' , 'tuactdur24' ) ]
resp_df <- resp_df[ c( 'tucaseid' , 'tufinlwgt' , 'tulineno' ) ]
rost_df <- rost_df[ , c( 'tucaseid' , 'tulineno' , 'teage' , 'tesex' ) ]Distribute travel-related activities (tutier1code == 18 from the lexicon) based on their second tier code:
act_df[ act_df[ , 'tutier1code' ] == 18 & act_df[ , 'tutier2code' ] == 99 , 'tutier1code' ] <- 50
act_df[ act_df[ , 'tutier1code' ] == 18 , 'tutier1code' ] <-
act_df[ act_df[ , 'tutier1code' ] == 18 , 'tutier2code' ]Sum up all durations at the (respondent x major activity category)-level:
act_long_df <- aggregate( tuactdur24 ~ tucaseid + tutier1code , data = act_df , sum )
act_wide_df <-
reshape( act_long_df , idvar = 'tucaseid' , timevar = 'tutier1code' , direction = 'wide' )
# for individuals not engaging in an activity category, replace missings with zero minutes
act_wide_df[ is.na( act_wide_df ) ] <- 0
# for all columns except the respondent identifier, convert minutes to hours
act_wide_df[ , -1 ] <- act_wide_df[ , -1 ] / 60Merge the respondent and summed activity tables, then the roster table, and finally the replicate weights:
resp_act_df <- merge( resp_df , act_wide_df )
stopifnot( nrow( resp_act_df ) == nrow( resp_df ) )
resp_act_rost_df <- merge( resp_act_df , rost_df )
stopifnot( nrow( resp_act_rost_df ) == nrow( resp_df ) )
atus_df <- merge( resp_act_rost_df , wgts_df )
stopifnot( nrow( atus_df ) == nrow( resp_df ) )
# remove dots from column names
names( atus_df ) <- gsub( "\\." , "_" , names( atus_df ) )
atus_df[ , 'one' ] <- 1Save locally
Save the object at any point:
# atus_fn <- file.path( path.expand( "~" ) , "ATUS" , "this_file.rds" )
# saveRDS( atus_df , file = atus_fn , compress = FALSE )Load the same object:
# atus_df <- readRDS( atus_fn )Survey Design Definition
Construct a complex sample survey design:
library(survey)
atus_design <-
svrepdesign(
weights = ~ tufinlwgt ,
repweights = "finlwgt[0-9]" ,
type = "Fay" ,
rho = ( 1 - 1 / sqrt( 4 ) ) ,
mse = TRUE ,
data = atus_df
)Variable Recoding
Add new columns to the data set:
# caring for and helping household members is top level 03 from the lexicon
# https://www.bls.gov/tus/lexicons/lexiconnoex2021.pdf
atus_design <-
update(
atus_design ,
any_care = as.numeric( tuactdur24_3 > 0 ) ,
tesex = factor( tesex , levels = 1:2 , labels = c( 'male' , 'female' ) ) ,
age_category =
factor(
1 + findInterval( teage , c( 18 , 35 , 65 ) ) ,
labels = c( "under 18" , "18 - 34" , "35 - 64" , "65 or older" )
)
)Analysis Examples with the survey library
Unweighted Counts
Count the unweighted number of records in the survey sample, overall and by groups:
sum( weights( atus_design , "sampling" ) != 0 )
svyby( ~ one , ~ age_category , atus_design , unwtd.count )Weighted Counts
Count the weighted size of the generalizable population, overall and by groups:
svytotal( ~ one , atus_design )
svyby( ~ one , ~ age_category , atus_design , svytotal )Descriptive Statistics
Calculate the mean (average) of a linear variable, overall and by groups:
svymean( ~ tuactdur24_1 , atus_design )
svyby( ~ tuactdur24_1 , ~ age_category , atus_design , svymean )Calculate the distribution of a categorical variable, overall and by groups:
svymean( ~ tesex , atus_design )
svyby( ~ tesex , ~ age_category , atus_design , svymean )Calculate the sum of a linear variable, overall and by groups:
svytotal( ~ tuactdur24_1 , atus_design )
svyby( ~ tuactdur24_1 , ~ age_category , atus_design , svytotal )Calculate the weighted sum of a categorical variable, overall and by groups:
svytotal( ~ tesex , atus_design )
svyby( ~ tesex , ~ age_category , atus_design , svytotal )Calculate the median (50th percentile) of a linear variable, overall and by groups:
svyquantile( ~ tuactdur24_1 , atus_design , 0.5 )
svyby(
~ tuactdur24_1 ,
~ age_category ,
atus_design ,
svyquantile ,
0.5 ,
ci = TRUE
)Estimate a ratio:
svyratio(
numerator = ~ tuactdur24_5 ,
denominator = ~ tuactdur24_12 ,
atus_design
)Subsetting
Restrict the survey design to any time volunteering:
sub_atus_design <- subset( atus_design , tuactdur24_15 > 0 )Calculate the mean (average) of this subset:
svymean( ~ tuactdur24_1 , sub_atus_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( ~ tuactdur24_1 , atus_design )
coef( this_result )
SE( this_result )
confint( this_result )
cv( this_result )
grouped_result <-
svyby(
~ tuactdur24_1 ,
~ age_category ,
atus_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( atus_design )Calculate the complex sample survey-adjusted variance of any statistic:
svyvar( ~ tuactdur24_1 , atus_design )Include the complex sample design effect in the result for a specific statistic:
# SRS without replacement
svymean( ~ tuactdur24_1 , atus_design , deff = TRUE )
# SRS with replacement
svymean( ~ tuactdur24_1 , atus_design , deff = "replace" )Compute confidence intervals for proportions using methods that may be more accurate near 0 and 1. See ?svyciprop for alternatives:
svyciprop( ~ any_care , atus_design ,
method = "likelihood" )Regression Models and Tests of Association
Perform a design-based t-test:
svyttest( tuactdur24_1 ~ any_care , atus_design )Perform a chi-squared test of association for survey data:
svychisq(
~ any_care + tesex ,
atus_design
)Perform a survey-weighted generalized linear model:
glm_result <-
svyglm(
tuactdur24_1 ~ any_care + tesex ,
atus_design
)
summary( glm_result )Replication Example
This example matches the “Caring for and helping household members” row of Table A-1:
hours_per_day_civilian_population <- svymean( ~ tuactdur24_3 , atus_design )
stopifnot( round( coef( hours_per_day_civilian_population ) , 2 ) == 0.47 )
percent_engaged_per_day <- svymean( ~ any_care , atus_design )
stopifnot( round( coef( percent_engaged_per_day ) , 3 ) == 0.217 )
hours_per_day_among_engaged <- svymean( ~ tuactdur24_3 , subset( atus_design , any_care ) )
stopifnot( round( coef( hours_per_day_among_engaged ) , 2 ) == 2.17 )This example matches the average hours and SE from Section 7.5 of the User’s Guide:
Download and import the activity, activity summary, respondent, and weights tables:
actsum07_df <- atus_csv_import( "https://www.bls.gov/tus/datafiles/atussum_2007.zip" )
resp07_df <- atus_csv_import( "https://www.bls.gov/tus/datafiles/atusresp_2007.zip" )
act07_df <- atus_csv_import( "https://www.bls.gov/tus/datafiles/atusact_2007.zip" )
wgts07_df <- atus_csv_import( "https://www.bls.gov/tus/datafiles/atuswgts_2007.zip" )Option 1. Sum the two television fields from the activity summary file, removing zeroes:
television_per_person <-
data.frame(
tucaseid = actsum07_df[ , 'tucaseid' ] ,
tuactdur24 = rowSums( actsum07_df[ , c( 't120303' , 't120304' ) ] )
)
television_per_person <-
television_per_person[ television_per_person[ , 'tuactdur24' ] > 0 , ]Option 2. Limit the activity file to television watching records according to the 2007 Lexicon:
television_activity <-
subset(
act07_df ,
tutier1code == 12 &
tutier2code == 3 &
tutier3code %in% 3:4
)
television_activity_summed <-
aggregate(
tuactdur24 ~ tucaseid ,
data = television_activity ,
sum
)Confirm both aggregation options yield the same results:
stopifnot(
all( television_per_person[ , 'tucaseid' ] == television_activity_summed[ , 'tucaseid' ] )
)
stopifnot(
all( television_per_person[ , 'tuactdur24' ] == television_activity_summed[ , 'tuactdur24' ] )
)Merge the respondent and summed activity tables, then the replicate weights:
resp07_tpp_df <-
merge(
resp07_df[ , c( 'tucaseid' , 'tufinlwgt' ) ] ,
television_per_person ,
all.x = TRUE
)
stopifnot( nrow( resp07_tpp_df ) == nrow( resp07_df ) )
# for individuals without television time, replace missings with zero minutes
resp07_tpp_df[ is.na( resp07_tpp_df[ , 'tuactdur24' ] ) , 'tuactdur24' ] <- 0
# convert minutes to hours
resp07_tpp_df[ , 'tuactdur24_hour' ] <- resp07_tpp_df[ , 'tuactdur24' ] / 60
atus07_df <- merge( resp07_tpp_df , wgts07_df )
stopifnot( nrow( atus07_df ) == nrow( resp07_df ) )Construct a complex sample survey design:
atus07_design <-
svrepdesign(
weights = ~ tufinlwgt ,
repweights = "finlwgt[0-9]" ,
type = "Fay" ,
rho = ( 1 - 1 / sqrt( 4 ) ) ,
data = atus07_df
)Match the statistic and SE of the number of hours daily that americans older than 14 watch tv:
result <- svymean( ~ tuactdur24_hour , atus07_design )
stopifnot( round( coef( result ) , 2 ) == 2.62 )
stopifnot( round( SE( result ) , 4 ) == 0.0293 )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 ATUS users, this code replicates previously-presented examples:
library(srvyr)
atus_srvyr_design <- as_survey( atus_design )Calculate the mean (average) of a linear variable, overall and by groups:
atus_srvyr_design %>%
summarize( mean = survey_mean( tuactdur24_1 ) )
atus_srvyr_design %>%
group_by( age_category ) %>%
summarize( mean = survey_mean( tuactdur24_1 ) )