index.Rmd

---
title: "Using the EPI CPS microdata extracts in R"
author: "Ben Zipperer"
date: "EARN Webinar | 29 August 2024" 
output:
  xaringan::moon_reader:
    css: [default, metropolis, metropolis-fonts, my_metropolis.css] 
    lib_dir: libs
    nature:
      highlightStyle: github
      highlightLines: true
      countIncrementalSlides: false
      beforeInit: "macros.js"
---

```{r global_options, include=FALSE}
# This is a way to set options for all code chunks at once
# Note that you can also dynamically control options by setting them to a value
# DPI setting increased for Word output, cairo is anti aliasing
knitr::opts_chunk$set(
  include = FALSE,
  tidy.opts = list(width.cutoff = 60), 
  tidy = TRUE
)
```



# Outline
## 1. Introduction to Current Population Survey
## 2. Set up R / Rstudio to use EPI CPS
## 3. Example usage with sample data
## 4. Download and use the actual data with R

---

# Where we're headed
## Write a R script that uses the EPI CPS extracts
  * to calculate the number and share of workers paid less than $20/per hour 
  * in your state in 2023
  * by gender

---
# Resources
### Slides for this talk

https://economic.github.io/zipperer_2024_webinar_epiextracts

### EPI CPS extracts documentation

https://microdata.epi.org/

### epiextractr package

https://economic.github.io/epiextractr/


---
# Current Population Survey data

### CPS background

* One of the most important surveys about the US labor market
* Monthly interviews of about 60,000 households conducted by the Census Bureau
* Source of monthly unemployment rate data
* Also a very good source of hourly wage data

### How to access the data?

* [Bureau of Labor Statistics](https://www.bls.gov/cps/) has easy to use aggregations
* But if you need a custom analysis, especially by state, you need the microdata
* Microdata are the individual survey responses
* You can download the microdata from the [Census](https://www.census.gov/data/datasets/time-series/demo/cps/cps-basic.html) (very cumbersome)
* Or you can use the [EPI CPS extracts](https://microdata.epi.org/) (much easier)
* [IPUMS](https://cps.ipums.org/cps/) is another great source for CPS microdata

### EPI CPS extracts documentation

https://microdata.epi.org/

---
# CPS data structure

### CPS Basic 
* Every month households are asked questions about demographics and employment and other "basic" items
  - This data the **CPS Basic**. Available monthly back to 1976

### CPS ORG
* About 1/4 of those people are asked questions about wages/earnings and union status
  - This is the **CPS ORG**. Available monthly back to 1979ish.
  - We're going to focus on this today.

### CPS supplements
* There are other CPS "supplements" that ask additional questions
  - Annual poverty rates come from the **CPS March supplement**
  - Voting and registration supplement, biennial in November

---
# How do you use the EPI CPS extracts in R?

## Open R/Rstudio and install required packages

**[epiextractr](https://economic.github.io/epiextractr/)**: makes it easy to use and download EPI CPS extracts
```r
install.packages(
   "epiextractr", 
   repos = c("https://economic.r-universe.dev", "https://cloud.r-project.org")
)
```

**[tidyverse](https://www.tidyverse.org/)**: general data processing tools

```r
install.packages("tidyverse")
```

**[usethis](https://usethis.r-lib.org/)**: helpful for some configuration we'll do later
```r
install.packages("usethis")
```
---
# Want to do other analysis with the CPS?
## Wage or union analysis: CPS ORG with `orgwgt`
```r
load_org(2023, orgwgt, ...)
```

## Employment analysis: CPS Basic with `basicwgt`
```r
load_basic(2023, basicwgt, ...)
```

## Data sources and weights
More on data sources and weights: https://microdata.epi.org/methodology/faq/#which-sample-weight-variable-should-i-use

---
# Thank you!
## Slides
https://economic.github.io/zipperer_2024_webinar_epiextracts