CurrentPopulationSurvey

About

CurrentPopulationSurvey.jl allows users to easily download & parse U.S. Census Bureau CPS microdata files for the 2007 - present time period (earlier years are coming in future releases). This package supports the Tables.jl interface so you can easily convert to a tabular structure of your preference (e.g. DataFrame).

Resources

• About the CPS: https://www.census.gov/programs-surveys/cps/about.html
• Files and data dictionaries: https://www.census.gov/data/datasets/time-series/demo/cps/cps-basic.html

Recommendations

I recommend that you familiarize yourself with the variables in the data dictionaries before calling cpsdata so that you can decide on a subset of the total available variables for parsing. One year's worth of data is roughly 5GB - 7GB so narrowing this down (by selecting only the variables that you need) will improve efficiency when working with the data.

Useage

This package exports a single function cpsdata:

cpsdata(year::Int, month::Int[, vars::Vector{String}])

Download/parse CPS microdata files for a given year & month, optionally retaining only the variables specified. There are hundreds of variables so specifying only those that you need will significantly increase efficiency when working with the data.

Arguments

• year::Int: the year for which you want to obtain CPS data.
• month::Int: the month for which you want to obtain CPS data.
• vars::Vector{String}: an optional argument specifying the variables in the microdata file that you would like to keep.

Examples

data1901 = cpsdata(2019, 1, ["HRINTSTA", "PWORWGT"])

If you want to work with the data as a DataFrame:

using DataFrames

data1901 = DataFrame(cpsdata(2019, 1, ["HRINTSTA", "PWORWGT"]))