This project investigates the labor market in the United States in 2020, where Industry is cagegorized by the North American Industry Classification System (NAICS) and occupation by the Standard Occupational Classification (SOC) System. Our primary data source is from U.S. BUREAU OF LABOR STATISTICS and our parameters of interest are total employment and mean hourly wage.

There are four sections of analysis centered around our parameter of interest:

• First section: we categorize our data by industry according to the NAICS and analyze the distribution of hourly wage within each industry.
• Second section: we categorize our data based on the six-digit 2018 SOC code.
• Third section: we group the data by state and focus on the spatial pattern of total employment and income level across states.
• Fourth section: we combine employment data with estimated automation probability score (Frey & Osborne, 2017) and investigate the relationship between salary, employment, and the predicted chance of automation across occupations.

• Build the image for this project by typing:

docker build . -t pj

• Then start RStudio in your web browser by typing:

docker run -e PASSWORD=<some_password> --rm -v $(pwd):/home/rstudio/ -p 8787:8787 -t pj

• Once the Rstudio is running connect to it by visiting https://localhost:8787 in your browser. Log in with username rstudio and the password you entered after PASSWORD=.

• There are two shiny interactive apps for this project. shiny_treemap.R visualizes total employment and mean hourly wage within a selected sector, while shiny_barplot.R visualizes the two across states within a selected occupation. To start either one, type: shiny::runApp('/home/rstudio/<name of the shiny script>') in RStudio's terminal to start the shiny app. Then the interactive plot will be shown in a pop-up.

• Type make report.pdf in the terminal to create the final report.
• Each R script generates the figures for a particular section. If interested in the analysis in a particular section, run load_data.R first, and then the R script for that section.

Dataset link and reference:

1. Salary Data: https://www.bls.gov/oes/special.requests/oesm20all.zip
   from: U.S. BUREAU OF LABOR STATISTICS

2. Automation Probability Data: https://www.kaggle.com/andrewmvd/occupation-salary-and-likelihood-of-automation
   from: The future of employment: How susceptible are jobs to computerisation?
   DOI: 10.1016/j.techfore.2016.08.019
   Journal: Technological Forecasting and Social Change
   Volume: 114
   Author: Frey, Carl Benedikt and Osborne, Michael A.
   Year: 2017
   Pages: 254–280

3. North American Industry Classification System (NAICS)
   https://www.census.gov/naics/

4. Standard Occupational Classification (SOC) System
   https://www.bls.gov/soc/