cpsvote

cpsvote provides an automated method for downloading, recoding, and merging selected years of the Current Population Survey’s Voting and Registration Supplement, a biennial large-(N) survey about voter registration, voting, and non-voting in United States federal elections. The package provides documentation for appropriate use of sample weights to generate statistical estimates. The package includes multiple vignettes that illustrate different applications.

Sources

All years of data come from the National Bureau of Economic Research (NBER), and the CPS is administered by the US Census Bureau and Bureau of Labor Statistics (BLS).

Process

A few main purposess of this package are to:

1. Download the CPS data (cps_download_data) and documentation (cps_download_docs) from NBER

2. Load the raw, numeric CPS data into R (cps_read)

3. Help you apply the correct factor labels to this numeric data (cps_label)

4. Access a “basic” version of the CPS that we have included with the package (cps_load_basic)

For help with these, see the documentation for these functions and the vignettes. Additional vignettes are provided with examples of some basic analyses that are possible with the CPS.

Instructions for Alpha Testing

Basic Use Information

1. Install the Package
   1. If you are a tester, you can install the package using remotes:

   remotes::install_github("Reed-EVIC/cpsvote")
   library(cpsvote)

   2. If you are a primary developer / contributor and are using a localized version control system, create your version control project using this URL: https://github.com/Reed-EVIC/cpsvote.git.
   1. devtools::load_all() will load the package and all the associated functions.
2. Quickly view some CPS data
   1. There is far more data included in the full CPS than we could include here, but there are two options to view a subset of the CPS data with ease.
      1. Running data(cps_sample_10k, package = 'cpsvote') will load a 10,000 row sample of selected CPS columns into your environment.
      2. Running cpsvote::cps_load_basic() will download the data, and provide you all rows and a selected set of CPS columns, and allow you to save a copy locally for speed in the future.
3. Download the CPS Files
   1. cpsvote will download the CPS files directly from a federal data repository (http://data.nber.org/data/current-population-survey-data.html for years 1994 - 2018). By default cpsvote will create a new subdirectory in your working directory named “cps_data” and store the compressed data files in that location.
   2. If you choose to download data files and store them in a separate location, cpsvote can only process files that retain the original file name. See the “Data Description” vignette for a list of the expected file names.
   3. cpsvote::cps_download_data(path = "dir", years = seq(start,end,sequence)) will download data
      1. Example: cpsvote::cps_download_data() will download all data files and place the files in the default download path (“./cps_data”).
      2. Example: cpsvote::cps_download_data(path = "~/mydata/CPS", years = seq(2000, 2014, 4)) will download the 2000, 2004, 2008, and 2012 CPS and place them in the path “~/mydata/CPS”.
   4. cpsvote::cps_download_docs() downloads all documentation files, and places them in “./cps_docs”
   5. cpsvote::cps_download_docs(path = "dir", years = seq(start, end, seqquence) allows you to control the target directory and years for the documentation.
4. Import the Data Into R and Create a Data Frame
   1. cpsvote performs a series of data transformations that produce a set of comparably coded variables across all of the survey years. These variables are listed in the “Data Description” vignette.
   2. cpsvote::read_cps(dir = "dir", years = seq(start,end,sequence)) will read CPS data files located in “dir” into memory. A reminder that the file names must contain the 4-digit year associated with their data in order to be read easily.
      1. Example: cps_vote <- cpsvote::cps_read() will read all data files (1994 - 2018) from the default directory (“./cps_data”), and merge into a single data frame called “cps_vote”.
      2. Example: cps_vote <- cpsvote::cps_read(data_dir = "~/mydata/CPS", years = seq(1994, 2000, 2)) will load CPS data from 1994, 1996, 1998, and 2000, assuming the data files are stored in “~/mydata/CPS”, and merge into the object “cps_vote”.
5. Attach proper labels to the numeric data
   1. By default, the CPS comes with numeric data values. These correspond to factor codes as detailed in the survey documentation. The cps_label function applies given factor codes to this numeric data in order to make it human-readable.
6. Attach proper turnout modifiers
   1. Certain steps have to be taken to properly reweight the CPS for voter turnout analyses. The functions cps_recode_vote and cps_reweight_turnout help to redo these codings to appropriately calculate voter turnout.

Recommended First Steps

The steps shown below install the package locally, download all CPS files from 1994 - 2018, load the files into memory, and results in a data frame with CPS Voting and Registration Supplement data from 1994-2018.

1. remotes::install_github("Reed-EVIC/cpsvote")
2. library(cpsvote)
3. cpsvote::cps_download_data() (Only necessary if using the package for the first time; after that, the files will already be downloaded and you can skip this step)
4. cps_vote <- cpsvote::cps_read() Will read all available years into memory, merging them in a data frame named cps_vote

Advanced Use: Manually Adding Variables to the Output Dataset

Adding variables at this time is tricky, and we can’t guarantee yet that it will work smoothly. However, if you wish to modify cpsvote to add variables from individual years of the CPS, you can make edited versions of the included objects cps_cols and cps_factors. To identify the names, column locations, and factor levels for new variables, please refer to the documentation for individual years of the CPS (cpsvote::cps_download_docs()).

For an example, let’s say that I wanted to add annual family income in 2006 and 2008 to the default data. First, I would define the column positions and factor levels for income and add them to the default cps_cols and cps_factors.

library(cpsvote)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

income_cols <- data.frame(
  year = c(2006, 2008),
  cps_name = "HUFAMINC",
  new_name = "INCOME",
  start_pos = 39,
  end_pos = 40,
  stringsAsFactors = FALSE
)

income_factors <- data.frame(
  year = c(rep(2006, 16), rep(2008, 16)),
  cps_name = "HUFAMINC",
  new_name = "INCOME",
  code = c(1:16, 1:16),
  value = rep(c("LESS THAN $5,000",
                "5,000 TO 7,499",
                "7,500 TO 9,999",
                "10,000 TO 12,499",
                "12,500 TO 14,999",
                "15,000 TO 19,999",
                "20,000 TO 24,999",
                "25,000 TO 29,999",
                "30,000 TO 34,999",
                "35,000 TO 39,999",
                "40,000 TO 49,999",
                "50,000 TO 59,999",
                "60,000 TO 74,999",
                "75,000 TO 99,999",
                "100,000 TO 149,999",
                "150,000 OR MORE"), 2),
  stringsAsFactors = FALSE
)

my_cols <- bind_rows(cps_cols, income_cols)
my_factors <- bind_rows(cps_factors, income_factors)

Then, I read in the 2006 and 2008 CPS data using my new column data, and factor it with my new factor data.

cps_income <- cps_read(years = c(2006, 2008),
                       cols = my_cols) %>%
  cps_label(factors = my_factors)
#> No new data files downloaded
#> Reading 2 data file(s)...
#> 2006 file read
#> 2008 file read
#> Warning in cps_read(years = c(2006, 2008), cols = my_cols): The column
#> names provided by the CPS do not refer to the same question across all
#> years. Be cautious that you are joining columns which correspond across
#> years.

Now we can look at the unweighted breakdown of income by year.

janitor::tabyl(cps_income, INCOME, YEAR)
#>              INCOME  2006  2008
#>    LESS THAN $5,000  2800  2647
#>      5,000 TO 7,499  2224  1963
#>      7,500 TO 9,999  2188  2164
#>    10,000 TO 12,499  3447  2992
#>    12,500 TO 14,999  3057  2815
#>    15,000 TO 19,999  5046  4596
#>    20,000 TO 24,999  6352  5945
#>    25,000 TO 29,999  6833  6192
#>    30,000 TO 34,999  7213  6901
#>    35,000 TO 39,999  6662  6068
#>    40,000 TO 49,999 10231  9951
#>    50,000 TO 59,999 10604  9855
#>    60,000 TO 74,999 12607 12488
#>    75,000 TO 99,999 14291 13838
#>  100,000 TO 149,999 11881 12886
#>     150,000 OR MORE  8257  8936
#>                <NA> 39562 40562

Examples of CPS data analysis