Pesquisa Mensal de Emprego (PME)

Build Status Build status

Contributed by Dr. Djalma Pessoa <pessoad@gmail.com>

The Pesquisa Mensal de Emprego (PME) is the monthly labor force survey covering the six largest Brazilian cities.

  • One table with one row per individual within each sampled household.

  • A complex sample survey designed to generalize to the civilian population of Brazil’s six largest cities.

  • Released monthly since March 2002.

  • Administered by the Instituto Brasileiro de Geografia e Estatistica.

Simplified Download and Importation

The R lodown package easily downloads and imports all available PME microdata by simply specifying "pme" with an output_dir = parameter in the lodown() function. Depending on your internet connection and computer processing speed, you might prefer to run this step overnight.

library(lodown)
lodown( "pme" , output_dir = file.path( path.expand( "~" ) , "PME" ) )

lodown also provides a catalog of available microdata extracts with the get_catalog() function. After requesting the PME catalog, you could pass a subsetted catalog through the lodown() function in order to download and import specific extracts (rather than all available extracts).

library(lodown)
# examine all available PME microdata files
pme_cat <-
    get_catalog( "pme" ,
        output_dir = file.path( path.expand( "~" ) , "PME" ) )

# 2016 only
pme_cat <- subset( pme_cat , year == 2016 )
# download the microdata to your local computer
lodown( "pme" , pme_cat )

Analysis Examples with the survey library

Construct a complex sample survey design:

options( survey.lonely.psu = "adjust" )

library(survey)

pme_df <- readRDS( file.path( path.expand( "~" ) , "PME" , "pme 2016 01.rds" ) )

# throw out records missing their cluster variable
pme_df <- subset( pme_df , !is.na( v113 ) )

pop_totals <- unique( pme_df[ , c( 'v035' , 'v114' ) ] )

prestratified_design <- 
    svydesign( 
        ~ v113 , 
        strata = ~ v112 , 
        data = pme_df ,
        weights = ~ v211 , 
        nest = TRUE
    )

pme_design <- 
    postStratify( prestratified_design , ~ v035 , pop_totals )

Variable Recoding

Add new columns to the data set:

pme_design <- 
    update( 
        pme_design , 

        one = 1 ,
        
        # calculate whether each person is at least ten years of age
        pia = as.numeric( v234 >= 10 ) ,

        # determine individuals who are employed
        ocup_c = as.numeric( v401 == 1 | v402 == 1 | v403 == 1 ) ,
        
        sexo = factor( v203 , labels = c( "male" , "female" ) ) ,
        
        region = 
            factor( 
                v035 , 
                levels = c( 26 , 29 , 31 , 33 , 35 , 43 ) , 
                labels = c( "Recife" , "Salvador" , "Belo Horizonte" , 
                    "Rio de Janeiro" , "Sao Paulo" , "Porto Alegre" )
            )
    )
    
pme_design <-
    update(
        pme_design ,
        
        # determine individuals who are unemployed
        desocup30 = as.numeric( ocup_c == 0 & !is.na( v461 ) & v465 == 1 )
    )
        
pme_design <-
    update(
        pme_design ,
        
        # determine individuals who are either working or not working
        pea_c = as.numeric( ocup_c == 1 | desocup30 == 1 )

    )

Unweighted Counts

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

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

svyby( ~ one , ~ region , pme_design , unwtd.count )

Weighted Counts

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

svytotal( ~ one , pme_design )

svyby( ~ one , ~ region , pme_design , svytotal )

Descriptive Statistics

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

svymean( ~ vd25 , pme_design , na.rm = TRUE )

svyby( ~ vd25 , ~ region , pme_design , svymean , na.rm = TRUE )

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

svymean( ~ sexo , pme_design )

svyby( ~ sexo , ~ region , pme_design , svymean )

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

svytotal( ~ vd25 , pme_design , na.rm = TRUE )

svyby( ~ vd25 , ~ region , pme_design , svytotal , na.rm = TRUE )

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

svytotal( ~ sexo , pme_design )

svyby( ~ sexo , ~ region , pme_design , svytotal )

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

svyquantile( ~ vd25 , pme_design , 0.5 , na.rm = TRUE )

svyby( 
    ~ vd25 , 
    ~ region , 
    pme_design , 
    svyquantile , 
    0.5 ,
    ci = TRUE ,
    keep.var = TRUE ,
    na.rm = TRUE
)

Estimate a ratio:

svyratio( 
    numerator = ~ desocup30 , 
    denominator = ~ pea_c , 
    pme_design ,
    na.rm = TRUE
)

Subsetting

Restrict the survey design to teenagers:

sub_pme_design <- subset( pme_design , v234 %in% 13:19 )

Calculate the mean (average) of this subset:

svymean( ~ vd25 , sub_pme_design , na.rm = TRUE )

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( ~ vd25 , pme_design , na.rm = TRUE )

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

grouped_result <-
    svyby( 
        ~ vd25 , 
        ~ region , 
        pme_design , 
        svymean ,
        na.rm = TRUE 
    )
    
coef( grouped_result )
SE( grouped_result )
confint( grouped_result )
cv( grouped_result )

Calculate the degrees of freedom of any survey design object:

degf( pme_design )

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

svyvar( ~ vd25 , pme_design , na.rm = TRUE )

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

# SRS without replacement
svymean( ~ vd25 , pme_design , na.rm = TRUE , deff = TRUE )

# SRS with replacement
svymean( ~ vd25 , pme_design , na.rm = TRUE , deff = "replace" )

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

svyciprop( ~ ocup_c , pme_design ,
    method = "likelihood" , na.rm = TRUE )

Regression Models and Tests of Association

Perform a design-based t-test:

svyttest( vd25 ~ ocup_c , pme_design )

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

svychisq( 
    ~ ocup_c + sexo , 
    pme_design 
)

Perform a survey-weighted generalized linear model:

glm_result <- 
    svyglm( 
        vd25 ~ ocup_c + sexo , 
        pme_design 
    )

summary( glm_result )

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 PME users, this code replicates previously-presented examples:

library(srvyr)
pme_srvyr_design <- as_survey( pme_design )

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

pme_srvyr_design %>%
    summarize( mean = survey_mean( vd25 , na.rm = TRUE ) )

pme_srvyr_design %>%
    group_by( region ) %>%
    summarize( mean = survey_mean( vd25 , na.rm = TRUE ) )

Replication Example